Had I been present at the creation, T would have given some useful hints for the better ordering of the Universe. e Wise, 1221-1284 The chessboard is the world; the pieces are the phenomena of the Universe; the rules of the game are what we call the Laws of Nature. The player on the other side is hidden from us. Thomas Wuxley, 1825-1895 Man builds no structure that outlives a book. Eugene Fit& Ware, 1841-1911
Edited by
J* rookline Technologies,Ballston Spa, New York, USA
and Dept. of Geology, Union College, Schenectady, New York, USA
JOHN WHLEY &SONS, LTD
Copynght 0 2002
John Wiley & Sons, Ltd, Baffins Lane, Chichester, West Sussex PO19 IUD, England
Phone (+44) 1243 779777 E-mail (for orders and customer service enquiries):
[email protected] Visit our Home Page on www.wiley.co.uk or www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copynght, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd., 90 Tottenham Court Road, London WlP OLP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons, Ltd., Baffins Lane, Chichester, West Sussex PO19 1UD. England, or emailed to
[email protected], or faxed to (+44) 1243 770571. Othr Wiley Editorial Ofices John Wiley & Sons, Inc., 605 Third Avenue, New York, NY 10158-0012, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Pappelallee 3, D-69469 Weinheim, Germany John Wiley & Sous Australia, Ltd., 33 Park Road, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd., 2 Clemenli Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada, Ltd., 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1
British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 0 471 49315 5 Typeset in 10/12 Times by Dobbie Typesetting Limited, Trident Rouse, Tavistock, Devon Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.
tio To the memory of
John Herbert 919-1985 Wise, vigorous, effective advocate of the relevance and value of scientific research in industry. His strong belief in the synergetic interaction of Principles and Practice in the field of metallurgy impelled him to assemble an innovative, diverse staff at General Electric, and to inspire independent exploration that benefited both science and engineering.
This Page Intentionally Left Blank
Contents
............................................. Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preface to 1995 edition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ListofAcronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Crystal Structure Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Some Intermetallic Families Defined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List of Contributors
I STRUCTUREA
COMPOSITION
............................ ..............................
1 Structural Classification and Notation Jose' Lima-de-Faria 2 Amalgams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cezary Guminski 3 Beryllides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Loren A . Jacobson, Robert J . Hanrahan, Jr., and James L. Smith 4 Precious Metal Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ira M . Wolf 5 Rare-Earth Metal Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maria L. Fornusini, Franco Merlo, and Marcella Pani 6 Zintl Phases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Slavi C. Sevov
NDING AND STABILITY
................................
7 Spectroscopic Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Esther Belin-Ferre' 8 Effects of Pressure on Stability and Properties . . . . . . . . . . . . . . . . . . . . . . . Y. V. Levinsky 9 Magnetic Phase Diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. M. Cadogan 10 Calculation of Phase Diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Didier de Fontaine
vii
xi XiX
xxi xxiii
xxxi xxxv
1 3
21 37
53 85 113
133 135 153 165
185
Contents
viii
. . . , . . . . . , ... ... . . . , .......,. ... . . ...... . ., .....
PROPERTIES AND PHENOMENOLOGY
209
11 Free Surface Structure and Properties Mathias Goken 12 Color and Optical Properties . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . S. G. Steinemann, W. Wolf, and R. Podloucky 13 Effects of Mobile Species . . . . . , . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marie-Louise Saboungi and David L. Price 14 IonTracks . B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert L. Fleischer 15 Thermal Defects and Diffusion . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . Woljgang Sprengel, Markus A. Mdler, and Hum-Eckhardt Schaefer
211
PV MECHANICAL PROPERTIES
...,,.......,............. ....
231 245 263 275
29 5
16 Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dilip Shah and Eugene Lee 17 Fatigue .................................................. N . S. Stoloff 18 Solution and Defect Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert L. Fleischer 19 Strain Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . George T. Gray 111 and Tresa M . Pollock 20 Dislocations in Quasicrystals . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Renhui Wang and Chengzheng Hu 21 Twinning and Mechanical Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . M . H. Yoo 22 Staclcing-Fault-Type Interfaces and their Role in Deformation . . . . . . . . . . . V. Paidar and V. Vitelt
297
P L I ~ A T ~ O . . .~. ~. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
469
23 High-Temperature Structural Applications . . . . . . . . . . . . . . . . . . . . . . . . Harry A. Lipsitt, Martin J . Blackburn, and Dennis M . Dimidzik 24 Structural Applications for General Use . . . . . . . . . . . . . . . . . . . . . . . . . . . Vimd K. Sikkn and Seetharama C. Deevi 25 Magnetic Refrigeration . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . K. A. Gschneidner, Jr. and V. K. Pecharsky 26 Niobium Silicide High Temperature In-Situ Composites . . . . . . . . . . . , . . . . B. P. Bewlay, M . R. Juckson, and M . F. X . Gigliotti 27 Coating Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . P. K. Watta, J . S. Burnell-Gray, and K . Natesan
47 1
325 351 361 379 403
437
50 1 519 541 56 1
...........................
589
28 Casting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Paul A. McQuay and Vinod K. Sikka
591
Contents
29 Forming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Appel. H . Kestler. and H . Clemens 30 Powder Metallury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Seetharaman and S . L . Semiatin 31 Thin-Film Deposition and Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Ramanath. H . S. Goindi. and D . B . Bergstrom 32 Bulk Amorphous Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . R . B . Schwartz 33 Sulfidation Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . K . Natesan and P . K . Datta
CH TECHNIQUES
..............................
ix
617 643 663 681 707
721
34 Novel Synthesis Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Naresh N . Thadhani 35 Nanostructured Intermetallics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Suryanarayana 36 Defect and Atomic Process Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . M . I . Baskes 37 Molecular Beam Epitaxy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hock Min Ng and Theodore D . Moustakas
723
.............................. Commercial Impact on the US Economy . . . . . . . . . . . . . . . . . . . . . . . . . . John V. Busch and AIan C. Goodrich Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
789
VIII MISCELLANEOUS TOPICS 38 39
JunLing hi. Chao Liu. and Zhihong X u 40 Computer Design of Materials with Artificial Intelligence . . . . . . . . . . . . . . . Nadezhda N . Kiselyova 41 Alloy Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S. Naka and T. Khan 42 Intermetallics on the Internet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J . H . Westbrook
749 765 779
791 797 811 841 857
Indexes Authorindex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Subject index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Compoundindex .............................................
Contents of Volumes 1 and 2
.........................................
875 959 1019 1037
This Page Intentionally Left Blank
List of Contributors F. Appel Institute for Materials Research GKSS, Research Center D-21502 Geesthacht Germany
Bernard P. Bewlay General Electric - CR+ D Schenectady NY 12301 USA
Michael I. Baskes Los Alamos National Laborator) MST-8, NS G755 Los Alamos NM 87545 USA
Martin J. Blackburn School of Engineering University of Connecticut North Eagleville Road Storrs CT 06268 USA
Esther Belin-Ferre CNRS LCPMR-UMR 7614 11 rue Pierre et Marie Curie 75231 Paris Cedex OS France
J. S. Burnell-Gray Surface Engineering Research Centre University of Northumbria at Newcastle Newcastle upon Tyne NE1 8ST UK
Daniel B. Bergstrom Intel Corporation Hillsboro OR 97123 USA
John Busch Ibis Associates, Inc Reservoir Place, Suite 164 1601 Trapelo Road Waltham MA 02451 USA
xi
xii
List of Contributors
J. M. Cadogan School of Physics The University of New South Wales Sydney NSW 2052 Australia
Dennis M. Dimiduk USAF Wright Laboratory WL/MLLM 2230 Tenth Street Suite 1 Building 655 Wright Patterson Air Force Base OH 45433 USA
Helmut Clemens Institut ffir Metallkunde Universitit Stuttgart D-70 174 Stuttgart Germany
Robert L. Fleischer Department of Geology Union College Schenectady NY 12308 USA
P. K. Datta Surface Engineering Research Centre University of Northumbria at Newcastle Newcastle upon Tyne NE1 8ST UK
Maria L. Fornasini U.d. Genova Via Dodecanesco 31 116146 Genova Italy
Didier de Fontaine Department o f Materials Science and Engineering Evans Hall University of California Berkeley CA 94720- 1760 USA
Michael F. X. Gigliotti General Electric - CR + D Schenectady NY 12301 USA
Seetharama C. Deevi Chrysalis Technologies, Inc. 4201 Commerce Rd Richmond VA 23234 USA
Harmeet S. Goindi Rensselaer Polytechnic Institute 110 8th Street Troy NY 12180 USA
..I
List of Contributors
Mathias Goken Department of Materials Science University of Saarland Building 43B PO Box 151150 D-66041 Saarbriicken Germany
Robert J. Hanrahan Los Alamos National Laboratories Los Alamos NM 87545 USA
Alan Goodrich IBIS Associates, Inc Reservoir Place, Suite 164 1601 Trapelo Road Waltham MA 02451 USA
Chengzheng Hu Department of Physics Wuhan University Wuhan 430072 China
George T. Gray 111 Los Alamos National Laboratory Group MST-8 MS G755 Los Alamos NM 87545 USA
Melvin R. Jackson General Electric - CR + D MB-223 Schenectady NY 12301 USA
Karl A. Gschneidner, Jr Ames Laboratory Iowa State University Ames, IA 50011 USA
Loren A. Jacobson Los Alamos National Laboratory MS G-770 Los Alamos, NM 87545 USA
Cezary Guminski Department of Chemistry University of Warsaw Pasteura 1 02092 Warsaw Poland
Heinrich Kestler Plansee AC Technology Centre A-6600 Reutte Austria
xn1
xiv
List of Contributors Jose Lima-de-Faria
ONERA 29 Avenue Divise Leclerc BP 72 Chitillon 92322 France
Instituto de InvestigaGlio Cientifico Tropica Centro de Cristalografia e Mineralogia Alameda D. Alfonso Henriques, 41-40 Esquadro 1000-123 Lisboa, Portugal
Nadezhda N. Kiselyova A A . Brtikov Institute of Metallurgy and Materials Science Leninskii Prospect 49 11 7334 Moscow Russia
Harry A. Lipsitt 1414 Birch Street Yellow Springs OH 45387 USA
Eugene S. Lee Pratt & Whitney M/S 114-41 400 Main Street East Hartford CT 06108 USA
Chao Liu Laboratory of Computer Chemistry Chinese Academy of Science PO Box 353 Beijing 100080 China
Y. V. Levhsky Moscow Institute for Fine Chemical Technology Prospect Vernadakogo 86 113571 Moscow Russia
Paul McQuay Howmet Research Corporation 1500 S. Warner Street Whitehall, Mf 49461-1895 USA
JunLing Li Laboratory of Computer Chemistry Chinese Academy of Science PO Box 353 Beijing 100080 China
Franco Merlo U. d. Cenova Via Dodecanesco 31 I16146 Cenova Italy
List of Contributors
Theodore D. Moustakas Molecular Beam Epitaxy Laboratory Boston University 8 St Mary’s Street Boston, MA 02215 USA
V. Paidar Institute of Physics Academy of Science of the Czech Republic Na Slovance 2 18040 Praha Czech Republic
Markus Miiller Institut fur Theoretische und Anpwandt Physik University of Stuttgart Pfaffenwaldring 57 50 Stuttgart Germany
Marcella Pani U. d. Genova Via Dodecanesco 31 I16146 Genova Italy
Shigeshisa Naka ONERA 29 Avenue de la Divise Leclerc Chiitillon 92322 Cedex France
Vitaly K. Pecharsky Ames Laboratory 242 Spedding Iowa State University Ames, IA 50011 USA
Ken Natesan Argonne National Laboratory, D212 9700 South Case Avenue Argonne, IL 60439 USA
Raimund Podloueky Dept. of Physical Chemistry University of Vienna Liechtensteinstrasse 22A/I/3 A- 1090 Wien Austria
H. M. Ng Physical Science and Engineering Lucent Technology Bell Laboratories 600 Mountain Avenue Murray Hill, NJ 07974 USA
Tresa Pollock Dept. of Materials Science and Engineering University of Michigan 2300 Hayward Street Ann Arbor, MI 48109-2136 USA
XV
Xvi
List of Contributors
David L. Price Argonne National Laboratory Argonne IL 60439 USA
Venkat Seetharaman Materials and Processes Engineering Pratt and Whitney 400 Main Street East Hartford, CT 06108 USA
G. Ramanath Department of Materials Science Rensselaer Polytechnic Institute Troy, NY 12180 USA
S. Leo Semiatin
Marie-Louise Saboungi Centre de Recherche sur la Matihre Divise CNRS 1B rue de la Ferollerie 4507 1 OrlCans Cedex 02 France
Slavi C. Sevov Department of Chemistry and Biochemistry University of Notre Dame Notre Dame, IN 46556 USA
Ham-Eckhardt Schaefer Institute of Theoretical and Applied Physics University of Stuttgart Pfaffenwaldring 57 Stuttgart D-70569 Germany
Dilip M. Shah Pratt and Whitney MS 114-45 400 Main Street East Hartford, CT 06108 USA
Ricardo B. Schwarz Structure/Properties Relations Group MSTIG755 PO Box 1663 Los Alamos, NM 87545 USA
Vinod K, Sikka ORNL Oak Ridge, TN 37831-6093 USA
USAF Materials and Manufacturing Directorate Wright-Patterson Air Force Base OH 45433-7817 USA
List of Contributors
James L. Smith Los Alamos National Laboratory STC MS K763 Los Alamos, NM 87545 USA
School of Materials Science and Engineering Georgia Institute of Technology 771 Ferst Drive, NW Atlanta, GA 30332-0245 USA
Wolfgang Sprengel Institut fur Theoretisch und Angewandt Physik Universitiit Stuttgart Pfaffenwaldring 57 D-70550 Stuttgart Germany
Vaclav Vitek Department of Materials Science and Engineering University of Pennsylvania 3231 Walnut Street Philadelphia, PA 19104 USA
Samuel G. Steinemann Institut de physique de Ia Matikre Condenske
Renhui Wang Department of Physics Wuhan University Wuhan 430072 China
Codoz 14
CH 1025 Saint Sulpice Switzerland
Norman Stoloff Department of Materials Science and Engineering Rensselaer Polytechnic Institute Troy, NY 12180 USA
Jack H. Westbrook Brookline Technologies 5 Brookline Road Ballston Spa, NY 12020 USA
C. Suryanarayana
Walter Wolf Materials Design s.a.r.1. 44 av. F.-A. Bartholdi 72000 Le Mans France
Department of Mechanical, Materials and Aerospace Engineering University of Central Florida Orlando FL 328 16-2450 USA
xvii
xviii Ira M. Wolff Technologies Group Conor Pacific Environmental Technologies Inc. Suite 920 1500 West Georgia Street Vancouver BC V6G 226 Canada
Zhihong Xu Laboratory of Computer Chemistry Chinese Academy of Science PO Box 353 Beijing 100080 China
List of Contributors
Man H. Yoo Metals and Ceramics Division, MS 6115 ORNL, PO Box 2008 Oak Ridge, TN 37831-6115 USA
reface
The world at present groans under a load of new publications on every branch of science and art, with which no former period of our literary annals can for a moment be compared. The most assiduous students, unable to peruse a thousandth part of the works which are daily soliciting their attention, are quite perplexed and distressed about what to choose and what to reject. This we have frequently found to be the case with ourselves, and while debating the question in our own minds, have lost, in doubt and uneasiness, the time we meant to set apart for practical manipulation. Impressed, therefore, with the unspeakable disadvantages that result from the circumstancesjust stated, and anxious to save others, in some degree, from that unpleasant dilemma in which we have ourselves been so often placed, we have resolved on the present publication, which we hope will to a very great extent accomplish the useful object we have in view. With what judgment, however, the design has been formed, and with what skill it has been executed, it becomes not us to determine - that question, to the result of which we are deeply alive, remains now with a higher tribunal.
It is disconcerting to note that, in fact, the above words were written, not by us, but by another metallurgist almost 150 years ago.* Talk about dejd vu all over again! To assuage, in our own field, the difficulties summarized above was indeed the primary objective of our original 1995 work, Intermetallic Compounds: Principles and Practice, and we could not have better exprcssed our feclings. One might suppose that the extensive treatment and coverage of the 75 chapters of the predecessor volumes 1 and 2 of our treatise would have exhausted such a narrow and, in the view of many, esoteric subject. This is not so! Interest and activity in the field remain high as indicated by the numerous sub-topical conferences and symposia that have occurred since the publication of the first two volumes and by the still increasing level of activity in new research (approximately 1900 literature additions noted in 2000 by the Permuterm Subject Index). Accordingly, in planning the present volume, we had no difficulty in defining many new topics not previously treated (new intermetallic families, new means of assessment of bonding and stability, new properties and phenomena, new applications, new practical processes and phenomena, and new research techniques). Other topics introduced for the first time include: economics, design of intermetallic-based materials, a compilation of data sources, and opportunities for information retrieval from the Internet. Altogether these new topics comprise 27 chapters in the present volume. In other instances (eight chapters) subjects previously treated were reprised, because of the intense current activity and great importance of their topic matter. In the case of seven more chapters, dealing with mechanical properties and phenomena, although their subjects were previously alluded to in one or more chapters, now totally new approaches were undertaken. A final innovation in the present volume is the inclusion in the front matter of a section by one of the editors (JHW) called ‘Some IntermetallicFamilies Defined’, intended as an assist to all readers. Four planned chapters did not eventuate due to over-commitment of the invited contributors and were reluctantly dropped from the book, In recruiting contributors for the present volume, we deliberately chose not to include any authors from volumes 1 and 2 so as to encourage, and profit from, new viewpoints, knowledge, and contacts. As before, we consulted carefully with many people in order to identify authorities for each subject. We sought experts wherever they were to be found, and our international set of 71 authors turned out to be almost equally divided between the United States and 13 other countries (Austria, Australia, China, Czech Republic, France, Germany, Italy, Poland, Portugal, Russia, Switzerland, Union of South Africa, and the United Kingdom). Having recruited contributors for the project, we then strove to assist them in achieving clarity and thoroughness - from outline, i o draft, to final manuscnpt. The contributors cooperated superbly, and we thank them for their hard work and high achievement.
* From the Preface to the First Edition of James Larhn’s Brass and Iron Founders Guide, H. C . Baird & Co.. Philadelphia, 1853. xix
xx
Prefuce
Their affiliations at the time of writing are shown on the title page of each chapter. Since several of these have changed by the time of publication, their current postal addresses are shown with their photos in the List of Contributors, pp. xi-xviii. Just as with volumes 1 and 2, this volume was planned to be an aid to both scientists and engineers. Together with the earlier volumes, this volume can serve as a base for those who wish to know about interrnetallics as an area in which to begin research or for those who wish to exploit the often unique properties of intermetallics in practical applications. Equally, it is a resource for workers who are already active in the field and need, or wish, to expand their knowledge of related science or practical technology. We also expect that many chapters are appropriate source matter for special-topic or seminar courses at the advanced undergraduate and various graduate school levels, Authors were advised to strive for clarity and, while providing copious references to the literature in their field, to present their chapter in such a way that it could stand alone and be fully comprehensible. Each author was asked to set forth the principles of his or her subject in terms that are meaningful to scientists and engineers who are not speclalists in the author’s field, and then to progress to include knowledge that workers in their own areas would wish to have. Concluding sections of most chapters give the authors’ critical assessment of the state of their subject and of where they believe further effort is merited. As an assist to readers, the acronym list in the front matter consolidates new entries from the present volume with those already compiled in volumes 1 and 2. Similarly, the subject, author, and compound indices constitute a consolidation from all three volumes. We have benefited from outstanding secretarial help by Jean Conley of Union College during the three years of this project. Assembly and publication of the final product were eased for us by the continuing efforts and cheerful good counsel of David Hughes, Helm McPherson, Susan Barclay, Samantha Hartley, and Susan Lambert at John Wiley.
J. H. WESTBROOK, Bullslon Spa, New Y o ~ k R. L. FLEISCHER, Schenectady, New I70rk
Preface to t e 1995 Edition
Intermetallic compounds were last comprehensively reviewed in 1967 in a volume that was edited by one of us (JHW). At that time the field was described as of spccial interest because it was undergoing ‘exponential proliferation’. That trend continues to the present. The number of intermetallic entries in the Permuterm Subject Index shows a doubling period of less than nine years, having reached roughly 1880 entries per year in 1993. Apart from scholarly interest, intermetallics have now become of substantial commercial significance; for some, such as Ni,Al, world-wide use is in the 1000s of tons; for others, for example 111-V semiconducting compounds, although the quantities employed are not in tonnage numbers, their value as vital components of electronic circuits is in the billions of dollars. From the 1967 book we remind the reader that ‘Thc first published paper dealing with intermetallic compounds appeared in 1839, and more than sixty years elapsed before.. .the first review paper by Weville in 1900. However, new results were then appearing so rapidly that fifteen years later two books were printed, devoted exclusively to this subject, one by Desch in England and one by Giua and Giua in Italy’. More recently, conference volumes that deal exclusively with intermetallics but typically only within specific, limited sub-topical subject areas have become common. The scope of the present work is as broad as that of its 1967 predecessor. However, the increased volume of activity in intermetallics and the increased significance of their applications have necessitated an expansion from the 27 chapters of the earlier work to the 75 chapters of the present treatise. First, what are intermetallic compounds? Generally, such a compound is a structure in which the two or more metal constituents are in relatively fixed abundance ratios and are usually ordered on two or more sublattices, each with its own distinct population of atoms. Often substantial or complete disorder may obtain, as a result of low ordering energy or the intervention of some external agency, for example extreme cooling rates, radiation, etc. Deviations from precise stoichiometry are frequently permitted on one or both sides of the nominal ideal atomic ratios, necessitating a partial disorder. Here we include as intermetallic compounds all metal-metal compounds, both ordered and disordered, binary and multicomponent. Even the metal-metal aspect of the definition is often relaxed by including some metal-metalloid compounds, such as silicides, tellurides, and semiconductors. We believe this inclusion is appropriate since the phenomenology of many such compounds is nearly identical to metal-metal ones, and they provide useful examples of priiiciples, properties, and practices. The burgeoning literature on intermetallics and the lack of a comprehensive single source of up-to-date descriptions of where we are, what we need to know. and what we can do with intermetallics created the incentive for the present pair of volumes. This work was planned to provide state-of-the-art assessments of theory, experiment, and practice that will form a solid base for workers who wish to know more than their own particular area. Each author was asked to set forth the principles of his or her subject in terms that are meaningful to scientists and engineers who are not specialists in the author’s field, and then to progress to include knowledge that workers in their own areas would wish to have. Concluding sections of most chapters give the authors’ critical assessment of the state of their subject and of where tlicy believe further effort is merited. This work is divided into two volumes in order that each be of manageable size. The first, on the theme Principles, is directed at the science of intermetallics - how do we understand their formation, structure and properties? The Practice volume considers commercial production and engineering applications of intermetallic compounds. The reader who browses carefully will recognize that the immediacy of the practice described ranges from hoped-for use, to beginnings of use, to actual commercial application - depending on the specific subject. Some of the hoped-for uses are fated never to be realized, but the authors have aimed to reveal what the obstacles are so that the reader may make his or her own assessment (and possibly providc a solution!). xxi
xxii
Preface to the 199.5 Edition
We conferred carefully with many people in order to identify authorities for each subject; having recruited contributors for the project, we then strove to assist them in achieving clarity and thoroughness from ouiline to draft to final manuscript. The contributors cooperated superbly, and we thank them for their hard work and high achievement. We sought experts wherever they were to be found, and our international set of nearly 100 authors turned out to be almost equally divided between the United States and 14 other countries. Manuscripts have in fact come from all inhabited continents. We planned this work as an aid to both scientists and engineers. It can serve as a base for those who wish to know about interrnetallicsas an area in which to begin research. Equally it is a resource to workers who are already active in the field and need, or wish, to expand their knowledge of related science or practical technology. We expect that many chapters are appropriate source matter for special topic or seminar courses at the advanced undergraduate and various graduate school levels. It is hoped that passage of the next 25 years will reveal some influence of this treatise on the further development of this field. As an assist to readers we have provided in the following pages a consolidated acronym list and some crystallographic tables. Nomenclature for crystal structure types is often complex, and some of the authors have introduced their own. Generally we have asked authors to includc both o f two commonly used types of symbols as they introduce structures. The two-part table following this preface lists many of the common types - by Strukturbericht symbol, prototype name (termed a structure type), and Pearson symbol. Strukturbericht symbols are only partly significant and systeniatic: A’s are not compound structures but consist of a single lattice of atoms (except for A15!); B’s are equiatomic ordered structures; C’s have 2-to-1 atomic abundance ratios, DO’S340-1. Structure type compounds are the specific ones used to designate a particular structure. Thus B2 compounds are also referred to as CsCl compounds. Many structures are better known to metallurgists and mineralogists by names other than the formula of the structure type chosen by crystallographers, e.g. Laves, fluorite, Heusler, etc. Such names have been added in selected cases. The Pearson symbols tell the crystal symmetry and the number of atoms per unit cell. Thus, B2, CsCl has a primitive (P) cubic (c) structure with 2 atoms per cell and hence the Pearson symbol (cP2). The Pearson designation is informative, but it is not necessarily unique. Although there is only one cP2 structure, W a r s and Calvert list two cP4s, three cF129 and twenty-two hP9s. Thus to be definitive, both the structure type and the Pearson symbol need to be given, or the Pearson and the Strukturbericht symbol. The index in each volume includes the subjects in both volumes of this work, in order that the reader may be able to locate any subject that is addressed. Although the purpose of such combined indices is not to induce the owner of a single volume to purchase the other, it possibly may help to reduce the barrier to such action. We havc bcnefited from outstanding Secretarial help during the three years of this project, first by Phillis Liu, then Constance Remscheid at General Electric, finally Nary Carey at Renssalaer Polytechnic Institute. We appreciate the hospitality of the General Electric Research and Development Center during the inception and middle period of preparing these volumes. Assembling the find product has been eased for us by the continuing efforts and cheerful good counsel at John Wiley of Jonathan Agbenyega, Irene Cooper, Philip Hastings, Vanessa Lutman and Cliff Morgan.
J. H. WESTBROOK, Bullston Spa, New York R. L. FLEISCHER, Schenectudy, New York
Upon these considerations, we have been induced to undertake the present extensive work, the purpose of which is to instruct rather than to amuse; in which nothing will be omitted that is elegant or great; but the principal regard will be shown to what is necessary and useful. - Isaac Ware, 1756
2D 3D 6D
ASA ASB
two-dimensional three-dimensional six-dimensional
ASRP ASTM
angular correlation of annihilation radiation angular correlation of positron ACPAR annihilation radiation atomic environment AE AES Auger electron spectroscopy AET atomic environment type AF anti-ferromagnetic atomic force microscopy AFM AI artificial intelligence AIM argon induction melting ALCHEMI atom location by channeling-enhanced microanalysis atomic layer epitaxy ALE ALICISS alkali-ion impact collision ion scattering spectroscopy air mass AM active magnetic regenerator AMR Advanced Materials Technology, Inc. AMT AN atomic number atomic probe AP AP atomic property APB anti-phase boundary anti-phase domain APD avalanche photodetector APD APE atomic property expression APFIM atom-probe field-ion microscopy APW augmented plane wave antireflection AR ARIPES angle resolved inverse photoemission spectroscopy angle-resolved photo-emission ARPES spectroscopy angle-resolved ultraviolet photoARUPS emission spectroscopy AMPS angle-resolved X-ray photo-emission spectroscopy ACAR
ASW BC bcc BCS bct BDTT BH
BIS BM BSCCO BSE BT BW BZ CAM CANDU CAP CAS CAT CBED CBLM
cc
CCD CCGSE CCIC CCMAI
xxiii
atomic sphere approximation anti-structural bridge (mechanism) advanced sheet-rolling process American Society for Testing and Materials augmented spherical wave bond charge body-centered cubic Bardeen-Cooper-Schrieffer body-centered tetragonal brittle-to-ductile transition temperature buried heterostructure bremstrahlung isochromatic spectroscopy Bowles-Mackenzie (theory of martensitic transformation) bismuth-strontium-calcium-copperoxide back-scattered electrons Bhatia-Thornton ('partial structure factor for liquid alloys) Bragg-Williams (theory of ordering) Brillouin zone c-axis modulated Canadian deuterium-uranium (power reactor) consolidation under atmospheric pressure Chemical Abstracts Service computer-assisted tomography convergent-beam electron diffraction cluster Bethe lattice method cluster center charge-coupled device concentric-circle grating surfaceemitting (laser) cabled conductor in conduit crystal chemical model of atomic interactioiis
List of A c~onyms cubic close-packe~ continuous cooling transformation
direct chill (casting) direct configurational averaging dominant diffusing species density functional (theory) distributed feed-back density-f~nctionaltheory double hetero (junction) double hexagonal close-packed
compact disc cluster e~paiision(technique) ~ontinuo~ electron-beam s accelerator facility dHVA
duplex layer coating directional levitation zone remelting density of states duplex displace~entper atom demonstration poloidal coil diamond pyramid hardness d e ~ ~ g Q nquasi~~ystal a1 dense random packing directional solidification dime~sio~ally stable anode diEerent~a1scanning calorimetry d i s ~ l a c e ~ eshift n t complete di~erent~al thermal analysis
~ n t ~ ~ n a t i oCommission ii~l on tant initial state couiiter-~ravitylow-pressure inert-atmosphere (investment casting)
colossal ~agnetoresistance coordinator number
e/a of comp~terl e ~ r ~ i n g ) coefficie~tof performance coardination polyhedron coheren~poteiitial approxii~ation critical resolved shear stress clie~~sorptioi~ complex stacking fault ~oincide~~ce-site lattice
chemical v a p r deposition luster variation method o ~ i t i n u ~ uwave s cold worked coiicentr~tionwave illiams method (theory of phase transfoima ti ons)
ESCA
FC fcc FGC fct
~ a t ~ a n a~ ~ e~ ~system ~~ e n~ t ~ ~ ~ t ~ l e - t o - b rtransition ittle teniperature direct c ~ ~ r e n ~
elcctron/atom (ratio) e m b e ~ ~ ~ d -method ato~ electron beam physical vapor deposition equal channel angular extrusion effective cluster interaction embedded-cluster method ekctro-optic directional coupler electrodischarge machining ener~y-dispersiveX-ray (spe~troscopy) electron energy-loss spectroscopy electric-field gradient electro~otiveforce eEective pair interaction electron spectro scopy for chemical analysis electrospar~depos~tion extrinsic stacking Cault electroslag remelting (copper) electroly t extended ion fine structure field cooled face-centered cubic fatigue crack growth face-centered ~ e t r a ~ o n a ~ f o r e i g n - a ~ odiffusion ~ finite-element analysis ~ n i t e - e l ~ ~method ent
List of Acronyms
Fusion Engineering International Experimental Magnet Facility field effect transistor field ion microscopy full-potential linearized augmented plane wave full-potential linearized augmented Slater-type orbital full-potential linearized muffin-tin orbital fully lamellar, narrow (spacing) fully lamellar, wide (spacing) ferromagnetic foreign objects and debris figure of merit Fabry-Perot (laser) Finnis-Sinclair (potential) phase transformation field warmed full width at half maximum fluxing followed by water quenching in fused silicon tubes floating zone fusion zone
HDDR
IC IC IC ICSD ICSU
GS GT GTA
gas atomization gainx bandwidth (product) grain boundary Gorsky-Bragg-Williams (free energy) gross domestic product glass-forming tendency generalized gradient approximation galvanostatic intermittent titration technique grazing incidence X-ray spectroscopy gas metal arc (welding) gravity metal mold giant magnetoresistance generalized perturbation method generalized random-phase approximation ground state Goody-Thomas (electronegativityj gas tungsten arc (welding)
HAZ HB HBT HCF hCP HD HD
heat-affected zone horizontal Bridgman heterojunction bipolar transistor high-cycle fatigue hexagonal close-packed hybrid deposition hydrogen decrepitation
FENIX FET FIM F-LAPW F-LAST0 F-LMTO FLn FLw FM FOD FOM FP FS FT FW FWHM FXWQFS FZ FZ GA GB GB GBW GDP GFT GGA GITT GIXS GMA GMM GMR GPM GRPA
HDZ HEMT HIP HOMOLUMO HPT HPT HPTB HR HREM HRTEM HSCT HTS HVEM HVOF HVTEM IAE IAET IBM
IDOS IEM IGC IHPTET ILS IM IMC IMC IMPATT IPES IPM IPS 1QC ICR IRR ISF ISM IT ITER
xxv hydrogenation-decompositiondesorption-recombination heat-and-deformation zone high-electron-mobility transistor hot isostatic pressing highest occupied molecular orbitallowest unoccupied molecular orbital heterojunction phototransistor high-pressure turbine high-pressure turbine blade high resolution high-resolution electron microscopy high-resolution transmission electron microscopy high-speed civil transport high-temperature superconductor high-voltage electron microscopy high-velocity oxy-fuel high-voltage transmission electron microscopy irregular atomic environment irregular atomic environment type International Business Machines Corporation intermetallic compound integrated circuit investment cast inorganic crystal structure database International Council of Scientific Unions integrated density of states interstitial electron model Intermagnetics General Corporation integrated high-performance turbineengine technology invariant line strain ingot metallurgy intermetallic compound inverse Monte Carlo (method) impact ionization avalanche transit time inverse photo-emission spectroscopy independent-particlemethod invariant plane strain icosahedral quasicrystal infrared internal rate of return intrinsic stacking fault induction skull melting (positive) inner tetrahedron International Thermonuclear Experimental Reactor
xxvi IUCr IUPAC
List of Acronyms
IV
International Union of Crystallography International Union of Pure and Applied Chemistry intermediate valence
J-FET
junction fiela-effect transistor
KHN KKR
Knoop hardness number Korringa-Kohn-Rostoker (bondcalculation method) Khantha-Cserti-Vitek (deformation model) potassium titanyl phosphate Kear-Wilsdorf (dislocation locking mechanism)
KSV KTP KW LA LAPW LAST0 LCAO LCF LCT LCW LD LDA LEG LED LEED LEISS LWC LKKR LM LMC LME LMTO Ln LNT LO LPCVD LPE LPPS LPS LPT LRO LSDA LSI MA MAPW MASC
longitudinal acoustic (wave) linearized augmented plane wave linear augmented Slater-type orbital linear combination of atomic orbitals low-cycle fatigue large-coil task Lock-Crisp-West (radiation analysis) laser diode local-density approximation liquid-encapsulated Czochralski (technique) light-emitting diode low-energy electron diffraction low-energy ion scattering spectroscopy Large Hadron Collider layered KKR (structure calculation) lattice mismatch liquid-metal cooling liquid-metal embrittlement linearized muffin-tin orbital lanthanide liquid-nitrogen temperature longitudinal optical (wave) low-pressure chemical vapor deposition liquid-phase epitaxy low-pressure plasma spraying long-period superstructure low-pressure turbine long-range order local spin-density approximation large-scale integration mechanical alloying modified augmented plane wave metal and silicide composite (a specific Nb,Ti,Hf,Cr,Al,Si alloy)
MB MBE MBT MC MCE MCS MD MEAM MEE MESFET MF MFM MFTF MG MH MIG MISFET MJR MLR Mm MMC MMC MN MO MOCVD MOS MOSFET MOVPE MQW MR MR MRI MRSS MRT MS DOS PSR MT MTD MVA MXPS NAICS NASA NASP ND NET
Martinov-Basanov (electronegativity) molecular beam epitaxy metal-base transistor Monte Carlo magnetocaloric effect Monte Carlo simulation molecular dynamics modified embedded-atom method migration-enhanced epitaxy metal Schottky field-effect transistor mean field magnetic force microscopy Mirror Fusion Test Facility miscibility gap metal hydride metal-inert gas (welding) metal-insulator-semiconductor field effect transistor McDonald jelly roll (superconducting cable construction) multilayer reflector misch metal metal-matrix composite metal mold casting Mendeleev number magneto-optical metal-organic chemical vapor deposition metal-oxide-semiconductor metal-oxide-semiconductor field-effect transistor metal-organic vapor phase epitaxy multiple quantum well magnetoresistance magnetic refrigerator magnetic resonance imaging maximum resolved shear stress orthodontic NiTi alloy Microsoft disk operating system muon spin relaxation muffin tin martensite transformation diagram million volt-amperes monochromatized X-ray photo-electron spectroscopy North America Industry Code System National Aeronautics and Space Administration (USA) National Aerospace Plane N-dimensional next European torus (fusion device)
xxvii
List of Acronyms
NMI NMR NN NNH NNN NOR NPV NSR
normal hydrogen electrode nano indentation neutral impact collision ion-scattering spectroscopy National Maglev Initiative (US) nuclear magnetic resonance nearest neighbor nearest-neighbor histogram next-nearest neighbor negative OR (logic operator) net present value notch strength ratio
OAZ OB OD ODR ODS OEXC oh OIFZ OIM ORNL OT
oxidation-affected zone occupied band outer diameter oxygen dissolution reaction oxide dispersion-strengthened opto-electronic integrated circuit octahedron optical-imaging float zone (process) orientation-imaging microscopy Oak Ridge National Laboratory (negative) outer tetrahedron
NHE NI NICISS
PA PAC PAM PAS PBC PBN PBR PBT PCM PCM PCT PD PDF PDOS PECVD
primary annealed perturbed angular correlation plasma-arc melting positron-annihilation spectroscopy periodic bond chain pyrolytic boron nitride Pilling-Bedworth ratio permeable-base transistor phase-change material point-charge model pressure-composition-temperature phase diagram pair-distribution function phonon density of states plasma-enhanced chemical vapor deposition PES photo-emission spectroscopy PFC planar flow casting PGM platinum-group metals PH Pearson’s Handbook PHACOMP phase computation (program) PIGA plasma-melted, induction-guided, gas atomization PKA primary knock-on atom PL photoluminescence PM powder metallurgy PM paramagnetic
PMTC
PVD PVDF PWHT PZT
phenomenological martensitic transformatioii concept periodic number type of photothyristor partial pair distribution function parts per million path-probability method Paidar-Pope-Vitek (L12 hardening model) plasma-rotating-electrode process Pearson symbol polysynthetically twinned phase transformation phenomenological theory of martensite crystallography physical vapor deposition polyvinyl difluoride post-welding heat treatment lead zirconate titanate (ceramic)
QC QCSE QENS QFD QN QSD QW
quasicrystalline or quasicrystals quantum-confined Starke effect quasi-elastic neutron scattering quantum formation diagram quantum number quantum structural diagram quantum well
RBS RC RCS RCP RD RD RDF RDS RE REP RF RHE RHEED
Rutherford back scattering ribbon comminution replacement-collision sequence relative cooling power rate dependent rolling direction radial distribution function rate-determining step rare-earth (metal) rotating-electrode process radiofrequency reversible hydrogen electrode reflection high-energy electron diffraction reactive hot isostatic pressing rate independent radiation-induced ductility rigid-ion model Ruderman-Kittel-Kasuya-Y oshida (electronic interactions) refractory-metal intermetallic composite root mean square remote plasma-enhanced chemical vapor deposition
PN PnPn PPDF PPm PPM PPV PREP PS PST PT PTMC
RHIP RI RID RIM RKKY RMIC rmS
RPECVD
xxviii
List of Acronyms
RRR RS RS RSP RSS R-T RT
residual resistivity ratio rapidly solidified reaction synthesis rapid solidification processing resolved shear stress rare-earth transition-metal (compounds) room temperature
SAD SAED SAGBO SAM-APD
selected-area diffraction selected-area electron diffraction stress-assisted grain-boundary oxidation separate absorption and multiplication avalanche photodetector simple cubic semiconductor standard calomel electrode self-diffusion specific damping capacity spin-density wave scanning electron microscopy superlattice extrinsic stacking fault stacking fault stacking-fault energy spin glass self-propagating, high-temperature synthesis semi-insulating VLSI self-interstitial atom self-interaction correction stress-induced martensite secondary-ion mass spectrometry superconductor-insulatorsuperconductor superlattice intrinsic stacking fault static inductance transistor semimetal second-moment approximation shape-memory alloy shape-memory effect superplastic forming scanning probe microscopy superparamagnetism superconducting quantum interference device short-range order solid-state amorphizing reaction structural stability diagram superlattice stacking fault Atlas of Crystal Structure Types scanning transmission electron microscopy scanning tunneling microscopy
sc
sc
SCE SD SDC SDW SEM SESF SF SFE SG
SHS SIjVLSI STA SIC SIM SIMS SIS SISF SIT SM SMA SMA SME SPF SPM SPM SQUID
SRO SSAR SSD SSF STA STEM STM
STN
sv sw sx
TA TB TCP TD TD TDFS TE TE TEC TEG TEM TEP TGW TIP TK TM TM TMA TMD TMF TMS TO TOF TPA TRIP TS TSRO tt TTS TTT UB UES UHF UHV ULSI UNESCO UPS UPS USAF
usw
Scientific-Technological Network (of CAS) Sodani-Vitole version of the Paidar et al. model spin waves single crystal transverse acoustic (wave) tight binding topologically close-packed thoria dispersion transverse direction temperature dependence of the flow stress thermo-electric transverse electric (field) thermo-electric cooler thermo-electric generator transmission electron microscopy triethyl phosphene Teatum-Gschneidner-Waber (atomic radius) thermally induced porosity Takeuchi-Kuramoto (dislocation locking mechanism) transition metal transverse magnetic field Ti-MO-Al (alloy) theoretical maximum density themon~echanicalfailure The Metallurgical Society (of AIME) transverse optical (waves) time of flight two-photon absorption transformation-induced plasticity tensile strength topological short-range ordering truncated tetrahedron tubular tin source time-temperaturetransformation unoccupied band universal equation of state ultra-high frequency ultra-high vacuum ultra large-scale integration United Nations Educational, Scientific, and Cultural Organization ultraviolet photo-electron spectroscopy ultraviolet photo-emission spectroscopy United States Air Force ultrasonic wave
List of Acronyms
xxix
WQFS
water quenching in ftised silica tubes Wigner-Seitz (cell) Winterbon-Sigmund-Sanders (model of irradiation damage) weight parts per million
UTS
ultimate tensile strength ultraviolet
VAR VAR VCSEL VEC VGF VGS VHF VHN VIM VINITI
vacuum-arc refined vacuum-arc remelting vertical-cavity surface-emitting laser valence-electron concentration vertical gradient freezing Van Gogh’s sky very-high frequency Vickers hardness number vacuum induction melting Russian Institute of Technical Information vapor-liquid-solid very large-scale integration vapor phase epitaxy Vickers pyramid number vacuum plasma spraying vacuum ultraviolet
uv
VLS VLSI VPE VPN VPS
vuv
WB WGPD WLR
weak beam waveguide photodetector Weclisler-Lieberman-Read (theory of martensitic transformation)
ws wss
wt.ppm
xuv
X-ray absorption near-edge structure X-ray (photo) absorption spectroscopy exchange correlation exothermic dispersion (synthesis process) X-ray emission spectroscopy X-ray inspection module X-ray photo-electron spectroscopy X-ray photo-emission spectroscopy X-ray diffraction extreme ultraviolet
YAG
yttrium aluminum garnet
ZFC ZIF
zero-field cooled zero insertion force
XANES XAS
xc
XDTM XES XIM XPS XPS XRD
This Page Intentionally Left Blank
tructure Nomenchture* Arranged Alphabetically by Pearson-Symbol Designation Pearson symbol cF4 CF8 cF12 cR6 cn4 cF32 cF52 cF56 cF68 cmo c m 12 cF116
c12 cI16 c128 cI32 c140 cI52 cI54 cI58 cI76 c180 cI96 cIl62 cP1 CP2 cP4 cP5
Prototype
Struktufbericht designation
cu C (diamond) NaCl (rock salt) ZnS (sphalerite) CaF, (fluorite) MgAgAs AlCu,Mn (Heusler) BiF, (AlFe,) NaTl AuBe, SiO, (B cristobalite) Cu,Mg (Laves) CuPt, UB,, Al,MgO, (spinel) c03s4
COJ, Sb203(senarmonite) Fe,W,C (q carbide) NaZn,, Cr23C6
Mn,,Th,,Cu,,Mg,Si, (G-phase) W cou Th,P* CoAs, (skutterudite) Ge,Ir, pu2c; Cu,Zn, (y brass) Fe,Zn,, ( y brass) Sb,Tl, ciMn (pphase) Cu,,Si, Md;O, AILi,N2 Mg,,(A1,Zn)49 ciP0 CSCl AuCU, ReO, AIFe,C (perovskite) CaTiO, (perovskite) Fe,N
Space group
Pearson symbol
Fmh Fd!m Fm3m Fdzm Fm3m Fdzm Fmzm Fm2m Fd3m F43m FdJm Fd 3-m Fm2c Fm3m Fdzm Fdzm Fm3m Fdzm Fd 3m Fm2c Fm3m Fm3m Im3m 1213 I43$ Im3 Im3m I23d I2zm Im2m Im3m 143m Zd3d Ia!
cP6 cP7 cP8 cP12 cP20 cP36 cP39 cP52 hP1 hP2 hP3
hP4
Ag*O CaB, Cr,Si (Bw) FeSi Cu3VS, (sulvanite) FeS, (pyrite) NiSbS (ullmanite) BMn BaHg,I Mg2Zn11 Cu,A14 (y brass) HgSn€-,o Mg
wc
AlB, Cdl, FezN LiZn, YSe ciLa BN C (graphite) NiAS
ZnS (wurtzite) hP5 hP6
hP8
10;
Im? Pm3m Pm3m Pm2m Pm3m Pmzm Pm3m P43m
Prototype
hP9
hPlO hP12
Ni,Al, CaCu, CoSn C%Te HgS MoS, Ni,In Na,As Ni,Sn TiAs CrSi, Fe,P WgZn SiO, (high quartz) Pt,Sn, cus MgZn, (Laves) SiO, (p tridymite)
Strukturbericht Space designation group ~3 02, A15 B20 H24
c2 8f)i
A13 02,
08, 083
Af A3 Bh C32 C6 L’3 ck
A8 A3’ Bk A9 B81 B4 D52
D513 D2d
B35
ch B9 C7
B82 0018 Do19
B, C40 C22 86
C8 DSb
818 C14 C10
pn3m Pmlm Pm3n P213 P43m Pa3 P213 P4122 Pm3~ pm3 P43m P6lmmm P6311wp2c P6m2 P6lmmm Plml P63tmmc P6s/mmc P3121 P6+~nc P6~lmmc P63/mmc P63lmmc P63mc P3ml Phl P6lmmm P6/mmm P6/mmm P3,21 P63lmmc P6smmc P63/mmc P63lmmc P63/mmc Pc222 P62m P3 P6222 P63lmmc P63/mmc P63lmmc P63/mmc continued
*Adapted (with additions and corrections) from ASM Handbook, Vol. 3, 10th ed, ASM International, Materials Park, OH.
xxxi
xxxii
Crystal Structure Nomenclature
Arranged Alphabetically by Pearson-Symbol Designation (contmued) Pearson symbol hP14 hP16 hP18 hP2O hP24 hP28 hR1 hR2 hR3 hR4 hR5 hR6 hR7 hRlO hR12 hR13 hRl5 hR20 hR26 hR32 mC6 mC8 mC12 mC14 mC16 mP12 mP20 mP22 mP24 mP32 mP64 oc4 oC8
oc12 oC16 oc20 oC24 oc28
of% oF40 oF48 OF72 oF128 0112 0114
0120
on8 0140 oP4
aP6 oP8
Prototype W*B, Mn&, Ni,Ti Al,C,Si A1,FeMgtSi, Mg,N1 Fe Th, -dS,, CU,P MgNi, (Laves) Co,AI aHg or BPo aAs aSm NaCrS, BiZTe, N G CaSi, NiS (millentel ' Al,C, MoA cyAI,O, (corundum) BaPb, Fe,W, (p-phase) B4C HoAI. Cr,Ali CuPt AuTe, (calaverite) CUO (tenonte) ThC, SNi,Sn4 FeKS AgAuTe, (syfvanite) ZrO, As,SA CO AI, FeksS ASS (realgar) BSe aSe
NU CaSi ctGa CrB I P (bkck) ZrSi, BRe, PdSn, PdSn, AI Mn T!Si Mn4b CuMg, GeS, aS
SiS, TaA AI,U
?%:f
AuCd
Strukturbericht designation
Space group P63lmmc P63/mcm P63/mmc PGmc P62m P6222 P63mc %/m P63cm P63/mmc P&l_mmc R2m R3m Rjm R3m R3m R32 R2m Rzm R3m R3_m Rjc R3m R2.m R3m R3m RJm R3m C2jm c2/c c21c C2/m a l C
P2/c P21Ic P2dC P21Ic P21lc P21lC P2llC p211c Cmcm Cmmc Cmca Cmcm Cmca Cmca Cmcm Cmcm Aba2 Aba2 Cmcm Fddd Fddd Fddd Fdd2 Fddd Ibam Immm Imma Ibam Imma Pmma Pnnm Pnnm Pnma Pbnm
Pearson symbol oP8
oP12
oPl6
Prototype
BCu Ti Feb GeS SnS MnP TiB CozSi, NiSiTi (E-phase) CozSi HgCh Al3Ni AsMn3
E&
oP20
CdSb CuS,Sb (wolfsbergite) Fe$ (cementite) cr3c2
oP24
oP40 tI2
Sb,O, (valentinite) AuTe, (krennerite) CuFe S (cubanite) Nis Ifmillerite) TiO, {brookite) CrG
t I4
116
tI8 tIlO
w:
MoSi, ThH, AI,Ti ALBa
Strukturbericht Space designation group Pmmn Pilmfl Pnma Pnzcn Pnma Pnma Pnma Pbnm Pmnb Pnma Pmmn P411t71 Pbca Pnma Pnma Pnma Pccn Pma2 Pn-ma R3m Pbca Pnma I41mmm I4lmmm 141lamd I4/mmm R3m I4/mmm I4jmmm I4lmmm 14/mmm
14/m
I4lmcm 1411gmd
t112
14
2114 2116
I4/mmm 142d 142m I4/mcm I41lamd I4lrncm I41mmm I4/mmm I4/mcm I4/mcm
?I18 1126 1128
2132
14
tP2 tP4
tP6 tPl0 tP16 tP20 tP30 tP40 tP50
PNP AuCu CuTi, yCuTi PbO Pb,Sr PtS Cu,Sb PbFCl TiO (rutile) $b4Pt Si,U, PdS B Th
b-J
crCrFe AI,Cu,Fe Zn,P, YB
I4/mcm P4lmmm P42.12 P4/mmm P4lmmm P41nmm P4/nmm P4/mmm P42/mmc P4lnmm P4/nmm P42lmnm P4lnbm P4lmbm P42/m P4/mbm P42lmnm P4zlrnnm P4/mnc P4zlnrnc P42lnnm
xxxiii
Crystal Structure Nomenclature Arranged Alphabetically by Stmkturbericht Designation StTukturbericht designatlon Aa Ah A, Ad Ar A, Ah Ai Ak AI
A1 A2 A3 A3' A4 A5 A6 AI A8 A9 A10 A1 1 A12 A13 A14 A15 A16 A17 A20 Bo Bb B, Bd B.? BA= B33) BB Bh Bi Bk Bi Bm B1 B2 83 B4
B 1 B82 B9 B10 B11 B13 B16 B17 B18 B19
Prototype aPa PU RNP PNP HgSn,,,
YB
aP0 BPO aSe PSe cu
w
Mg aLa C (diamond) BSn In aAs YSe C (graphite) aHg or BPo aGa aMn (X-phase) PMn I* Cr,Si @W) CCS P (black)
aU cou o w n CaSi ?Nisi CdSb CrB MOB
wc
TiAs BN ASS (realgar) TiB NaCl (rock salt) CSCl ZnS (sphalerite) ZnS (wurtzite) NiAs Ni,In HgS (cinnabar) PbO yCuTi NiS (millerite) GeS PtS (cooperite) CuS (rovelite) AuCd
Pearson symbol
Space group
t12 tP30 oP8 tP4 hP1 tP50 CPl hR1 mP64 mP32 cF4 CI2 hP2 hP4 CF8 t14
I4lmmm P42lmnrn Pnma P4212 P61mmm P42/gnm Pm3m R3m P21Ic P2LlC Fm>m Im3m P63lmmc P63/?mc Fd3m 141lamd I4/mmm R3m P3121 P63jmmc R3m C-mca 143m P4132 Cmca Pm3n Fddd Cmca Cmcm n,3 P3 Cmmc Pbnm Pbca Cmcm 141!amd P6m2 P63/mmc P63/mmc P21 lc Pnma Fmzm Pm3m F43m P63mc P63mmc P63mmc P3121 P4/nmm P4/pmm R3m Pnma P421mmc P63lmmc Pmma
ti2
hR2 hP3 hP4 hR1 oC8 cI58 cP20 OC8
cP8 oF'l28 oC8 OC4 cI16 hp9 OC8 oP8 oP16 OC8 tI16 hP2 hP8 hP4 mP32 oP8 CF8
CP2 CF8 hP4 hP4 hP6 h P6 tP4 tP4 hR6 oP8 tP4 hP12 OP4
Stmkturbericht designation B20 B26 B21 B29 B3 1 832 B33(= Bf) B34 B3 5 B31 Ca cb
CC C,
c, c h
ck
c1 clb
c2 c3 c4 C6 c7 C8
c9 c10
c11, cl l h c12 C14 C15 Cl5b C16 C18 CI 9 c 2I c22 C23 C28 C32 c33 c34 c 35 C36 c37 C38 C40 C42 c43 c44 C46 c49 c54 DO, DO:
Prototype FeSi CuO (tenorite) FeB SnS MnP NaTl CrB PdS CoSn TlSe Mg,Ni CuMg, ThSi, PdSn, ThC, Cu,Te LiZo, CaF, (fluonte) MgAgAs FeS, (pyrite) Ag2O TiO, (rutile) Cdl, MoS2 SiO, (high quartz) SiO, (p cristobalite) SiO, (@ tridymite) CaC, MoSi, CaSi,
FeS, (marcasite) aSm TiO, (brookite) Fe,P Co,Si, NiSifi (E-phase) HaCI, AIB, Bi2Te, AuTe, (calaverite) CaC1, MgNi, (Laves) Co,Si Cu,Sb CrSi, SiS, ZrO, GeS, AuTez (krennerite) ZrSi, TiSi, PCu,Ti SiU,
Pearson symbol
Space group
cP8 mC8 oP8 oP8 oP8 cF16 oC8 tP16 hP6 1116 hP18 oF48 2112 oC24 mC12 hP6 hP3 cFl2 cF12 cP12 cP6 tP6 Iz P3 hP6 hP9 cF24 hP12 tI6 tl6 hR6 hP12 cF24 cn4 tIl2 oP6 hR3 oP24 hP9 oP12 oP12 hP3 hR5 mC6 aP6 hP24 oP12 tP6 hP9 oI12 mP12 OF72 oP24 oc12
P213 C2/c Pnma Pmcn PnFa Fd3m Cmcm P42/m P6/mmm I4lmcm P6222 Fddd I4llamd Aba2 C2jc P6/mmm P63/mmc Fm3m F43m Pg3 Pn3m P42!n?nm P3m 1 P6,lrnmc P6222 Fd3m P&/mmc I4lmmm I4/%mm R3m P63lmmc Fd3m F43m I4/mcm Pnrm R3m Phca P62m Pnma Pmnb P6/mmm R3m W m Pnnm P63 lmmc Pbnm P4lnmm P6222 Ibam p211c Fddz Pma2 Cmcm Fddd Pmmn I4jmcm
On4 oP8 tA6
continued
xxxiv
Crystal Structure ~
~
~
~
~
~
l
a
t
~
~
r
Arranged A ~ p ~ a b e ~ i bc ya StruktL~rbericht ~~y ~esignation~ c o ~ ~ i ~ ~ e ~
trukturbericht des~~nation
Prototype
Ir,Si AsMn, Ni3P CoAs, (~kutterudite) BiF,, AlFe, ReO, Fe,C ( ~ e m e ~ ~ i t e ) BaS, Na,As Ni,Sn A 1,Ni
Ni,Ti ~ o ~ i , A1,U PdSn,
M nu,
czAl,O, ( ~ o ~ u n d u m ) La203 r\/4[n,0, b,O, (seaarmontite) Sb,S, ZnP, Cr,C, Sb,O, (valentinite) Ni,Al, GNi,Sn, Ta43, A4C3 CO,%
Pearson symbol tIl6 0P16 1132 cI32 cF16 cP4 02'16 0Pl6 hP8 11P8 0P16 hP24 118 tIl6 hP16 tll0 0120 o c20 CPlO tP20 oF40 /IRIS tf1O tI26 tI28 hP6 cP36 cF52 tI18 oC28 cP7 cFI 12 rPlO hP1O cI40 h R.5 mP20 hRIO hP5 cI80 CfsO
0P20
tP40 0F20 oP20 hP5 me14 of14 hR7 cF56 cI28
Space group
Strukturbericht designation
Prototype
Co,Al, Mg,,(A1,Zn),g Ge7Ir3 Ga,Mg,
Pearson symbol m P22 ell62 cI40 0128
hP14 hR7 MO,& hP20 Th,S,, tp32 CrA W,Si, t132 cI52 Fe,Zn,, Cu,Zn, y brass cI52 cP52 Cu,Al, cF116 Cr,,C, Fe,W, (@-phase) hI113 c176 Cu,,Si, Mn,Si, hP16 eF68 CO$, CrgAlg hR26 Co,Al, hP28 oP40 CrF3 Fe,Th, hP20 PbFCl tP6 FeAsS mP24 0Cl6 AgAuTe, (sylvanite) mP12 CuFeS, ( ~ h ~ l c o p y ~ - i tIt6 ~~) CaTiO, (perovskite) cP5 Al,CdS, tll4 Al,Cu,Fe tP40 hP18 Al,FeMg,Si, Mn,Al~~i hP26 AlLi,N, c196 CuFe,S, (cubanite) oP24 cF112 Fe,W,C (q carbide) Al,C,Si hP18 cP12 NiSbS (ullrnanite) mC16 FeMS, hR4 NaCrS, CuS,Sb (~ol~sbergite) oP16 AI,MgO, (spinel) cF56 Cu,VS, (sulvanite) cP8 @u,FeSnS, 1116 cP5 Fe,N CuPt, cF32 AuCu tP4 AuCuII 01'40 CuPt hR32 AuCu, cP4 AlFe,C (perovskite) cP5 iiCuTi tP2 ThH, 116 AlCu,Mn (Heusler) cF16 cI54 Sb,TI, hP3 FC,N CuTi, tP4
~ , &
I
Space group
~
tetrahedra~~y coordinated compounds based on the wurtzite B3 (cF8) structures and all other coinpounds with structu stacking or substitutional or vacancy defect variants. There may or may not be ordering on cation and/or anion sites. Examples of both norinal ternary, mid quaternary adamantane compounds include: In ,Te, (*), and AgGd,InTe,. See tetrahedral structures. a
S
a
alloys, binary or ~ u l t i n a r ywhere , one of the components is merc~ry.They may be moiwphase or mlaltipliase, solid, liquid/solid, or liquid. Thus they may be, or contain, i~itermetallicsbut not necessarily so.
,
structure^, g e o ~ ~ e t r i c aidentical ll~ with the ~ u o r i t es~ructure(c the positions of positive and negative ions interchanged. Thus, whereas in Buorite (CaF,) each Ca' is surrounded by eight F- ions at the corners of a cube and each F-- ion by four Ca', ions at the corners of a t e t r a h ~ d r o ~in; the a n t i - ~ ~ o r i t e structure Li, Na, and oxides, sulfides, selenides, and tellurides, the chalcogen ions are close-packed with the alkali ions in tetrahedral interstices. and tellurides would usually be classed with I anti-fluorides also exist, e.g. digenite, (Cu9U)' sites are empty.
complex crystalline phases c o n ~ a i ~ large i ~ g clusters of atoms in nearly perfect tetrahedral coordination that approximate the local structure of a quasi-crystal il is an a~proximantto the icosahedral ~ u a s i - c r ~ ~int athe l Frank-Kasper phases (c1.v.) with the (cI162) structure; an example i s ~g~,(Al,~n)49 erthollides are iiiter~etall~cs in which s i g n i ~ ~ ~deviations nt from etry are allowed; thus these compounds must have CO concentrations of substitutional (antisite) or vacancy dcfects. Ternary have a mare complex definition. *
eta-
electron phases (S.V.)~ stable at an e/a ratio of 3/2, which are bcc structures, usually disordered at high temperatures, but may become ordered, CsCl temperatures, e.g. Cu n. Other exaniples are Ag,In, Al,Ce,,and
eta-
complex cubic (cP20) phases with /3 Mn the Frank-Masper phases (q.v.). Example
ructure; they are related to
*We define only compound families whose structure and/or composition are not obvious (from the term itself) and exclude tradenames, e.g. Alnico, Nitirial, Tribaloy, ctc.
xxxv
Some I ~ t ~ ~F~~ilies ~ e ~~~~~d t ~ l ~ ~ ~
XXXVi
A,B compounds AB (cP8) structure, isomorphous with Cr30 (W,O was ); include: Cr,Si, previously thought to be a tungsten ~ o l y m o r ~ h examples Nb,Sn, and Ti,Pt.
a nickname for the C60polymorph of carbon in allusion to the resemblance of its soccer-ball-like structure to the geodesic domes designed by ~uckminsterFuller.
es
62
literally all compounds formed between metallic elemen I s and the chalcogens, those of column VIB of the periodic table. Only selenides, tellurides and polonides are usually classed as inter~etallic~; they may be either binary, e.g. ZnSe or niultinary~ e.g. NaVTe,. cornpounds with the chalcopyr~te(CuFeS,) structure, (t116), related to zinc blende and having two general compositions: 136,. e.g. CuAlTe,, or 245,, e.g. MgSiP,, ZnSnSb,, and CdGeAs,.
c
compounds whose electronic bonding is neither wholly ionic nor wholly covalent, but a blend of the two; see also Zintl phases. ses
ternary compounds of the type M ~ where~M is a ometal, X, ~ a chalcogen ~ (S, ~ , Te), and x is a number between 0 and 4. ~tructurallythey may be regarded as med by stacking of Mo,X, building blocks wherein the M atom occupies large (x=l) or sinall (x> I) interstices. Many Chevrel phases are type 11 superconductors.
es
e
y intermetallics having ordered arrangements of the a-Mn 58) where the occupancy o f four crystallograpliical~ydifferent by atomic size factors; examples include F e , ~ ~ r ~ , M and o~,
~ g, 7A 1~ 2. a family of ~ o m p o u n TEX,,(2 ~s 2 m > 1.E),discovered where T is a transitio~metal and is from Croups I1 TiSi, structure (oF24) forms the basis of a large family of tetragonal superstructures embracing 35 crystal structure types, some of which have >GO0 atoms per unit cell. The T atoms form a ‘chimney’ of y1 p s e u ~ o c ~within l ~ s which ‘ladders’ of tn pseudocells of X atom pairs are drawn up to an equal height. Electron concentration plays some role in control of this family, and ternary compounds exist where a third element may partially substitute for either T or X, r other discrete structures. Examples include RuaSn3, V17Ce31, Ru,,(d;a,.,,ce,.gP;>1,4, and (Mn, RCr0.2)29Si51. compounds with large framework or cage stru~tures,analogous to zeolites, which can acco~modateother atoms within the cavities; examples include Ma8Si,, and Na24Ge,,,. compoLinds that are strictly stoichiometr~cwith no disce~niblehomogeneit~range; see line compounds.
e co
s 8
a class of quasi-crystals (q.v.) which have 10-fold rotational symmetry in the quasiperiodic plane, while being periodic i third dimension (also called t-phases). (A1,Si)-Go-Cu, and Al-Mn-Pd systems, Decagonal phases are found in the A1 among others. one of a family of related tcp phases iiic~uding6, p, g, P, and R viewed as hybrids constructed of various arrangements of Cr3Si (hP12) blocks; the prototype phase, MoNi, (oP56) is the (hP7), and MgZn, merous ternary phases with this structure (cPll2) occur, for
Some r n t e r ~ ~ ~ tFamilies a l l ~ ~ Defined
xxxvii
compounds analogous to metal hydrides but based on deuterium, the heavier, stable isotope of hydrogen, with twice the nzass of ordinary hydrogen, phases which form at certain ratios of the number of valency electrons contri by all atom species present to the total number of atoms in the unit cell* ratios near or about the . Examples include: @U ,Sn (7/4). Using various d e ~ n i ~ i o of n svalence electrons, tlie concept has also been extended to transition iitctal compounds such as s i phase (q.v.). of exceptional stability (ty~icallym.p electronic structure correlation, e.g. such a compound was first discovered in the Ni-Ti-Si system at the e ~ u i a t o ~ i c compositioii, hence E-phase; atoms occupy three independent site sets in the C0,S.I s t ~ ~ c t u so r e that ordered ternary alloys o Examples include: NiTiSi, CoGe’V, LalrAl a compound of different structure, ZrZn,,, is
one of the Humery electron phases e/a=7/ 4 (actual 1.65 to 1.89). Ex LiZn4. Other unordered hcp phases with a soniewhat larger c/a ratio and less restricted e/a range (1.22-1.83) are known as zeta phases (q.v.). S
aeei %
ees9
compounds isomorphous with Fe3W,C (q carbide) stabilized ternaries, e.g. Ta,Pt,O and silicides, e.g. Ni,T ternaries, e.g. Ni, considered as filled or partially fille
a type of one-dime~isiona~ quasi-crystal of mult~layered,related GaAlAs, usually formed by molecular beam epitaxy, where the ~ l t e r ~ aof t ~the o~ and B layers follows the Fibonacci rule (a numeric sequence in which each member is the sum of its two predecessors 1,1,2,3,5,8,13.. .).
etures
those structures with large voids into which other ato electronic s ~ ~ u c tcan ~ r ebe inserted, thus ‘filling’ or ‘s Examples include tlie Chevrel phases (q.v.)s Hf5Sn,C structure, filled gainina brass, e.g. Tl,Sb,, filled skutt fulleranes, e.g. Na,7,1n,97Z2 where Z=Ni, Pd, or
ppropriate size and
topologically close-packed (tcp) structures a c c o ~ i ~ o d aatoms t i ~ ~o~f ~ i ~ e rsizes ~nt
ses
See Laws phases. all-metal c o ~ ~ o u n dbased s 11) s t ~ u c t u ~exam on the c60 ful e~ Pd, or Pt), -LiMg include: Na961i~97Z, (Z=Ni, Pd, or at), Na172 where the @ symbols indicate onionskin-lik~c o n ~ e n t r ~ c (Mzmixed Li and A,(CT6, intercalation compounds formed between C6*fullerite (buckyballs) and an alkali metal, where the alkali metal ions can be a c ~ Q m ~ o ~ aint ethe d t e t r a ~ ~ dor r~1
~
~
xxxviii
Some ~ ~ t e r ~ n ~~t aal ~l i i~ DeJined l ~ e ~ octahedral interstices of the fullerite. Strictly speaking, these are not true I a kind of carbide. However, analogous true IMCs exist (see fulleranes).
G-phases
ordered ternary compounds of the Mn,,Th, D8, (cFll6) structure. The first-known example was dubbed 'G-phase' because of its presence as an embritting precipitate, 011 grain boundaries o f Ni-base high-temperature alloys. Other examples Ni16Ti6§i7. include Co16Hf6Si7CoI6Zr6Si7, and Cu16 electron compounds or Hume- othery phases (q.v.) with an e/a ratio of 21/13 and (~152)crystal structure. They are built up from 26-atom clusters in various ng arrangements. The architecture of these clusters consists of an inner tetrahedron (4 atoms), an outer tetrahedron (4 atoms), an octahedron (6 atoms) and a cubo-octahedron (12 atoms), sites which inay be variously occupied. Examples include: CusZn 8, gamma phase (cI52); Cr,A18 (h 26); Cu9A14(cP52) and Cu41Snll(cF416). these phases are covalently bonded coinpounds, with a zinc blende (sph~leriteor wurtzite) structure, formed between elements lying equally to the left and right of Group IQ elements, thus 111-V, 11-QI, and I-VII compounds (4.v.).
ses
hexagonal analogs of perovslcite phases with formula type T, Tztransition metal, M=non-transition metal, and X is a metalloid; examples include Ti,AlC, Ti,lnC and Cr,AlC. ternary sulfides, selenides, and tellurides of the defect tetrahedral class with one ordered vacancy per formula unit. Int e~et al l i cexamples include: ZnA12Se4(cF5&), HgGa,Te, (tll6), CuSbSez (oP16), and CdIn,Te4 (tI14). nds or half-filled (also known as sem Whereas Heusler all0 formulation, struc interpenetrating fcc sublattices, the half-Heusler phases possess the same cubic structural motif but in them, as indicated by their XYZ fornzulation, half of the X sites are empty in an ordered array. The resulting narrow bands and gaps in the energy spectra lead to semiconducting, half-metallic conducting, paramagnetic or ferromagnetic behaviors. So mising thermo-electric rm in the same system, as is sometimes properties. Even if both XYZ the case, it does not imply a c id-solution between them. Examples of half-Heusler phases include T1 hexagonal or rhombohedral phases formed by various stackings of Laves and CaCllrs layers. Examples include: CeNi, (hP24), Ce,Ni7 (hP36), and Co,,Sm, (hR24).
ounds
a class of IMCs whose low-temperature properties are characterized by a huge magnification of the density-of-states of charge carriers near thc Fermi level and the appearance of quasi-pa~ticleswith eKective masses of the order of 102-103 times the free electron mass; in some cases they have been treated as 'concentrated Kondo systems'. They usually contain a rare-earth or actinide element with a partially filled 4f or 5f electro shell; they are divided into two subclasses: those with no~~integral valence ) and those with nearly filled valence (#). Examples include: UBe13*, U6Fe, URu,Si,*, CeA12#, CeNi," CePdAl, CeAs, and NbNiSn. compounds which crystallize
(cF16), some of which s; other (non-magnetic)
Some ~ ~ t e ~ m e t uFumilies l l ~ c Dejned
ses
s
phases which form at characteristic ratios of valence electrons to atoms, 312, 21/13, and 7/4. See electron phases. Beta brass, ganiina brass, epsilon pliase, and zeta phase are such phases in the Cu-Zn system. so~~etimes called i-phases; a major class of ~uas~-crystals whose c o m ~ o ns inotif is a group of 20 slightly distorted tetrahedra packed around a comnioxi vertex; the struct~rehas no transl~tionalperiodicity. i(CuFe), i(TiMnSi), and i ( ~ t ~ S i ) . compounds in which metals play both cation and anion roles; normal valency rules are followed; they are densely packed, highly symmetric $tructure~with little or no directional bonding; stoichiometry is very closely observed with no signi~cant homogeneity range; and properties exhibit characteristic behaviors of ionic bonding. Examples are Mg,Si, Fe,NiN, and e (see normal valence ~~mpounds).
cs
iO
xxxix
ternary coinpounds whose prototype is Fe3AlC,, which may st~ucturesstabilized by interstitial carbon or as a perczvskite examples are: Ti3AlC, Fe,NiN, and Mn,AlC.
haws
a series af structurally related phases, carbides, borides, oxides, and all-metal prisms, icosahedra, intermetallics formed from M,M’4 building and octahedra, variously filled andlor o,W,, (hP34). F&IfgRe,
ases
topologically close-packed compounds formed by stacking of rows of icosahedra (or layers o f pentagonally coordinated atoms in (1 phases (q-v.), but here with 5, 6, 8, 9, or 10 layers, g(NiC42, and ~ ~ ( ~ n , A g ) , * lattices (both metallic solid solutions and IMCs) coiitaini~ a small con~entration of ma~neticimpur~tiesin a non-magnetic matrix. A cha~~cteristic ~ i ~ i m inL the ~ i ~ resistivity vs. 2”’is shown by these systems at very low temperatures; below the m i n i ~ u m the , resistance increases logarithmically and then becomes temperature i~dependent. Interesting pheno~ena that have been de~onstrated by these materials: the possibility to become s~perconductingand yet not possess a superconducting energy gap; to become superconductors at one tenip~ratureand at lower te~peratureto come out of the superc~nductin~ state. E dilute Fe in Cu, dilute Cr in Cu, dilute Fe in Au, dilute Ge in (La (La, Ge)B,. OV CO
s
compouii~sformed, not from a liquid alloy, but by orderi terminal solid solution or intermediate phase, e.g. Cu3Au, Fe,
~ ratio close-packed AB, compounds, whose stability derives from both a t o size three basic structu~~s: and electron c o ~ c e ~ t r a t i and o ~ , posses (hP24). They may be (liP12); MgCu, C15 (cF24); and MgNi, formed by the stacking of two kinds of la one puckered h ~ m o ~ e n e o layer u s of atoms of different sizes, designated by T, and the other a 3636 k Thus the structural formulas of the common Laves phases are: [[CU,][M~]](~~)~’, and [ [ ~ i 2 ] [ M g ~ ~ ( S ~true r ~ )tures c h . derivalive fro ~ultilayerstacking (see Kornura phases), ordering in ternary or ~ u a t ~ r ~ a r y compounds, or deformation. Other examples are: Al,iZr, FeSi FeNiTa (cF24); and NbZn,, Cu6Ga4Mg, (hP24). The iii derivative type are the AuBe, (cF24) structures and the CaCu,
Some I
~ F ~ ~ iDefined ~l i ~ ~
formed by a combination of orde include: PdBe,, UNi, (cF24) and 8
~
~
~
n. Examples of this subclass
compounds that are strictly stoichiometric with no discernible range of composition (see Daltonides). of the marcasite structure (q.v.) named after the mineral loellingite, (oP6), having a e/b cryst~~llo~rapliic ratio of about 0.485, lower than that of the other marcasites (-0.615). Examples include: As,Os, CrSb,, and RUP,. alloys (either ordered solid solutions or conventional IMCs) exhibiting stable periodic displacements in the ordering scheme by interposition o f antiphase b o u n ~ ~ r i ethus s , extending the unit cell. The period may be in one, two, or three directions in the antiphase domains and rnay vary from 2-10 as a function of teniperature and composition; it rnay even be non-integral due to mixing of two or more regions with integ of 1-d LPS in ordered alloys include CuAu 21, Au3Cd, and in ordered alloys include: Cu,Pd,-,, Au3+Zn, and Au3Mn. Examples of conventional IMCs with LPS include ZrSi2, @e,U, and Nb,Ga,,. belong to the F r a n ~ - ~ a s p efamily r (q.v.) of t o p o l o ~ i c a close-packed ~l~ structures with prototype A l , ~ N b 4 ~(0~ i 3 ~ this group, named after the mineral marcasite, FeS, derivative from the NiAs type. Examples include: As2 those in which a commensurate or incoinrneiisurate modulation is superposed on the basic ordering scheme. Examples are: Ta,-,Nb,Te, and Zr, superlattices). one o f a faniily of related tcp phases, including 6 , p , 6,P, an as hybrids constructed of various arr~ngementsof Cr, gZn, (hP12) blocks; the prototype for mu is the rhombo ,, and ~ a , M ib19. i ~ ~ ~ other examples are: C o ~ ~Fe,SiRe,,
a compound is called a normal valence compound if, under the assumption that the cations 63 formally transfer all their valence electrons, their number is just correct for all the anions A to be able to complete their octets without sharing electrons. Thus in normal valence conipounds one finds neither cation-cation nor anionanion bonds nor lone electron pairs on the cations (see ionic inte~etallicsand octet compounds). a populous group of coinpounds where electronic factors, as well as size factors, play a role; they rnay be binaries as in the prototypes ,(hP16) and VV5Si3 D8, (tI32), but iiiore often ternaries: C, N,0, Gd,Ce, and CHo,Si,, or even all metal ternaries, et Q:
also known as normal valence compounds (q.v.), a subgroup of valence compounds nce electrons per AB unit; examples 452ternary ~ o m p o u n also ~ s occur, e.g. nGeAs,. See also normal valence compounds and
a transition phase displacive and ordering r e a c t ~ Q ~ $
f certain Ti alloys undergoing both
a family of ternary s~perstruct~re of D
is AlNbTi,. The structure is oC16, a of Ti and Nb on mixed sites of the
~
Some ~ n t ~ r m e t a lFamilies li~ Defined
xli
hexagonal planes; alternatively, it may be regarded as related to the structure through the intermediate B 19 (op4) phase. AlNbTi, is remarkable among IMCs for its relative ductility. one of a Pdmily of related tcp phases, including 6, p, U, P, and R phases, that can be viewed as hybrids constructed of various arrangements of Cr,Si A15 (cP8) , Zr,Al, (hP12) blocks; they have the (oP56) structure; an example is
ases
intermetallics whose structures derive from the 3-d analog of Penrose tilings. quasicrystals and pentagonal intermetallics.
llics
those with a 5-fold symmetry in their x-ray diffraction patterns, the first such discovered being i(A1Mn). This pentagonal character is also manifested in pentagonal Facets on or pentagonal channels through quasicrystals (q.v.) and related structures. It has been shown that such structures relate to Penrose tilin which are non-periodic ways of tiling a plane or 3-d space. It can also be demonstrated that all 3-d Penrose tilings result from the projection of part of a 6-d lattice onto a 3-d space. A currently proposed model postulates that 3-d quasicrystals are formed from arrangements of a single type of building block, an atomic cluster, where neighboring clusters overlap by sharing atoms. Examples of peiitagoiial intermetallics include: Al,,Co4, Al,Mii, and Cu9,,TeSb,. d after the mineral or L1; (Fe,AlC), (cP5), p from cubic-close-packed layers; the A atoms fill some of the octahedral holes in the ccp stacking and are thus surrounded by six X atoms. Although the majority of perovskites are oxides or , or 0 can stabilize metal atoms in this structure, e.g. and O,PtV,. There are also cases where only sub-sto of the interstitial are needed at the A sites to stabilize the rnetal atoms in this structure, e.g. NZn2Ti, or CGe4Co12,or where non-adherence to the A formulation can be accounted for by random inter-substitution af A and X atoms on their normal sites. rnetastable upe er conducting phases formed by rapid quenching or ~igh-p~essure synthesis of a combjnation of noble metals with group elements or of two different group B elements and having an e/a ratio of 4-5. They have the PO (cP1) structure. Examples include: AuBi,, Pd3Sb17,and SnTe,,
es
literally all compounds formed between metallic elements and those of column V of the periodic table. Only arsenides, antirnonides, and bismuthides are usually classed as intei-metallics.
s in these compounds, homocatenation or polymerization occurs via linkage of like ions to form ‘polyanions’. Thus in polyanionic compounds the number of valence electrons transferred by the cations is i n s u ~ c i e for ~ t the anions to complete their octets, and ions of the higher valence element tend to cluster, forming bonds between themselves. For polya~ioniccompounds we have CsPb (Pb4 tetr~hedra), NaGe (Ce, tetrahedra), and CaSi (Si zig-zag chains). In polycat~oniccompounds such as GeAs, Case, and GaTe, more valence electrons are available than necessary for octet completion, so the catio (GaSe) or have lone electron pairs at polycationic compounds (principally cha stackings (not close-packed) of four close-packed layers of atoms in triangular istances between chalcogenide atoms are such that van der Waals
xlii
Some ~ ~ t e ~ m~ e~~ ~~ ~i~ l le ii ~~ e n~ ~e ~ bonds exist, and the separation of cation pairs from those in an adjacent layer is such that the compounds are not metallic conductors but semiconductors. The polytypic stacking types include ABC as in hex GaSe, rhoinbohedrat GaSe or complex stacking as in monoclinic C stable or metastable phases characterized by non~crystallographic orie~ta~ional symmetry and quasi-periodicity~ these comprise: icosahedral compounds (q.v.), decagonal compounds (e.g. t-phase), and even one-diniensional compounds. one of a family of related tcp phases, including 6, p , CT, P, and R phases viewed as hybrids constructed of various arrangements of Cr,Si Zr,Al, (hP7), and MgZn, 614 (hP12) blocks; the prototype phase (hR5 Co-Cr-Mo system; other examples are Fe, iV,, Mn,Ti, A13&4g2,, and Ni3SiV6. phases possessing orientational as well as translational disordering. ~nterinet~l~ic examples include NaSn and CsPb.
a more complex d~velopmentof the Frank-Kasper family (q.v.1 which consists of arrangements of ftised, high-coordination polyhedra rather than fully interpenetrating polyhedra as m the Frank-Kasper phases. The family, studied principally by S. 0. Samson, is of partic~ilarinterest because of the very large unit cells with > 1000 atoms. Exttmples include: Mg NaCd,, and Cu,Cd3.
a phase first found in Fe-Cr alloys, t e t r a ~ o ~ tcp, al
(tP30), a member of the Frank-Kasper family where full or partial ordering can occur on three of the five sets of crystallographic positions in the structure. Examples include: FeCr, CrRe, AlMoNb, Ir4Moll,and Ni,SiV,, compounds whose structure and stability are dominated (but not exclusively) by geometric packing considerations; examples include Laves, Frank-Kasper, and trigonal prismatic phases (q.v.). intermetallics isomorphous with the cobalt-rich mineral skutterrudite, (~#~Ni,Fe)As,, with cIJ2 structure. They include the analogous Pt-group arsenides, binary and ternary antimonides, and certain phosphides~e.g. IIrAs,, CoSb,, FeNiSb6, and PdP,. spinels are mostly oxides and sulfides whose general structural formula is AtB,O[G,T (where t and o mean tetragonal and octahedral coordination respectively and c, cubic closest-packing) and with (cF56) structure, but inany other, more nearly intermetallic, isomorphous compounds e or Te, e.g. Al,CuSe, and CuCr,Te4, and even all metal compound A atom occupies tetrahedral sites are tetrahedral sites and some A atoms in octahedral site general structural formula o f the inverse spinels is Bt (also known as fast ion conductors or mixed conductors) compounds in which a significant fraction of the current is carried by ions as well as electrons. Most such compoun~sare oxides, chalcogenides~or iodides, but intemetallics such as LiAl and Mg3Bi2also belong to this class. See decagonal phases. carbides, borides, or carbo-borides that are stabilized Cr,,C, structure types; exaniples are Crl,Fe,NiC,, Ni20Ti,B,, and IFe23C3B,. those in which every atom has four neighbors s u r r o u n ~ i nit~t e t ~ ~ h e ~ r adefect lly~ tetrahedral structures have some vacant coordination sites. ln such M,N,
Some ~ n t e ~ ~ n e t Fumilies ~ l l i c Defined
xliii
compounds, ~/~ = (eN-4)/(4--eM)where eNand eMrepresent the number of valence electrons on each atom. Examples of normal tetrahedral 1 ~ include ~ 1iiSb s and ZnAs, in the binaries, Cu,SiTe3 and Cu,AsSe, in the ternaries, and ~nIn,GeAs4 and ~ g C d ~ I n T ein, the quaternaries, Defect tetrahedral struct~resinclL~defor e3, Iii&s4Te3, HgGaTe,, and CuSbSe,. ee also ndainantane structures. phases comprising atoms of unequal size in which pscki packing in various ways. Examples include sigma, Lwes, a r4Al,, and various hybrids of these.
~ses
binary phases c o ~ p r i s i natoms ~ with radius ratios subst~~tially greater than one3 the larger atoms forming a triangular prism, ceiitered by the smaller atoms. They are found p r e d o ~ ~ i ~ a nin t l yborides, carbides i n t e ~ e t a l i ~ cexist. s Dy5Ru2 (mCZS), Th
s
phases which require vacancies for stabilization of the structure for either lattice strain or electronic reasons. They are of three types: those whose stoichio deviation is fulfilled by'the presence of vacancies on the minority atom sites, Nil-,Al and Pdl-, In, botli(cP2); those where the vacancies form an ordered superstructure of a simpler basic structure, e.g. As,Cd, (tII6O) derived from the CaF2 structure (32 vacancies per unit cell); and those where a cer vacancies is required but no defect ordering occurs, e.g. In,Te, (cP52). In the latter two types, the vacancies may occur on either or both of the two metal sites. In at least one conipound, PtCe1.619 phases with ordered and disordered vacancies co-exist. compounds in which all atoms, either accept, provide, or share valence electrons to obtain stable octet con~gurations,ns2np6; they may arise either by ionic interaction (a~eptance/donationof electrons) or covalent interaction (sha electrons); most compounds exhibit a hybrid or intermediate type of bo Examples include: Pb e, Cu,Mg, KGe, and CuSi,. Normal valence compounds are those where the ratio of total valence electrons per anion=& Tn a broader sense, valeiice compounds also include those (polyionic valence compounds, q.v.) where this ratio i s 8 (polyanionic compounds) and where it is r compounds). ternary silicides and germanides, A4B4C,, tI60 structure, prototype where A is a transition metal, B can be Mn, Fe, CO, Ni, or Cu, and C is Other examples are: Nb4C04Ge7 and Fe4Hf4Ce7; they have crystal characteristics similar to both Frank-Kasper and Laves phases (q.v.). electron phases with disordered hexagonal close-packed structure occurring over the e/a range of 1.22 to 1.83 and with c/a=1.633. Examples include: ~ u ~Ag,Al, ~ a , and Auslii. lsostructural related phases with c/a=1.57 and more restricted e/a range (1.65-1.89) are known as epsilon phases (4.v.); metastable zeta phases are commonly formed by rapid quenching; many of them are superconductors. phases formed by metals from opposite sides of the periodic table, characterized by ~on~pletely filled electronic orbitals, normally a full octet shell; their bondin be ionic, metallic, covalent, or mixtures thereof, and they CLlr in a variety of crystal structures. Examples iiiclude: NaTl, Mg2Si, &1 3, MgLiSb, arid ~~~gAuSn.
xliv
Some In fermptcdlic Families DefiBed the 12 compounds of the group IIB elements (Zn, Cd, and Hg) with the group VIB chalcogens (0,S, Se, and Te); only the selenides and tellurides would be classed with intermetallics. Many of these and their alloys are of interest for electronic applications.
CQ
a class of covalent semiconductor materials formed as a combiiiation of group IIIB elements (B, Al, Ga, In, and Tl) with group V elements (N,P,As,Sb, and Bi). The B, N, and P compounds are not normally regarded as intermetallics, although their structure and properties may be similar.
the equiatomic chalcogenides formed by combination of the group 1V elements (Si, Ge, Sn, and Pb) with the chalcogens ( S , Se, PO,and Te); only the siilfides would not normally be regarded as internietallics. potentially semiconducting cornpounds formed between group VB and VIB elements. Intermetallic examples include: Bi,Se, and Sb,Te3.
Q ~ Y f~ ~Ssome r
of the define
American crystallographer (1942- ) who worked with the Shoemakers at California Institute of Technology. is
French physical chemist (1 748-1 822) ~ a i ~ t a i n that, e d under certain ~ircu~stances, there could be variable combining ratios in chemical reactions. American physical chemist (1919- ), long-time member of the faculty at Uiiiversity of California-~erkeleyand staff member at Lawrence iation Lab.; espoused ctronic structure in the and promoted Engel's unconventional view of the role o stability of IMCs. French mineralogist and chemist (1931- ). At CNRS, ennes, France, known for his synthesis of ternary molybdenum sulfides, which have striking superconducting properties. English chemist (1766-1 8441, enunciated the chemical law of definite ~roportions; hence IMCs which admit no deviation froin stoich~o~etry are called Daltonides. ~anish-Ainericanphysical chemist (1904- ) who proposed rules for the existence of particular phases and crystal structures dependent upon electronic structures, not only the average number of unpairecl electrons available for bonding, but also involving roles for the s, p, d, and f electrons. Italian-American theoretical physicist (190 1-1954) who studied electronic structure of a toms (Fermi surface) and who argued from quantum-mechanical principles that in certain cases at low temperatures huge magnifications of the density of states of charge carriers should occur. Italian mathematician ( w 1170-1230), also known as Leonard of Pisa. Published several books and was arguably the most outstanding niatliematician of the Middle Ages. He popularized the modern decimal system of numbers and is best known for conceiving the mathematical series, the value of each tern of which is equal to the sum of the two preceding terms, IY,,+~=U~, iU,.
~r~~~
British physicist (1911-1998), served in scientific intelligence at the British Defence Research Establislimeiit during the war years and iininediately thereafter, professor of physics at University of Bristol for most of his career; worked on dielectrics,
Same ~ ~ t e r ~ e t ~F al l~i ic ~ i e ~ ~~~~~~
xlv
dislocation theory, and polymers in addition to his collaboration with understanding the structures of complex IMCs. American electrical engineer (1896-1972) who first determ~ned the crystal structures of MgCu, and MgZn, while at Carnegie Institute 1927, and later worked for the Navy Dept, Bu. Ordnance, and
ria
Ameri~anarchitect and designer (1895-1983) who conceived and ~ u i l tgeodesic domes, the structure of which was seen by crystallographers as analogous to the structures of the C6*polymorph of carbon, hence ‘Buc~yballs’. Geman physical chemist (1887-1958), at University of ~ u n i c hstudied the nature of bonding, especially in inorganic compounds. German inorganic chemist and crystal~ographer(1915- ) at the University of Geman inorganic chemist and crystallogra~herat TechnischeHochschule, German i~dustrialist(1866-1947) at ~sabellenhutte,GmbH in ~ i l l e n ~ u who r g ~ in 1914 serendipitously discovered certain nonferrous alloys which were magnetic, e.g. Cu-Mn-Al, although containing no ferromagnetic elements. Nfeusler’s son, 0 years later determined that the magnetic behaviors derived from the c o ~ p o u n dC , u ~ M n ~hence l , Heusler alloys.
am
English metallurgist (1899-1968), long-time member of the fxulty of University of Oxford, a ~ t h o of r several books on the structure of metals and alloys; showed that electron concentration controls the structure of many intermeta1lic phases. American crystallographer (19 15- ), at General Electric R D Center for most of his career, ~ursuedX-ray and neutron diffraction studies of crystals, es~ecially complex intermetallic compounds. Japanese crystallographer, originally at Osaka City ~niversity and later at Hiroshima University and Tohoku University. Japanese theoretical physicist (1917- ) who was able to account for anomalies in the electrical resistivity, specific heat, and magnetic properties of certain alloys and intermetallic compounds.
kov, olai
ovic
Russian metallurgist and physical chemist (1 860-1 941), studied phase equilibria in salt systems and alloys, recognized metallic phases of variable c o ~ ~ o s i t i o n (~erthollides)and ordering of solid-solution alloys. German-Swiss crystallographer (1906-1978) who affiimed that the existence of certain intermetallics and their crystal structures could be understood in terms of ion of dense packings of atoms of different sizes and specific proportions, We now understand that both size and electronic effects are operative. ~ u s t r i a n metallur~ist and crystallographer (19 1 1- ) contri~uted ~ ~ ~ e r o u s publications on structural, mineralogical and metallurgical chemistry. r ~tiling patterns, which ~ n ~ l~ ia ts~ ~e m a ~ i c(1931i a n ), worked on the g e o m ~ t of turned out to be the basis for the atomic arra~igeme~t in ~uasicrystals. Swedish-American crystallographer (19 17-1 993), at California Institute of Tecliiiology, known especially for studies of ~ n t ~ ~ ~ e twith all~ i ~ si cells ~ (>n1000 ~~to~s). German physicist (1868-1951). For most of his career, professor of theoretical physics at University of Munich; explained the role of the electron in the metallic state.
German chemist (1898-1941) at Technisclie Hochschule, Dai-instadt, proposed a rule di~erentiatingintemetallics whose bonding is essentially ionic or heteropolar and which possess ionic-type crystal structures from those whose bonding and structures are more typically metallic. ques and otlier input to these definitions were contributed by R. W. Cahn, J. Lima-de-Far~a, Pearson, and R. M. Waterstrat. Any remaining errors are the responsibility of the author J. El.
Socrates: ‘But why should we dispute about names when we have realities of such importance to consider?’ Claucon: $Why, indeed, when any name will do which express the thought of the mind with clearness?’ 0 427-347 BC
. . . a man that seeketh precise truth, had need to remember what every name he uses stands for; and to place it accordingly; or else he will find himselfe entangled in words, as a bird in lime-twigs; the more he struggles, the more belimed. s 1588-1679 There can be nothing clearer or more conve~i en~ for the purpose of setting one’s ideas in order and for conducting an abstract discussion, than precise definitions and inviolable lines o f demarFation. 0 ~ t ~ 0 184.5-1923
~
~ ~ l en ~ e ~n~~t e~ upun t~ ~r ~~ ~r ~t solo. i ~~ te ~ o much can elements do when nothing cre
11 things began in order, so shall they end, and SO shall they begin again, according to the ordainer of order and mystical mathematics of the 2
. . . all the work
of the crystallographers serves only to demonstrate that there is only variety everywhere where they suppose ~ n i ~ o r m .i t.~. that in nature there is notliing absolute, nothing perfectly regular.
Chapter 1 Structural ClasslJication and ~ o t a ~ i o ~ Josk Lima-de-Faria Centro de Cristalograjia e Mineralogia, Instituto de Investigaq?o Cientgca Tropical, Lisboa, Portugal
1. Introduction Similar to chemical elements, the inorganic crystal structures call for some display scheme, such as a map or a table may provide, if their individual organization and mutual relationships are to be easily recognized. The systematic knowledge of such organization is synonymous with structural classification and, to move within such a large domain, a ‘vehicle’ should be used which would facilitate the recognition of these relationships, namely, a good notation. The aim is to understand how the chemical elements are linked together in order to reveal the rules, laws, tendencies, principles, etc., of the architecture of inorganic crystal structures. 2. The Structural Classification of Inorganic Compounds Several attempts have been made in the past to present general classifications of the crystal structures (Table 1). Laves (1930) was possibly the first to propose a classification of crystalline structures, as a kind of extension of the structural classification of silicates to all crystal structures. Niggli (1945), Gdrrido and Orland (1946), Bokii (1954), and De Jong (1959), proposed similar classifications. In all of these attempts only a general approach was proposed but was not applied systematically to existing compounds. In 1976, Lima-de-Faria and Figueiredo made the first systematic approach to a general classification of inorganic structure types, and applied it to 782 structure types, which corresponded to approximately
5200 compounds (Table 6). Not only chemical composition but also, and mainly, the crystal structure determines the properties of a compound. This is why structural classification is the natural classification of inorganic cornpoundy. Later, Hawthorne (1983) also suggested a structural classification of crystal structures based on the polymerization of the coordination polyhedra with higher bond valence as a consequence of the application of the bond valence theory to inorganic structures, and considered four main categories: clusters, chains, sheets and frameworks. In 1989 Jensen proposed a classification based on Mdchatschtki (1947) and Niggli’s (1945) ideas, also considering four main categories: molecular, chain, layer and framework, and developed a complex notation, the crystal coordination formulas. Kitaigorodskii (1955) studied the organic compounds and considered only four categories of organic structures: molecules, chains, sheets and frameworks. Consequently the classification of inorganic structures by Lima-de-Fark and Figueiredo (1976), which acknowledges five main categories, namely, atomic, group, chain, sheet and framework, may be applied not only to inorganic structures, but also to crystal structures in general, either inorganic or organic. The term coordination structure used by Bokii (1954) and by Dc Jong (1959) characterized certain structures that are homodesmic (structures in which all bonds are generally similar in kind, though not identical), the particles of which are surrounded by adjacent particles more or less evenly (De Jong, 1959, p. 144). They considered as an example the halite
Intermetallic Cornpounds:Vol. 3, Principles and Practice. Edited by J . W.Westbrook and R. L. Fleischer. 02002 John Wiley & Sons, Ltd.
Structure and ~ o ~ p o ~ ~ i t i ( ~ n
4
General classification of crystal structures Laves 1930 kikdlld§
Chains Sheets Frameworks
Niggli 1945 Isolated particles Molecules radical^ Chains Sheets Frameworks
Garrido and Orland 1946
Bokii 1954
De Jong 1959
Isolated groups
Coordination Islands
Chains Sheets Frameworks
Chains Sheets Frameworks
Coo~~ination Molecules Radicals Chains Sheets Frameworks
wever, this term presents a certain framework class, for which they selected, as an example, the cristobalite structure of i02. This ambiguity disappears if halite is considered close-packed structure, and cristobalite a framework r fact balite is a closest packing s t r ~ ~ c tAs ~ ~a e~ .a t t e of of the C1 atoms with Na atoms occupying the hedral voids; and cristobalite is a framework of tetrahedra. lose-~acked structures are dense tures, and frameworks are less dense; they are open structures, containing large voids. The structural classificat~oiiof inorganic compounds i s based on the ~ o ~ d - s ~ rdistribution e ~ ~ t h and also on the directional character of the bonds. These factors define the so-called struct~rulunits, which are the parts of the structure more tightly linked together internally ed in various ways to form different The main structural cate close-packed (UP utomic), where the structural units are i n d ~ ~ ~ datoms ual - g ~ o ~ p( ~ s ~ ~ b b e ltrian~les, ls, rings, small chains, etc.) - ~ ~ ~ ~chains n i t e i ~ ~ n~ i t ~ e ~ ~ ~ s -
I
-~
~ u ~ e ~ o ~ * ~ s .
For a more detailed description of our structural c~assificatio~~ see Lima~~e-Faria and Figueiredo (19’76) or Lima-de-Faria (1994).
3.
In order to relate easily the inorganic structures we use structural formulas, that is, chemical formulas to whkh is added the structural information. The main structural factors to be indicated are: the structural unit, i t s category and way of packing, a i d the c o o r ~ i ~ a t i oofn the atoms:
Lima-de-Faria and Figueiredo 1976 Atomic (or close-p~cke~) Croups Gbains Sheets Frameworks
the structural unit is placed within square brackets the category of the s t ~ u c t unit ~ ~ (x) r ~is~placed within curly brackets just before the symbol of the structural unit the packing ofthe ~~ructurul unit (y) is indicated as a superscript at the right-hand side of the structural unit the c o o r ~ i ~ ~o tj ithe ~ ~atoms ~ s (a, 8, y , 3 , . . .) are represented as ~uperscripts~within small square brackets, at the right-hand side of the syinbol of each chemical element.
The category of the structural unit is expressed by its dimensionality with the aid of the symbol 00, for the structural units of infinite diinensionality like chains, sheets or framewor~s.The ~~rresponding symbols are { 1CO 1, (2 CO 3 , and (3CO) respectively. They correspond to the Machatschki syinbols &, A) but written in a com~uter-frien~ly way. For finite dimens~oiiality we use the symbol instead of writing zero infinity, {g) for group, and for the structural units of individual atoms of the close-packed structu~esno symbol is used. When there are no i~terstitialatoms and the structural formula starts with the infinite dimensionality symbol, the curly brackets can be omitted. The structural formula of a compound corresponds therefore to its cheinical formula, complemented with the structural information mentioned above. Accordingly, the general structural notation for a c a ~ p o u n d
(A, A,
For instance, for pyrite, FeS,, the structural forinula is Feo&) [SzIc,meaning that the sulphur atoms farm S, groups which pack to ether in such a way that the centers of gravity of th pairs of sulphurs form a facecentered-cubic lattice, that is a cubic closest packing c, and the iron atoms occupy the voids of this packin with an octahedral coordination 0. The structural formula for halite is Na0[CIIc. The structural units c h are formed by the individual Cl atoms, ~ ~ i are
T a ~ 2l ~Various
chemical formulas and corresponding
of higher dimension. Examples of structural formulas are given in Table 2. The cubic body-centered packing is re~resentedby Bb, where B means the nature of the layer of this packing parallel to (010) (Figure 1); and the symbol b, the stacking at a half of the side of the plane unit cell of this layer. The various sym~olsfor the stacking of layers are given in Figure 2. 21 is More complex layers are the formed by interconnected triangles and squares, with tetragonal symmetry, in the proportion o f two triangles for one square. An example is Cuacb f meaning the kind of stacking. R2I is formed by rows of triangles and rows 2 to 1 (Figure I}. An For inore commo hexagonal closest packings, and their co~binations c/h, and also for the cubic ~ody-cente~ed packing the symbol for the kind of layer is omitted, and ch, . . . b are used.,The Laves phases can be as decomposed In T layers means puckered, the dot
structural formulas Structural formula
Chemical f o ~ u ~ a Cu (Copper) NaCl (Halite) Mg,SiO, (For~terit~) AI,MgO, (Spinel) TiCaO, (Perovskit~) S (Sulphur) FeS, (Pyrite) CaMgSi,O, ( ~ ~ o p s ~ d e )
KAl~(OH),Si~AIO~o (~uscovite) C (Diamond) KSi,AlO, (Microcline)
packed in c, the cubic closest packing, with Na atoms occupying the octahedral voids 0. The category of the structural unit (in this case the close-packed} i s not des~gnated,because, by convention, it only need be represented for structural units
( w
N 2'
T
T
K
i ~ ~ 1r eSome homogeneous and other ~eterogeneou~ layers (after L~~a-de-Faria and ~igueiredo,1978)
6
Structure and Composition
C
e 2 Cicneratized stacking symbols. T stands for triangular tesselntion; Q, R and 0 mean quadrangular, rectangular, and ue unit cells respectively (Lima-cte-Faria, 1994; adapted from Lima-de-Faria & Figueiredo, 1976). The arrows indicate the d i s ~ l a c e ~ eofn the ~ ~uccessivelayers from the origin of the unit cell
means heterogeneous layers of atoms of different sizes, and T triangular tesselation) and a K layer (the kagomk layer). These two layers and K have only one way of stacking together, but the assemblage (T?' K) may be stacked in two diff'erent ways, c or h, as in the erefore to the more comnion Zn,Mg, and Ni,Mg, correspond [[Cud [Mg]]@"k, [[Zn,] [Mg]J@K)h,
The structural. formiila of GaCu, is [[Ca] [Cu,]lT, where T corresponds to another heterogeneous puck-
ered layer [CaCuJT and K. to the kagonik layer [CL@ (Figure 1). Some compounds, like CrB and MO tures based on puckered simple hexagonal packings of Cr or MO, with boron atoms in prismatic voids. The packing is puckered because of the relative displncement of certain parts. These packings are denoted by x Ts, where w means puckered by d i s p l a c e ~ ~ e ~ t . Some crystal structures like stibnite, mosaic of linked close-packed blocks (Wellner, 19SS>, They are called r e c o i ~ b i ~ a ~structures ~on (Lima-de-
Structural ClusslficatioPz and Notation
Faria et al., 1990) and the corresponding structural foriiiula is [Sb2S3J#3 where # means linked close-packed blocks, or a recombination packing (previously represented by c t h, Lima-de-Faria, 1994). However, this notation does not describe the way the blocks are linked together and needs some improvement, possibly based 0x1 the work on sulpbosalts by Makovicky (1985, 1997a,b). According to Parthi: and Chabot (1984) and Partlit: (1990), some strul;tures may be decomposed into slabs of other known structures. Examples are CrzAl B2 and Ce Co4B. However, these prticular structures can also be described as a stacking of layers. In fact these structures may correspond to the structural formulas: AICb B! [Cr,lRl1 and [[Ce] [CO,] JJ3]]T1c respectively (Figure 3), where R11 is a layer with rows o f tnangles and squares in the proportion 1 to 1, parallel to the sheet, and k and K are the layers already described for CaCu, running perpendicular to the plane of the sheet. The frameworks can be imagined as decomposed into special layers, the so-called connected units, because they are not individual layers but are linked to the rest of the structure. An example is diamond, C, where the connected units are puckered hexagonal layers fi of carbon atoms, and the corresponding structural formula is 3i;c[CtjHc. When the layer description.is not applicable, because the packing is three-dimensional? as in garnet,
7
C a p Ali SiTs[012]*7the thre~-dimen§ioiialpacking is represented by an asterisk *, In fact the garnet structure has a high packing coefficient of approximately 60%, which is com~arableto that of forsteriie, Al; Mgt[OJh, with 70°/0, and Eorstei-ite is certainly based on a hexagonal closest packing of the oxygens, Other kiiids of structural units also tend to pack together as closely as possible. When the ~ t ~ u c t u r ~ 1 units are groups, they tend to behave as organic molecules. The molecu~esstack in such a way that the . bumps in one are inserted in the hollows of a n ~ t h e rA coiiiplete study of the packing of* finite iiiolecwIcs, infinite molecular chains, and i ~ ~ nmolecular ~ t e sheets in organic structures has been made by (1955). This study may be applied to the categories of inorganic structural units. For the coordination of the atoms two levels of symbols have been proposed: complete and simplified. Each complete synibol gives the total number of atoms coordinated to a certain atom and the type of coordination ~oly~iedron, indicated by lower-case letters. The symbols for the most common coordirration polyhedra are presented in Table 3. For the sake qf s ~ ~ p ~ ~ c athe t i ~xtructural Pz . f ~ ~ ~ usually includes only the already inentioazed ~~~~~r~ st~Ucdzrrcll~~aCtors. However, it may be useful in mrtain cases to add more ~ t r u c t u r ~information, l either to describe more coinpletely tlie structure, or to compare
CeCo,
3 ~ e c o ~ p o s i t i oofnsome structures into slabs of other known structures aiid the co~respondings t r u c ~ ~ rformulas: al (a) AI B,,,(ParlhC, 1990) -+AICbBg[Cr,]R"s (layers parallel to the plane o f the sheet); (b) CeCo,B ( ~ ~ and a Ghabot, r ~ ~ ~ -+[[CeB,]I [Co,lK[CeCo,]' [Co3IK](layers perpendicular to the plane of the sheet), corresponds to Ce2Go,B,, or [[Ce] [CoJ [ the several square brackets meaning heterogeneous packing of Ce, CO and €3 atoms ( ~ e p r o ~ uwith c e ~permission) UP(:
Symbols for common coordination polyhedra (adapted from Lima-de-Faria et al., 1990, reproduced by permission of TUCr) Coordinatioti polyhedron around atom A Single neighbor Two atoms colliiiear with atom A Two atoms non-collinear with atom A Triangle coplanar with atom A Triangle non-coplanar with atom A Triangular pyramid with atom A in tbe ceiiter of the base T~t~d~iedron quare coplanar with atom A Squarc lion-coplanar with atom A Pentagon coplanar with atom A Tetragonal pyramid with atom A in the center of the base Trigonal b i p y r ~ ~ i d Octahedron Trigoiial prism Trigonal antiprism ~ e n t a ~ o nb~pyramid al Monocapped trigonal prism Bicapped trigonal prism Tetragonal prism Tetragonal antipri~ni Cube Anticube ~odecahedronwith triangular faces He~agonalbipyram~d Tricapped trigonal prism Cuboctahedron Anticuboct~~edron (twinned cuboctahedron) Icosahedron Trunca ted tetrahedron Hexagonal prism Fran~-Kasper polyh~drawith: 14 vertices 1.5 vertices 16 vertices
Complete symbol
~ i m p l i ~ esymbols d
tr 2
s*
*
0
P aP
ch acb do
CO
aco E
“Also [4s], ** or sq.
complex structures. For instance, if we consider do[I2Ihwe do not know hew the Cd atoms are distributed in the octahedral voids, and the structural formula may be confused with that of rutile Tio[827h. In rutile the Ti atoms are distributed in rows and in CdIz they alternately fill these layers. To complete this structural infor~ationone could write Cdo@[I2lh, the /a meaning alternately filled layers, and Tio’r[O,]h, the /r meaning rows of filled voids. It has also been recognized that when a structure is not close-pack~d,but belongs to another structural category, it can still be sliced in layers, and, in most cases, the stacking of the layers corresponds to one of the generalized stackings c or h, that is ABC or ABA, as in the closest packings. Even with more complex structures such as the barates where the atoms form simple icosahedra BI2 or truncated
icosahedra B,,, the stacking in c or h applies (Pearson, 1972). For instance, in the case of U the structural formula is U (g>[B,J coilsidering the €Ilz icosahedra as though they were spherical atoms in cubic closest packing (~ullenger and Kennard, 1966) (Figure 4). The fulleranes, e.g. Na1721n,,,Z, where 2 = Ni,Pd,Pt, although much more complex, also present similar truncated icosahedra of In7* (Sevov and Corbett, 1996) (Figure 4), which also behave as large spheres packed in hexa packing. The Subcommittee on Nomenclature of the International Union of ~ryst~llographyhas s t ~ ~ e e o ~ ~ the e ~used oef ~ ~ t ~ ~ ~c tou rr ~ l~in ~order l ~ s , to facilitate the progress of crystal ~ ~ e ~ ~Every ~ s t r ~ worker in this field is advised to adhere to such recommendations and to make a point of using
~ t r ~ c t u r Classjficntion nl and Notation
9
Figure 4 (a) Cubic closest packing (ABC) o f B,, groups (icosahedra) and (b) truncated icosahedron of boron atoms ( S u l l e ~ ~ e r and Kennard, 1966): (c) truncated xcosahedron of In7, in fulleranes (Sevov and Corbett, 1996, reproduced with p e r ~ i s s ~ oofn Acadeiiiic Press)
structural formulas whenever they can be substituted for conventional chemical formulas alone.
e Axes of the Unit Cell There is a tendency to represent crystal structures on the basis of projections along the crystallographic axes. However, in many cases, these axes have no structural
interest and the projections along them may render structural understanding difficult. Many structures which were not considered as closepacked have been found really to be close~~acked structures. For instance, hodgkinsonite, (OH),, in projection along the crystallog (Figure 5) does not disclose that it is based on a slig~tly distorted hexagonal closest packing of the oxygens and hydroxyls with Mn in octahedra1 voids and Zn and Si in tetrahedral holes. The corre§~on~ing structural
Packing r e ~ r e s e ~ t a tand i o ~projection along the crysta~~og~aphic axes of the structure of hodgkrnsonite (after Wyckof-, 1968)
formula is by Rentzeperis, in 1963, and further described in terms of a condensed model by Figueiredo (1976); the structure being built by equal layers (see Lirna-deoore (1 992) considers that the structure representation is very important, and has claimed that he has discovered many structures based on close packings that previously had not been described as such (Moore, 1995). Another curious example is tolbachite, CuCI,. yckoff was a crystallographer very interested in and acquainted with close packings, but did not notice that tolbachite i s a close-packed mineral (Wyckoff, 1963). The same happened with Burns and Hawtharne (1993). This structure is in fact based on a slightly distorted, cubi~-closest-packingof' the C1 atoms, with Cxi atoms occupying o ~ ~ a h e d rvoids a ~ in alternate layers, and should be considered as a distortion derivative of cadinium chloride, Cdo[CIJc. These s t r ~ ~ t u rcorrespond es to the same atomic arrange~ent, but have been considered different because they were
regarded under different projections (Figure 6). A sphere-packing model of cadniiwxn chloride, and the use of the ICSD (Inorganic Crystal Structures Database) and the 'Diamond' com~uterprograms, organized by Bergerhoff (1995, 1997), have helped to confirm this relationship. This is a typical example where the distortion by lowering the syi~metry,and the corresponding projections alon the crystallograpliic axes, niasks the real structural nature of the compound.
~ o r ~ a l lthere y are several inorganic compounds reIated to a given one, and they are considered as pertaining to the same hmily, e.g. the perovskite family (Lima-de-Faria, 1994, p.74), or to an homologous series such as the plagisclases. There are various processes or mechanisms which enable one to relate one struct~re to another, either by sli distortion, by replacement of one kind of atom by
me 6 Structural reprcsentation of the tolbachite structure, Cu"[Cl,]", C2/m: (a) after Wyckoff, 1963; (b) after Burns and wtharne, 1993 (Reproduced with permission of the Mineralogical Society of America). (c) Packing representation of the structure of cadmium chloride, Cd" [Cl$, R3m, after Wyckoff, 1963. These three representations correspond to the same structural arrangement; tolbachite i s only a distortion derivative of Cd"[Cl,]C
S~ructuraL~ L a ~ ~ s i ~ cand a t iNotation o~
two or more kinds of atoms, by insertion of atoms within the voids of the packing of the structural units, or even by subtraction of certain atoms. The more important degrees of similarity among crystal structures have been defined by the IUCr Conimission on Nomenclature in a paper by Lima-dent Faria et al. (1990). Structures with i n v a r ~ ~e~uivalent positions, the same space group and the same occupied equivalent positions normally lead to the same structure type; but, if the e~uivalentpositions are not invariant, the variables inay drastically change the coordination and give rise to a completely different atomic arrangement. Three main de~reesof similarity among structures have been considered in the definitions that follow in rank order: ~ . ~ # p o i ~have t ~ Z : the same symmetry (space group), and the same occupied equivalent positions - i s o c o ~ ~ g ~ r ~ t i oare n a zisopointal and the atoms have the same coordinations - isotypic: are isoconfigurational and have the same structural units packed in the same way. -
For more detailed considerations on these three definitions see Lima-de-Fa~iaet al. (1 990). There are many inorganic compounds with the same structure, and we are mainly interested in dealing with the different atomic ar~angements, therefore we assemble the isotypic compounds (with the same ~ e The structure) under the concept of ~ t r u c t ~type. number of structure types is obviously much less than that of the inorganic compounds, which will greatly facilitate our study. For instance there are approximately 200 compounds pertaining to the halite structure type, NaOEGl]". Halite itself, or NaC1, is called the p r o t o t y ~ eor a r c ~ e ~ y(or p ~even ari~totypein the sense of Megaw, 19731, that is the structure chosen to represent the structure type, and the 200 compounds are its popu~a~ion. When comparing structures differing only by the values of their variable parameters, that is, isopointal
11
structures, in order to assemble them in structure types, we have to consider limits of this variability that do not change the coordination of the atoms. According to Bergerhoff et al. (19991, to each pair of isopointal structures there corresponds a value of the difference of their coordinations, expressed by a parameter called A. The value of h will det er~i neif the structures should be considered i~ocon~gurational and will measure their similarity. A nomenclature for the structure types was first proposed by Ewald and Hermann in the ~ t r u k t ~ r ~ e richt in 1931, however, this notation has not been much used due to the fact that it was not self-explanatory~A l , A2.. .B1, B2, B3. .. are arbitrary letter codes, the meaning of which has to be memorized. In 1965 Hellner proposed another symbology based on the lattice complex concept, which is useful for very simple compounds, but this notation rapidly becomes very complicated for more complex compounds. Many chemists have already used the chemical formula of the prototype to symbolize the structur~type, but this is still not self-explanatory, and it is impossible to imagine the corresponding atomic arrange~entunless we are very familiar with the particular structure type. In 1965 Lima-de-Faria suggested the use of general structural formulas for the symbols of the structure types, and the concept was further extended when in 1976 Lima-de-Faria and Figueiredo (1976) proposed structural formulas of the prototype structures (Table 4).
There are some relationships which are of special interest, and among them the s ~ r u c ~ u~~ ae l~ ~ v ~are~ t i ~ ~ e of particular importance. We shall consider first the derivatives in the sense of uerger (1947), which include two categories, distortion and substitution
Various symbol~smsproposed for the structure types
cu M& NaCl ZnS (sphalerite) Al,Mg04 (spinel) Mg,SiO, (olivine)
Ewald and ~ e r m a SB ~ n (1931)
Hellner (1 965)
A1 A3
F or (c) H E or (h) H F+F' F + F" F;;,+D, T (h)nC,,+00, 1/2, 12xy;AzI11/4,1/4, E;
1 B3
HI 1 1332
Lima-de-Faria (1965)
Li~ia-de-E;ariaand Figueiredo (1976)
xc
[cur
Xh AoXc A")(" A'jB'XC, A'jB'X;
P%Ih
NaO[C11" Znt[SIc AliMgL[04]' Mg$3it[04]"
12
Structure and Composition
derivatives. A distortion derivative of a certain comThe domain of synthetic inorganic crystal structures pound corresponds to another compound that can be has become more and more complex, leading to atomic derived from it by a slight distortion. A s u ~ ~ ~ ~ t u f i arrangements on difficult to describe in simple terms. The structural not at i o~concept i s still at the beginning deriv~tiveis obtained by replacing one kind of atom by two or more atoms. However, there is another stage, and ~ ~ p r o v e m ewill ~ t sbe necessary in order to category that Buerger did not consider, the defect make it possible to use such notation when dealing derivatives, which are derived by subtraction of some with more complex structures. atoms, generating vacancies. These vacancies are important because they are fundamental to defining the packing. An example of a defect derivative of perovskite, Tio[CaQJC, is dzhalindite, Ino[O(OH),IC, which belongs to the R e 0 [ ~ 0 , l cstructure type. When considering two structures one of which is a distortion derivative of the other, the distortion With the structural classification and the correspondderivative is the one with less symmetry, and the ing notation we can order the inorganic structures. other is called the basic structure. To decide which is However, we realize that many structures have the the s t ~ c t u r ewith less symmetry we need to know how same atomic arran~$ment,that is, belong to the same to i~z~asurethe s y ~ ~ e t rofy a crystal structure. structure type, and instead of ordering the inorganic Attempts to measure the symmetry of a structure structures we ordered the inorganic structure types, were once based on group-subgroup relations, but this which greatly simplified our work. gave rise to contradictory results. A definition of the Work on this problem, first approached by Lima-demeasure of the symmetry of a crystal structure was Faiia { 1965) and Lima-de-Faria and Figueiredo presented by Lima”de~Faria(1988); it corresponds to (1969), led to a table reproduced as Table 6, containing the ~ ~ ~ t ~ l iofc ithe t y general p o s ~ t i ~ofn the space 782 inorganic structure types, corresponding to group. This definition is the natural extension to space approximately 5200 crystal structures (Lima-de-Faria groups of the one adopted for point groups. In fact the and Figueiredo, 1976, 1978). This table was presented multiplicity of the general form in point groups in improved form in a book by Lima-de-Faria (1994). corresponds to the multiplicity of the general position In the horizontal direction, five structural categories in space groups. For further details on this definition are shown in their order of increa§in~complexity; see ~ima-de~Faria (199 1 and 1994). vertically, various structural formulas are listed, from simple atomic ratios of binary compounds to the more complex formulations of ternary and higher order compounds. One can notice that most of the intermetallic compounds are located within the close-packed category of structures, on the left-hand side of this table. As Belov (1947) said: ‘In spite of the variety of The structural notation we recommend has been the mineral crystalline world, the whole “mineralogiapplied to approximately 70% of all the structure cal game” just reduces to various modes of filling prototypes described in the table ‘Crystal Structure gaps in uniform close packings with the various ~omenclature’which is presented at the beginning of corresponding patterns’. The same happens with the n ~~ ~i , c and the book, ~ n ~ e ~ C~ o ~~ p~ oa ~ ~Principles intermetallic compounds. We realize that most of the ~ r ~ c ~vol. ~ c 1,e edited , by Westbrook and Fleischer, classified prototyp~sincluded in Table 5 are based on 1995 (Table S). Out of the 220 listed structure close packings, approximately 80%. However, among prototypes 155 have been classified: 124 are closethe chemical elements, there are some structures packed structures,‘ and 31 are of the other structural which have higher dimensionality, e.g. groups, chains, categories. sheets, and frameworks: sulphur (g)[S,], selenium 1 There i s a difference between close-packed and closest~ ] ]tin ~ ~ 1300) ~, (1 CO] [Set2]], arsenic { 2 o o ) [ A ~ [ ~ ~and packed structures, The close-packed structur~sinclude all the [SntINHC. Also for AmB, com~ounds,including interstructures which are not group, chain, sheet, or framework metallics, there are among the non-close-packed structures. Therefore they include the closest packed, the category such examples as CoAs,-Co: (g}[As& body-centered cubic-packed, simple hexagonal and simple (skutterudite), ~ ~ S b ~ ~ o o { g (~drcasite), ~ [ S b 2 ~ ~ ~ cubic packed, etc.
13
Structural Classification and ~ o t ~ t i o ~ Table 5 A ~ p l i ~ a t of ~ othe n s t ~ c t u r notation a~ to the intemetallic protot StPukturbeberieht designation A, A, A, Ad
4
A, A,
4
A, A, A1 A2 A3 A3' A4 A5 A6 A7 A8 A9 A 10 A1 1 A12 AI 3 A14 A15 A16 A17 A20 Be Bb
Be
4
4 Bm
Structure type
Prototype
Pearson symbol 812 tP30 OP8
t P4 hP1 t P50 CP1 hR1 mP64 mP32 c cdz
hP2 hP4 C B
fI4 f I2 hR2 hP3 hP4 hR 1 OC8
s presented in the table 'Crystal structures nomenclature'
Space group
Strukturbericht designation B1 B2 B3 B4 B81 B8,
B9 B10 311 813 B16 B17 B18 319
B20 B26 327 B29 B3 I 332 B33 I334
c158
m5
CP20
B37
OCIS
cP8
OR28 oC8 OC4
cI16 hP9 OC8 OPS oP16 oC8 tB 6 kP2 hP8 hP4 mP32 oP8
'h
cc c e
C, Ch
Ck C1 cl h c2 C3 424 C6 C7 C8 c9 c10
Structure type
Prototype
estbrook and Fleischer, 1995) Pearson symbol
Space group
c n CSCl cP2 ZnS (s~hale~ite) C B ZnS ~ ~ r t z i t ~ ) hP4 NiAs hP4 Ni,h hP6 HgS (cinna~ar) hP6 Pb0 tP4 yCuTi tP4 NiS ~ ~ 1 1 ~ r ~ t hR6 e ~ CeS aP8 tS (cooperite) CuS (rovelite) AuCd oP4 FeSi CP8 GuO (tenorite) mC8 FeB oP8 oP8 oP8 cF16 CrB oC8 PdS tB16 CoSn hP6 TlSe tIl6 hP18 oF48 t112 oC24 PdSn, ThC, mc12 Cu,Te hP6 LiZn2 hP3 cH2 cF12 CP12 cP6 ~g,O Ti02 (mtile) tP6 hP3 h S Si cF24 SiO, (8 t ~ ~ ~ ~ hP12 ~ t e )
contimed
14
Stmkturbericht d e s i ~ a t ~ o n $tr~cturetype Cl 1, C l 1, CI 2 CL4 cl5 C15, C16 Cl 8 Cl 9 c2 1 c22 c23 C28 c32 c33 c34 c35 C36 c37 C38 C40 c42 c43 e44 C46 c49 6-54 DOa DO, DO: D0d
DO, DO2
Do3 f309 DO1 1 DO17
Do18 Do19 0020 DO21 DO22
rototype
arson mbol if6 lr6
Space group
Strukturbericht ~esignation
Stru~turetype
rototype
Pearson symbol tri fi hPli 4
trio
[CU, Sbf" [Cr Siz]" Si'[S,fc ZlF[QJQS
3 00 [GetSd Auo[TeJh
pi SiJTd [Cu, Tilh [Si UJ [Ir,Sir
hP12 cn4 cn4 $112 oP6 hR3 oP24 bzP9 oP12 oP12 hP3 hR5 AuTe, ~ c a ~ a v e ~ t ~nzc4 } CaCl, oP6 ~ g(Laves) ~ j hP24 ~ Co,Si oP12 Cu2Sb rP6 CrSi, hpst SiS, of12 mP12 ZrQ, OF72 GeS, AuTe, ~ k r e n ~ e r i ~ ~oP24 ) ZrSi, OC12 TiSi, on4 PCu,Ti oP8 SiU, tfl6 Ir,Si oP16 1132 er32 cFI 6 cP4 oP16 oP16 BaS, faP8 Ma,As hP8 oP16 ~ ~ , S n hP24 tI8 AI,Ti r116
OI20 OC20 LPlO
tP20 oF40 hRl5 t1lO ti26 t128 hP6 cP36 cF52 tIl8 oC28 cP7 CF112 tPl0 hPl0 cI40 hR5 mP20 hRlO hP5 CI80 C
0
LP40 OP20
O r i4
hR7 cF56 cf28 cFll4 tP3O eP39 mP22
Space group
~ t ~ ~ cClassification t u ~ ~ l and Notation
Strukturbericht designation
Structu~etype
35
~rototype Mg32tA1,Zn)49 Ce,Ir, 5
Th731, @r,%
%G
Fe,W6 (p-phase) Cu,,Si, Mn,Si, CO&* Cr,AI, Co,G1 Cr,C, Fe,Th,
cooco',[s,i"
,
PbFCl FeAsS MgCuAl, AgAuTe, (sylvanite) CuFeS, (chalcopyrite) CaTiO, (perovskite) A1,CdS4
Pbl911[F]IC1](Q2'.Qlff'f Feo{g)[AsSch)y
AgoAu"[TeJ" Cut Fef[SJ' Tio[CaOJ Alt,Cdf[SJ
Pearson Space s y ~ ~ o l group ell62 cI4140 on8 hP14 hR7 hP20 tI32 t132 cI52 c152 cP52 cF116 hR13 cf76 hP16 cF68 hR26 hP28 oP40 hP20 hP28 tP6 mP24 oC16 mP12 t1l6 cP5 tIl4
Strukturericht esignation
S t ~ c t u r etype
ro totype Al,Cu2Fe
FeKS, NaCrS, Cu,VS, (sulvanite) Cu2FeSnS4 Fe,N AIFe,C (perovskite) ThH2 Fe,N CuPt, AuCu CuPt AuCu, GCuTi A ~ C u 2 ~~ n ~ e Sb2T17
CuTi,
Pearson symbol tP40 hP18 h~~ c196 on4 cF112 hP1S cP12 mC16 hR4 oP16 cF56 cP8 tI16 cP5 cP5 $16 hP3 cF32 t P4 AR32 cP4 tP2 u cF16 ~ ~ d54 tP4
Space group
~
r
~
Structure and ~ o ~ p o ~ i t i o n
16
Note: symbols A, €3, and R represent interstitial atoms; X, Y, and Z represent packing atoms
f
N:'
[o&
01'
[Cu Mg Sn Cl]'
Structural Classification and Notation
17
Table 6 (continued)
P HOMOOENEOUS
4ETEROOLNEOUS
HOMOGENEOUS
Sirnplr and fiomwaiih
S,rnp\o
n
**.I
I
I
K1
*I'
,,*I
...
18
Structure and Composition
Table 6 (continued)
I
I
I
I
I
1
Struct~ralCluss$cntion and Notation ~ b ~ ~ ~ - 1~00b)[Pb"O,] ~ 3 n 1 ~ (minium) and As,S,--.f200)[As~~]S~] (orpiment). We therefore conclude that there is a great tendency for inor~anicc o m ~ o u ~to~.form s close-~ackedstructures. Only when special direc~ionalbonds (normally covalent) exist, does the structure tend to form a less dense, open str~cture,giving rise to large voids within of the structural units. also notice that the kayer descr@ton is u ~ p l i c ~ b lto e most i ~ o r g u ~ i~c ~ t r ~ c enabling ~~res a simple description of the structural atomic arrangement. Tlirs is a consequelice of the fact that ia general nd in m a ~ ycases less than I5 A (Lima-de~Faria,19 Another important CO s that the generalized s t ~ c ~siyn~~b o cl and ~ h are ~ r e ~ o m i n a nand t , are even applicable to complex structures such as borates and fulleranes. Furthermore, the layer descr@tion obeys certain ~ s i m ~rules ~ l e (Lima-de-Faria, 1978). The other ~mportantfeature to notice is a kind of p o l y ~ e r i ~ ~ process t ~ o n qf the ~structur~l units, condensing atoms into groups, those into chains, chains into sheets, and eventually into frameworks. This process is very clear in the silicates through the polymeri~atio~ of chains into sheets of tetrahedra (Belov, 1956; Liebau, 1956) and later generalized by Lima-de-Faria and ueiredo (1976). It also applies to octahedra (Moore, 1974) and to other polyhedra. This process of poly~erizationis inherent to the basis of the structural classi~cationwe propose. Many structures are derivatives, either by distortion or substitution of certain basic structures. This means that the v ~ r io~~ ta ~t o ~aircr u n ~ ~ ~ eisnnot t s so large as might appear. These relationships will reduce the number of structure types to be inemorized, and many others can be simply derived from them. According to van Spronsen (1969) 'the periodic system of the chemical elements has passed through three d i ~ ~ r e nstages: t that of initiation, that of ~he~omenological development and that of theoretical development'. The table of inorganic structure types will possibly have to go through similar stages. We are now on the first stage, that of initiation, and much work has still to be done.
I am very grateful to my wife, Natasha, and to Ana Luisa Cunha for revision of the text and also for helping me with the use of the computer.
19
My friend Arnaldo Silvkrio, as usual, made himself available to peruse the ~ a n ~ ~ s c rfor ~ pwhich t , 1 tkank him most heartily.
Belov, N. V. (1947). Structure of Ionic Cr,vstuls a ~ t d~ e ~ a l j i ~ Phases. Izd. Akad. Nauk SSSR. Moscow. Belov, N. V. (1956). Essays of structural niincralo ~ ~ 10, ~ 10-3 l Russian). ~ ~ nSb.e (Lvov), Bergerhog, G., Berndt, M., Br~~ndenbur~, (1999). Concerning inorganic crystal structure types. Acta
Bokii, G. B. (1954). Ilatroduction to Crystal Chemistry. Russian original: I%date17stvo Mo~covs~o English translation: United States Joint Service, New York, 1960. Buerger, J. M. (1947). Derivative crystal stru~ures.J. C ~ ~ Physics, 15, 1-14. Burns, P C., and ~awthorne,I?. C. (1993). Tolbachite, CuCI,, the first exaniple of Cu2+octahe~raIlycoordirzated by Cl-. Amer. ~ i ~ i ~78,r 187-189 . , De Jong, W. F. (1959). General ~rystallography.A San Francisco. C o ~ p e n d i Freeman, ~~. Ewald, P. P., and Hennann, C. editors (1931). S t ~ ~ ~ t u r ~ e r i c j z t (for 1913-1928), Akade~ische ~erlagsge$ellschaft M.B.H., Leipzig. Figueiredo, M. 0. (1976). Private communication on condensed models. Garrido, J., and Orland, J. (1946). Los r ~ ~ la ~o s ~~ ~- ~ ~ c ~ fina de 10s cristales. Dossat, Madrid. Hawthorne, I;. C. (1983). Graphical enumeratio~ of polyhedral clusters. Acta Cryst., Hellner, E. (1958). A s~ructuralscheme for sulp~ideminerals. f. Geology, 66, 503-525. Hellner, E. (1965). Descriptive symbols for crystal structure types and h o ~ e ~ t y p based es on lattice complexes. Acta Cryst., 19, 703-712. Jensen, W. B. (1989). Crystal coordination formulas. In The Structup~~s of Binary C o ~ p o u n d s(eds F. D. G. Fettifor). Elsevier Science Publishers. Kitaigorodskii, A. 1. (1955). Organic Chemical C r y s ~ a l l o ~ r a p(in ~ y Russian). Press of the Academy of Sciences of the ISRSS, Moscow. English translation and revision (1 961). Consultants Bureau Enterprises, New York. Laves, F. (1930). XVI. Die B a u - ~ u s a ~ e n h ~innerhalb ~ige der Kristallstrukturen. 1 Teil. Zeit. Krui., Liebau, F. (1956). Bemerkungen der Systeniatik der Kristallstruk turen vo rnit hochkondensiert~n Anionen. Chim. Phys 6, 73-92. Lima-de-Faria, J. (1965) derivation of inorganic close-packed structures. AX and AX, compounds, sequence of equal layers. 22%. Krisf., 1
20
Structure and Composition
Moore, P. B. (1992). Betpakdalite unmasked, and a comment on bond valences. Aust. J . Chem., Moore, P. B. (1995). Closest-packed Franklin-Ogdensburg: Kepler's gift o f the snowflake (private comm~~ication}. Niggli, P. (1945). Grundlagen der Stereochemie. Verlag Birkhauser, Basel. French translation Leks Bases de 10: StdrPochinzie, Dunod, Paris (1952). Parthk, E. (1 990). ~ l ~ ~of e~ n~ o~r gs~s~ruc~ural nzc ~~e~zs~ry. Linia-de-Farm, a. (1994). S ~ r ~ c ~~~~ rn ae ~r a l oAn ~~. First edition (page VIII-70) Pety-Lnncy (Geneva): Introducrion. Kluwer Academic Publishers, Dordrecht. Katharina Sutter Parthi Editor. Lima-de-Faria, J., and Figueiredo, M. 0. (1969). A table Parthi, E., and Chabot, B. (1984). Crystal structures and IC close-packed structure types. crystal chemistry of ternary rare earth, transition metal Figueiredo, M, 0. (1976). bodies, silicides and homologues, In Handbook on the, ~lassi~cation, notation, and ordering in a table of ~ h y s ~ and ~ . sChemistry of Rare Earths, Vol. 6 (eds K. A. inorganic structure types. J . Solid State Chem., 16, 7-20. Cschneidner, Jr., and L. Eyring). Elsevier Science Lima-de-Faria, J., and Figueiredo, M. 0. (1978). General Publishers, Amsterdam, p. 188. chart of inorganic structural and building units. Pearson, W. B. (1972). The Crystal Chemistry and Physics of Garcia de Orta, SPrie Geologia, Metals and Alloys. Wiley Interscience, New York. Lima-de-Fana, J., Hellner. E., Li Rentaeperis, P. J. (1963). The crystal structure of and ParthC, E. (1990). Nomenclature of inorgaiiic hodgkinsonite Zn,Mn [(OH),SiO,]. Zeit. Krist., 119, structure types. Report of the International Union of 117-1 3s. ~ r y s t a l ~ o g r a p ~ y~. o m ~ i s s i o non Crystallog~aphic Sevov, S . C., and Corbett, .?. D. (1996). A new indium phase Nomenclature, Subcommittee on the Nomenclature of with three stuffed and condensed fullerane-like cages: Inorganic Structure Types. Actu Cryst., A46, 1-1 1 . Na,,,ln,,,Z, (Z= Ni, Pd, Pt). J . Solid State Chern., 123, ac~dtschki~ F. (1 947). Konstitutio~s~ormeln fur den festen Zusldnd, ~ ~ n a t sChenz., ~ h . 77, 333-342. 344-370. Makovicky, E. (1985). The building principles aiid Sullenger, D. B., and Kennard, C. H. L. (1966). Boron classification of sulphosalts based on the SnS archetype. crystals. S c i e n t ~ cA ~ e r ~ ~ u n , van Spronsen, J. W (1969). The Periodic System of Chemical akovicky, E. (1997a). Modular crystal chemistry of Elements. Elsevier, New York. sulpliosalts and other complex sulphides. EMU Notes in Westbrook, J. H., and Fleischer R. L. (eds) (1995). Mineralogy, 1, chapter Intermetallic C o m ~ o u ~~rin~iples ~s, and Practi~e,Vol. 1 . Makovicky, E. (1997b). Wiley, Chichester, UK. approaches. EMU Nor Wyckoff, R. W G. (1963). Crystal Strucfures, vol. 1. Megaw, H. D. (1973). Crystal Striactwr.es. A Working Second edition. John Wiley Interscience Publishers, New aunders Company, P h ~ l a d e l p ~ a . York. oore, P. B. (1 974). Structural hierarchy among miiierals Wyckoff, R. W. G. (1968). Crystal Structures, vol. 4. John containing octahedrally coordinating oxygens, 11. N . JD. Miner. Abh., 120, 205-227. Wiley Interscience Publishers, New York.
Lima-de-Farm, J. (1978). Rules governing the layer organi~~tion of iiior~aniccrystal structures. Zeit. k'rzsf., 1-5. c-Faria, J. (1988). The hierarchy of symmetry. Presented at the Xl European (Vienna). Abstract published in Lima-de-Faria, J. (1 99 1). On the p the symmetry of crystal structures. Garcia de Orta, SPrie
er
Mankind was probably acquainted with mercu even in prehistoric times. It is certain that Hg compounds were well known in the Mediterranean area in the fourth century before Christ, but probably Hg had been applied in China, India and Egypt even earlier (Barnes and Bailay, 1972; Weeks, 1968). Metallic Hg possesses unique physical properties (Guminski, 1992). It is a relatively noble metal with and boiling poiiits of all metals (at 0.101325 MPa) of - 38.8290 and 356.623 "C, respectively. Gaseous Hg consists almost entirely of moiioatomic species. The element loses its metallic character above the critical point at 1492°C and 151 MPa, Solid Hg exists in several allotropic forms labelled ay /3, y and 6. Gamma Wg appears only after inducing a tensile strain. The hexagonal bHg is slowly formed at pressures > 24 GPa at room temperature; thermodynamic equilibrium aHg % 6Hg is estin~ated at 30 t- 6 GPa. The rhombohe~rala H g is stable at normal and moderately elevated pressures at above - 194 "C. The tetragonal PElg is formed at lower temperatures by applying high pressures or by hammering at -268 "C. The aHg % PHg transition is of martensitic nature.
Hg is a rather rare element, seldom found in the native state. The most frequently inet minerals are: HgS (cinnabar), HgSe (tieinaniiite), HgTe (cologSe (onofrite) and solid amalgams of Ag (m~schellandsb~~~ite, ~araschachner~te, schach-
nerite, and weishanite) and Au (weishaiiite). Small amounts af Hg c o ~ p ~ u n are d s formed in volc~ni~al1y active regions. All these minerals are insoluble in
performed from larger amounts of the low-grade ores (typically 0.3 to 3.0% Hg). The usual way of obtaining Hg from HgS consists af roasting its concentrate in air. The conceiitrates can be aiso roasted with CaO or scrap Fe. ~ h e m i ~ a l pure ly recovered by distillation at reduced pressure. Further purification may be attained by pro1 tion while mixing liquid Hg with a solution of ~ g ~ ( ~ ~~b s e~ q u~e n t) ~ . Hg is recommended. Governments in some countries still treat Hg as a strategic mate~ial, production estimates may be erroneous. production of Hg is of the order of ~ 0 0 tons/year. 0 ~
Despite the significant toxicity and scarce o c c u r r e ~ ~ e , Hg and its alloys have always played 8 very important role in the developnieiit of science aiid technology. Although numerous endeavors of a l c ~ e ~ i sto t $transform Hg into Au were fruitless, nevertheless they 'Much rarer minerals include the ainalgams of Gu (kolymte), Pb ( a l t ~ a r k ~ t eand ) , Pe (poterite).
~ n t e ~ ~ e t aCornpounds: ll~c Vol. 3, Principles und Practice. Edited by J. H. Westbrook and R. L. Fleischer. 0 2 0 0 2 John Wiley & Sons, Ltd.
22
Structure and Composition
contributed a lot to the collection of important information for experimental physics, chemistry, and nzedicine. At the turn of the nineteenth~twentieth centuries, the alloys of Hg (also known as amalgams) started to be objects of intensive scientific investigation. Phenomena o f mutual solubility, activity coe~cientsof ~omponents,heats of dissolution and other thermodynamic parameters, viscosity, diffusion and chcmicdl reactions occurring in liquid Hg were widely invest~~ated. Generally, amalgams are good model systems for liquid alloys, and their features may be easily and exactly ineasured at room and near-room ue to its noble character, the clean and removable surface when liquid, the high overpotential for hydrogen ion djscharge, and the possibility to dissolve many metals, Hg was a real driving force in electro~hemsitry as an econoniic electrode material. Electrocheinists were able at last, with liquid mercury electrodes, to verify theoretical equations which had not been c o n f i ~ m eexperimentally ~ by means of solid electrodes. ~lectroanalysisstill continues to profit from olarography by Heyrovsky in the 1920s (and many related techniques later on). Stripping analysis with the use of hanging drop and thin Fbm Hg elcctrodes is an important tool in the trace determination of metals at 10-6--10-9 mol/did (Galus, 1994). The 100-year-old Weston cell: sat./Hg-Cd (12.5 mass% Cd) which contains Cd amalgam on the right-hand side is the best with which to calibrate potentiometric sjistems because of its stability and a very low temperature coefficient (see Westbrook, 1995).
itio ~ m a l ~ ~are m ~s e ~ as ~ alloys e d where one of the compone~tsconsists of Hg. Simple amalgams contain combiiied with Hg, and complex in more than one metal alloyed with . The name amalgarn is used for either mono-phase (homogeneous) or multi-phase (heterogeneous) syss exist a solid, solidlliquid or liquid2 forms. stable molecules have ever been detected in the gas phase in measurable concentrations, therefore gaseous amalgams ought rather to be treated as a mixture of atoms or as dimers (M2). Melting one or more metals with Hg may lead to formation of either 2See p. 26ff. for discussion of short-range order in liquid mercury alloys.
I
10 I
20 I .
30 I
40 I
50 i
80 I
70 I
80' I
A t o m i c number ~ i ~ 1 ~ S~lLibility r e of elements in mercury at 25 "6.Circular and triaiigular symbols represent data from exact deterinitiations and minimum detection levels respectively; open symbols the formation of an IMC; filled symbols nonformation of an IMC
solutions or defined intermetallic compounds (lMCs) as well as their mixtures. The degree of solubility of NI in liquid Hg and the tendency of M to form a stable IMC with Hg are the two major factors that should be taken into account in amalgam classification. The selected solubilities of elements in Hg at 25 OC are collected in Figure 1, and nurncrical values for metals are to be found in Guminski and Galus (1986) or Guminski (1989a). The solubility values denoted by circles and triangles come from exact d e t e r ~ i n a t i o ~ and s rninimum detection levels, respectively. Eventual formation of an IMC or its non-formation in a denoted by empty and full symbols, respectively. The solubility changes with the atomic number of an element reflect changes o f the physical properties of elements throughout the Periodic Table. Taking into account properties of simple amalgams in the sense of: the type of phase diagram for a Hg-M system (Massalski et al., 1990), the solubility of M in Hg, the heat of the dissolution process, the activity and diffusion coefficients. and the kinetics of electroreduction of M"+ on a Hg electrode with the corresponding a ~ ~ a l ~faom~ a t i o(Gumins~i, ~ 1989a), one may formulate a classification of simple amalgams into four groups (Guminski, 1989b). This is not the first attempt at an arrangement, since various generalizations have been done several times before, but this classification has the most general character, is based on the most contemporary knowledge of amalgam
Amalgams features, and is of essential service for practical applications of amalgaiiis,
The first group contains: alkali metals (also ammonium aiid substituted ammonium radicals), Mg and alkaline-earth metals, lanthanides (Ln's) and actinides (An's), All these metals dissolve at 25 "C in Hg to levels higher than 10-3~molYO(with the exception of a few heavy Lns). Due to the large difference in electronegativity of M and Hg, these metals form IMCs of partly ionic bonding. The most stable IMC formed in a g-M system melts congruently at a temperature higher than the corresponding melting point of pure M (with the exception of Th, U, and heavy Lns which possess quite high melting points). In the case of IMC melting points being higher than the boiling point of Hg at normal pressure, one should consider the M-Hg phase diagranis established at an applied pressure to keep the com~onents in the condensed state; see the example of the Ba-Hg phase diagram shown in Figure 2. The saturated liquid a m a l g a ~ are s always in eq~~libriuni with their IMCs in . The majority of the ~~~s are line compounds. ctrode processes at ~ ~ - ~ / interfaces M n are relatively fast; they are fastest for the alkali metals and slowest for Lns and Ans. The amalgam half cell H g - ~ / ~ npotentials + are well defined, and these amalgams may be effectively used in technological processes connected with electrochemical reactions. Use of aprotic solvents and a strictly 0-free atmosphere is recommended. Formal potentials of pure metals EfMiMn+ are always more negative than the n + . experimental a ~ a l g a m potentials E f ~ g - ~ , MThe solubility values are typically lower than those predicted by the theory of ideal solutions. Since the dissolution process is markedly exothermic, preparation of these amalgams by electrolysis or cementation is suggested instead. The liquid homogeneous amalgams deviate strongly froni Raoult's law, activity coefficients of these metals in their dilute solutions (referred to pure metals) are very small, and the liquid phase contains M exclusively in the form of MHg, molecules, but not o f free atoms. This situation is also reflected in the difference between diffusion coe~cients predicted f the simple theory of Sutherlan~-Einstein) and e x p e ~ ~ e n t a l l measured y values; the experimental values are always smaller than the theoretical ones. +
a
20
40
60
I
I
I
1
80
w
ure 2 Assessed Ba-Hg phase diagram at c o ~ ~ t ~ ~ i i ~ e d pressure (from J . Phusc &'@I.)
The second group c or d valence electrons (Al, Ga, In, Cu, Ag, Au, Zn, and Cd). All the soluble in liquid Hg to levels higher than 10 -3 mol% at room temperature. weakly. Even if an it displays a certain decomposes much below the meltin the example of the T1- g phase ~ i a ~ r asmh o ~ nin Figure 3. These amalgams are simply prepared by direct dissolution of the metals, ceme~itatio~, or cation electroreductio~.~nthalpiesof the are small, negative, or positive. slight deviations from Raoult's law (positive or negative) are observed, and the activity c o e ~ c i ~ n t s o f the oscillate around unity. The formal ~~otentials EfHg-M,Mn+ and EfMiMn+ half cells are placed closely together on the potential scale, and the electr~redL~ctions of these cations are the fastest elect~odeprocesses known. The predicted and ex coefficients are quite similar; only and T1 (which form relatively st show measured values lower than those from the predictions. The M-Hg interactions typically have metallic character.
.3
Ps
The third group contains high melting d metals and p elements of aiiiyhoteric character (V, Nb, $a, Cr
24
Structure and Composition
TL w e 3 Assessed Hg-TI phase diagram (from J. Phase Equil.j
g at room temperature are at concentrations below 10-3mol% and in some cases at immeasurably low levels. The experimental values are always lower than those predicted for the ideal solution. The Ce- g phase diagram as a typical representative is shown in Figure 4. Practically pure metals are the saturatin~solid phases. These amalgams may be obtained by electroreduction or chemical reactions, The products are frequently oversaturated by several orders of magnitude. However, quite large amounts (up to 20mol%) of Cr, Fe, or CO have been introduced into Hg by electrolysis, forming products with an initial consistence of milk, then cocoa with
aoo
I
L M.P.
0
Ge-Hg phase d i ~ g r (from a ~ J. Phase ~ ~ ~ i 1 . j
increasing concentration of M, then cream, butter, and finally a plasticine-like paste. Therefore formation of such colloidal dispersions must not be treated as evidencing good solubility of these metals. The dissolution of metals in this group is always an endothermic process. Aquo-cations of these elements are poorly reducible at a FEg electrode (slowly, with a large overpotential). Since these metals do not undergo a stable solvation by Hg, their diEusion coefficients should be similar to those predicted by the SutherlandEinstein equation. Activity coefficients of the metals in their amalgains are much higher than unity aiid the deviations from RaouWs law should be positive. Due to the very low solubility of the majority of these metals, we can experimentally verify only a limited number of features of the selected a~al gam s.The MHg interactions have a dominantly Van der Waals character.
The fourth group of amalgams contains the rest of the d metals and probably Be. Their solubilities are lower than 10-3mol% at room temperature; the experimental values are frequently higher than those predicted by ideal-solution theory. These amalgams can be prepared by direct contact of the elements at elevated temperatures. Often a poorly soluble film of the reaction product (M-Hg) inhibits the reaction progress at lower teinperatures. These amalgams are effectively obtained by the electroreduction, cementation, and chemical reactions. The saturated amalgams are in equilibrium with their, most frequently linear, IMCs which are stable up to 600°C at normal pressure. As an example the Ni-Hg phase diagram is shown in Figure 5. The thermal effect of the dissolution process is the algebraic sum of the significant endothermic surmounting of the cohesioii forces in pure M and the exothermic energy of the IMC (M-Hg) formation^ finally the heat of the dissolution in this group may in some cases be endothermic as well as exothermic. Activity coefficients of these metals in their amalgams are lower than unity. The cation electroreduction processes on Hg electrodes are rather slow and occur at moderate overpotentials. The diffusion coeflicients of known metals are much lower than those calculated from the ~utherla~d-Einstein theory. Negative deviatioiis from Raoult’s law are expected for these amalgams; however, they have not been confirmed experimentally due to the low solubility of these metals. Bonding between Hg and M has a metalliccovalent character.
25
0 M.P.
4400
0
30
10
50
70
90
1200
. 9000 s!
400
200 0
M.F?
0
20
40
60
80
100
i Figure 5 Assessed Hg-Ni phase diagram at 0.101 MPa (from Phase Diagrams of Binary Nickel Alloys)
From our definition, the systems of Hg with nonmetals should not be named amalgams. Nevertheless, such systems may be divided into two categories. Noble gases, 13, C, Si, N, P, and As do not interact spontaneously with Hg and, due to their very low solubility (with the probable exception of P), may be safely included in the third group of amalgams. 0, S, Se, Te, F, C1, Br, and I form stable ionic, rather linear, compounds with Hg which melt above the corresponding melting points of the elements; see the exaniple of the Hg-I phase diagram shown in Figure 6. Since solubilities of such compounds In liquid Hg at room temperature are quite limited, these systems should rather be included with the fourth group of the simple amalgam classification. One could also divide the amalgams according to the s, p, d, and f valence electrons of the metals, but such systematization would be much less informative for practical and theoretical goals. This is ftilly obvious when one would compare for example Be and Ba, Mn and Re, or Ge and Pb amalgams of metals belonging to the same groups of the periodic table.
~~~~r~ 6 Assessed Hg-E phase diagram at constrained pressure sufficient to keep the coiiiponents as liquids (from J Phase Eguil.)
The easiest way of amalgam p r ~ p a r a ~ i oisn direct solution of M in Hg, if the corresponding solubility i s sufficient and the surface is not chemically passivated. The kinetics of the process may be increased by increasing the temperature. 5.2 ~
~
~
~
~
~
c
~
e
Electrochemical reduction of metallic ions on a cathode from aqueous or non-aqueous solvents as as from molten salts allows the introduction of both soluble and insoluble metals into the Deposition of a M away from the very seldom observed; therefore this be quite universal. Some amalgams may be obtained by simultaneous reduction of their soluble salts in a solution. Hg2' may be reduced on noble metal (Ag, Au, Pd or Pt) electrodes with formation of their amalgams. A i ~ m o n i uand ~ alkylammonium radical amalgams are most effectively obtained by electroreduction of airunonium and a l k y l a ~ m o n i ~ions m on a Hg electrode.
5.3 ~ Depending on the kind of metal and available technical possibilities, ainalgams may be prepared by means of any of five groups of methods (Kozlovskii et al., 1971).
e
e
~
~
e
n
t
~
~
Cementation is almost as effective as the ele~trolytic method, but its application makes estimation of the amalgam concentration more difficult; there are problems as well with the comp~etenessof the reaction and
~
~
26
StrLictzive and Composition
the product purity. of preparation: La3
+
+ Na-Hg-,
ere are two examples of this type
La-Hg + Na
+
5.
hemical rea~tionsmay be used for preparation of some kinds of amalgams. Carbonyls of Fe, Ni, CO,MO decomposition at 300 "G In sequent amalgam formation. metallo-organic compounds may also lead to the amalgam formation as in the following example:
Also hydrides of several metals may react with Hg or its compounds:
Taltacs (2000) described an ancient mechanochemical reaction for recovery of mercury from native cinnabar, rubbed with vinegar in a brass mortar with a brass pestle. In the opinion of this author the brass metals are partly transform~d into sulphides and partly dissolved in Hg in the course of this process.
6, ~ t ~ ~ ofc Amalgams ~ u ~ e Amalgams have been investigated with the use of the following physico-chemical techniques: thermal analysis, vapor pressure, density, viscosity, surface tension, hardness, i~iagnetics~sceptibility,diffusion coefficient, calorimetry. enif, electroanalysis, conductance, metallography and x-ray, electron or neutron diflraction. These properties and methods allow for distingLiishing between homogeneous and heterogeneous systems, because formation of a new phase generally produces an abrupt change in the parameters measured. Every techni~ueinforms us to some extent about a particular phase structure. Structures of solid amalganis have mainly been investigated by diffraction techniques. Hg
and M may combine in various stoichiometries from MEZg,, (as in BaMg,,) to HgM,, (as in MgIn,,). The crystal structures and lattice parameters of IMCs have been collected by Villars (1997). It is rather difficult to ascribe certain crystallographic structures to selected stoichiometries of amalgam IMCs. Such general rules could only be observed for Ln-Hg and probably An-Hg IMCs (Iandelli and Palenzona, €979). Structures of liquid amalgams have been deduced from a variety of investigations (see also Singh and March, 1995). For example, in the Hg-Na system, analyzing the thermodynamic functions, Ace*, A H , and A F y , and the concentration Auctation function, Scc(0)= (S2AG/SC2)-' where 4 G is the Gibbs energy and C is concentration), the structure Pictor a J r ) from diffraction experiments, magnetic susceptibility, viscosity, vapor pressure, electrical resistivity and molar volume versus composition for the liquid alloys at 100400 OC, one may find unequivocal iriformation about the existence of a large amount of this liquid amalgam, both below and above the liquidus line (Tamaki et al., 1982; Kozin et al., 1977). The S,,(O), AGex,A H , A F X ,the electronic part of the magnetic susceptibility, the electrical resistivity and the molecular volume, all show a niininmm but viscosity a maximum, at about 33 mol% Na (corresponding to Hg,Na) in the concentration dependence of these functions. Similar conclusions may be drawn from measurements for the Hg-Tl, Hg-In, Hg-K, and Hg-Cs systems where minimal values of the structureinforming functions point to the typical compositions of the Hg-M complexes foiined in the corresponding liquid phases. Around room temperature and in dilute amalgam solutions, information about the liquid amalgam structures may also be obtained from analysis of diffusion coefficients (Guniinski, 1989a) and emf measurements (Korshunov, 1990). metric experii~ents with amalgam cells one may estimate the thickness of the Hg solvation sphere around a dissolved M atom. In case of the alkali metals, such values range between 1.2 (for Li) to 1.8 (for Cs); and, in the case of the alkaline-earth metals, they are between 1.9 (for Ca) and 2.1 (for Ba) atom layers of Hg around these atoms. Quite similar numbers may be deduced from analysis of the diflusion coefficients of these metals. A starting point for an analysis of the diffusion coefficients in a liquid amalgam medium at high dilution is the Sutherland-Einstein equation:
D, = kT/4nrmrl,,
A vnalgams
relating the diffusion coeEcient of the metallic solute
27 I
D M ,the temperature T, the radius of the diffusing entity TM and the viscosity of the dilute amalgam qHg (the last one being practically equal to the viscosity of pure Hg). A correlation between selected experimental values of D M and YM is shown in Figure If one knows D M , one may estimate the radius of the particle that diffuses in Hg. It was demonstrated that such metals as Zn, Cd, Hg (self-diffusion), Al, Ga, In, Ge, Sn, Pb, Sb, Bi, Cr, and Fe diffuse in Hg in the form of naked atoms because the calculated radius from equation (1) is equal to or very near to the effective radius of pure M. All the metals mentioned are known to interact very weakly with Wg, and most of them do not form IMCs or intermediate phases with Hg. Atoms of the alkali metals, akaline-earth metals, lanthanides, and part of the transition metals are solvated by Hg atoms to different degrees, and their experinieiital diffusion coefficieiits are always lower than would be predicted with the help of equation (1). In several cases (Hg,Li, Hg,Na, Hg,Ca, Hg,Tb, Hg, &fn, Hg, 25Ag), the formula estimated from the experimental value of the diffusion coeEcient is equal to the formula of the solid phase (IMC) being in equilibrium with the saturated liquid amalgam at room temperature. However, some metals show explicitly larger or lower solvation numbers than would be deduced from the corresponding binary Hg-M phase diagrams. Foley and Reid (1963) measured diffusion in concentrated T1 amalgams between 0 and 40mol% Tl. These authors observed a distinct decrease of the mean diffusion coefficient near 25 mol% T1 which may indicate that Hg,Tl complexes dominate in the liquid amalgam at room temperature. The coniposition of the equilibrium solid phase in the Hg-Tl phase diagram is Hg5TI2,(28.5 at%), and thus is quite similar. ‘Viscosity and thermodynainjc functions of T1 amalgams similarly show extremal values at compositions from 2528 m01% T1.
1.a
1.
2.
Figure Dependence of the reciprocal diffusion coefficients of metals in liquid mercury on the effective radius of metal at 25 “C
metals with the Hg solvent, two limiting cases will be considered for the solutes: (i) metals that do not interact insofar as could be experimentally detected; and (ii) cases where they interact to such extent that an IMC, M,MI is formed in the Hg medium. The first investigations on this field were performed by Tamniann and Jander (1922) who discovered fomation of Bi3Ce4, Cd-Gu, CeSn2, C L I Z ~AuZn, , and Mg,Sn in relatively dilute amalgams, but detected no IMCs in the Ag-Cd and Ag-Zn amalgam systems. The nature of such compounds has been discussed for years, but now we find an. overwhelming amount of proof that the compounds are very seldom soluble but are typically precipitated in Hg ( ~ u ~ i n s k 1986a). i, The correspon~ingequilibria may be described according to the following schemes:
+ yM’-Hg 4 M x M & - ~pre~ip~tation g xM-Hg + yM”-Hg % M,M&-Hgsoluble product xM-Hg + yM’-Hg It.i:~ ~ ~ soluble ~ - W g
xM-Hg
Although many quarternary systems coiitaining Hg have been i~vestigated, so far most o f them are pseudobinary systems of the type HgX-MX (where X and X’ are non-metals). The corresponding phase diagrams are collected by Tomashik and Grytsiv (1982). However, we are concerned in this paragraph only with systems where Hg has but two metals introduced into it. In addition to interactions of solute
product % precipitation
(2) (3)
(4)
The equilibrium (4) seems to be the most general case that should be true for all reacting systems. However, the results of many experiments show unequivocally that the conce~trationof ~ ~ M ~ soluble , ~ Hmolecules g
28
Structure and Composition
is undetectably low, and most systems in fact obey equilibrium (2). Accordingly, the equilibria (2) and (3) may be expressed as the solubility product: (5) or as the equilibrium constant:
K= [M*M~l/[M]x~M’]y
(6)
The equilibrium concept of formation of solid IMCs in ternary solid-liquid alloys, as expressed by equation (5), was originally proposed by Hume-Rothery (1936) and applied by Zebreva to amalgams (1958). Using various physico-chemical and e~ectroa~alytical techniques, about 7 ~ M - ~ ’ - Hsystems g in the Hg-rich corners of the phase diagrams have been quantitatively investigated (Gumi~ski and Galus, 1992). It was observed that the TMCs formed in Hg solution have simple formulae and their compositions correspond to the most stable compounds in the M-M’ binary systems, as for example: AuCd, AuCu, AuMg, AuMn, AuSn, Pb, BiK,, CdSb, Cu,Sn, , and SbZn. There are CuZn, InSb, n ~ e r o u systems s where two or even more IMCs are formed, as for example: Au,Cd-AuCd, Au,In-AuIn, AuSn-AuSn,, Cu,Sn-CuSn, Cu,Zn-CuZn-CuZn,, and Ni~n-NiZn,. It appears that I-Ig seldom forms the third compoiient of an IMC being precipitated, as for example Mn,Zn,oHg, or HgTlTe,. Here is a typical example of the analysis of the solubility data of CdTe and Cd,Hg,-,Te alloys being introduced into liquid Hg. HgTe and CdTe are almost line compounds which are completely miscible in the solid as well as in the liquid state. Several authors measured the solubility of CdTe in pure Hg (at 250600 “C), in Te amalgam (at 230-530 “C) and performed thermal analysis at higher temperatures (>730 “C). Guminski (198617) showed that KgdTe= [Cd][Te] best described the equilibria in dilute Cd-Te amalgams; see Figure 8. Most of the reliable data lay on or very near to the straightforward relationship of log&, vs. I/??. The positive deviations from this line in the high temperatur~ range are probably due to signi~cant changes of the activity coefficients of the components in the Hg-poor melt. One may consider a t h e r ~ o d y n a m icycle ~ and compare the c o r r ~ ~ ~ o n d ti hn ~g r m o d y n a ~functions ic for the formation of an IMC in a Hg medium and in a related binary system. Such an analysis was performed by Guminski (1986a) on 23M-M”-Hg systems taking into account the enthalpic eEects of reactions:
M + H g ~ M ~ I - I g AHM
(7)
t
0‘1
0.01
TW4x 1
-1
Figure 8 Dependence of the CdTe solubility or the square root of its solubility product in Hg vs. reciprocal temperature ( ) A. V. Vanyukov, I. 1. Krotov and A. I. Ermakov, I d. Nauk SSSR, Neorg. Mater., 13 (1977) 815; ( 0 ) J. Y. Wong (1980), as cited by C. H. §U, P. IS.Liao, T. lectron, Mater., 11 (1982) 931; J. Electron. Mater., 13 (1984) L. Schmit, C. J. Sperschneider Trans. Electron Devices, 27 (1980) 24; (A) A. V. Vanyukov, I. J. Mrotov and A. I. Ermakov, tzv. Akad. Nauk SSSR, Neorg. Mater., 14 (1978) 657; (+) F. R. Szofran and Lelioczky, J. Electron. Mater., 12 (1983) 713; ( x ) P. J. Tung, J. Electron. Mates.,
Equations (7) and (8) denote M and M’ dissolution; (9), precipitation of an IMC in Hg; and (lO), formation of an IMC in the corresponding binary system. Balancing equations (7-10) we may write:
A~ a l g a ? n s
+ yAllpf,l + A H ~ ~ ~ ~ ( H g )
29
Ti, Mn, Fe, Ni, Zn, Cd, Al, Ga, In, Tl, Pb, Sb, Th, and U by means of these procedures. Hg can also be useful - AHMXM~~ (1 1) in metal separation. For example, a Ce-Fe alloy, used in lighter flints (see Chapter 31, Vol. 2), is d e c o ~ ~ o s e d If hH[,,) approaches 0, then we have an additional in liquid Hg at elevated temperatures; then Ce proof that an IMC formed in the Hg medium and in dissolves the Hg, and Fe forms an undissolved scum. the corresponding binary system are the same. HowAfter filtration and evaporat~onof Hg from the filtrate ever, without further investigations we must not to recover the Ce, and from the scum to recover the Fe, ascribe a non-zero enthalpic effect, AH(Hgl,to either both the Ce and the Fe are ultimately separated. formation of an IMC with Hg as the third component, Kozin (1964, 1970) described the essential technical f o ~ a t i o nof a soluble form of an IMC, or surface details important to electrore~ningof metals to high f, crystals by Hg, The AH(Hg)values purity (> 99.9999%) by means of polarization of the approach 0, within experiniental error, for the most and M, amalgam cells: M, M’, M”, . . . Hg/M2~/M-Hg exhaustively investigated compounds, such as: AuCd, M’, M”’, . . . Hg/M2+/M.Electrolysis carried out with AuCu, AuIn, AuSn, AuZn, CdTe, CuZn, LiSn, the use of direct current brings about subsequent Ni,Sn,, and PdZn. Assuming the validity of equation deposition of M or its amalgam on the right side. M’ (1 1) one may predict experimentally unknown A H M ~ M ; and M” are impurities which remain in the Hg phase, if values for several INICs, as - 1222 15 kJ/mol for they are more noble than M, and are in the electrolyte AuMg or - 105.7$. 1.6 kJ/rnol for KPb. phase in the form of ions, if they are more electroactive than M. When the process is continuously repeated several times, then the M is considerably enriched at every stage of the process and the content of impurities decreases. If an M amalgam is finally formed instead of M, Hg is distilled off at high temperature and reduced Hg was used even in antiquity for extraction and pressure. Applying this inethod of multistage electrorefining of Au and Ag from ores as well as for silvering lysis Kozin (1964, 1970) obtained Cd, Ca, In, Tl, Pb, and gilding; see Chapter 24, Vol, 2 and Weeks (1968) and Bi, all at a purity higher than is obtained from the for more details. Knowledge about the safe perforzone melting technique. mance of these processes is very essentials3 The recovery of noble metals from ores may be based on either dissolution in Hg with subsequent formation of IMCs with them (as in the case of Ag and Au) or even surface wetting (as in the case of Pt-like metals). Amalgams prepared by reacting liquid mercury with Amalgams of Zn or Fe have been used for the wetting various metals can also create starting materials for of Pt, Pd, Rh, Ir, and Cu which are present in the powder metallurgy operations. The metals which have autogenous Pt ore. It was found that a process is most high solubility in liquid Hg (Ag, Au, Zn, Cd, Sn, Pb, effective when the input amalgam contains 5 mass % Bi, Mg, Ca, Ba, Li, Na, and K), give up Hg at high Zn and the electrolyte of pH2-3 contains 2mol/ temperatures and form macrocrystalline and poorly dm3NaCl (Shemyakin et al., 1979). Hg may also be sintered sponges. The slightly soluble or insoluble exploited for extraction and refining of various other metals (Al, Ti, Zr, V, Ta, Cr, MO, W, Mn, Fe, CO,Ni, metals. Differences in their solubilities and normal Pt, and U) after similar treatment form fine powders electrode potentials in Hg are the primary basis of the whose particle sizes depend on the formation (or not) corresp~ndingtechnologies. Jangg and Bach (1961) of TMCs by these metals with Mg. The particle size described methods of preparation of the alkali metals, increases with time and temperature during the process. From a fresh Fe amalgam one obtains 3Quite recently this author spent three weeks at a provincial particles of 2-5 pm diameter, and after 1 h of conuniversity in Peru gwing lectures on hydromet~llurgy.All ditioning the amalgain at 300 “C the diameter increases participants in this course were especially interested in Au recovery from ores with the help of Hg, asking for very to 25 prn. If a metal forms an IMC with Hg (the fourth specific details. They were fortunate to learn how to improve a group of amalgams) then the metal particles are even primitive apparatus for Au extraction so that the Wg could be smaller after the Hg isolation (Jangg, 1965). Finely safely reused for the next process and would not poison their divided transition metals from this process are valuable environment. One may therefore deduce that this old and primitive method of Au acquisition is still in practice there. catalyzers for chemical synthesis. AH(HgI = xnHM
30
Structure and Composition
There are many examples of the useful formation of alloys and defined IMCs with the amalgam method. Two metals introduced into Hg may react inimediately lloy may be formed after the g is distilled oE. In this way powders of CO-Fe A13Ni2,Al,Ni, A N , Cu-Zn, nd Bi-Mn (Kozlovskii et al., alloys with Fe, CO, Ni, Ag, and Au (ICirchmayr, 1965), rare earth metals, Mn, Ni, i and Ge (Mayer et al., 1967) have been prepared.
After successful enrichment of OLi isotope in a Hg phase from an equilibration of Li amalgam with a %i+ and 7Li+ mixture, the phenomenon was utilized for separation of H and D, 14M and 15N, 39Mand 41K (Kozlovskii et al., 1971). The interphase exchange between the ionic solutions aiid amalgams compared to the aiialogous exchange with solid M is quite fast, a characteristic essential for practical separation kinetics. ~ i x t u r e s of radioisotopes are effectively separated by this method. Amalgams of the transamericium elements were the first alloys investigated. The radiopolarograp~ymethod allows for reduction on a Wg electrode of Cm, Bk, Cf, Es, Fm, Md, or No ions to the metallic state at concentrations as low as 10-8 to 10-11m01/dm3. The partial excess Gibbs energies of formation of these amalgams were estimated from the po~arographichalf-wave potentials by
amounts of these metals. Liquid alloys of Ln, such as La,Ni, were proposed as possible getters for tritium (3H) in a fusioii reactor operation, Such alloys have low equilibrium pressure, large capacity, and their surfaces are not regeneration may be performed to the following scheme: 5 3H2+ La,Ni-+(La 3W2),Ni Hg
absorption at 530 “C
+ (La ’ H , ) , N ~ ~ ~ a ~ ~+i 5- W ,H2g amalgamatioii at 250-300°C i
+ Hg
regeneration at 400-430°C
as reported by Carstens and David (1983).
~~v~st~gatio~s
The ab0ve”mentioned formation of stoichiometric AuZn in liquid Hg was first utilized by Tammann and Kollinanai (1926) as a kind of potentiometric titration of Au (present in Hg) by Zn introduce^ into it). After detailed investigation of the Hg-rich corners of the Au-Zn-Hg system, (Guminski and Galus, 1971), the Pd-Zn-Wg system (Dergacheva et al., 1978) and the Pt-Zn-Hg system (~um i nskiet al., 1983), the K,, values of AuZn, PdZn, and PtZn, were determined to be 7 x 10-*0nio12/dmA,4 x 10-9 iiio12/dniA,and 1 x 10-14m013/dm9at 25 ‘C, respectively. The fast precipitation of these IMCs in Wg was used to determine the Au diffusion coefficient (Guminski and Galus, 1977) and concentrations (solubilities) of Au, Pd, and Pt in Hg (Guminski, 1991) based on amperometric, voltammetric, and chronopotentiometric experiments. All these noble metals are not electroactive at negative patentials in an amalgam form, but their concentrations and diffusion coefficients are precisely traced by electroactive Zn reacting . formation of vigoro~s~y with them in a Hg m e d i u ~A IMCs in Hg may sometimes complicate electroanalytical processes and especially the stripping analysis; the most typical disturbance is caused by reaction of Cu and Zn on a Hg electrode. Therefore one should be very careful during aiiterpretation of the voltainmetric curves which should have purely di~usionalcharacter in oxidation steps. The catalogue of possible interactions between metals being introduced into Hg was compiled by Guminski and Galus (1992).
Reduction r e a c t i o ~with ~ the use of a ~ a l g a ]nave ~s occasionally been used to obtain chemical csnipounds (Smirnov, 1970). The classical example is the reaction of Na-Hg (obtained at the g cathode from electrolysis of NaCl solution with C1, by-product at the graphite anode) with H28:
+
2H20 2Na-Hg+H2
+ 2NaOH + 2
where pure NaOH and gaseous H are produced; the Hg is then reused after the reaction is complete. Hydroxides of alkali metals, alkaline-earth metals, Ln, An, and A1 are formed in a similar way. The analogous reactions of Na-Hg with alcohols lead to formation of alkali metal alcoholates and hydrogen. The action of 0,, CO,, NO,, ClO,, SO, and Na2S4on Na-IIJg leads to the formation of Na,O,, Na2C204,NaNO,, NaGlO,,
Na2S204,and Na2S, respectively. Na amalgams are frequently utilized in organic chemistry for hydrogenation, reduction of hydroxide-, carboxyl-, nitro-, nitrozo-, azoxy-, hydrazo-, oxime-, halogen-, sulphur-, or arsenic-containing groups. In some cases even the C-C or C-N bondings may be broken. In the reaction of alkyl chloride with Na-Hg, the very poisonous mono- and di-substituted allcyl and aryl compounds of Hg are formed. The HgC $- ion is probably the most ubiquitous. Substantial amounts of Hg are consumed in paints while lesser quantities are used for production o f agricultural and medical chemicals (Vcnetsky, 1981).
Froni a historical perspective several applications of anialgams werc once significant. An alchemist at the enry VI discovered that Cu rubbed with Hg acquired a silvery tint; soon, on the King’s order, Cu coins in masses were coated with Hg counterfeiting the Ag coinage (Venetsky, 1981). Early electrical machines used Hg or amalgams to establish good contacts; especially Hg-In alloys with a certain admixture of 0 have the advantage of well wetting all metallic surfaces at temperatures even below 0 “C (Westbrook, 1986). Hg was applied to join Sn foil to glass in forming ~nirrors.~ The first successful tungsten lamp lament was prepared by ~ ~ t r u d i nBi-Cd-Hg-W g alloy from which the Bi, Cd, and Hg were subsequently removed by heating (Coolidge, 1965). Ag, Cu, and Sn amalgams and combin~tionsthereof have been used as dental restorative materials for almost 500 years and are still important today (Waterstrat and Okabe, 1994). Due to their highly toxic properties, where possible Hg and its alloys are substituted by other materials. However, since Hg is easy to purify to a spectacular degree, it is still used as the working fluid in therniometers, barometers, electric resistance standards, circuit breakers, switches, diffusion pumps, seals, lasers, and electrode materials in pure and applied electrochemistry. Miiiimal interaction of constructional materials with Hg is quite essential for long-time employment of such instruments. Partial solutions, IMC formation at the solid-liquid interface, grain boundary grooving, embrittlement, and wetting have been investigated in several papers (Barlow and Pla~ting,1969; Bennett, 1977; Lahiri and Gupta, 1980; Sudarshan et al., 1984). Hg has a higher thermal 4See insert, above right,
Little Willie from the nzirror, Licked the mercury all off, Thinking in his childish error, It would cure the whooping cough, At the funeral, Willie’s mother Sadly said to Mrs. Brown, “Twas a chilly day for Willie When the mercury went down”. This anonymous bit of doggerel, in two short verses, manages to allude to three different qualities of mercury. fin D. NcCord, What Cheer, Longmans, 1945
conductivity coefficient, boiling point and critical point than water has. These are quite promising factors in the case of using Hg as a circulating coolant in thermal reactors. Hg was used successfully in a fast breeder nuclear reactor coolant, in a turboelectric power conversion system for space craft, in (Fleitman and Weeks, 1971), and as a ~agnetohydrodynamic working fluid (Branover and Lesin, 1995). Fleitman and Weeks (1971) performed and summarized many efforts to establish the ~orrosion resistance of metallic and non-metallic materials in Hg between 250 and 700 “C. Pure Fe, and graphite best withstand the Hg at forming IMCs with Hg such as Ti, Zr, Mn, Ni, and Pt are not recommended for application in construction, although these metals show quite resistance to Hg, even for long tiine temperature. Steels and other alloys sliow preferential leaching of Cr, Ni, and Mn as well as other readily soluble constituents at high temperatures (Suzuki and Mutoh, 1992). it was found that addition of Ti or Zr to liquid Hg inhibits its corrosive action due to formation of highly protective Ti-C-N or Zr-C-N films on the steel surfaces. Cu, Pb, or Sn dissolved in Hg accelerate the corrosion. Graf et aE. (1985) found that amorphous 78Ni-8%-14B glass showed quite good corrosion resistance to Ng attack at room temperature. The Cd-Hg-Te system and many related systems ( Hg-Se and M-Hg-S) have been frequently investigated (Tomashik and Grytsiv, 1982). Much of work has been devoted to the Cd-Hg-Te materials. Solid solutions formed between CdTe and HgTe are semicond~ctors with an energy gap increasing between -0.3 and 1.6eV as a nearly linear function of the CdTe molar fraction content. These materials are used in the Fabrication of infra-red detectors in both military and civilian applicatioiis such as surveillance and in space. Iii order to obtain these alloys with desired electronic
32
Figure 9 Margaret Weiler's research on mercury-cadmiumtelluride photovoltaic detectors has helped Lockheed-Martin TJR Imaging Systems develop arrays such as this, in which each 60 ~ m - s q u amesa-~elineated ~~ pixel is capable of detecting on the order of IOpm infrared light with w~velength~
properties, their growth, optical, dielectric, thermodynamic, t h e ~ o p h ~ s i c and a l mechanical properties, solid state diffusion, defects, and band structure have been widely studied. The corresponding information has been collected in numerous papers, separate books, databases aiid conference proceedings (Mercury Cadaterial Properties Database, 1994; Capper, 1994; Procee~~ngs, 1989). As an example of significant achievement in this field Figure 9 shows a Cd-Hg-Te plio tovoltaic device made according to the accomplishments of M. H. Weiler (Young, 1999). Small amounts of Hg are introduced into glass tubes of fluorescent lamps because Hg vapors emit radiation with high efficiency. To decrease the toxicity of Hg vapors and to adjust the light characteristics, several substitute alloys (Hg-Ni, Hg-Na, Hg-Zn, HgCd, Hg-Sn, Hg-Pb, and in the form of solid pellets were proposed and applied instead. After heating the amalgam, they are decomposed to give the proper Ilg vapor pressure, The amount of Hg intro~ucedcan also be reduced by employing sintered pellets o f a porous Cu-Fe-Hg alloy containing Hg in its capillaries. The Cu-Fe a ~ a l g awas ~ prepared by electrolysis, then tempered at 250°C for 1h, and the Hg excess filtered off and pressed away. These alloys are readily reproducible lamp fillings, they show reduced environmental contaminat~onand also possess useful ferromagnetic features (Schuster, 1986; Guminski, 1990a). Tubes filled with gaseous mixtures of Ar, Ne, and Hg gave lighting of various colours
depending on the gas composition, tube geometry, and mode of the discharge (Lee et al., 1997). When large amounts of Fe or CO, exceeding their saturation levels by a million times, are introduced into Hg (most easily by electrolysis), the still liquid product possesses durable inagnetic properties at room temperature, To prevent an agglomeration of the Fe and CO particles of a colloid-like nature, addition of Sn, Ga, or Na to the amalgam was found to be effective. Then Sn or Ga form monolayers of IMCs on the Fe or Go particles that do not significantly decrease the magnetic properties of the liquid magnets, keep their viscosity low, and inhibit the particle growth experienced on typical crystallization of these metals. An excess of Sn or Ga leads to the destruction of the magnetic moment of the particles (Windle et al., 1975; Keeling et al., 1984). The Ga-Hg (20vol.%) alloy was selected to investigate the stability of a metallic dispersion under microgravity conditions in the Space Shuttle. The Ga-Hg binary system shows an extended niiscibility gap. The dispersed state of Hg was recorded on x-ray sensitive film. The sample was cooled at different cooling rates from 220"C, where the sample was homogeneous. It was observed that the required homo~enizationtime of the sample at 220°C was half as much in space. When cooling the sample into the miscibility gap, precipitation of Hg droplets occurs rapidly by ~eterogeneousnucleation at the Ga surfwe. Supercooling o f the melt appears to be less than 20 "C. At higher cooling rates convective, non-diffusional material transport is likely to contribute to Hg particle growth. Hg droplets near the alloy surface seem to be stationary (Otto, 1984). 8.6
~ ~ s t i ~ ~
This author and his colleagues have been asked many times to prepare so-called red mercury and were offered a very good remuneration, even without giving a recipe. No metallic amalgams are knowii to possess a red colour (see Stein~mannet al. this volume). One can imagine that a red-coloured Hg could arise from an adsorbed film of a red dye on its surface, but such a product would be highly expensive and for what? Any serious information about red mercury could not be found in the open scientific literature. The recent reports by Orlov (2001) and Ladika (2001) put more light on this matter. It seems the most reliable explanation that the name Red Mercury was given to the amalgams o f the transuranium elements. They are relatively well soluble in mer~ury,easily a ~ ~ l g a ~toa t e
33
Amalgurns
form IMCs, and the Elg medium attenuates the radioactivity, making transportation and handling potentially easier. Pu-Hg intermetallic is a reddish powder according to Ladika (2001). A capsule filled with Red Mercury is hot and contains the dangerous stuff inside; therefore the adjective Red seems to be ap~ropriatein the figurative sense as well.
Problems of contamination and poisoning by Hg have been known for centuries. The action of Hg and its cornpounds is frequently not immediate but generally causes irreversible damage in human organs. Due to the high volatility of imtallic H g and the relative ease with which Hg compounds decompose at elevated temperatures, all operations with Elg, amalgams, and Hg compounds must be performed in tightly closed containers placed in a well-ventilated system, in case of any failure of the apparatus. Therefore it is fundamental to collect and keep products containing Hg in absolutely safe places. Even very low concentrations of Hg and its compounds present in water are absorbed by some kinds of bacteria, then accumulated in the bodies of water animals, and may finally be consumed by hui~ans,causing serious diseases. Recovery (recycling) of Hg from waste materials may be carried out by thernial decomposition and es distillation of Hg. Selection of proper t e ~ ~ e r a t u r for such processes should be done only after studies of the corresponding binary and ternary phase diagrains collected in ~ a s s a l et ~ ~nl.i (1990), Toiiiashik and Grytsiv (1982), ~ ~ l ~of eAlloy t i Phase ~ Diagrams, and Jourizal of Phase ~ ~ ~ i l i b r i a . orgstedt et al. (1994) proposed a special method of Hg extraction from waste batteries: steel, iZln-Hg/ KOH/HgO,C,steel. Such batteries, having been cracked open, were treated with liquid Na in an Ar i i ~ red~ictionof Zn and ~ ~ t ~ ~ o s pNhR~~rde~. i t i acaused Hg from their oxides to their metallic states. The NaZn amalgam, after separation from the solid bodies, was electrolyzed in a solid electrolytic cell: Hg-Zn-Na/ beta alumina/Na. In this way the Na was almost quantitatively separated from the Hg (but contaminated by Zn and other metallic impurities). The electrolytic Nn may then be reused, and the only byproduct of the process is Na,O as a scum on the amalgam.
oys to
The question as to whether other low-melting alloys possess properties similar to the am al ga~s,is easy to answer affirmatively; however, our experimental ~ ~ o w l e about d g ~ them is of ~ o ~ s i d e r alesser ~ l y degree. al Guminski (1990b) compared the e ~ p e r i ~ e n tsollxbilities of metals in liquid, low-melting metals such as Gta, In, Sn, Pb, Bi, Li, Na, K, demonstrated that metallic solutes, well soluble in one metallic solvent, are usually well soluble in another one. Similar observations were established with sparingly soluble metals. In every metallic solvent the solubilities, ordered according to the atomic number of the metallic solute, form dependencies similar to that shown in Figure 1. Exceptions are observed for Lns and Ans which are well soluble in p metals but not in the alkali metals. Ln and An do not form IMCs with the alkali metals, but the p metals do. Analyzing diffusion coefficients of various solutes in dilute solutions of 20 liquid metallic solvents ( ~ u m i n s ~ , 1995) one may observe that if an IMC is formed in a binary system solute-solvent, then its diAFusion coefficient is lower than predicted by Equation (1). If no specific interaction between a solute and a solvent exists, then the diffusion coefficient is near to the prediction of equation (1). It was further shown by Guminski (1996) that taking into account the type of binary phase diagram, the solubility, the heat o f the dissolution process, and the activity and diffusion coefficient of metals in liquid Al, Ga, In, or T1 may be divided into four groups in the same ~ a n n e as r was described for the binary amalgams in section 4 of this chapter. The first paper dealing with specific i ~ t ~ r R c ~ini o ~ i ~ liquid ternary metallic systems was published by Tamniann and Schafmeister ( I92S), who observed that partition of a third metal between two im~iscible liquid metals is not in general equal. Concerning solutions of some metals in liquid Ga (sometimes named ‘gallams’)Kozin et al. (198 1) observed that nine metallic solutes, dissolved in p o ~ ~ ~l y~ ~ i sGa-Hg ~ibl~ alloys at 35 “C, always showed a higher concentration in the Elg phase than in the Ga phase (the highest partition coefficient was observed for Mn (= 2 x 105) and the lowest for Sn (= 2). In monographs devoted to Ga (Yatsenko, 1974) and In (Yatsenko, 1987), their author collected experimental data on the binary phase t i these ~ ~ ~ l i t d ~ a ~ asr well ~ ~ as ~ 0~11 sthe c ~ ~ ~ ~ a of metallic solvents with pure metals and alloys. By g such as W, Re, analogy to Hg, the high ~ e l t i n metals
e, Ta, Nb, and OS are the most resistive to corrosive dissolution in both Ga and In (700-1200 "C). Soine struct~ralalloys show etching of the more soluble metals by these liquid metals. Additions of Ti or A1 inhibited corrosion which was also diminished by the on of Cr protective layers. Ceramic materi, AlN, A1203,SiO,, graphite, and ZrO, are als not ed by Ga and In at temperatures lower than 850 "C. Examples of interactions similar to those observed in amalgams may be found for other liquid metal solvents. ~ a r t m a n n et al. (1959) observed precipitation of Cu,Sb, Cu3Sn, and Cu,As in liquid osa et al. (1981) showed the appearance of In3Sb, InSb, and Zn4Sb3complexes in liquid In-SbZn alloys. Solution chemistry in the liquid alkali metals also shows some similarities to amalgams. However, the specific influence of oxygen on the solubilities and corrosion of the transition metals in liquid Na, K, Rb, and Cs has a spectacular character, as does the presence of nitrogen in liquid Li (Borgstedt and Guminski, 2000). The presence of 0 in the Cr-Na system leads to the formation of NaCr02, which is more soluble than pure metallic Cr in liquid Na. Si and N introduced into liquid Li react with precipitation of Li,SiN3, whereas solid Fe,P introduced into Li is decomposed into dissolved P and Fe precipitated in Li. u m m i ~ gup, the amalgams do not seem to be a peculiar kind of alloy, and many features first observed in amalgams have been later discovered in solutions of other lowmelting metals.
Barlow, M. and Planting, P. J. (1969) Z. Metalll~.,60, 292. Barnes, J. W. and Bailay, E. H. (1972) ~ o r l ~ M i n i25, n ~49. , Bennett, 3. E. (1977) Microstr. Sci., 5, 395. Borgstedt, H. U., Guminski, C. and Perk, Z. (1994) Abstraeks of ~ a p e r sp ~ e s c n ~ eat d the 6th Intl. Symp. on Solublity Phenomena, Buenos Aires, Argentina, p. 0-12. Borgstedt, H. U. and Guminski, C. (2000) Monatsh. Chew.., 131, 917. g ran over, H. and Lesin, S. (1995) ~ i q u i~ ~e t a Systems, l H. U. and G. Frees, eds. Plenum, New York, p. 377. Cammarota, V. A. (1975) ~ ~Facts and ~ Problems, e ~U S . ~ Bureau of Mines. Capper, P, (1 994) Pro~ertiesof narrow gap cad~~um-based cornpourin's. EMTS Data Series, LEE, U.K., 640pp. Carstens, D. H. W. and David, W R. (1983) J. Nuclear Mat", 115, 203. Coolidge, W. D. (1965) Sorby ~ e r ~ t e n n ~ S ayl ~ ~ o on s ~theu ~ History of ~ekullurgy.C. S. Smith, Ed., Cordon & Breach, New York, p. 443.
David, F., Sarnhoun, IS.,Lougheed, R, W., Dougan, R. J., Wild, J. F., Landrum, J. H., Dnugan, A. D. and Hulet, E. K. (1990) S u d i o c h ~ Acta, ~ . 51, 65. Dergacheva, M. B., Kozin, L. F. and Panova, N. L. (1978) VINTTI Dep. 3595-78, Moscow. Fleltman, A. H. and Weeks, J. R. (1971) N L L CEng. ~ . Design, 16, 266. Foley, W. T. and Retd, L. E. (1963) Can. 1.Chem., Galus, Z. (I 994) Fundamental~sof ~lectroche~ic~il Analysis, 2nd ed., Horwood, Chichester and PWN, Warsaw. Graf, K. H., Lohmann, W. and Ribbens, A. (1985) Rapidly ~ u e n c h e d~ e t a l s ,Steeb, S. and Warlirnont, H., eds. Elsevier, Amsterdam, p. 1411. Greenwood, N. N. and Earnshaw, A. (1984) C h ~ ~ ~ i soft rthe y E l e ~ e ~ tPergainon, s, Oxford, p. 1395. Guminski, C. (1986a) Z . Metaflk, 74, 87. Guminski, C. (1986b) 9. ~ e . s ~ ~ - C o m ~ ~116, L15. 1. Guminski, C. (1989a) J. Mater Sci., Guminski. C. (1989b) J . Mater. Sci. 5. Guminski, C. (1990a) Technical Report to POLAM (Light Sources Producer), Warsaw, Poland; unpublished. Gurninski~C. (1990b) Z . met all^,, 81, 105. m ? ~ 16 on Guminski, C. (1991) J . ~ ~ s ~ ~ " C o Met., Guminski, C. (1992) J Phase Equil., 13, 657. Guminski, C. (1995) ~ i~ e t ~a Systems, l ~ €3.i U. Borgstedt ~ and G. Frees, eds., Plenum, New York, p. 345. Guminski, C. (1996) Abstracts of papers prcrscnted at the 7th Intl. Symp. on Solubility ~l~enow.ena, Leoben, Austria, p. 0-28. Gurninski, C. and Galus, 2.(1971) Bull. Acad. Pol. Scz., Ser. Sci. Chim.,19, 771. Guminski, C. and Galus, 2;. (1977) J. ~ f e c t r o a n aChem., ~. 83, 139. p. 1395. Guminski, C. and Galus, Z. (1986) Metals in ~ e ~ c u ~ ~ Solubility Data Series, Vol. 25, C. Hirayama, ed., Pergamon, Oxford. Guminski, C. and Galus, 2. (1992) I n t e r ~ ~ ~ aCl lo~~c p o u n ~ in Mercury, Solubility Data Series, Vol. 51, J. G. Osteryoung and M. N. Schreiner, eds., Pergamon, Oxford. and Galus, Z. (1983) J . Hartmann, H., Ensselin, IF. and Wunderlich, F. (1959) Z. E r ~ b e r ~ b aMetallhuttenw., u 12, 374, 437, and 537. Hurne-Rothery, W. (1936) Phil. Mag., 22, 1013. Iandelli, A. and Palenzona, A. (1979) ~ u ~ d b oono the ~ Physics and ~ l ~ e m ~ sof ~ rthe y Rare ~ a r t h s , K. A. Cschneidner and L. Eyring, eds., Vol. 2, Ch. 13, NorthHolland, Amsterdam. l Janng, G. (1965) et all., 18, 442. Janng, G. and Bach, H. (1961) ~ u e c k ~ ~ i fand ~er A m a l g a ~ ~ ~ e t a l l u r gi~ne ~ a ~ d b ~ i cder h Technische ~lektrochemie, Band 1, Teil 1, Akademie Verlag, Leipzig, 592. Keeling, L., Charles, S . W. and Popplewell, J, (1984) J. Phys. F., 14, 3093. Kirchmay~,H. R. (1965) Z. Metallk., 56, 767.
Korshunov, V. N. (1 990) Amalgam Systems, Moskovskii Unsversitet, Moskva, Ch. 2 and 9 (in Russian). Kozin, L. F. (1964) Ph.ysi~o-cher~~ca1 F~~ndamen~als oj Anzalgam Metallurgy, Na~il~a, Alina-Ata (in Russian). Kozin, L. F. (1970) A ~ a ~ g a~m e t a l ~ u r gTekhnika, y, Kiev (in Russian). Kozin, L. F., Nigmetova, R. Sh. and Dergacheva, M. B. (1977) T1~ermodynamic.s of Binary Am&am Systems, Nauka, Alma-Ata (in Russian). Kozin, L. F., Sarmuzina, R. G. and Popova, T. V (1981) Tr. Inst. Org. Kutal. Elektrochim. Kaz. Akud. Nauk, Kozlovskii, M. T., Zebreva, A. I. and Gladyshev, V. Amalgams and Their ~ p ~ l i c a t i ~ )Nauka, ns, Alma-Ata (in Russian). Ladika, S. (2001) Science 292, 64. Labiri, S. K, and Gupta, D. (1980) J . Appl. Phys., Lee, J.-Ch., Kang, D.-Y., Park, D.-H., Ham, J.-K., and Aono, &$.-A. (1997) Proc. 5th lnt. Con$ P~iip.Appl. Dielectr. Mater., Institute of Electrical aiid Electronic Engineers, New York, 1, 154. Luborsky, F. E. (1961) J. Electrochern. Soc., 108, 1138. Massalski, T. R., Subrainanian, P. R., Okainoto, €3. and Kacprzak, L. (1990) Binary Alloy Phase D~agr~ms, 2nd ed., ASM, Materials Park, OH. Mayer, I., Shidlovsky, 1. aiid Yanir, E. (1967) J. Less-
Singb, R. M. and March, N. H. (1995) in rnter~nefallic Co~t~ouP i ~r ~i n~~: ~ ~and l e ~Practice, Vol. 1, Wiley, Chichester, IJK, p. 661. Siniriiov, V. A. (1970) Reduction with A m a l ~ a Khimiya, ~~, Leningrad (in Ruswan). Sudarshan, T. S., Lim, M. H., Nefley, P. L. and Thornpson, J. E. (1984) J. Appl. Phys., 56, 2236. ture Suzuki, T. and Mutoh, I. (1992) High T ~ ~ F ~ ~ e r aCorroszon of Advanced ~ c t t e r z ~and l s Protecfive ~ o u ~ i n gV. s , Saito, Ea. Onay and T. Maruyama, eds., Elsevier, Amsterdam. Takacs, L. (2000) J. of ~ e t ~Jan. l ~p., 12. Tamaki, S., Waseda, U,,Takeda, S. and Tsuchiya, Y. (1982)
Tammann, G. and Jander, W (1922) Z. Anorg. Chem., 105. T a m ~ a n n G. , and Scbafmeister, P. (1925) 139, 219. Tammann, G. and Kollman, K. (1926) Z.Amrg. Ctzem., 269. ~s Tomashik, V. N. and Grytsw, 'v. I. (1982) State ~ i a g r a m of Systetm Based on A"Bvl S e ~ ~ c o ~ ~ ~ C~~mI~ounds, uci~3r Naukova Dumka, Kiev (in Russian). Venetsky, S. (1981) Tales about ~ e t a l s , Villars, P. (1997) Pearson's Handbook of Alloys, ASM Intl,, Materials Park, OH. Mercury Cadmium Telluride Material Properties Database Waterstrat, R. M. and Okabe, T., Ch.27 in Interme~allic (1999) Advanced Materials and Processes Technology Compounds: Principles and Practice, J. HI. Westbrook and Information Analysis Center, 201 Mill Str., Rome, N.Y. R. L. Fleischer, eds., Wiley, Chichester, UK, Vol. 2, Nugent, L. J. (1975) J. Znorg. Nucl. Chem., 34, 1767. p. 575. Orlov, V. A. (2001) Addressing the Challenge of Illicit Weeks, M. E. (1968) Discovery c?f' the E l e ~ ~ n7th t ~ edn. ~ , J. Niiclear Trafficking, in Symp. How ro Harnzonize Chem. Educ., p. 46. Peaceful Uses of Nuclear Energy and N o n - P r o l ~ ~ ~ r u ~ i o nWestbrook, ~, J. H. (199.5) Ch. 31 in I n t e ~ ~ e t a l l iCom~ound~s: c Japan Atomic Industrial Forum, Tokyo, Japan, 12 pp. ~ r ~ n c i p l eand s Prctctice, J. H. ~ e s t b r o o kand R. L. Otto, G. H. (1984) European Space Agency Rep. ESA SPFleischer, eds., Wiley, Chichester, UK, Vol. 2, p. 645. 219, p.43. Westbrook, J. H. (1986) Encyc~o~edia of ~ ~ ~ eScience r ~ ~ l s Proceedings of the 1989 U S . Workshop on the Physics and and ~ngzneering.M. B. Beyer, Ed., Pergamon, Oxford, Chemistry of Mercury Cadmium Telluride and Related p. 2548. 11-VI compounds, 3-5 October 1989, San Diego, CA. Windle, P. L., Popplewell, J. and Charles, S. W. (197.5) IEEE Rosa. C. J., Sonimer, F., Rupf-Bolz, N. and Predel, B. (1981) Trans. Mugrzet., 11, 1367. Z . M e t ~ l l k72, . ~ 47. ns ~ e ~ ~ l Yatsenko, s. P. (1974) ~ctllium- ~ n t e r a c ~ ~ owith Samhoun, K. and David, E; (1979) J . Inorg. Nucl. Chem., Nauka, Moskva (in Russian). 357. ~ ~~pp~ications, ~ e s Yatsenko, S. P. (1987) I n d i u ~- ~ r o ~ e and Schuster, W. (1986) 4th Intl. Symp. of Science and Technology Nauka, Moskva (in Russian). of Light Sources, U. of Karlsruhe, Karlsruhe, G e ~ a n y , Young, P. Irtdustr. Phys. 1999, no.4, 40. p. 263. Zebreva, A. 1. (1958) Vesfn. Akacl. ~ a Kaz.~ SSR, ~ no..1 1 , Sheniyakm, V. S., Rrik, K. A., Bukman, S. P. and Karasev, M. A. (1979) Zzv. Akad. Kaz. SSR, Ser. K h m No. 3, 73. 88.
This Page Intentionally Left Blank
Chapter 3 eryElides Loren A. Jacobson, Robert J. Hanrahan, Jr. and James L. Smith Los Alamos National Laboratory, Los Alamos, N M , USA
1. Introduction Forty years have passed since the first major investigation of beryllide intermetallic compounds was conducted, and this early work was based primarily on the high melting points and excellent oxidation resistance associated with these materials. This work, which was sponsored by the United States Air Force over the time period March 1956 to December 1961, was performed by the Brush Beryllium Company (Booker et al., 1962). Oxidation rate results were reported for various beryllide compounds including TaBe,,, Ta,Be,,, ZrBe,, and Zr,Be,, over the temperature range from 2300" to 2750 "F. Ternary systems of beryllium and silicon with molybdenum, niobium, tantalum, tungsten and zirconium were also studied. A mixed phase material with the stoiclziometry MoBeSi proved to be the most promising ternary system, but very little follow-up work was done to exploit the outstanding high temperature oxidation resistance and strength properties that were measured for these beryllium-containing intermetallics. G. V. Samsonov (1966) published an extensive compilation of beryllide properties. This article, which was condensed from a larger monograph, discusses the beryllides as they form with the elements in groups across the periodic table. Some phase diagram information is presented, and there are 179 references to other work. One revealing statement in this review has to do with the peculiarities of the electronic structure of the beryllium atom. The normal electronic configuration of beryllium is 1s22s2 but in solid beryllium and its compounds there can be a single s to p promotion, giving a configuration of ls22s2p, and it is the sp configuration that brings about the
formation of covalent bonds between beryllium atoms, particularly in the lattices of its intermetallic compounds (Samsonov, 1966). In the late 1980s, renewed interest in the beryllide intermetallics was generated in part by the National Aero-Space Plane project. In 1995 a review volume by Dudley and Desai (1995) compiled the data available up to that time. This review is very thorough and presents the available property data including mechanical properties and oxidation behavior, as available, for compounds beginning with CrBe, and ending with Zr,Be,, + Ta,Bel,, arranged in alphabetical order. The presentation is detailed, including many charts and graphs, and the reader is referred directly to the review for information on specific compounds. Updated tabular data on many beryllides is given in Tables 1 to 3. Other unusual physical and structural properties that these compounds exhibit have contributed to the renewed interest in the beryllide intermetallics in the 1990s. For example, beryllide intermetallics with actinide metals exhibit unusual superconducting properties based on a solid-state entity called a heavy fermion, more generally described by Aronson and Coies (1994). Several other efforts to investigate the structure and properties of beryllides have also been conducted with the aim of developing better understanding of the behavior and potential of these materials. This chapter will first describe the various types of beryllide intermetallics that have been characterized, then will summarize the information that is presently available on mechanical, chemical, thermal, magnetic, and other physical properties of these materials. Methods of synthesis will be treated, and the chapter will conclude with some possibilities for application of beryllide intermetallics toward the
Intermetallic Compounds: Vol. 3, Principles and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 0 2 0 0 2 John Wiley & Sons, Ltd.
38
Structure and Composition
Table 1 Physical constants of beryllides of Group IVRVIIB elements Beryllide
Crystal Structure Lattice Parameters Pearson Symbol a, c (A) -__ CrBe, hP12 4.260, 6.988 CrBe,, tI26 7.23, 4.173 HfBe hP3 3.787, 3.1 59 HfBe hP6 4.534. 3.471 cF112 10.00 HfBei3 a-Hf2Be,, hR1Y 7,494, 10.93 hP38 7.44. 7.38 P-Hf,Be,, MnBe, hP12 4.231, 6.909 MnBe, 5.931 cn4 MnBe,; tI26 7.276, 4.256 4.89 Mo,Be cP8 MoBe, hP12 4.433. 7.341 MoBe,, tI26 7.271, 4.234 11.636 cFI84 MoBe2, tPlO 6.49, 3.35 Nb,Be, NbBe, cm4 6.535 MBe, hR12 4.561, 21.05 hP6 NbBe, N/A tI26 7.376, 4.258 NbBe,, hR19 7 409, 10.84 NbzBe,, ReBe, hP12 4.354, 7.101 ReBe, N/A N/A ReBe,, N/A NIA cF24 N/A Re0 9zBe1, 11.54 cF184 ReBe,, 6.010, 4.89 t112 Ta,Be 6.50, 3.32 tPlO Ta,Be, 6.51 TaBe, cF2.4 4.53, 20.95 hR12 TaBe 7.334, 4.267 TaBe,, tn6 hR19 7.388, 10.74 Ta2Be,, , CP2 2.94 TiBe (metastable) 6.450 TiRe, cn4 4.49, 21.32 TiBe, hR12 TiBe,, 7.35, 4.19 tn6 7.392, 10.79 hR19 a-Ti,Be,, 7.36, 7.30 hP38 fi-Ti,Be,, VBe, hP12 4.394, 7.144 t126 7.278, 4.212 VBe,, 4.446, 7.289 hP12 WBe, 7,362, 4.216 tI26 WBe,, cF184 11.628 WBe2, 3.82, 3.24 ZrBe, hP3 4.564, 3.485 ZrBe, hP6 10.047 C F I 12 ZrBe,, 7.548. 10.997 hR19 Zr,Be,,
Melting Temperature Specific Gravity (K)
solution of several challenging technological problems. Finally, some general comments are made regarding the toxicity of beryllium, and beryllium-containing alloys and compounds. In the year of writing, 2000, this is one of the primary reasons why this promising class of materials has not been more thoroughly investigated.
-21 10 - 1610 1603+60 1858+ 10 1963+ 10 2043 70 2043 & 70 N/A N/A N/A 1173&100 2300 & 200 1973 (?) 1573 I863 19033.30 2353 f 50 N/A 1945+5 2073-50 N/A N/A N/A N/A N/A > 1300 N/A <2070 N/A 2123 +28 2263 t28 N/A 1623& 50 1723+50 1823+60 1943+50 1943+50 N/A N/A c 2520 <2020 N/A 1510 1750 2200$.28 2255 & 28
+
4.34 2.44 8.322 6.006 3.93 4.66 4.66 4.494 N/A 2.40 8.425 6.12 3.02 2.51 NIA 5.28 4.72 N/A 2.88 3.28 11.632 N/A N/A N/A 3.38 N/A 13.270 9.79 8.19 4.18 5.05 N/A 3.27 3.01 2.26 2.43 2.42 3.834 2.37 10.2 4.57 3.27 4.32 3.60 2.72 3.05
Ref. 3,4 5 6.31 6,3 1 6,31 6,3 1 6,31 3 Okamoto and Tanner (1987) 5 7 396 9,lO 7,ll 12,153 6,13,30 13,145 Okamoto and Tanner (1987) 5,30,60 14,30,73 3 Okamoto and Tanner (1987) Okamoto and Tanner (1987) Okamoto and Tanner (1987) 11,15 16 6 5,6 6 3,4,60 6,60,73 Okamoto and Tanner (1987) 3,4,31 6,3 1 6,17,31 6,17,31 6,17,31 3 5,17 3 5 7,11 18 19 20,32,60 19,21,73 (Numbered references from Dudley and Desai (1995))
2. Types of Beryllides and their Structure
2.1 MBp,, This compound has a cubic structure, based on the ZrZn,, analog. The Pearson symbol for this structure is cFI84 and the space group is Fd3m. Examples are the compounds WBe,, and MoBez2.
39
Beryllides Table 2 Physical constants of reported beryllides of group VIII elements __
Beryllide CoBe CoBe,, (metastable) &(Co2Be17?) p' (metastable)
c Co5Be2, Fe,Be FeBe, FeBe, eFeBe, FeBe,,, IrBe IrBe, IrBe, 1r2Be17
Ni,Be (metastable) p' NiBe Ni,Be,, OsBe, (metastable) OsBe, Os,Be,7 Os,Be OsBe Pd,Be Pd,Be, Pd,Be, PdBe PdBe, PdBe,, Pt,,Be PtBe Pt,B% PtBe, PtBe,, RhBe RhBe, RhBe6.6 Rh2Be17 Ru,Be, RuRez Ru3Be1, Ru,Be,, Ru,Be,, RuBe,, RuBe,
Pearson Symbol
Lattice Parameters
cP2 tI26 hP19 CR
2.611 7,237,4.249 4.128.10.717
cR?
2.658,3.16 8.3783,13.7652 7.6136& 1.525
hP96 cI52 & c9-416 cFl6 hP12 cn4 hP19,h P48?
Melting Temperature (K)
Specific Gravity
6.307
N/A
N/A
2.49
4.221,6.848 5.890
4.632 3.298
4.193,10.89
5.38
2.611 7.62& 1.53
6.248 3.688
11.342 4.221.10.94 11.350,10.628,8.4803
6.586 5.27
2.819 5.994 7.271,4.251
8.54 4.670 3.169
CP2 cI52
2.80
N/A
~$24
6.004 7.237,4.252 2.7397
4.521
tetragonal
cP2 hP7 tP
cP2 cI52 & cF416 hP* hP12 ell60 hP? mP7 N/A N/A N/A
CP2 cn4 1126
N/A
hP*
1126 cP2
NIA
N/A
hP*
hP7 cI80
hP12 CP
cl160 hP?
4.191,10.886 4.203.10.90 1.142 5.96,9.18 11.03 11.337 4.2036.10.90
hP*
mP*
2.2 MBe13
This compound has a face-centred cubic structure with 112 atoms in a unit cell, the binary prototype of which is NaZq3. The Pearson symbol for this structure is cF112 and the space group is Fm%. The metals Mg,
N/A N/A
3.60
4.158 3.54 N/A N/A
Ca, Pu, Sb, Sc, Sr, Zr, Hf and Y form compounds with beryllium that have this structure, as do all of the lanthanide and all of the actinide series metals investigated to date. This structure is illustrated in Figure 1, which also depicts various coordination arrangements for the different types of atoms.
Structure and Composition
40
Table 3 Physical constants of rare-earth and selected actinide beryllium compounds Beryllide Pearson Lattice Melting Specific Symbol Parameters Temperature Gravity (A) (K)
- _ _ _ _ _ _ I
CeBe,, DyBe13 ErBe,, BuBe,, GdBe,,
HoBe,, LaBe,, LuBe,,
NdBe,, NPB% PrBe,, PuBe,, ScBe, ScBe,, SmBe,, ThBe,, ThBe,, TmBe,, UBc,, YBe,, YbBe,3
cFll2 cFl12 cFt 12 cFI12 cF112 cF112 cF112 cF112 cF112 cFll2 cFl12 cF112 hP6 C F I 12 cFl12 cFl12 cFl12 cA12 cF112 cF112 cF112
10.376 10.237 10.209 10.291 10.280 10.221 10.456 10.173 10.355
3.060 3.463 3.551 3.280 3.361 3.509 2.979 3.686 3.128
found in the scandium, niobium, hafnium and zirconium systems. Other MBe, compounds with AuBe, (cF24) structure include tiose where M = CO, Fe, Pd, and Pt. 2.6 MBej This type of compound exists in a number of different crystal structures. In the Cu-Be system, the structures have Pearson symbols cF24 and hP3. In the Ti and Nb systems the compound has a crystal structure with the Pearson symbol hR12. These compounds have not been very well investigated.
N/AN/A 10.368
3.076
2.7 MBe,
N/AN/A 4.55,3.500 10.102 10.304 10.252 10.395 10.193 10.256 10.238 10.184
2.382 2.089 3.249 3.404 4.169 3.588 4.371 2.552 3.650
These phases are generally Laves phases, and can have all three of the crystal structures that are members of this group, Cubic C15 (cF24), Hexagonal C14 (hP12) and C36 (hP24). Most of the work on these phases has been related to their potential for storing hydrogen, and their unusual magnetic properties, as will be discussed later on. Representatives are found in the Ti, Cr, Cu, Fe, Hf, Mn, MO, Nb, Ta, Ti, V, W and Zr systems.
2.3 MBe,,
This type of compound is the D2, body-centered tetragonal in structure with 26 atoms in a unit cell, and with a binary prototype of ThMn,,. The Pearson symbol for this structure is t126. The metallic elements Ag, Au, Cr. Mn, MO, Nb, Pd, Pt, Ti, V and W are known to form beryllides with this stoichiometry and structure. The structure is shown in Figure 2. 2.4 M2Be17
This compound can exist in two crystal structures, a hexagonal lattice with the Pearson symbol hP38 (i.e. with 38 atoms per unit cell), and a rhombohedral structure, hR19. Two compounds, Ti2Bel7and Hf2Be17 are known to exhibit both structures, while Nb, Ta and Zr beryllides of this type appear to exist only in the hR19 form.
2.8 MBe These compounds generally have the CsCl or B2 crystal structure. The most extensively investigated compound with this structure is NiBe, which will be discussed more thoroughly later. The Be-Co and BeCu systems also have stable compounds with this composition. Another example of this structure was found in the Ti-Be system where a metastable ordered B2 phase formed on devitrification of amorphous alloys that were obtained by quenching melts of approximately the eutectic composition between the metal and the nearest stable beryllide, TiBe, (Tanner and Giessen, 1978). Two additional metastable compounds of the MBe stoichiometry were found in the Zr-Be and the Zr-Hf systems, where the ordered structures were determined to be the orthorhombic B,, or the CrB structure type (Tanner and Ray, 1979; Tanner, 1980).
2.5 MBe, This type of compound, with the CaCu, type D2, structure with a Pearson symbol hP6 has been observed in beryllium metal as an impurity phase, but has not been investigated as a single-phase intennetallic. Representatives of this compound are
2.9 Other Compounds and Phases There are several other beryllide intermetallics, none of which has been extensively studied. Such compounds include Nb,Be,, Mo,Be, BeFe, (D03, cF16 cubic), compounds in the Be-Co and Be-Fe systems with
Beryllides
41
Figure 1 Metal-beryllium,, structure. A. The overall arrangement of the 112 atoms in the unit cell. Note the FCC symmetry for Type I Be atoms. Atomic sizes are not to relative scale. B. Coordination of Type I Be atoms about the metal atom. Atoxrnc sizes are drawn to scale. C. Coordination of Type I1 Be atoms about a Type I Be atom. Because of the difliculty of visualivng this coordination, D. shows the same arrangement of atoms with the Type II Be atoms at half their actual diameter. E. Coordination of Type 11Be atoms about the metal atom. Agam it is difficult to visualize the actual arrangement, so F. shows the coordination with half of the Type II Be atoms removed. In G the metal atom has been scaled to the size o€ the hole within the polyhedron of Type I1 Be atoms. The radius ratio for the metal atom to the hole radius IS 1.1/1.9
42
Structure and Composition
Ni-Al-Be system, compositions based on the L12 phase Ni,(Al,Be) also showed the presence of Ni solid solution and the B2 phase NiBe (Matsuo et al., 1994). Good room-temperature ductility was found for the alloy Ni-20 at%Al, 10 at%Be. Much more useful work could be performed to investigate multicomponent beryllide phases and alloys based on them. 3. Properties of Beryllides 3.1 Mechanical Properties Figure 2 The MBe,, (tI26, I4/mmm) structure illustrating the body-centred tetragonal symmetry. Note the similarity to the MBe,, structure shown 111 Figure I
uncertain stoichiometry, and compounds in the Be-Pd, -Pt and -Ru systems. An unusual feature of the MBe,, and the MBe,, compounds is the extensive solubility of other elements that can form one or another of these phases, probably due to the stiffness of the beryllium sublattice cages, which can accommodate other metals. It was observed that dilute alloys of beryllium with titanium in the 1-2 weight percent range, processed by centrifugal atomization, had unexpected impurities of vanadium, zirconium and molybdenum, some of which originated with the starting materials, and some from the process (Jacobson et al., 1991). The impurities were all concentrated in the particles of TiBe12,and no other phases were found, despite the fact that the lowest metal concentration beryllide for zirconium is ZrBe,,, and for molybdenum, the lowest metal concentration phase is MoBe,,. Other observations of this phenomenon were made by Brimhall et al. (1992) who found that in sputter deposited alloys with both Nb and Zr, which form NbBe,, and ZrBe,, respectively, the compound phase was always MBe12 when the Nb concentration exceeded the Zr concentration, and was MBe,, when Zr was in the majority. It was also observed that on annealing at elevated temperatures the phases (Nb,Zr)Be,, + (Nb,Zr),Be17 or (Nb,Zr)Be,, +(Nb,Zr)2Be,, were present. In no case was the coexistence of (Nb,Zr)Be,, and (Nb,Zr)Be13observed. Samsonov (1966) refers to work on the ternary UTh-Be system in which there is a continuous phase between UBe,, and ThBe13.One can perhaps assume that this observation would hold for other pairs of elements in MBe,, and MBe13 intermetallics. Interesting work on a multi-phase system containing two phases that involved beryllium showed that in the
3.1.1 Elastic Properties
It should be noted that the elastic moduli of the beryllide intermetallics are quite high and compare very favorably with other high-temperature intermetallics (Fleischer et al., 1989; Fleischer, 1991). The low densities of these compounds, coupled with the high elastic moduli, result in truly remarkable specific moduli, generally higher than most other intermetallic compounds. While the characteristic of high specific modulus may be useful for stiffness-limited structural applications, it may be counterproductive in the context of fracture toughness, which is generally thought to be proportional to the reciprocal of the elastic modulus, and in the prospects for lowtemperature ductility, for which the possibility of a large Burgers’ vector and a high shear modulus may lead to more brittle than ductile behavior. 3.1.2 High-Temperature Strength
The refractory metal beryllides of the compositions MBe13,MBe12and M2Be,7have unique combinations of properties that include good high-temperature strength, oxidation resistancc, and low density. For example, the strength (modulus of rupture) of Ta,Bei7 at 1060°C is 420MPa, and its density is 4.99Mg/m3. At the same temperature the strength of Incone1825 is only 42 MPa, at a density of approximately 8 Mg/m3. The strength to density ratio for the beryllide is thus about 16 times that of the nickel-base alloy, at temperature. One of the high-strength nickel-base alloys, Inconel 718, has a comparable strength to density ratio, but at a lower temperature of 760°C. Unfortunately, the beryllides have poor ductility and toughness at ambient temperature, which is believed to be related to their complex crystal structures, that were described above. Thus, their potential usefulness in engineering structures is presently quite limited.
Beryllides An interesting exception is the CsCl (B2) structure beryllide, NiBe. Samples of different stoichiometry were prepared by casting, and hardness was measured as a function of composition (Nieh et al., 1989). A minimum was found at the stoichiometric composition, similar to the aluminides NiAl and CON, which have the same crystal structure. Some roomtemperature ductility was inferred from the fact that there was no cracking associated with hardness indentations. Later work on extruded NiBe (Pharr et al., 1991) showed that the strain to failure at room temperature is 1.3% in tension and 13% in compression, with failure controlled primarily by the cohesive strength of grain boundaries. At temperatures above 400°C strains of 30% in compression could be achieved before failure, and the strength levels at elevated temperatures compared favorably with those of NiAl, but they were somewhat lower than CoAl strengths (and CoAl is brittle at room temperature).
3.1.3 Hardness und Ductile-Brittle Transition Temperature Several investigations have been performed on the relationship between hardness and test temperature for many of the interesting beryllide mtermetallics. Fleischer and Zabala (1989) measured the hardness of Nb2Be17,ZrBe,,, NbBel, and TiBe,,, and at 1000 "C the hardnesses were 9.5, 6.0, 5.8 and 4.7GPa, respectively. These hot hardness levels compare favorably with many other intermetallic compounds that are presently being investigated for possible high temperature application. For example, the compound NbCr, has a hardness of about 6.5GPa at 1000°C (Fleischer and Zabala, 1990). The data of hardness vs. temperature for these beryllides did not give a clear indication of a ductile-to-brittletransition temperature (DBTT), but from an observation of the absence of cracking associated with the hardness indentations, the temperatures were estimated to be 1050, 840, 1000 and 840"C, for the four compounds in the order of decreasing hardness listed above. An investigation of the hardness of three of these same beryllides, and substituting TaBe12for TiBe,,, was also performed at different temperatures (Nieh and Wadsworth, 1990). The materials were prepared by vacuum hot pressing, and hardness samples were sectioned directly from the hot pressed materials. These investigators found that by plotting the logarithm of the hardness vs. the reciprocal of temperature, a clearer indication of a ductile to brittle transition could be determined. Table 4 below compares the DBTT results of the above
43
investigations, along with those from Bruemmer et al. (1993).
3.1.4 Toughness and Creep
These high values for DBTT are undoubtedly sufficient to discourage the application of these complex crystal structure beryllides in low teniperature structural components. The estimate of 750 "C for the DBTT of VBelz by Nieh comes from a more extensive study of this compound which measured room temperature properties, including fracture toughness, as well as elevated temperature creep behavior (Nieh et al., 1992). The fracture toughness was estimated from the microcracking associated with a hardness indentation and found to be 0.84 MPa rnl/,. Steady-state creep rates were compared for vacuum hot pressed and hot isostatically pressed material for different stress levels at 1100°C and 1150 "C. At low strain rates the stress exponent was found to be near 4, consistent with dislocation climb as the rate-controlling mechanism. This is similar to the stress exponent found for other intermetallics such as TiAl, Ti,A1, NiAl and CoAl. At higher stresses the power law appears to break down. At low strain rates the creep activation energy was found to be 270 kJ mol- l, which again is close to the activation energies observed for the other intermetallic compounds. Nieh and co-workers (Nieh et al., 1993) have also investigated the creep behaviour of the beryllide NbzBe17.Again using the microcracking associated with a hardness indentation, the room temperature fracture toughness was found to be 1.1MPam1!2. For this compound, the stress exponent for steady-state creep was found to be close to 3, a value that is associated with dislocation glide as the rate-controlling mechanism. The activation energy was determined to be approximately 575 kJ mol-', a value which is higher than that found for other beryllides. It was not possible Table 4 Ductile-brittle transition temperatures for several beryllide intermetallic compounds Compound
TiBe,, NbBe,, Nb,Be,, ZrBe,, TaBe,, VBe,,
DBTT, "C DBTT, "C DBTT, "C (Fleischer and (Nieh et al.) (Bruemmer et al., 1993) Zabala, 1989) 840 1000 1050 840
770 815 740 790 750 (est.)
700 955
690
44
Structure and Composition
to associate this activation energy with any specific diffusion process because of the lack of data for the compound. Observation of the absence of intergranular crack formation after 15% deformation in creep was also cited as evidence for dislocation glide as the rate-controlling mechanism. This brings us to the concluding part of this discussion of mechanical properties of beryllides, in which we summarize the results of several research efforts done jointly by Battelle-Pacific Northwest Laboratories, and Washington State University (Bruemmer et al., 1992; Henager et al., 1992, 1993; Sondhi et al., 1993). These studies have shed some important new light on deformation mechanisms and dislocation structure in the niobium beryllides, NbBe12 and Nb2Be17.For the former compound, dislocations were observed to be partials bounding planar faults, with slip systems that were identified as 1/2(101){121}, 1/2(101){101) and 1/2(100){011>.The partial dislocations were related to the possible phase transformation from NbBe,, to Nb,Be,, (the latter structure is very closely related to the former) and could account for enhanced dislocation mobility and the promotion of high-temperature deformation. Atomistic modeling has been performed for a[lOO] edge and screw dislocations, and an a/2[1001 screw dislocation (Sondhi et al., 1992). Other results of these investigations compare favorably with prior measurements; in particular it was found that the room-temperature fracture toughness of NbB,, was 4MPam1/,, which is higher than that measured by other investigators. In general, it was found that hot isostatic pressing (HIP) gave better properties for this compound; but intermediate temperature tests revealed that intergranular enbrittlement may be present in HIPed material. Twinning modes in vacuum hot-pressed and HIPed NbBe,, were also investigated by these workers (Charlot et al., 1991), who found that twins and extended faults have (101) habit planes and (011) twinning directions, while shorter range faults have an (001) habit plane. This extensive collection of work should be consulted by future investigators into the mechanical behavior of the MBe12 intermetallic compounds. 3,2 Oxidation and Thermal Properties
The oxidation behavior of beryllides is enhanced by the fact that beryllium forms a protective oxide, BeO, which has two allotropes of which only the alpha phase is usually observed. Doychak (1994), Brady and Doychak (1999) provide a comparison between some
thermodynamic and oxidative measurements of beryllium and other metal/oxide systems. Among the beryllides are some of the highest melting intermetallic compounds, as well as some of the lowest density refractory metallic phases. These densities range from 2 to 4Mg/m3. Despite the attractive properties of beryllides, only a handful of studies on the oxidation of these compounds has been conducted since the 1960s. In addition to the serious safety issues which must be dealt with in order to work with beryllides, there are two technical reasons which are often cited to explain the dearth of research into the oxidation of beryllium compounds. Historically, the majority of the research was conducted on complex line compounds, all of which were brittle to very high temperatures. This may have contributed to a decision not to pursue these efforts further. Most significantly, however, the few studies which were published on the oxidation of beryllides found significant problems of intermediate temperature disintegration and/or oxide volatility in the presence of water vapor. The reaction between beryllium oxide and water vapor at elevated temperatures (Stuart and Price, 1964) may be simplified as: BeW) -I-H20(g) = Be(OH),(g) This reaction has also been implicated in the accelerated oxidation of beryllides at intermediate temperatures (a 'pest' reaction, for example see Doychak, 1994), and at high temperatures (Aitken and Smith, 1962: Perkins, 1963; Dudley and Desai, 1995). A review of the effects of moisture on the oxidation of beryllides is provided by Hanrahan (1999). Based on both the early data and recent work on the compounds TiBe,,, CrBe2, and alloys of NiAl and Be, he concludes that the effects of water vapor on compounds that form continuous Be0 scales is only significant at temperatures above 1100"C. The rates observed are comparable to the Si0,-H,O reaction (Opila, 1999), hence in stagnant environments and/or low partial pressures of moisture, beryllides might be successfully employed at considerably higher temperatures. Nonetheless, this may be considered the upperuse temperature for beryllides exposed in moist environments. Why should we be interested in beryllides today? The simplest answer is that for some applications, there are no other materials with the properties of beryllium and the beryllides. The problems that have been observed in studies of the oxidation of beryllides were based on a very limited amount of work, which did not eliminate possible improvements through
45
Beryllides
alloying additions. In fact, some early reports show that minor alloymg additions could result in dramatic improvements (Paine et al., 1963; Perkins, 1963) although this information has somehow been overlooked in recent years. Finally, the beryllides deserve further investigation in order to understand the properties of beryllium modified intermetallics. Such materials may very well represent the last and best available options for improving both the mechanical and oxidative properties of this class of compounds. Hanrahan et al. (1998,1999) have demonstrated that alloys based on NiAl with beryllium substituted for aluminum will form a previously unreported protective oxide, chrysoberyl BeOAI,O,. This phase is orthorhombic, with an oxygen sublattice that is very similar to that of alpha alumina. Unlike alumina, chrysoberyl is only observed in the highest density allotrope, therefore it is not surprising that the alloys that form continuous chrysoberyl scales actually demonstrate significantly better oxidation resistance than even binary NiAl at temperatures up to at least 1100°C. The addition of Be to NiAl also improves the creep resistance and hardness of NiAl, although it does appear to increase the DBTT (Levit, Hanrahan and Noebe, unpublished work).
3.3 Magnetic and other Physical Properties Beryllide intermetallics have many unusual properties, and this includes electrical and magnetic properties. For example, the compound TiBe, was claimed by Matthias et al. (1978) to be an itinerant antiferromagnet. The ‘itinerant’ qualifier was based on the fact that the elements that make up the compound display no local magnetic moments in their metallic alloys and compounds. However, the cubic (cF24) Laves phase, TiBe,, does have an inverse magnetic susceptibility that increases with increasing temperature, shows metallic behavior in its increase of electrical resistivity with increasing temperature, but was later recognized not to be an antiferrornagnet.The compound has been studied with ternary alloying additions such as copper (Giorgi et al., 1979) and gallium (Giorgi and Stewart, 1982). Modeling of the structure predicts further anomalous behavior of TiBe, in high magnetic fields (Yamada and Terao, 1998). The misinterpretation of TiBe, as an antiferromagnet occurred because it was one of the first of many compounds in which at very low temperature, the electrons exhibit highly correlated behavior; that is, no simple model can predict the behavior.
The most extreme example of a highly correlated electron compound also contains Be. The compound UBe,, has been found to be an unconventional superconducting material (Ott et al., 1983). Earlier work on the magnetic properties of UBe,, had assumed that because of the large U-U distance in this compound of 5.13 A, the uranium ions had a tetravalent configuration with two well-localized Sfelectrons. The specific heat of the compound was observed to increase with decreasing temperature from above l S K , and strong diamagnetic signals were observed below 1K. However, the possibility of bulk superconductivity was discarded, and the results were ascribed to precipitated uranium filaments in the bulk samples. It was subsequently found by Ott et al. (1983) that UBe,, was indeed a genuine superconductor, albeit a very unusual one. Single crystals were grown by slowly cooling U and Be in a flux of aluminum. The temperature dependence of the resistivity of the single crystal sample is shown in Figure 3, also from the work of Ott et al. The magnetic susceptibility measurements were consistent with an electronic system that could be described as a Fermi liquid. (A Fermi liquid is just the next step after the simple free electron gas, which shows that interactions between electrons are making the properties more complicated in the same way that a liquid is more complex than a gas.) And that the presence off electrons is essential for the occurrence of superconductivity in UBe,,. But it was the fact that the heat capacity increased to a value three orders of magnitude higher than all but one other superconductor (CeCu,Si,) that led to intense study of
*(.
0.5
0
0
I
0
50
I
1.5
I
2
26
I
......**.,.
3
I
100 150 200 Temperature [Kj
I
250
300
Figure 3 The temperature dependence of the resistivity of single crystal UBe,,, from the work of Ott et al. (1983)
Structure and Composition
46
UBe,, around the world. Study of other compounds with similar properties suggested that the Fermi liquid had effective particles called quasiparticles, which replaced the electrons; such particles had extremely large effective mass, and hence these solids were called heavy Fermion materials. Soon after the Ott letter was published, additional investigation was made of other beryllides of actinide elements - NpBe,,, Npo,,,Uo 32Be13 and PuBe,,, and the same kind of heavy Fermion behavior as was observed for UBe13was noted for all these systems (Stewart et al., 1984). However, as an additional f electron is added to UBe,, by substituting Np for U, the superconductivity is suppressed, while the heavy Fermion behavior remains, and some form of itinerant-electron magnetism appears. The additional f electron for Pu leads to a Kondo-resonance type of behavior. These phenomena were summarized by Fisk and his colleagues in a later publication (Fisk et al., 1988) in which they described more of the unique features of the low-temperature normal state properties of heavy electron systems. Other related work on MBe,, compounds includes the magnetic properties and specific heat measurements of GdBe,, made by Besnus et al. (1996). The role of the Be in TiBe, and UBe,, is probably best thought of as an inert metallic spacer. Because Be has only Is, 2s and sometimes 2p electrons, these electrons cannot hybridize with d or f electrons. So when the 3d electrons of Ti or the Sfelectrons of U are spread out by the Be atoms, whose electrons cannot hybridize with those of the heavier metal, it is similar to making an insulator out of these atoms with a partially filled electron shell that is likely to possess a magnetic moment. But the Be still conducts electricity. So in a few compounds, this competition between spreading atoms out to make them magnetic but forcing them to stay electrically conducting produces properties that fascinate physicists. Only Be can perform this trick in a binary compound. It takes a ternary to do it without Be, as in CeCu,Si,. Beryllium here is so important because it has so few electrons that they cannot hybridize with d or f electrons.
growth rate of the intermetallic compound YbBe,,, f o m n g between molten Yb and solid Be, is governed almost exclusively by the diffusion of beryllium through the compound toward the liquid, and the diffusion of vacancies from the liquid toward the compoundberyllium interface. Conversely, for the lower atomic number rare-earths such as Ce or Pr the growth of the intermetallic is principally via diffusion of the lanthanide. Intermediate lanthanides such as SniBeI3grow by diffusion of both species. In all cases, lattice diffusion of both species occurs by vacancy motion on the Be sublattices. Therefore, the concentration of vacancies within a particular compound controls the mobility and hence the growth rate of the intermetallic compound. The mobile species on the other hand, is dictated by the valence structure of the particular lanthanide, and its ability to occupy (albeit temporarily) beryllium sites. Another work that deals with diffusion and beryllide intermetallics is that of Brimhall and Bruemmer (1992). They investigated the reactions between the beryllide NbBeI2and refractory metals MO,Ta, and W. Also the ceramics Sic, MoSi, and AI@, were studied. Diffusion couples were prepared between NbBe,, and the other materials, and in no case was the diffusion of other metallic elements into the beryllide observed. Rather, it was primarily the diffusion of Be from the intermetallic into the other materials that governed the interfacial reaction. with some evidence that the compound Nb2Be17 was formed due to the depletion of beryllium in the intermetallic phase. It was observed that Si from the silicon-bearing ceramics diffused into the beryllide. The extent of the reactions that were observed was considerable, and since this study was done with the possibility in mind that the beryllide could ultimately be combined with the other materials in a composite; it was concluded that some form of reaction barrier would have to be used in order to limit the degradation of the interface. It seems clear that the high mobility of beryllium at elevated temperatures is a factor that contributes to the reactivity of beryllium and high melting point beryllides with other refractory materials and molten metals. (Conversely, this also explains why relatively small alloying additions of beryllium can result in the growth of continuous beryllium oxide scales in high temperature exposure.)
3.4 D@usion
There is work at Los Alamos on the reaction rates between liquid rare-earth metals and beryllium (Kanrahan et al., 1997). The differences in diffusion behavior between the higher atomic number rare-earths (Er, Yb) and the lower members of this grouping (Ce,Nd, Sm) have been clearly demonstrated. For example, the
4. Methods of Synthesis
4.1 Powder Reactions
Most of the work on the high beryllium intermetallics (M,Be,, and up) has been performed on materials
47
Beryllides
synthesized by mixing and reacting elemental powders. This form of synthesis has the advantage of forming fine-grained, robust samples for study. However, because the reaction is usually done by grinding up the initial product and reacting and grinding it one or more additional times, there is usually a fairly significant amount of oxide included in the compound sample. One recent example of such a procedure is contained in the work of Nieh et al. (1993). The measurements are nearly always complicated by the presence of this oxide, and, to some extent, by other beryllide phases that can form when the intended stoichiometry is not maintained. 4.2 Liquid and Liquid-Solid Reactions Because of their low-temperaturebrittleness, it has not been very straightforward to prepare beryllides by melting a stoichiometric mixture of the constituents. However, it has been demonstrated that large crystals of some beryllides can be grown from a molten aluminum solution, as described below. Studies of the reaction between liquid metals and solid beryllium have demonstrated that the beryllide intermetallic phase can form, but so far this method has not been used for the preparation of specimens for further measurements of mechanical or electrical properties. The compound TiBe, can be prepared easily by arc melting. Like most C15 structures, it is quite brittle, and likely to shatter as a result of thermal shock. So making very small samples or flipping large samples and remelting while still hot are necessary. Apparently the vapor pressure of the Be is well suppressed by oxide on the surface. If there is excessive oxygen in the beginning constituents, the black powder that comes off is not Be0 but TiO, which is critical if one is trying to compensate for weight loss. If arc-melted TiBe, is melted and slow cooled in a Be0 crucible in He or Ar gas, the resulting huge grains can be cut out as unstrained single crystals. However, significant Be losses do occur on arc melting of the higher beryllides such as MBe,3. This might be explained by the lack of a protective oxide, however the results of Paine et al. (1963) show that these compounds, while reacting somewhat more rapidly than some of the lower beryllides, do form continuous Be0 scales and exhibit somewhat protective oxidation. The loss of Be upon melting of the MBel, compounds is therefore interpreted as demonstrating that the activity of Be in these compounds is considerably higher than in the lower beryllides, and may indeed be higher than that of pure Be (Smith and Hanrahan, unpublished observations).
The MBe13compounds are easily prepared by using molten aluminurn as a solvent. Stoichiometric quantities of U and Be can be dissolved in aluminum in an outgassed Be0 crucible heated to 1200°C and slow cooled to the melting point of AI. Then the A1 is dissolved in a concentrated NaOH solution, and faceted single crystals remain. The ratio of UBe,, to A1 can be as low as 1:lO. However, the quality of the crystals can be improved by adjusting the atomic ratio of U:Be:Al. A slight excess of Be helps them not to stick to the crucible. High ratios of A1 help to avoid the modest substitution of A1 on the Be sub-lattice, which is easily seen by measuring the lattice parameter. Slower cooling yields larger crystals with fewer regions of pure A1 trapped inside the large voids that form with fast cooling. So a 200-400 hour cooling of U:Be:Al in a ratio such as 1:14:200 is close to optimal. 4.3 Thermal Spraying and Sputtering
Plasma spraying of beryllium has become a wellestablished technology, and is a principal candidatefor the fabrication of first-wall components of a fusion reactor (Castro et al., 1996). There should, in principle, be no significant technological barrier to the plasma spraying of beryllide intermetallics. One reference was found for work on flame spraying of zirconium beryllide as a protective coating for Zircaloy-clad reactor fuel rods. The coating was said to improve the steam-oxidation resistance and the high temperature oxidation resistance of the rods (Unknown, 1988). The use of sputtering for the preparation of thin coatings of beryllides has been reported by Brimhall et al. (1992a). The as-sputtered coatings were initially amorphous, and the beryllide phases were formed upon elevated temperature heat treatment. 4.4 Mechanical Alloying Only one reference to mechanical alloying as a method of preparing beryllide intermetallics was found (Chou et al., 1992). The authors investigated compositions corresponding to NbBe,, and NbzBe17,starting with the appropriate ratios of elemental powders. After 72 hours of ball milling in a tungsten carbide vial, using tungsten carbide balls, the resulting powders were amorphous. The desired beryllide phase, along with minor amounts of other phases, was obtained after annealing the amorphous mixture for 4 hours at 1000 "C. In both cases, some contamination from the tungsten carbide was noted.
48
Structure and Composition 5. Opportunities for Application
5.1 Refractory Structures
The high melting temperatures and thermodynamic stability of many of the beryllide intermetallic compounds has encouraged their study as hightemperature structural materials. Unfortunately, the brittleness and low toughness of the high beryllides has not given a great deal of encouragement to the possibility of developing useful engineering materials for application even at high temperatures. However, the compound NiBe has revealed potentially useful ductility and toughness, and it is likely that useful structural materials could be developed from this compound or its alloys with NiAI. As has been discussed above, the high mobility of beryllium at elevated temperatures also suggests that there will be great difficulty associated with the use of beryllides in refractory composites; developing useful coatings containing beryllide phases may be possible.
5.2 Hydrogen Storage It has been found that three intermetallic compounds of beryllium form definite hydride phases. These compounds are TiBe,, ZrBe, and HfBe, (Maeland and Libowitz, 1983). The compound TiBe,, when heated in H2 to 375 "C at atmospheric pressure forms a limited solid solution of composition TiBezHo,os, which involves a small expansion of the cubic TiBe2 lattice. When an applied pressure of 1.5 x 104kPawas used, a hydride of approximate composition TiBe,H, was formed. However, when the pressure was lowered to atmospheric pressure, all of the hydrogen was released, indicating that the hydride was unstable, with a dissociation pressure between IOOkPa and 1.5 x 104kPa. The compound ZrBe, reacted spontaneously with hydrogen at atmospheric pressure and room temperature, to form the phase ZrBe,H, S. The dissociation pressure at room temperature was too low to measure accurately. The hydrogen content of the hydride appears to increase with increasing pressure: at 13kPa the composition was found to be ZrBe,H,,. The compound HfBe, behaves similarly to ZrBe,, but the hydrogen content is less, at a composition HfBqH,.,. Other beryllium compounds including VBe2 and ZrBes were investigated, but not found to have significant hydrogen absorption. The same was true of HfBes and TiBe,. The conclusion was that although the hydrogen storage efficiencies of the three Group IV di-beryllides were not as great as
BeH2, they still compare favorably with some of the hydrides of other intermetallics that are being investigated as hydrogen storage media, particularly on a weight percentage basis (Schlapbach, 1994). The hope was expressed that additional investigation would uncover practical ways to lower the dissociation pressure of TiBe2H, or raise that of ZrBe2H,,5so that the dissociation pressure of the hydride would fall in a range suitable for practical use. The possibility of developing a useful hydrogen storage material using a beryllide has not been investigated recently. Study of the structure of the ZrBe,H,,, phase showed that deuterium occupied tetrahedral sites in a partially ordered configuration at T = 12K, and continuously transformed to a disordered structure at T = 298 K, in which it occupied trigonal bi-pyramidal sites (Hauback et al., 1995). Hydrogen diffusion in this phase has also been investigated, with the interesting finding that hydrogen diffusion is restricted to two dimensions within individual powder particles (Kimmerle et al., 1998). Practical hydrogen storage using beryllide intermetailics could become useful. 5.3 Oxidation Resistant Coatings
Beryllide compounds may form protective oxides based on Be0 or chrysoberyl. These may be applied to service conditions up to at least 1000°C in moist exposure and possibly as high as 1500°C under dry conditions. The resistance of beryllia- or chrysoberylforming alloys to more complex environments (such as combustion environments) has not been investigated to date. The simple beryllium oxides are among the very few systems (the others being alumina, silica, and chromia) which can be grown as protective layers on metals, hence it is inevitable that some of the beryllides will be used at least as oxidation resistant coatings or bond coats on materials for high-temperature oxidation resistance, if not as structural materials. 5.4 Other Applications
The primary importance of the actinide-beryllium compounds so far has been their impact on condensed-matter physics (where they have been very difficult to understand). Now, many years after their special properties were discovered at Los Alamos, they are briefly discussed in introductory textbooks but not explained, because they are still not understood. There are some possible practical applications. As simple refrigerators are needed that can go below l K , the
Beryllides record-setting large heat capacity of UBe,, would be useful in the regenerator of a closed-cycle refrigerator. Similarly, the enormous thermopower of UBe,, can be used for thermoelectric cooling, for example in Peltier refrigerators that work close to 1K. Such low temperatures are not needed in many applications, and the thought of using U and Be in commercial products puts fear into design engineers. However, in time computers will need these temperatures in very tiny volumes, and U and Be will be used.
5.5 Ternary and Higher Order Beryllide Intermetallics It has often been observed that, in order to achieve significant improvements in the properties of intermetallics, particularly with regard to ductility, one of the most attractive approaches is to alloy with elements that may substitute for one or both of the elements in a binary compound. The expectation is that by introducing compositional defects the compound may more easily accommodate antisite defects and vacancies, possibly activating additional slip systems. This approach has only been investigated in a few beryllide systems, despite the fact that there are numerous possible systems in which ternary alloying may be tested. Many of the binary beryllides exhibit mutual solubility (e.g. the MBe,, compounds), however, in these systems the substitutions tend to be principally on the M sublattice so it is unlikely that any significant changes in physical properties will be achieved. There are, however, a large number of ternary beryllides which have been identified which may represent substitution on more than one sublattice. In addition to the B2-B2 NiAl-NiBe system mentioned above, particularly noteworthy are compounds based on the C14 and C15 Laves phases. One of the first of these identified was Be4AlFe, which has been observed as a precipitate in commercial beryllium (Rooksby, 1962) and in welded beryllium (Cotton and Field, 1997). This is a C15 phase, similar to the unusual C15 Be,Fe. The broad solubility range and stoichiometry (Okamoto and Tanner, 1987) of the latter compound (and the adjacent C14 Be,Fe) suggests that the structure accommodates both antisite defects and constitutional vacancies. Assuming this holds true for the higher order C15 beryllides, there may be a route to test the possibility of the ductilizing mechanism outlined above. Aldinger and Petzow (1979) list some 14 C14 and 3 C15 ternary beryllides (and 74 compounds in all) identified as of 1977. As of the present date, however,
49
the potential for these compounds as structural materials has remained largely unexplored. Beryllides have also been investigated as reinforcing second phases in other intermetallics. Tien et al. studied Fe-40% AI reinforced with TiBe12, ZrBe,,, TazBe17,and Nb,Be,,. They noted remarkably good compatibility between the matrix and reinforcing phases which they attributed to the formation of a reaction zone which acted as a ‘diffusional lock’ minimizing further reaction. Taksugi et al. (1986) investigated Ni,A1 modified with Be. They reported improved ductility and high-temperature strength, attributed to both solid-solution strengthening and (at higher Be levels) second-phase precipitates. Mishima et d. (1994) investigated even higher Be content alloys in the same system, and produced a series of NiBe-Ni,Al alloys. Finally, Hanrahan et al. (2000) have investigated the NiAI-NiBe system. In the middle of the NiAl-NiBe quasi binary system remarkably fine lamellar microstructures were developed which may indicate a route to improved mechanical properties. These few studies serve to illustrate the broad range of higher order beryllide systems which remain to be investigated.
6. General Comment As a family of materials, the beryllide intermetallic compounds have many very interesting properties. Unfortunately, the toxicity of beryllium for some percentage of the human population has made the safety of working with these materials a primary concern, and has discouraged investigators from working with these compounds. Thus, the many potential contributions that beryllides could make toward improved performance, particularly of hightemperature structural materials, has not been realized. We hope that the understanding of the factors that govern the toxic nature of beryllium will improve, even to the extent that chronic beryllium disease can eventually be eliminated. Then the beryllides will be able to take their long overdue place among the materials making significant contributions to the performance of high technology systems.
7. Acknowledgments The authors wish to acknowledge the support of the U.S. Department of Energy under Contract W-7405ENG-36 with the University of California. LAJ also
Structure and Composition
50
wishes to acknowledge Dr David Fanning for his help with the IDL computer program by which the views contained in Figure 1 were generated.
8. References Aitken, E. A. and Smith, J. F. (1962). Oxldation. J. Nucl. Mats., 6, 619. Aldinger, F. and Petzow, G. (1979) Constitution of Beryllium and its Alloys. Beryllium Science and Tecknology (eds D. Webster and G. J.). London, Plenum Press, 1, 235-305. Aronson, M. C. and Coles, B. R. (1994). Heavy fermion compounds. Intermetallic Compounds: Principles and Practice (eds J. H. Westbrook and R. L. Fleischer). New York, Wiley, 1, Principles, 21 1-223. Besnus, M. J., Fraga, G. L. F. and Schmitt, D. (1996). Magnetic properties and specific heat of GdBe,,. J. Alloys and Compd~.,235, 59-61. Booker, J., Paine, R. M. and Stonehouse, A. J. (1962). Investigation of Intermetallic Compoundy for Very High Temperature Applications, Technical Report No. WADDTR-60-889, part JI. Brady, M. P. and Doychak, J. (1999). High Temperature Oxidation and Corrosion of Intermetallics. Weinheim, Germany, Wiley-VCH. Bnmhall, J. L. and Bruemmer, S. M. (1992). Compatibility of hgh temperature materials with Be,,Nb. Scripta Met. et Mater, 27, 1747-1752. Brimhall, J. L., Charlot, L. A. and Bruemmer, S. M. (1992). Phase stability in Be-Nb and Be-Nb-Zr intermetallics. Mater. Sci. and Eng.. A152(1-2), 76-80. Bruemmer, S. A., Charlot, L. A., Brimhall, J. L., Henager, C. H. and Hirth, J. P. (1992). Dislocation structures in Be,,Nb after high-temperature deformation. Philosophical Magazine. A, 65(5), 1083-1094. Bruemmer, S. M., Arey, B. W., Brimhall J. L. and Hirth, J. P. (1993). Hot-hardness comparisons among isostructural Be,,X intermetallic compounds. J. Mater. Res., 8(7), 1.550-1 557. Castro, R. G., Stanek, P. W., Elliott, K. E., Youchson, D. L., Watson, R. D. and Walsh, D. S. (1996). The structure, properties and performance of plasma-sprayed beryllium for fusion applications. Phydca Scr&ta, T64,77-83. Charlot, L. A., Brimhall, J. L., Thomas, L. E. and Bruemmer, S. M. (1991). Twinning relationshp in Be,,Nb, Scripta Metall. et Mater., 25, 99-103. Chou, T. C., Nieh. T. G. and Wadsworth, J. (1992). Structural evolution in niobium beryllides dunng mechanical alloying. Scripta Metall. et Mater., 27, 881886. Cotton, J. D. and Field, R. D. (1997). Microstructural features of cracking in autogenous beryllium weldments. Metall. and Mater. Trans. A , 28A (March 1997), 673480. Doychak, J. (1994). Oxidation of Intermetallic Compounds. New York, Wiley.
Dudley, R. D. and Desai, P. D. (1995). Properties of Intermetallic Alloys III. Beryllides and Miscellaneous Intermetallic Alloys, West Lafayette, Indiana, Metals Information Analysis Center. Fisk, Z., Hess, D. W., Pethick, C. J., Pines, D., Smith, J. L., Thompson. J. D. and Willis, J. 0. (1988). Heavy electron metals: new highly correlated states of matter. Science, 239, 33-42. Fleischer, R. L. (1991) Alloys for Future Structural High Temperature Intermetallic Compounds. International Symposium on Intermetallic Compounds, Sendai, Japan, Japan Institute of Metals. Fleischer, R. L., Gilmore, R. S. and Zabala, R. J. (1989). Elastic moduli of polycrystalline high-temperature binary intermetallic compounds. Acta Metall., 37(10), 2801-2803. Fleischer, R. L. and Zabala, R. J. (1989). Mechanical properties of high-temperature beryllium intermetallic compounds. Metall. Trans. A , 2QA(7), 1279-1282. Fleischer, R. L. and Zabala, R. J. (1990). Mechanical properties of diverse binary high-temperature intermetallic compounds. Metall. Trans. A, 21A, 27092715. Giorgi, A. L , Matthias. B. T., Stewart, G. R., Acker F. and Smith, J. L. (1979). Itinerant ferromagnetism in the C-15 Laves phase: TiBe,-,Cu,. Solid State Comm., 32(6), 455458. Giorgi, A. L. and Stewart, G. R. (1982). Expansion of the Laves phase TiBe, by addition of gallium with anomalous results. Solid State Comm., 44(1 I), 1465-1469. Hanrahan, R. J., Jr. (1999). Molsture Effects on tke Oxidation of Beryllides, High Temperature Corrosion and Materials Chemistry: Per Kofstad Memonal Symposium, P. Y. Hou et al., eds. Honolulu, HI, The Electrochemical Society. Hanrahan, R. J., Jr., Butt, D. P., Chen, K. C., Taylor, T. N., Maggiore, C. J. and Thoma, D. J. (1999). High Temperature Oxidation of Ni,(AI,Be),. Elevated Temperature Coatings: Science and Technology 111, TMS, Warrendale, PA, p. 305. Hanrahan, R. J., Jr., Chen. K. C. and Brady, M. P. (1998). The Effects of Berylkm Additions on the Oxidation of Nickel Aluminide and Titanium Aluminide Based Intermetallics, The Electrochemcal Society, Pennington, NJ. Hanrahan, R. J., Jr., Zocco, T. G., Thonia. D. J., Jacobson, L. A., Lowery, J. L. and Pereyra, R. (1997). Surrogate Studies of the Pu-Be Reaction II, 21st DOE Compatibility, Aging and Stockpile Stewardship Conference, Albuquerque, NM. Hauback, B. C., Fjellvag, H. and Maeland, A. J. (1995). Temperature-induced structural changes in Be,ZrD, studied by powder neutron diffraction. J. Alloys and Compds., 224(2), 241-243. Henager, C. H., Bruemmer, S. M. and Hidh, J. P. (1993). Strength and toughness of beryllium niobium intermetallic compounds. Mater. Sci. und Eng. A, 17Q(1-2), 185-197.
Beryllides Henager, C. H., Jacobson, R. E. and Bruemmer, S. M. (1992). Elevated temperature mechanical properties of Be,,Nb. Mater. Sci. and Eng. A , 153(1-2), 416-421. Jacobson, L. A., Stanek, P W.. Castro, R. G., Elliott. K. E. and Martin, P. (1991). Unpublished work. Kimmerle, F.. Majer. G . , Kaess. U,, Maeland, A. J., Conradi, M. S. and McDowell, A. F. (1998). NMR studies of hydrogen diffusion in ZrBe,H, 4' J. Alloys and Compdr.. 264(1-2), 63-70. Maeland, A. J. and Libowitz, G. G. (1983). Hydrides of berylliuni-based intermetallic compounds. J. LessCommon Met., 89, 197-200. Matsuo, T., Hosoda, H.. Miura, S. and Mishima, Y . (1994). Phase Stability, Microstructure and Mechanical Properties in the Multi-Phase Alloys Based on the LI,Ni,(Af,Be). Materials Research Society Symposium Proceedings, Boston, MA. Matthias, B. T., Giorg, A. L., Struebing, V. 0. and Smith, J. L. (1978). Itinerant antiferromagnetism of TiBe,. Phys. Left., 69A(3), 221. Mishima, Y., Miura, S., Suzuki, T., Matsuo. T. and Hosoda, H. (1994). Design of ductile Ni,Al being two phase with NiBe. Alloy Modelling and Design (eds G. M. Stocks and P. E. A. Turchi). Warrendale, PA. TMS, pp275-282. Nieh, T. G. and Wadsworth, J. (1990). Hot indentation tests of refractory metal beryllides. Scripta Metall, et Mater., 24, 1489-1494. Nieh, T. G., Wadsworth, J., Chou, T. C., Owen, D. and Choksht, A. H. (1993). Creep of a niobium beryllide, Nb2Be,,. J. Muter. Res.. 8(4), 757-763. Nieh, T. G., Wadsworth, J., Grensing, F. C. and Yang, J. M. (1992). Mechanical properhes of vanadium beryllide. VBe,,. J . Mater. SCL.,27, 2660-2664. Nieh, T. G., Wadsworth, J. and Liu, C. T. (1989). Mechanical properties of nickel beryllides. J . Muter. Res., 4(6). 13471353. Okamoto, H. and Tanner, L. (1987). Phase Diagrams of Binary Beryllium Alloys. Metals Park, OH, ASM International. Opila, E. J. (1999). Vanation of the oxidation rate of silicon carbide with water vapor pressure. J. Am. Ceram. Soc., 82(3), 625-636. Ott. H. R., Rudigier, H., Fisk, Z. and Smth, J. L. (1983). UBe,,: an unconveiitional actinide superconductor. Phys. Rev. Lett., 50(20), 1595-1598.
51
Paine, R. M., Stonehouse, A. J. and Beaver. W W. (1963). High temperature oxidation resistance of the beryllides. Corrosion. 20, 307-3 13. Perkins, F. C. (1963). Intermediate Temperature Oxidation of Beryllides, USAEC Report DRI, 2128. Pharr, G. M., Courington, S. V., Wadsworth, J. and Nieh, T. G. (1991). Deformation of an extruded nickel beryllide between room temperature and 820°C. J. Mater. Res., 6(12), 2653-2659 Rooksby, H. P. (1962). Intermetallic phases in commercial beryllium. J. Nuc. Mater., 7(2), 205-21 1. Samsonov, G. V. (1966). The chemistry of the beryllides. Recssian Chemical Reviews (Uspekhii Khrmii), 35(May), 339-361. Schlapbach, L. (1994). Intermetallic hydrides and their applications. Ch. 21. Intermetallic Compounds: Principles and Pracrice (eds J. H. Westbrook and R. L. Fleischer) New York Wiley, 2, Practice, 475-488. Sondhi, S., Hoagland, R. G. and Hirth, J. P. (1992). Atomistic modeling of dislocations in Be,,X compounds. Mater. Sci. and Eng. A, 152(1-2): 103-107. Sondhi. S., Hoagland, R. G., Hirth, J. P., Brimhall, J. L., Charlot, L. A. and Bruemmer, S. M. (1993). Deformation Mechanisms in Bel$ Compoundr. Boston, MA, USA, Materials Research Society. Slewart, G. R., Fisk. Z., Smith, J. L., Willis, J. 0. and Wire, M. S. (1984). New heavy fermion system, NpBe,,, with a comparison to UBe,, and PuBe,,. Phys. Rev B, 30(3), 1249-1252. Stuart. W 1. and Price. G. H. (1964). Reaction between Be0 and water. .I. Nucl. Mats., 14. 417-433. Takasugi, T., Imm, 0. and Masahashi, N. (1986). Improved ductility and strength of Ni,Al compound by beryllium addition. Scrzpta Metal., 20(lO), 1317-1322. Tanner, L. E. (1980). Stable and metastable phase relations in the Hf-Be alloy system. Acta Metall., 28(12), 180.5-1816. Tanner, L. E. and Giessen. B. C. (1978). Structure and formation of metastable phase M-TiBe. Metall. Trans. A, 9(1), 67-69 Tanner, L. E. and Ray, R. (1979). Metallic glass formation and properties in Zr and Ti alloyed with Be. 1. Binary ZrBe and Ti-Be systems. Acta Metall., 27( I l), 1727-1747. Tien, J. K., Vignoul, G. E. and Kopp, M. W. (1991). Materials for elevated temperature applications, Materials Sciences & Engineering A , A143, 43-49. Unknown (1988). Evaluation of Fuel Cladding Properties at High Temperatures, Electric Power Research Institute.
This Page Intentionally Left Blank
Chapter 4
Precious Metal Compoupz Ira M. Wolff Mintek, Randburg, South Africa*
1. Introduction Of all the classes of intermetallic compounds (ICs) jockeying for adoption as advanced industrial materials, those based on precious metals enjoy a surprisingly wide applications base and extent of conimercial adoption. Their success is based on circumstances of a historical, techno-economical and property-related nature. They were among the first actual examples of intermetallic compounds recognized historically, and also provided fruitful grounds for early research in the evolving science of intermetallics. Modern-day usage of the precious metal ICs continues to be governed by their unique properties on the one hand, and their relative scarcity 011 the other. This chapter confines itself to underlining some of the recent studies in which the precious metal ICs have surfaced, with emphasis on the more mainstream commercial developments in the PGMs (the platinum group metals Pt, Pd, OS,Rh, Ru and Ir) and also gold (although elements such as Ag and Re might arguably also have been included). Four classes of intermetallic compounds receive particular attention, namely: 0
0
0
The CsCI (B2) systems, of interest in hightemperature structural and coating applications, Compounds comprising the precious metals as a minor alloying addition, Precipitation-strengthened alloys, particularly the fcc/LI, compounds, which have drawn interest as analogues of the nickel-base superalloys,
*See List of Contributors for present address.
0
The coloured ICs, in which the cubic crystal structures predominate. In these, ornamental applications are often juxtaposed with their more prosaic industrial applications.
Brief consideration is given to the prevalence of precious metal ICs in other applications, and their wider exploitation is evaluated vis-a-vis the underlying market fundamentals governing availability and supply.
Reviews of our constitutional knowledge of the precious metal compounds are incomplete and somewhat dated, and have only been attempted by a few authors. Raynor (1976) has delineated the main classes of gold compounds, including the 312, 21/13 and 714 electron compounds, noting also the wide range of intermediate compounds identified in the literature. A more current description is in preparation (Rapson, in press). A very limited treatment of some of the PGM IC classes has been provided by Darling (1973) and Savitsky et al. (1978), and a compilation of published binary, ternary and quaternary equilibrium diagrams lists about 570 systems involving the PGMs (He et al., 1983, 1993). An unexpectedly large number of systems are absent from, or only partially determined in the literature, although a growing number of binary and ternary equilibrium phase studies have recently been added to the precious-metal knowledge base. This reflects emergent interest in the precious metals as they come to the fore in new applications. However, outside of a few key applications, thermodynamic
Intermetallic Compounds: Vol. 3 , Principles and Practice. Edited by J. H. Westbrook and R. L. Fleischer. @ 2002 John Wiley & Sons, Ltd.
and ~ i n e t ~know~edge c is poorly developed, and the underlyin~platform for product developnicnt consequently fa~lsshort in inaiiy respects, compared with most of the other rans sit ion metal systems. Some of the reasons for this state o f affairs are:
eiivironment in some app~ications (corrosion and oxidation), and their highly sekctive action on the environment in others (catalysis) provide most of their development impetus. It i s particularly the former that attracts attention in the field of structural intermetallics.
The precious metals are almost by definition scarcer ity spectrui~,and their restricted ~ ~ b u n d ~limits ~ i c ~ possible mass engineering e the products further above the An illustrative accouiit of historic intermetallic-based
eographic constra~~i~s. The largest proportion of the precious metals, particularly the PCMs, is sourced from just two geo-political regions (~outhernAfrica and the fornier Soviet Union), uring the ~ ~ m e d ~ apast t e years of economic isolation of the above countries, free markets not research these elements, but was actively undertaken to ence on these elements for (for example Fogg and
Global trends have nevertheless fuelled a growing a p ~ ~ i ~ ~ t base, i o n s particularly in roles where the precious metals have suites of properties not available to the designer elsewhere. Examples of such trends are: estricted automob~leemissions, widely enforced y legislation in most developed countries. More % respectively of Pt and Pd c o n s u ~ ~ t i oisn used in catalytic converters for emission control (Johnson Matthey, PZatinum t r a ~ ~ ~ t ~power t i o ngene~~tion, particularly in the land-based emissions; here the rature properties, particularly i ~ ~ r t n eto~ sattack, make the PGMs frontline materials for ~ ~ ~ t - g e ntion e r aalloys, In aircraft engines goals of lower noise levels, greater fuel e~ciencyand higher thrust-to- ass ratios, all in the wake of h i ~ ~turbine e ~ - temperatures, ~ x t e n ~ service ed intervals, and reduced ~aintenance (auto~otive component^ such as spark plugs and catalytic converters, spacecraft), Higlier quality standards, including purity (for example the ~pecialityglass industry). Other than the purely ornaiiieiital or investment attributes of the ~reciousmetals, their inertness to the
record, and are cited in applications as diverse as relig~ousicons (a gold amal~amdating to twelfth- or thirtee~th-century India), fish-hooks used by preColumbian South Americ Indians (AuCu and/or Cu,Au), imitation gold (C t), pen nibs and instruments of ~ ~ ~ , C uand ) ~metallic t , mirrors (Fe with large amounts of P the CsCl structure Several precious modern developments in the science and application of etallics. Two are worth singling out: the study of range ordering p h e n o ~ in ~ ~thea ~ u ~ and C ~ II systems, and the early observation of shape memory behaviour in the Au-Cd system. remain archetypes for ongo~ngstudies into st~uctural order. The first recorded account (c. 1905) of an alloy ~ardenableby slow cooling from elevated t e m ~ ~ r a t u r ~ s appears to be of an alloy of composition 64Au-12.5Ptl6.5Ag-7Cu, actually predating the discovery of precipitation age-hardening in ~ u r a ~ u m i nin 1911 (Chaston, 1971). During the 193Os, this came to be recognized as being the result of atoms taking up preferred locations in the lattice, often accompanied by dramatic shape deform~tion, and changes in the mechanical properties, hence order hardening. The concept of ordering has become central to many important industrial processes, such as the magnetic CO-Pt alloys, and the hardenable 18ct gold jewellery alloys (which are strengthened by a complex combination of order and age hardening, depending on composition). Pertinently, it is in the f u ~ d a m e ~ t a l studies of the Au-Cu and Pt-Cu systems where much of the unde~standingof ordered structures and i n t e ~ e tallic compound behaviour has evolved (see for example Warren (1965), Mitchell et al. (1973, Maluff and Cahn (1985) and Rapson, Chapter 26, Vol. 2).
~
Precious Metal Cornpounds Curiously, the ordering phenomenon in equiatomic Au-Cu, and the accompanying crystallographic changes, showed behaviour consistent with the phenomenological theory of martensitic transformations, which were becoming topical in the study of shape memory phenomena. Wayman and Harrison (1989) note that by 1950, martensites with twinned lamellar structures had been observed in FePt, CoPt, AuCd, fnT1, CrMn and BaTiO,. The evolutionary spread of studies following from Chang and Read's (1951) report of rubber-like behaviour (and shape recovery) in AuCd has been treated extensively by Warlimont and Delaey (1974) and Wayman and Harrison (1989). The first patent for a shape memory device was for an Au-AgCd alloy (Muldawer and Feder, 1961). Some of the other precious metal systems exhibiting crystallographic transformations of relevance to shape memory and related phenomena are summarized in Table 1. Thc cited works are offered as being representative of the great number of studies, rather than comprehensive. For a more detailed treatment, the reader is referred to Schetky (Chapter 25, Vol. 2) and Otsuka and Wayman (1998). Modern precious metal IC development has shifted in emphasis to the high-temperature structural intermetallics. Much of this work has paralleled research into high-temperature alternatives to the nickel-base superalloys, or augmented superalloys. The titanium and nickel aluminides make up the mainstream of structural IC research (for a more indepth treatment, see Chapters 1-5, Volume 2). The B2 nickel aluminide NiAl has received special emphasis owing to its excellent corrosion resistance, higher melting point and low density. In addition, the pNiAl structure is the operative phase in many aluminide diffusion coatings for oxidation resistance (National Research Council, 1996). Applications remain somewhat tentative, since the requisite niechanical properties (room-temperature toughness and hightemperature strength) have to date not been achieved by alloying or microstructural control. This has encouraged investigations into related systems that show promise of converging their superior properties with those of NiAl by inter-alloying. In this respect, the PGM-based B2 aluminides have attractive attributes. The precious metals also preside over another major attribute - notably the existence among them of an fcc crystal structure. Their substitutability for nickel makes them almost unique among the refractory transition metals, and forms the basis of attempts to provide augmented or alternative superalloys. The following sections examine progress in these fields.
55 4. The B2 Compounds
B2 aluminide structures have been reported for all six of the PGMs, although PtAl is not an equilibrium B2 phase at room temperature. RuAl and IrAl in particular have attracted a good deal of interest for high-temperature use, and common features in their binary phase diagrams have often led to parallel studies (Figure 1). The RuAl compound has recently been reviewed in the context of its potential applications (Wolff, 1997). and its main features are summarized below. The recurring interrelationship with IrAl is highlighted. 4.1 The RuAl System
Initial work by Fleischer (1992a) gave prominence to a small cluster of PGM-based intermetallics - IrNb (Llo), RuTa (Llo), RuSc (B2) and RuAl (B2)-all exhibiting credible room-temperature ductility. These compounds have enjoyed varying degrees of attention (see Chapter 11, Vol. 2), but RuAl in particular has derived development impetus from its singular combination of thermodynamic stability (Jung and Kleppa, 1992), toughness (Chapter 11, Vol. 2), corrosion resistance (McEwan and Biggs, 1996), oxidation resistance (Fleischer and McKee, 1993), electrical and thermal conductivity (Smith and Lang, 1995; Anderson and Lang, 1998) and amenability to substitutional alloying (Fleischer, 1993a,b; Wolff and Sauthoff, 1997). In addition, useful gains in oxidation resistance. high-temperature strength and cost-effectiveness can be achieved by ternary alloying. The structure-property evolution of RuAl can be traced from its early identification (Obrowski, 1963) and subsequent findings by Wopersnow and Raub (1979) that the compound exhibited unusual resistance to attizck in hot mineral acid environments, R U N was reported to remain immune to concentrated HNQ3. aqua regia, FeCl, and H F rmxtures, and this list was confirmed and subsequently expanded to include a range of other aggressive chemical media (McEwan and Biggs, 1996). In combination with useful oxjdation resistance up to about I 100"C, the corrosion resistance of RuAl has profiled it for exploitation in coatings, electrodes, and high-temperature structural applications (WoB, 1997). The early Ending (Wopersnow and Raub, 1979) that RuAl had only 'medium brittleness", especially in combination with the (Ru,AI) solid solution, was borne out by the classic 'chisel toughness' tests reported by Fleischer (Chapter 11, Vol. 2). While not
56
Structure and Composition
Table 1 Shape deformation and related transformations rn precious metal ICs Compound
System
Reference
TaRu, NbRu NbIr, NbRh
B2(~)-,Ll,@”)+orthorhombic (lp”)
TiPd
B2-+B19 (orthorhombic)
TiPd-Fe Ti,Pd,,-,Cr, AuCuAl
B2-+B19 B2-tB19 B2 +tetragonal
TiAu, TiNiAu AuTi, PdTi, PtTi AuCd AuZn Au-Cu-Zn AuCuZn, (Heusler phase)
B2-+B19[2H) B2-tB19
Ritter et al. (1964) Schmerling et al. (1970) Das and Liebeman (1975) Das et al. (1976) Fleischer et al. (1990) Fleischer et al. (1991) Fonda et d.(1998) Otsuka et ai. (1993) Golberg ei al. (19953 Otsuka and Ren (1999) Enami et al. (1986) Schwartz and Tanner (1992) Japanese Patent (1991) Jp 3013535 Wolff and Coitie (1994) Levey et ul. (in press) Wu and Wayman (1987) Donkersloot and Van Vucht (1970) Warlimont and Delaey (1974) Ridley and Pops (1970) Brook and Iles (1975) Maluta and Nagasawa (1984) Tadaki et al. (1990) Oshima et al. (1988) Muto et al. (1988) Yuasa et a2. (1994) Baranov and Barabanova (1995) Yi et ai. (1988) Semenova et al. (1995) Warlimont and Delaey (1974) Warlimont and Delaey (1974) Chen and Franzen (1990) Marezio et al. (1971) Asada et al. (1985) Oota et al. (1985) Jorda et al. (1988) Semenova and Kndryavtsev (1994) Kudryavtsev and Semenova (2000) Giessen et al. (1964)
Fe,Pt Fe,Pd FeRh + Ni,Pt,Pd,Ir RhTi TiNi-TiRu CuPd AuMn ‘0
VRu
46
B
B2
P
P-tB2-+L2, LI,-tbct
bee-bct b c c jbct-+monoclinic B2-t rhombohedtal+monoclimc B19’ B2-+fcc B2 4tetragonal tetragonal-+orthorhombic B2-t tetragonal
ZrRh, ZrIr
B2-tB19‘
TaRh
?+tetragonal
fully understood, various aspects of the unusual plastic flow behaviour of RuAl have since come to light. The ductility of RuAl lies somewhere between that of Fe& on the ductile side, and NiAl, where deformation is almost exclusively limited to (001) slip. RUM has had a greater number of slip vectors reported than its more brittle counterparts, NiAl and &AI, viz. {loo), (110) and ( I l l ) , thus satisfying the Von Mises criterion for homogeneous deformation (Fleischer et al., 1991; Lu and Pollock, 1999). It remains to establish whether these slip vectors are the result of multiple glide or the product of dislocation reactions, Sabariz and Taylor (1998) found that only ordinary dislocations of { 100) type (not superdislocations) operated in polycrystalline RuAl compressed at
4.2K and 293 K, although (1 Il)-type dislocationswere identified after compression at 1273K. If this is the case, it is difficult to understand the low-temperature ductility observed. From another perspective, atomistic modelling (see for example Lin et al., 1992) predicts:
(i) a lower antiphase boundary (APB) energy for RuAI in the (I 1I} direction (690 mJ/m2) than for NiAl (1000 mJ/m2); (ii) directional charge distribution along the (111) direction in NiAI; and (iii) substantial differences in the charge densities and hybridization states between NiAl and RuAL
1
57
Precious Metal Compound9 f
¶
Ai
RU
h
I
10
20
30
40
50
60
70
80
90
at.% Ir
Figure 1 Binary equilibrium phase diagrams of (a) the Ru-A1 system (after Boniface and Cornish, 1996) (b) the Ir-A1 system (after Hill et al., 199%)
58
Structure and Composition 2500
2000
1500
1000
500
Figure 2 High-temperature behaviour of PGM-base eutectic composites (data after WoH et al. (1997); Hill and WoE(1999))
Thus readier slip processes could by implication be achieved by lowering of the APB energy or modifying the hybridization andtor cohesive energy via ternary alloying between systems that exhibit unlike cohesive properties e.g. within the Ni-AI-Ku system. Studies have been carried out to evaluate the temperature dependcnce of flow stress, and strainrate sensitivities, of a range of RuAL-based alloys relative to those of NiAl (Eow and Pollock, 1998; Lu and Pollock, 1999). At RT and 77 K, RuAl exhibits a much lower temperature dependence than NiAI, in fact by up to an order of magnitude. Saliently, this is not affected by the presence of secondary phases. The question as to whether other ductile ruthenium alumimdes were to be found was raised by Paxton and Pettifor (1992). A trialuminide phase A1,Ru with DO, structure, was predicted to have room-temperature ductility on the basis of density-functional theory. Although its electroiiic structure was such that i t was highly improbable that it could be made cubic by alloying, AI,Ru, unlike other transition metal trialuminides, showed an ideal axial ratio. Subsequent experimentation unfortunately failed to confirm this, and a reassessment of the phase diagram found a
competing phase, AII3Ku4, to be the more stable (Manh et al., 1995).
4.2 The IrAI System
Like KuA1, the B2 IrAl compound is stable in a limited concentration range (48-52 at.%). Both compounds melt congruently, and exhibit a eutectic L-tIrAl (RuA1) + ITsJ(Et%). However, room-temperature hardnesses of over 1000 #" have been recorded for the binary IrAl, and it is considerably more brittle. Whereas RuAl is notable for its room-temperature toughness, IrAl stands out as a candidate for hightemperature strength. Separate studies have therefore focused primarily on its high-temperature capability and improvement of its room-temperature properties: a The Ir-A1 and Ir-Ru-A1 systems have been studied
by Hill et al. (1998a, 1999a). The B2 Ir,-xRuxAl pseudo-binary was found to extend in a continuous series with a very limited width. Room-temperature hardness of RuAl increases sharply if more than 40% of the Ru is replaced.
59
Precious Meifa1 Compounds
The Ir, -,Ni,Al series has been evaluated in compression up to 1400°C (Chiba et al., 1998). The effect of Ni is to improve ductility, but it decreases strength. The specific strength becomes independent of Ni content at temperatures greater than 1200 "C.Significantly, the 0.2% proof stress of IrAl is an order of magnitude greater than NiAl (0.2% PS at 800°C = 1200 MPa, and 350 MPa at 1200°C). Some lOat.% Ir increases the creep strength of NiAl by a factor of 4. Composite 1ameHar eutectic Ir + IrAl structures have been tested in compression to 1300 "C by Wolff and Sauthoff (1996b). The results are compared with those obtained for the Ru-Ni-A1 (mixed eutectic y(Ni) + (Ru) + BRuAl), and Ru f RuAl systems in Figure 2. The oxidation resistance of IrAl has been evaluated for: -1rAl (Lee and Worrel, 1989; Chou, 1990; Hill et al., 1999b) - I N + Si (Lee and Worrel, 1994) -IrAl+ B (Wolff and Hill, 2000) -1rAl + Ru (Hill et al., 2000) -1rAl + Ni (Hosoda et al., 2000a) tudies confirm exceptional high-temperature capability, superior to that of RuAl, but requiring Al-rich stoichiometries, or alloying with Si, for continuous protective scale formation at temperatures beyond about 1550 "C. Ni was reported to reduce mass-loss under isothermal oxidation up to 1400°C. Finally, an assessment of the Ir-based ICs would not be complete without noting the exceptional mechanical properties found among some of the other systems. The Llz compounds are considered in Section 6. Compounds based on the L1, structure additionally deserve mention. Studies by Fleischer et al. (1990) found a promising combination of room-temperature toughness, plasticity and strength among simple binary and ternary formulations such as IrNb, Ir~NbsoNi,,, Irz5Niz5TiS0, Ir&o,,Nb,,, and Ir2,Niz5Vs0.The latter, in particular, showed immense strength in compression, rising to 2500 MPa at room temperature.
4.3 &+fleetof Boron on B2 Compounds In further studies, Fleischer (1991, 1993c) also went on to demonstrate the effectiveness of small boron additions in enhancing the strength and ductility of RUM, This was related to a change in fracture mode from primarily intergranular to primarily transgranular, akin
to the now well-established boron ductilizing effect, featuristic of some other cubic systems, notably Ni,AI. While boron has a clearly demonstrated beneficial effect on the mechanical properties of RuAl, no apparent benefit to room-temperature compressional ductility or strength appears to be gained in the binary Ir-A1 system (Wolff and Hill, 2000). The extraordinarily high hardnesses to be found amongst the IrAl compounds may dictate that any ductilization effects on the part of boron are overridden at room temperature by the inherent resistance to slip. This is in fact the case reported for the B2 NiAl system; boron segregates to the grain boundaries, strengthening them and allowing transgranular cleavage, but the increase in hardness as a result of the boron addition counteracts any potential improvement (Chapter 38, Vol. 1). Support for this premise can be found in the study by Hill (1998), who concluded that boron additions to slightly aluminium-rich alloys in the pseudo-binary series RuxIr50~xA150 allowed reasonable toughness and ductility when the ruthenium content exceeded 20 at.%. The structural distribution of boron in the bcc systems RuAl and IrAl remains to be established. Fleischer (1993c) inferred from RuAl lattice parameter data that boron is more likely to occupy substitutional sites than interstitial sites. The existence of a B2 RUB phase suggests that boron would be likely to substitute for aluminium, but the resulting defect structure is a matter for conjecture. Phases of Is3 stoichiometry have also been reported, but exhibiting orthorhombic or hexagonal symmetries (Villars and Calvert, 1985). Boron additions of up to 0.5 at.% do not appear to influence the microstructure on the Ir-rich side of the Ry,Ir,_,Al series (Hill et al., 1998b). An unexpected finding is the debilitating effect that boron additions of O.Sat.% have on the oxldation resistance of RuAl- and IrAl-based compounds (see below) (Wolff and Hill, 2000). Reduced oxidation resistance in a boron-doped ternary (Ru, 1r)Al alloy has also been observed (Hill et al., 2000). 4.4 Ternavy Alloying
Various ternary substitutional alloying additions have been examined with a view to enhancing the strength and oxidation resistance of RuA1, or incorporating its favourable properties into allied systems. Cr, Si and Y (McKee and Fleischer, 1991), and Ir (Hill et al., 2000) have been shown to improve its oxidation resistance, although at the expense of room-temperature toughness. In the case of Ir, substitutional additions of greater than 20 at.% are required to make a significant difference to
60 30
25
w
20
@ 15 E
10
5
0 CO content, at.%
1800
30
1600
25 1400
20 ae
1200
@ 9
I.I
1000 15
800
600
10
400 5
200 0
0
0
10
20
30
40
60
70
COcontent, a t %
Figure 3 Cobalt substitution, showing the effect of a compliant two-phase necklace structure on the mechalucal properties of RuAt {a) RuAI-CoAl single-phase (b) RuAI-Co,&l, two-phase (after WoIff and Sauthoff, 1997)
61
Precious Metal Compounds
oxidation behaviour at temperatures up to 1300°C. Sc additions of 7-9 at.% improve both room-temperature and high-temperature hardness up to 1150"C (Fleischer, 1991, 1992b), and the alloy has only 'marginally unacceptable' oxidation rates at 1350"C, making it a possible contender for lower temperatures. Several studies have sought to substitute Ru in isostructural series based on (Ru, M)A1, where M has been represented by Fe, Ni and CO, and Ti has been substituted for A1 (Fleischer, 1993a. 1993b; Wolff and Sauthoff, 1997). This has been most successful in two-phase microstructures. Stoichiometries slightly rich in Ru, and Iying just beyond the single-phase RuAl field, can be formulated to give a so-called 'necklace structure' in melt-processed material. A thin film of the Ru-rich solid solution, taking the form of a 'divorced' eutectic, acts as a compliant layer, greatly assisting extrinsic crack-toughening processes. So, for example, up to 40 at.% of the Ru can be substituted in this way by CO without a significant deterioration in room-temperature mechanical properties (Figure 3). The Ru-Ni-A1 ternary, for reasons including those discussed above, has received rather more attention than the other B2 ternaries. The structure of the pseudo-binary Ru,-,Ni,Al series has not been conclusively clarified, although various arguments have been put forward to amount for a pronounced hardness maximum within the series (non-ideal solutioning behaviour) at room temperature. A contributing interest stems from the strong solid-solution strengthening response to either Ru or Ni in the respective NiAl or RuAI B2 phases. Unlike the relatively benign effect of CO or Ti on RuAl, hardnesses of close to 1000 HV can be achieved with substitutional additions of Ni to RuAl (Wolff and Sauthoff, 1996a). The interest in this is obvious, given the possibility of extrapolating this strengthening effect to higher temperatures, but the underlying mechanisms are unclear. The possible existence of a miscibility gap between the NiAl and RuAl phases has been investigated by, inter alia, Tsurikov et al. (1980), Chakravorty and West (1985), W O Eand Sauthoff (1996a), Hark et al. (1997), Hill et al. (1997), Sabariz and Taylor (1997, 1999), and Homer et al. (1998). In essence, early observations pointed to melt-processed material solidifying with a diphasic structure, being the NiAl and RuAl phases with a small lattice mismatch (cO.h%). Such mismatch strain might account for the hardening observed and, provided that a sufficiently fine mixture of these two phases could be achieved and sustained at
high temperature, coherency strains could provide a potentially useful strengthening mechanism. An alternative view put forward is that the sloping liquidus between the high-melting-point RuAl (-2050 "C) and lower-melting-point NiAl (1638 "C) produced a coring effect rather than a two-phase microstructure per se. Material processed via reactive powder processing routes (SHS or RHIP) shows no such coring, nor any evidence of a two-phase structure. The lattice parameters show a linear gradation between RuAl and N N , irrespective of processing route (Figure 4). Thus phase misfit strain does not appear to play a determinate role (Wolff et al., 1997). Curiously, the lattice parameter data in Figure 4 show a singularity corresponding to the RuNiAlz stoichiometry, raising the question of a possible ternary ordering effect, as occurs in the Nb-Ru-A1 and Ti-Ru-A1 systems. A comparison of modelled and experimental data in fact points to a Ni,,Ru2,Al,, alloy as exhibiting the maximum degree of order, i.e. approximately 80% (Harte et al., 1997). In another study by Sabariz and Taylor (1999), on material consolidated by reactive powder processing, peak splitting and broadening in XRD data was noted at the RuNiAl, composition. At the same time, hot hardness measurementsexhibited a departure from the linear variation with temperature at this point. Sabariz and Taylor provide microstructural and X-ray evidence for two co-existing isostructural B2 phases (lattice mismatch 1.3%) to account for this. Whether these observations are compatible with a ternary ordering effect requires further work for clarification. An additional consideration has been raised by the work of Munroe (1996) and Munroe and Kong (1996), who have pointed out that Ni also increases the hardness of B2-structured FeAl beyond that predicted by atomic misfit calculations, and reduces the softening that occurs on low-temperature annealing of this compound. They advance a defect hardening mechanism to account for this, speculating that the structure of the Ni site favours the formation of divacancies in the vicinity of the atom, which are effectively restricted in mobility. The existence of a hardness peak in other Ni-based pseudo-binary series, see for example Ru,Ni,-,Ti (Semenova et al., 1995), has led to the premise that this effect may be intrinsic to Ni in the B2 structure (Wolff et al., 1997).
-
4.5 Production
Given the high melting point of RuA1, its aggressive attack of refractories, and the volatility of its
62
Structure and Composition 0.300
0.298 0.296
I 0.294 ‘ L
,g 0.292 0,
0.298
0.288
a 0.286 0.284
0.282 0.280
I $
5
10
15
25 30 Ni content, at.%
20
35
40
45
Figure 4 Lattice parameter data For RuAI-NiAI alloys prepared by RWIP or melt processing, showing a singularity at RuNiAl, (after Wolff et a?., 1997)
constituents, manufactured quantities of RuAl were initially limited to the capacity of cold-crucible melting techniques. Even then, porosity and inhomogeneous microstructures rendered the product unsatisfactory. Bulk production of RuAl waited on the development of a powder metallurgical processing route, based on reactive hot isostatic pressing (RMIP) (Wolff, 1994, 1996; Ishiyama et al., 1994; Sabariz and Taylor, 1999). Exploiting the primary powder products of the PGM refineries, and the high enthalpy of formation of the PGM-aluminides, thermally efficient production routes at temperatures well below the melting point can be engineered to produce simple near-net-shape components (Figure 5). This production route follows similar industrial temperature and pressure regimes to those used in, for example, WCCO production, placing it within the realms of commercial feasibility. Other production routes for the B2 compounds have been discussed previously (Wolff, 1997). Techniques that remain no more than laboratory curiosities at this stage, but have provided interestiiig possibililies include the production of nano-phase crystallites by high-energy ball milling (Hellstern et aE., 1989), mechanical aUoying (Xu et al., 1994), and liquid solvent precipitation (Johnson and Garland, 1990).
4.6 Composite structures
Two-phase microstructures have been engineered to usefully enhance the mechanical capability of RuAIand IrAl-based materials, and their ternaries. ‘Ductilephase toughening’ is particularly effective in raising both the strength and toughness of these alloys (Wolff and Sauthoff, 1996b). Melt-processed materials around the eutectic composition exhibit classic lamellar eutectics, wth submicron interlamellar spacings. The high interface density leads to increased strength by providing obstacles to dislocation motion in the normal way, but also provides for extrinsic toughening (crack deflection). The fine scale of the lamellar phases does, however, appear to place as upper limit on the high-temperature strength; beyond approximately onehalf o f the melting point, deformation proceeds by interfacial sliding, with a concomitant loss of strength relative to coarser microstructures. Such structures take on the characteristics of superplasticity, and have been studied in, for example, the eutectic AI-PdA& system (Piatti and Bardy, 1981). The production of RLAI by reaction synthesis paved the way for modified microstructures, either by the incorporation of a dispersoid, using the intermetallic as a matrix material, or by exploiting the
Precious Metal Cornpounds
63
Figure 5 Cylinders of RuAl produced by reactlve hot isostatic pressing
intrinsic two-phase microstructure to grow a discrete in situ composite. In the latter, the unusual situation exists where the intermetallic matrix phase (RuA1) is more ductile than the solid solution phase (hcp Ru). An example is shown in Figure 6, where discrete particles of the Ru solid solution are embedded in the intennetallic matrix. The benefits to be had from the composite microstructure in terms of high-temperature strength are shown in Figure 7 for a simple binary (Wolff and SauthoR, 1999). The structure can be refined to a point where the high-temperature strength exceeds that achievable with the lamellar eutectic structure, but with a commensurate loss of lower-temperature ductility.
4.7 ~ i ~ ~ - T e m p e r aPerformance t~re
One of the more demanding service environments targeted for RuAI is that of the automotive spark-plug electrode. Requirements for longer service intervals (60000km in Europe and Japan, 100000 miles in the USA) have focused attention on the PGMs to provide the erosion resistance required (Raw, 1992). The potential of PGM-based iiitermetallic compounds as spark-plug electrode materials was recognized in an early patent (McGitl and Selman, 19781, but the almost universal brittleness of the ICs has doubtlessly prevented their adoption. Given a way to overcome the problems with mechanical robustness, the high thermodynamic stability, environmental resistance and intrinsic resistance to self-diffusion make the ZCs natural candidates. In this regard, RuAl-based materials have shown early promise (Steyn et al., 1994). Their ductility and thermal shock resistance combine with excellent electrical and thermal transport properties
Figure 6 Microstructures engmeered to show: (a) Necklace structure in RuAI-CoAI (after Wolff and SauthoE, 1997); @) lamellar eutectic structure in A1,1r,,Ru,9 (after Hill et aL, 1999a); (c} in situ eutectic composite structure in Ru-rich RuAI produced by reactive hot isostatic pressing (RHIP) (after Wolff et at., 1997)
(Lang, 1999), and environmental resistance. Endurance trials on simple binary compounds have demonstrated them to be well within the capability of meeting the less onerous 60000 km targets, with ternary and higher formulations showing potential for even greater capability.
64
Structure and Composition 1400
.............. .................................
1200
a loo0
P
g.
"
........... ......
.............
_.
......._... "
.I,
.-,,
__.
.
4F&MW(RHIP)
-a- R@ln (RHIP) -A-- RudI,, + B (RHIP)
."'
...... 4 RuAIn (RHIP)
.................................................................
800 ..............................................................
......................................................
s
2
400
.......................
..........................................................
I"
.I.......................
..
0 400
600
800
900
1100
loo0
1200
1300
Temperature, O C
Figure 7 Effect of melt-processing, Reactive Not Isostatic Pressing (RHIP) and composition on the hgh-temperature compression strength of RuAl-based alloys (data after Wolff and Sduthoff, 1999)
.......................................................................
4
.... L
2-
1.5-
1-
0.5
.... ....
-
0-
Figure 8 Erosion rates of RuAl spark plug electrodes as a function of composition (after Wolff and Hill, 2000)
Precious Metal Compounds
Spark-plug development work has also driven a number of investigations into related aspects of RuAI. One of the findings is that optimal erosion resistance is shown by two-phase compositions, around the 68 at.% Ru level (Figure 8). The governing erosion mechanism appears to be the preferential breakdown of the RuAl constituent, ostensibly by the cyclic formation and removal of an aluminium oxide, leaving a rutheniumiich residue at the surface (Wolff et al., 1997).Although elemental Ru exhibits excellent erosion characteristics; its intrinsic brittleness, as with the ICs, has tended to preclude it from application. The effective role of RuAl thus appears to be that of providing the fracture resistance to wrthstand the thermal cycling. Unexpectedly, as can be seen from the chart in Figure 8, the small amount of boron alloyed to the RuAl-based spark-plug electrodes to provide ductility during manufacture (for example, when the insert is crimped into the surrounding nickel sheath) dramatically debilitates the erosion resistance of R u ~ ~ AThis I~~. has been associated with an enhanced propensity for internal oxidation (Wolff et at., 1997). The adoption of ICs in applications like spark-plug electrodes presages requirements for large-scale manufacture, in this case wire, within narrow dimensional tolerances. Some progress has been made with metal powder injection moulding techniques. The (Ru,M)Al materials allow near-full density wire to be achieved with relatively coarse size fractions ( ~ 3 4 p r n )and sintering cycles akin to those of, for example, alumina (DuPreez and Wolff, 1999) (Figure 9). 4.8 Constitutional Studies Development work has been accompanied by a growing experimental base for the more fundamental properties. Recent attention has been given to phase
F i r e 9 Prototype spark plug with RuAl inserts. Also shown is Imm wire produced by PIM, and the associated microstructure
65
relationships in the Ru-AI, Ir-AI, Ru-Ni-Al, Ru-Cr-Al, Ru-Ir-Al and Ru-AI-Y systems (Boniface and Cornish, 1996; Hill et al., 1998a, 1999a; Horner et ai., 1998; Hohls et al., in press; Compton et al., in press). 5. Precious Metals as Minor Alloying Additions
In a number of systems the precious metals play a role in secondary phase formation, or modification of the intermetallic constituent. Examples of the following, discussed below, are noteworthy: e Additions of precious metals modify the structure, 0
0
or kinetics of secondary precipitate formation. Small additions of precious metals provide secondary phases for modifying properties, such as in precipitation strengthening. Alloys in which an intermetallic phase, as with the y' phase in nickel-base superalloys, constitutes the majority volume fraction, and in which the properties are augmented by the precious metal addition.
5.1 PGMs in Sigma Phase Formdon Small additions of PGMs (<0.2 wt.%) are made to passivatable alloys like ferritic stainless steels to effect so-called cathodic modification in reducing acid environments (see for example Higginson, 1989), as well as enhancing the high-temperature oxidation resistance (Wolff et al., 1998). The PGM addition promotes spontaneous passivation in the reducing acid, improving corrosion resistance by up to five orders of magnitude. Extraordinary gains in the hightemperature oxidation resistance can also be engineered. It has, however, been noted that the PGMs can accelerate the formation of the generally slow-forming tcp a-FeCr phase (D8,) at intermediate temperatures (Wolff et al., 1998). This observation leads to an application developed by Cortie et al. (1995) for high-Cr stainless steels, in which the BGM plays an active role in modifying the microstructure. When exposed to intermediate temperatures (<800 "C), ferritic stainlesssteels with Cr contents greater than about 20 wt.% become 'sigmatized'. Although this is generally debilitating to the corrosion and mechanical properties, a fully sigmatized alloy is readily and cost-effectively milled to a fine powder, which has use in thermal spraying, powder metallurgy, etc. Recognizing the propensity of small ruthenium and other additions to enhance the rate of sigma phase formation, fully sigmatized alloys can be made at intermediate annealing temperatures within minutes.
The s i w a phase in the final product is removed by an appropria~e~ i n t ~ r i ntreatment g above the solutioni~ig teiii~erature,with or ~ i t h o uthe ~ benefit of added s ~ g ~ ~ - p hretardants ase ( ~ o n g * ~eta al., n 1998).
oxide under ~put~ering ~ o n ~ ~and ~ ~ (iii)o ~~ is ~, A3~ d ~ r segregation to the unoxidized surface of the samples. It is worth making mention here of the improvements in ductility and fabri~abi~ity brought about by 1-3 at.% a~ditionsof Pd to Ni3A1(Vol. 2 Chapter 2). In this case, the results have been explained in ternis of a reduction in the ordering energy. Although Ir additions of up 5at.% reportedly Alloying additions to the gamma titanium alu~~iinides increase the h ~ r d e n I nrate ~ of Peal (€321, ostensibly seek to improve d~ictilityby one or more of the t y p ~ the ~ ~ ~ r e ds i s at ~ nin ~~an ~~ ~ n f o l l ~ w ~ nmge c h a n ~ s ~~so~ d i ~ sfip, e d ~ i c r o s t r ~ c t u r a ~by v a c a ~ c ~ ~defects, oxygen atmosphere at 1300°C is not i m ~ r o v e ~ re~nement,and ~recipitationof small amounts of the (Hosoda et crt,, ~ O ~ O b ) . vely ductile (bcc) @-phase (Chapter 4, Vol. 2). enium ~dditionsof up to 5 at. % have been found ve strong b e ~ ~ ~ s t a b i l ~ zrotes i n g in Ti-Al-Rti ~ ~ t e r n ~ ~ and t a l three l ~ ~ ~types , of martensite can be m the P-fieId, depending on Tke ~ x ~ e ~~ ti ~g hu ~~ ~~ el ~~ ~ i ~e c~ hr aapropern~i ~ ~~ ~e al., 1989). Alloys based on ties of the nickel-base su~era1loys rely on the 7Al- I Cr-2Nb alloys have precipitation of coherent, ordered ~ ~ ~ A l - t y(fcc pe also been found to have greatly ~ m ~ r o v ethennod L1,) particles in an fcc matrix. mechanical p r ~ c e s s i ~attribut~s g thanks to the addit~mperatureferritic alloys can be engi~eeredaround tion of 1 at.% Ru, which encourages a fine distribution coherent CsC1-type (bcc B2) particles in a bcc matrix, of equiaxed B2 particles (Chen and Wang, 1997). In One of the mare unusual examples i s to be found in addition to distributing strain more evenly, the B2 the Fe-Au-Cu system (~ahlgren,1967). Additions of phase reporte~lyprovides nuc~eat~on sites for static y 3.8at.% each of Au and Cu to iron were found to recrystallization during post heat treatments. result in ordered zones with structures (apparently ~artition~ lend ~ gsite o c c u p a ~ ~ofyRu in t ~ o - p h ~ s e AwGu in c o ~ p o s ~ t i o ~onsistent n~ with the CsCl a2+ y Ti-48at.%A1 alloys have been studied by atomstructure. This is unexpected, since none of the binary probe icroscopy ~ ~ P F(Kim f ~ and ) Smith, Au-Cu c o m ~ o ~ ~have ~ i d sthis structure. No apptications have been reported, 1997; 1999). In alloys with Rti additio~isof Precipitation-strengthened ferritic alloys based on Up u was observed to substit~teyreferentitslly for A1 in the ordered 1' phase, Ru was considered CsC1-type p ~ e c i ~ i ~in~ tFe es - ~ i - ~ l (- ~C o~ ~ ~ ete r o ~ al,, 1988) and Fe-Cr-Al-NI ( ~ a i l and ~ ~Pineau, r ~ 1982; i~g to follow IVb with regard to p a ~ t ~ t i o n behaviour, but to follow Cr with regard to s ~ ~ i f the t ~ (n ~~ * + ~ Zliu ~ /and ~ T'jong, 1997) have enjoyed more interest, but have been limited in appli~letiondue to c o a r s e n i ~of~ boundar~to the Af-rich side in these alloys. No the precipitates at teniperatures much above 700 "C. evidence of Ru segre~atioi~ to a2/y interfaces was NiAl-type p r e ~ i ~ ~ t could ~ t i o si~i1arly ~i be expected in found; nor did Ru appear to si~ificantlyaffect the Fe-Cr-A1 by alloying with Ru. partitioning of interstiti~lelements C and 0. Small additions (-0.2 wt.%) of Ru to ferritic idtoys Single-crystal studies of NiAl, to which 10at.% fr based on Fe-Cr-A1 have been shown to drama~ically was added, have d ~ ~ ~ o n s t r athe t e dbenefits for the flow improve passivity ia both aqueous and oxidation stress and creep resistance up to 1200 "C (Chiba et al., environments (WolR et al., 1998). Increasing the Ru 2000). Flow stresses about it factor of 4, 5 and 8 higher to 4 wt, % ( 2 at, %) provides an age-hardthan those of NiAl were ~ e p o r t ~ atdroom ~ e ~ p e r a t ~ r e addition , enable ~ y s t e with ~ , a fine, s e ~ i - ~ o h epre~ipitate r~~t 900 "C and 1200"C res~ect~vely. A change in the active dispersion in Fe-20Cr-5A1 and Fe-35Crslip plane was oEered to account for this pronounced ~ A1 nya and Wolff, 1999). ~ o w e v e rthe effect of Ir. ~ h i g ~ ~ t e ~ p e ~ ~47:1 ~ u rine these alloys is high, and chemical The role of Pt in e n h a ~ c i nthe found that the p~eci~itate species (in the over-aged oxidation r~sistanceof mQdi~edaluniinide coatings c o ~ d i t i ~ nis) a lower a de, p ~ o b ~ bthe 1 ~lesss Using has been ~ ~ a ~byi ~~ i e ~d ~ etd al.u (1993). symniet~ic(CS4 or C1I) phase, which displays /3-NiAl with (1101, s ~ n g ~ ~ ~st~~dies ~ ~ y ofs tPt-doped ~l little adva~~tage in i ~ i c r o s t ~ c t ustability ~al relative to (100) and ( I l l ) o r ~ e n t ~ t i o nPt~ , was noted to (i) ti sg.Al:Ru ratio in t ~it foro nits base-~etal~ o ~ ~ ~ t e ~r o~ va i~~the ~ ~ c i l i tA1,03 a t ~ ~ ~ ~ un ~(IOO),~ but~ hinder the bulk alloy closer to unity appears to successfully (1 10) and (11I), (ii) provide for a very stable and dense ~~~~~~
-
precious Metal Compounds
induce the B2 RuAl precipitate. Jackson (1991) has claimed a precipitation-strengthened FeCrAlU composition, with Ru additions of up to 20 at.%, and in which the B2 fRu,Fe)Al phase predominates. Patents clainiing the use of RuAl-containing coatings for high-temperature oxidation resistance have issued for the Ru-AI-Cr system (Jackson, 1990, 1998). These devolve from the synergy offered by Cr and A1 for oxidation resistance, and provide for Ru contents of up to 45 at.% to obtain the required microstructures.
5.4 Nickel-base ~ u p e r a l with ~ ~ sPGM Additions Second-generation superalloys, typically containing upwards of 11 different alloying additions, are approaching their asymptote in terms of microstructural stability, environmental resistance and hightemperature creep performance. The last recourse of alloy designers is to the (surprisingly unexploited among the transition metals) PGMs as bulk alloying additions. Although the so-called 'PGM concept' is more than two decades old in the patent literature (Selman and Midgley, 1975). increasing demands for greater Performance, fuel efficiency, and noise reduction have only in recent years propelled such alloys below the cost-benefit curve. (Other factors, particularly for alloys operational in land-based power plants, are the need to reduce emissions, and to cope with dirtier fuels than are typical in aerospace applications.) Overcoming the inherent resistance among alloy engineers to these costly additions has been greatly assisted by two developments: The success of Pt-modified aluminide coatings has inculcated an appreciation of the benefits to be had on the basis of total life-cycle costing. Furthermore, the interaction of these coatings with the substrate has necessitated an understanding of the effects of the Pt when diffused into the nickel-base substrate. This has laid the basis for an early database on the effects of Pt in the superalloy system. Secondly, the advent of the relatively scarce, strategic and expensive Re additions at levels up to 6 wt.% in the current superalloys created a defucto precedent for its PGM counterparts. According to the US Geological Survey (Feb. 1997), only 27200kg of Re were produced in 1996, for an average selling price of $1100/kg (or $33/oz). The USA is the largest producer (68%), but has to import to meet demand. Some 60% of Re production is used in superalloys. For comparative purposes, some 10 000 kg of Ru are taken up by the
67
market each year - the average price (1999) was about $3910~.It can therefore be seen that the PGMs are not as extraneous as might be supposed. Attention among turbine alloy engineers has focused mainly on Pt, Ir and Ru additions, and PGM-modified superalloys have been cited in numerous patents (see for example Coupland and Pearson, 1982; Hill et al., 1984; Duhl and Cetel, 1986; Walston et ul.,1995; and Putatunda et al., 1998). The key issue in achieving greater environmental stability is a combinahon of improved oxidation resistance and hot-corrosion resistance, normally only achievable with higher Cr contents (with implications for microstructural stability), or expensive MCrAlY-type coatings. However, PGM additions dramatically improve both the oxidation and hotcorrosion resistance without debilitating the mechanical properties. There are also a number of associated advantages, as enumerated by Corti et al. (1980): The fcc PGMs, particularly Pt, show a high degree of compatibility and substitutability with nickel. e While MO and Ta are more effective solid solution strengtheners up to about lOOO"C, at higher temperatures Pt shows clear solid-solution strengthening advantages over even M O and Ta (no data available for Re). e All the PGM-modified alloys show improvements in environmental stability over the standard superalloys, although their rankings differ. Cyclic oxidation resistance improves according to the ranking Ru < Rh < Pt
The reported partitioning behaviour of the PGMs between the gamma and gamma-prime phases is not consistent, and appears to be a function of composition. Guard and Westbrook (1959) provide empiricalrules of thumb to predict substitutional behaviour in Ni,AI, viz. o
6
The occurrence of a structurally related Ni3Xphase dictates Al-type substitution (and presumably Nitype substitution can be expected from elements that exhibit an isomorphous X,Al equivalent). Elements most like Ni in electronic configuration (i.e. they foim continuous solid solutions with Ni) will substitute for Ni; those Like AI for Al.
68
Structure and Composition
However, evidence has been advanced to show that the site preference can be modified by the alloying content and multicomponent influences (Ochiai et al., 1984). Of the six PGMs: All, with the exception of Ru, show complete solid solubility in Ni, with no compound formation (Ru and Ni dissolve up to 34.5 and 50 at.% of the other element respectively in their solid solutions at elevated temperatures), e Only Pt has had both a Pt3A1 phase, and a Ni,Pt (LI,) phase reported (Villars and Calvert, 1985), which may account for the inconsistency in its reported partitioning and substitutional behaviour. e
Coupland et al. (1982a) report strong partitioning of all the PCMs, with the exception of Qs, to the y'phase in simple model y/y" systems. A developmental alloy containing 4.5wt.% Pt was found by Coupland et al. (1980, 1982b) to have Pt substituting predominantly for AI. However, Ochiai et al. (1984) invoke a semi-empirical model based on the Miedema formula (incorporating electronegativity, electron density and size) to link the change in total bonding energy to predict the site preference in three-component systems. Both Pt and Pd are shown to have a preference for the Ni sites, in agreement with other investigators cited by them (see also Pd in Ni,AI, Vol. 2, Chapter 2). Similar substitutional behaviour was found for Pt and Pd in the L1, Ni3Ga, Ni,Ge and Ni,Si systems. At high Pt levels, the L1, structure is destabilized, and a ternary tetragonal (Ll,-variant) phase of the type Pt,CuAl forms at stoichiometries of Pt2NiAl (Kamm and Milligan, 1994). The partitioning behaviour of Ir similarly appears to be a function of composition. In Ni-18at.%Al-1, 3, or Sat.%Ir alloys, Ir has a preference to partition to the y phase. (Murakami et at., 2000). However, quaternary Ir-Ta-Ni-AI alloys induce Ir to partition preferentially to the Ll, {Ir,Ta) phase @U' et al., 1999). Atom-probe fieid-ion microscopy studies of the later-generation TMS-79 (Re-free) alloy reveal a slight preference of Ir atoms to partition to the matrix y-phase, but substitute for the A1 site in y' (Murakami et al., 1998), whereas Re-containing TMS-80 exhibits a slight preference for Ir to partition to the y' precipitates (Murakami et al., 2000).
The behaviour of Ru is Iess well documented. A study of the Ni,Al-Ni3Cr-Ni,Ru section of the Ni-CrAI-Ru system indicated a preference of Ru for the matrix y-phase (Chakravorty et al., 1985).
6. The Refimtory Srrpe~all~y~ The concept of supplanting nickel in the ubiquitous superalloys is not new. Analogues based on a gamma/ gamma-prime precipitation-strengthened structure have been sought amongst the higher melting-point refractory transition metals, but invariably within the constraints of much higher densities, poor roomtemperature ductilities, and very poor oxidation resistance. Furthermore, most of the refractory transition metals have open bcc structures, which, although capable of exhibiting coherent structures, exhibit inferior creep properties (Stoloff and Sims, 1987). As noted already, a number of the precious metals have fcc structures, in common with the Ni/Ni,Al system. A quest for alloy systems capable of performance at ultra-high temperatures (defined here as service capabilities at least 200 "C above those of existing nickelbase superalloys) ultimately focused attention on the PCMs, whose relative nobility and excellent environmental resistance complement their higher melting points and desirable crystal structure. A particular subset, developed in the superalloy paradigm of a coherent y/yf structure, are referred to here as 'refractory superalloys', in keeping with the terminology proposed by Yamabe et al. (1996).
6.1 Systems Based on I r Iridium has partmlar merit, based on its fcc structure, high melting point (2443 "C) and intrinsic resistance to deformation. Gamma-fa and gamma-prime L1, microstructures can be equilibrated in the highmelting-point systems Ir-Hf, Jr-Nb, Ir-Ta, Ir-Ti, Ir-V and Ir-Zr (Massalski, 1990). Of these, the fr-M, system has embodied the most impressive resistance to hightemperature deformation to date. Ir,Nb systems (Llz, Tm>24OO0C)have been variously investigated by Bruemmer et al. (1990), Wolff and Sauthoff (1996b), and Yamabe-Mitarai et al. (1997). The most noteworthy characteristic in the simple binary system is a compressive strength comparable with the strongest metallic materials identified to date at temperatures up to A proof stress of 212 MPa at 1800°C has been recorded for Ir-10 at.% Nb (this despite the phase diagram predictmg a single-phaseconditlon, suggesting that Nb also has a potent role to play as a solid-solution strengthening agent), and a compressive stress of over 300 MPa has been reported for Ir-15 at.% Nb at the same temperature. Improvements in high-temperature
69
Precious Metal Compounds
strength appear to correlate with higher degrees of positive lattice misfit (i.e. the precipitate lattice parameter is greater than that of the matrix), but this may not offer optimal creep resistance or resistance to microstructural coarsening at lower strain rates. Interestingly, peak compression strengths appear in the Ir-Ir3Nb system at roughly 50% volume fraction of precipitates (cf. 60-70% volume fraction y' in thirdgeneration nickel-base superalloys) (Yaniabe-Mitarai et al., 1997). Comparable properties have also been reported for two-phase alloys based on Ir-Hf, Ir-Ta and Ir-Zr (Yamabe-Mitarai et al., 1997) (Figure 10). Fundamental deformation studies of the L1, constituents of these systems have received comparatively limited study. In particular, there seems to be some conflicting evidence regarding the existence of a positive dependence of the flow stress on temperature.
A moderate yield strength anomaly has been reported for Ir,Nb in thin film studies (Bruemmer et aZ., 1990), and a pronounced increase in flow stress with temperature up to 800 "C has been noted in compression studies of polycrystalline Ir-25 at. %Nb (YamabeMitarai et al., 1999a). In the latter study it was also noted that, as in the case of Pt,Al, Ir,Nb shows a strong decrease in strength between - 196 "C and RT. In contrast, no anomalous hardening was observed in polycrystalline hardness studies (Gyurko and Sanchez, 1993). Recently, it has been reported that compositions around Ir,Nb also exhibit the characteristic 'boron ductilizing effect' in compression (Wolff and Hill, 2000; Gu et al., 2000). Small additions of boron improve strength and ductility, and change the fracture mode from predominantly intergranular to transgranular in nature.
1600 Ir-Nb Rh-Ta
Figure 10 High-temperature compression strength of Ir and Rh-base alloys (after Yamabe-Mitarai et al.. 1997)
70
Structure and Composition
The high-temperature properties of the EngelBrewer intermetailic, Ir,Zr (Ll,, Tm=2280"C) have been examined by Gyurko et al. (1992). (Engel-Brewer intermetallics are extremely stable compounds between high-melting-point transition metals from the second and third transition series, based on their electronic structure.) k3Zr showed no yield strength anomaly, and compared unfavourably with Ni,AI and Ir3Nb in terms of yield strength at elevated temperature (Ir,Zr had only half the resistance to deformation of Ir3Nb). However, its high ordering energy has been put f o w a d a a basis for excellent creep resistance. Furthermore. preliminary studies of a ternary L1, compound Ir, 71Nboz,Zro,os,show a high-temperature hardness plateau from 600 to 1200"C, outperforming both binary systems (Gyurko and Sanchez, 1993). The system thus shows considerable potential at ternary or higher alloying levels. Precipitation hardening and the associated defomation structwes in Ir-Nb and Ir-Zr have been studied by Yamabe-Mitarai et al. (1997; 1999b,c). Typical microstructures are shown in Figure 11, In order of increasing lattice misfit. A close correlation has been found between lattice mismatch, precipitate morphology, and ffow stress. In general, greater mismatches give rise to an increase in the compressional flow stress, but are associated with microstructural instabilities and d i s ~ o n t i n u o ~coarselung. ~ High mismatches (- 2%) induce plate-like precipitates (Figure I l e and f), while mismatches between 0.3-0.4% are associated with cuboidal precipitates (Figure Ilc and d). As a general observation, extraordinarily close parallels exist between these systems and their nickel-base counterparts in respect of their heat-treatment and deformation response, and much of what has been learnt in the past decades of superalloy alloy engineering appears to be transferable. Two major drawbacks of Ir - its high density, and its relative scarcity - act as incentives to explore at. least partial substitution. Replacement of some of the Ir by Ni in the Ir-l5at.% Nb system has been examined, with improvements in both the roomtemperature compression strength and ductility noted for Ni contents up to 20at.x (Gu et al., 1998). The yield strength at 1200"C was, moreover, comparable with that of the binary Xr-15Nb composition at these nickel levels. Significantly, the nickel appeared to promote an intergranular-to-transgraiiular transition in the fracture mode, Control of the inicrofracture and deformation processes can also be effected by varying the Nb, MO and C contents (Gu et al., 2000).
-
The above results have also led to an examination of the convergence between the Ir-base and the Ni-base superalloy systems. Studies of the IrlTa-Ni,Al and Ir,Nb-Ni,Al systems have been reported fyu et al., 1999, 2000). Strengths intermediate between those of the nickel-base and the binary Ir alloys at 1200 "C were recorded, but with considerably improved ductilities relative to the binary compositions.
6.2 Systems Based on Rk
Characterization studies of Rh-based systems have also been pursued, where the lower melting point is offset by a more attractive density and oxidation resistance (Yamabe-Mitarai et al., 1997). The assessed phase diagrams suggest that two-phase y/y' structures can be equilibrated in Rh-Am, Rh-Ce, Rh-Hf, Rh-Nb, Rh-Np, Rh-Ta, Rh-711, Rh-Ti, Rh-U, Rh-V and RhZr (Massalski, 1990). High-temperature capabilities much superior to nickel-base superalloy comparitors, but somewhat inferior to the Ir systems, were found for the Rh-15 at.% Nb, Rh-15 at.%X, Rh-15 at.% Ta and Rh-25 at.% Ta systems. Nickel has similarly been investigated as a subsDtutional addition in Rh-15 at.% Nb alloys (Gu et a[., 1999). Up to 30 at.% Ni can be accommodated, although for optimal benefits in compression strength, fracture mode and ductility, no more than 10 at.% Ni is used.
For mainly techno-economic reasons, discussed more fully at the end of this chapter, considerable effort has gone into evaluating Pt as a basis for a refractory superalloy. Platinum has a more modest melting point (1772"C), but has an edge in availability, environmental stability, and density, over Ir. Scope to develop useful (high-temperature) two-phase binary y/y' structures is more limited, being confined to the Pt-Hf, Pt-Ti, Pt-Cr, Pt-Zr and Pt-A1 systems. However, the existence of several L1,-type Pt3X phases (X = Ga, In, Mn, Sn, Zn, CO) with lower-temperature stability, provides scope for ternary or higher alloying to modify the y'-phase. There also exists the possibility of finding some coherency in other Pt3X systems, or even to stabilize an L1,-related structure of Pt3X by alloying (e.g. Pt3V (t18), Pt3Ta (mP48), and Pt,Nb (oP8). This finds concomitance in nickel-base alloys, where Ta and Nb both segregate strongly to, and stabilize, the y'
Precious Metal Compounds
71
Figure 11 Precipitate morphology in heat-treated Ir-1Sat.YoX,where X = (a) V, (b) Ti, (c) Nb, (d) Ta. (e,g) Hf, and (f.h) Zr. The lattice mismatch (L.M.) is shown in each case (after Yamabe-Mitarai e t al., 1997)
72
Structuw and Composition
O ! 0
200
4a0
1005
1200
14
Figare 12 Mechanicalproperties of some Ll, Pt,X compounds and Ir&b (after Wee et aL, 1980; Yamabe-Mitaraiet at., 1999a)
phase, despite the fact that Ni3Ta and Ni,Nb do not exhibit LI, structures). The results of limited high-temperature mechanical (mostly hot-hardness) studies of Pt3X systems are available, including the L1,sti-uctured Pt,Al, Pt,Ti, Pt,Jn, Pt3Sn, Pt,V, Pt3Cr, Pt,Ca and Pt3Mn (Wee and Suzuki, 1979; Wee et al., 1980). Some of these results are reproduced in Figure 12. From a practical consideration of melting points, the In, Sn, Ga and Mn systems are not very useful for high-temperature applications. It can be noted that the Pt,Al and Pt,Ti systems manifest the most promising hot-hardness behaviour amongst the Llz Pt compounds, sustained up to about 900"C, the limit of testing reported. In comparison with Ir,Nb, the Pt,Al single phase exhibits higher flow stresses in this temperature range (see Figure 12). A number o f Pt-Pt,X systems have been evaluated in terms of microstructural stability and oxidation resistance (Hill et al., 2001a). Three systems have enjoyed comparatively more attention, viz. Pt-Zr, PtHf, and Pt-Al. Intermetallic dispersion-strengthened Pt-Hf alloys have been proposed by Wang et al. (1996) and Fairbank et al. (1999). The latter have also
examined the phase equilibria in the Pt-Zr system, as well as Rh additions to both the Pt-Hf and Pt-Zr systems. Revised constructions of the Pt-Kf, Pt-Zr and Pt-Zr-Rh phase diagrams have been proposed, but no mechanical properties have been reported. Pt,Zr and Pt,Hf have also been examined as contact materials, based on their excellent corrosion resistance, hardness and electrical conductivity (see Pecora and Ficalora, 1977). The Pt-AI system has the advantage of a trackrecord, albeit in coating form. Platinum-modified coatings have become the protective aluminide system of choice for high-performance blades (Chapter 22, Vol. 2), representing one of the most adherent, oxidation- and hot-corrosion-resistant formulations identified to date - which provides a basis for believing that structural alloys based on Pt-AI will be similarly endowed. The Pt-AI system is made more complex by the existence of a low-temperature, tetragonal variant of the L1, Pt3Al phase (Oya et al., 1987; Bronger er aZ., 1997; and others). It nevertheless closely emulates the nickel-based systems in terms of electronic structure (Pt and Ni share the same group in the periodic table)
Precious Metal Compounds and the inherent capability of forming its own protective oxidation layer. The evidence to date shows that Pt,Al exhibits only mild manifestations of the so-called yield-strength anomaly. Generally, deformation in L12 compounds is understood to be governed by the intrinsic features of the cores of the screw dislocations. Pt3Alis classified as belonging to that class of L1, compounds where the screws are always sessile, militating against the yield anomaly, or 'Kear-Wilsdorf locking (Vitek et al., 1991). The compressive flow stress of Pt,AI single crystals has been measured in three different orientations ([OOl], [I231 and [ill]) over the temperature range from near liquid He to 1080 K (Wee et al., 1984). The study found: A sharp increase in flow stress with decreasing temperature below room temperature. This dramatic effect has received little attention in the literature, and its practical implications remain uncertain, Two distinct slip systems depending on orientation over the entire test range of temperatures, Constant to slightly increasing flow stresses above 400 K, depending on orientation. The strong low-temperature and weak high-temperatur; dependence of flow stress corroborates earlier hardness studies conducted by Wee et al. (1980). Two approaches to obtaining the coherent precipitation-strengthened microstructures have been considered. Conventional solutionizing and ageing practices, as followed in the nickel-base superalloys, are constrained by the shape of the assessed binary equilibrium phase diagram. Because the trajectory of the Pf, solvus line dictates a relatively small volumefraction of secondary phase ($), as per the lever-rule principle, ternary approaches have been examined as a means of modifying the shape of the Pt solid solution solvus line (Hill and Wolff, 1999). More usefully, composite eutectic structures can be exploited in the Pt-AI system as an alternative approach to achieving a fine phase mix (Hill et al., in press (b)). The as-cast eutectic structure following homogenization is characterized by a fine (< 1pm) discrete particle morphology that similarly shows a high resistance to coarsening. Mechanical testing shows these alloys to be among the strongest Pt-based alloys on record (see for example Figure 2), but to date results are considerably below the strengths achieved in the Ir-based systems. Given the well-known intrinsic resistance of Ir to deformation, and considering that single-phase Pt3AI exhibits higher strengths than the
73
Ir,Nb phase at intermediate temperatures (see above), the Pt solid-solution phase appears to be the critical factor governing the high-temperature resistance to plastic deformation. Further work examining solidsolution strengthening and/or microstructural refinement appears to be indicated. Developmental studies extending to the Pt-A1-X systems (X=Ru, Re, Ta, Ti, Cr) have confirmed the high resistance to oxidation which is a hallmark of the Pt-aluminide coatings (Hill et al., in press (c)), with compression properties above 1200 "C showing a clear ascendance over their nickel-base counterparts (Hill et al., 2001b). Creep behaviour remains to be evaluated. There is some evidence to suggest that the incorporation of other PGMs into the Pt-AI system has additional benefits for the oxidation behaviour of PtAI coatings, that may extend to the bulk alloys (Fisher et al., 1999). Rhodium acting in combination with platinum reportedly increases durability of the coating under oxidizing conditions. Iridium-modified coatings have been found to produce a thinner coating than the simple Pt-A1 system, but with rather less adherence.
6.4 The Occurrence of Short-Range Order (SRO) in Pt-based Alloys Perhaps one of the most intriguing and potentially useful properties to be found in the Pt-based alloys relates to the ability of single-phase Pt solid solutions with relatively small alloying additions to be strengthened by high-temperature treatments. Recent studies have linked this to short-range ordering (SRO) activity. SRO revolves around atomic order on a highly localized scale. Due to the microstructural scale of SRO, it is generally inferred from scattering experiments (XRD, neutron, electron) rather than direct microstructural observation. Although imperfectly understood, evidence for SRO has been offered in a number of systems. However, no unified microstructural description of this phenomenon, other than purely deterministic treatments (see Chapters 2 and 33, Vol. l), is available in the literature. Schonfeld (1999) recognizes four models of SRO, viz. The statistical model - in which SRO exists homogeneously in a one-phase equilibrium state, and in which the occupation of a neighbouring site fluctuates around some value, statistically greater than a random value,
74
Structure and Composition
Disperse order model - SRO is heterogeneous in a two-phase equilibrium state i.e. particles of a longrange ordered phase in a less ordered matrix (the equilibrium interface energy just compensates the elastic energy of the particles, ensuring a maximum particle size of 2 to 5 nm), e- ~ ~ c r ~ d o model m ~ ~- nwell-defined, well-ordered microdomains (a few unit cells in size) heterogeneously embedded in a random matrix, o Lattice defect model - non-homogeneous (melastable) heterogeneous variations in composition because of defects (vacancies, dislocations, etc.) introduced by plastic deformation. 0
The ageing behaviours of these systems are often similar to those of precipitation-strengthened alloys, in that the ordered domains can interfere with the passage of dislocations. With ageing, coarsening of the microdomains can occur, ultimately leading to peak hardening and subsequent overaging (StolofT and Dames, 1966). Recent observations of anomalous hardening in Pt-5 wt.% Ru and R - 5 at.% MO alloys quenched from temperatures up to 1300°C have been inferred to be indicative of changes in the structural order, although this was premised largely on the absence of any information to the contrary (Towle et al., 1998; Towle, 1999). However, in other studies, neutron-scattering experimeilts performed on Pt-10 at.% V and Pd-10 at.% V alloys quenched from 1100"C show satellite diffuse peaks at room temperature (Murakami and Tsunoda, 1999). These have been interpreted in terms of SRO concentration waves, which reflect the characteristic shape of the Fenm surfaces in Pt and Pd. Related studies, in which magnetic SRO has been inferred for Pt and Pd alloys with around 10 at.% Mn, Fe, Cr and Go, were also cited (Murakami and Tsunoda, 1999). The SRO phenomenon has also been studied in Pdbased alloys containing MO, W, Cu and Au (Kim and Flanagan, 1967; Lang and Doyle, 1996; and others). Substitutional additions of transition metals to the solid solution bring about resistivity anomalies that can be interpreted in terms of changes in the structural order above a critical temperature. In effect, SRO appears to be effective in increasing the roorn-temperature hardness of Pt and Pd alloys when they are quenched from above a critical temperature. Exploiting this mechanism to improve the room-temperature mechanical and wear properties of high-caratage jewellery alloys has been proposed (Towle, 1999; Lang et ad., 2000). Of wider interest is its possiblc effect on high-temperature mechanical properties. Tile generally limiting strength of the y solid
solution, rather than that of the y', is widely understood to be the governing parameter inhibilng greater resistance to plastic deformation in the nickel-base superalloys (Giamei et al., 1985). SRO in the Pt-based system potentially offers a mechanism whereby the Pt solid solution can be strengthened in situ at attractively high temperatures. The existence of such high-temperature SRO in nickel-chromium alloys has, in fact, been suggested by Nordheim and Grant (1953) and Akhtar and Teghtsoonian (1971), and has also been advanced as the reason that Re has such a potent effect on creep resistance in superalloys (Blavette et al., 1986). In the case of Pt, the phenomenon usefully applies to systems to which refractory alloying additions have been made, of direct interest to extending the superalloy paradigm. Experimental evidence for this mechanism in two-phase Pt-base systems is yet to be produced, but represents a compelling area for further research. 7. Coloured Inte~meta~e ~ornpoun~$
Investigations into the underlying origins of colour in metals predate even ir consideratioii of their mechanical and other properties (Petersen, 1989). Apart from gold and copper, intrinsic colour in metallic systems is to be found only among the intermetallic compounds. Systematic studies of their optical properties have provided a fairly good, if not comprehensive, basis for understanding the attributes which lead to colour, and various 'conditions' for the occurrence of colour have been proposed for different IC classes (see Chapter 12 by Steinemann et a2.). Thus, an upper limit on the valence electroii concentration has been put forward for the Zintl phases (MgAuSn, LiMgPtSn, etc.), while an electron-to-atom ratio i s stipulated for the WumeRothery phases (Eberz, 1983; Drews et al., 1986). Predominant amongst the coloured intermetallics are the cubic structures, which conform to the general requirement for a highly symmetrical crystalline structure in which the electronic band structure allows sp-d hybridization, with a covalent character, to prevail (Steinemann, 1994). A by no means exhaustive list of the some of the coloured ICs is given in Table 2. 7.1 The CaF2 or Hitorite Structiare
Colour is one of the more notable attributes of this class of intermetallics. It includes PtAl, (deep yellow), AuAI, (purple), and CoSi, (intense blue), all three of which have considerable industrial significance; PtAI2
Precious Metal Compounds
75
Table 2 Coloured compounds based on the precious metals
Compound
System
Colour
Reference
PdIn AuCuAl &AI, + Cu AuAI, AuEn, AuGa LiMgPtSn Li, ,MgPtSn LiMgPdSn LiMgIrSn Li,AuGa Li,AuIn Li,AuSn LL~AuP~ LiPd,Ge LiP4Sn Li,PdPb L1,PtSn Li,Pt,Sn, LiAuSb MgAuSn MgPdSn MgPdSb MgPtSn MgPtSb LiMgPdSn LiMgPtSn LiMgPtSb LiMgIrSn
B2 82 Cap2 CaF, CaF, CdF, Cubic L1,Bi-related
Purple-yellow Yellow-red Orange-pink-yellow Purple Blue Bluish Coppery red Dark red Red/violet Grey-blue Greenish yellow Greenish yellow Orange Violet Brown Yellow Yellow-grey Yellow Yellow Red-violet Violet Yellow Violet Dark red Dark red Violet Bright red Violet Grey-blue
Randin (1994) Levey et al. (in press) Huriy and Wedepohl (I 993) Cahn (1998) Cretu and Van der Lingen (in press) Cretu and Van der Lingen (in press) Eben (1993)
Cubic Li,Bi-related
Cubic Li,Bi-related Cubic Li,Bi-related Cubic Li,Bi-related Cubic Li,Bi-related Cubic Li,Bi-related
is a common constituent of modified aluminide diffusion coatings for high-temperature turbine blades; AuAl, is the basis of the infamous ‘purple plague’, to be avoided in soldering contacts in electronic circuit boards (arising from the interdiffusion of Au and Al) (Cahn, 1998); good lattice matching allows CoSi, to be grown epitaxially on Si (001) surfaces for semiconductor-metal-semiconductor transistors. The fcc structure of these compounds notwithstanding, all exhibit poor fracture toughness - a factor which has nevertheless not precluded their use in ornamental applications, 7.1.1 Platigerns
Caratage platinum alloys have not found wide favour in jewellery, owing to the traditional marketing ethos of platinum within the higher-purity alloy range (greater than 900 parts per 1000). Unlike gold, the alloys of platinum have no intrinsic colour. However, as is apparent from Table 2, a number of brilliantly coloured Pt iiitermetallicshave been identified, with Pt contents varying up to 77 wt.%. Platigeme is based on
Petersen (1989)
Ibid.
Ibid.
Ibzd. Ibid.
Ibid.
the fluorite PtAI, phase. which occurs as an intense yellow compound resembling gold in colour. Additions of copper allow colours ranging from yellow, through orange to pink to be obtained (XIurly and Wedepohl, 1993). A higher caratage white intermetallic compound (consistent with a hallmark of 900 parts per thousand) based on the Pt-AI-Ti system has also been developed vaylor and Biggs, 1996). The overriding attribute of the Platigem compounds i s their high hardness, which allows them to be polished to a high lustre, or even to be faceted (Figure 13). Indeed, the compounds are marketed as synthetic gemstones, and are set in conventional jewellery alloys as an inlay or gemstone (Wolff, 1995). Coloured compounds for jewellery are not the exclusive domain of platinum. The gold counterpart of Platigems, AuAl, (more probably Au6Al,,), or ‘purple glory’, has enjoyed similar applications. Meltprocessing of AuAl, allows the production of robust and durable synthetic gemstones (Figure 13), although high-pressure production techniques reportedly lead to room-temperature decomposition into constituent A1 and AuAl phases (Cahn, 1998).
76
Structure and Composition
Figure 13 Examples of the jeweller’s craft with coloured intermehllics (a) PlatigemB synthetic gemstones based on PtAI, (courtesy Min?ek/Lonrho Platinum}; @) Purple gold based on AuAl, (courtesy M. B. Cortie, Mintek); (c) Spangold@ based on the AuCuAl beta compound (courtesv Mintek/WGC). See also Figure 13 (colour plate section) between pages 870 and 871.
Alternative approaches to accommodating the intrinsic brittleness of AuAI, suggest themselves. Rapson (Chapter 26, Vol. 2) has inquired into a composite structure. The AI-AuAl, phase field would appear to offer the possibility of mixing a ductile AI solid solution with the brittle TC. The eutectic has been studied by directional solidification, but the colour characteristics and mechanical properties were not documented (Piatti and Pellegrini, 1976). Off-stoichiometric compositions to induce a compliant two-phase necklace structure proved to yield little advantage (Suss and Cretu, 1998). Another approach lies in the production of a surface coating of AuAI2. Sequentialfy layered deposits of Au and Al readily interdiffuse at room temperature to form compounds, although the phase formation sequence is problematic, not always yielding the thermodynamically most stable phase (Chapter 26, Vol. 2).
Following the discovery of shape memory in the (brittle) Au-Cd system, some effort was expended in
Precious Metal Compounds
finding gold-based shape memory alloys of sufficient ductility, caratage and colour to allow their use in jewellery. Applications such as clasps for stones have been cited (Brook, 1973; Brook and Iles, 1975; and Japanese Patent JP 30,113,535, 1991). The Spangold@ concept represents an interesting departure from conventional shape memory applications in that the crystallographic memory effect is exploited for decorative rather than thermomechdnical ends (Wolff and Cortie, 1993). The shape deformation accompanying the martensitic transformation is harnessed to give Spangold its characteristic glitter, or ‘spangle’ - heat treatment distorts a prior-polished surface of the parent phase, rendering a myriad of randomly tilted facets, or laths, that scatter incident light (Figure 13~). The convergence of the properties that make Spangold attractive for jewellery applications is somewhat serendipitous, Spangold was originally thought to be a derivation of the AuCu I (Ll,) ordered lattice, that is centred on a composition of Au-50 at.%Cu (75 wt.% Au, or 18 carat, which is the basis of the most widely adopted jewellery in western markets). The hardening associated with the ordering transformation in AuCu, so-called ‘French gold’, has long been recognized and exploited as a strengthening mechanism (Chaston and Sloboda, 1951). As in the shape memory alloys, the ordering transformation of the fcc lattice to a tetragonal AuCu I, and ultimately orthorhombic (AuCu 11) structure is accompanied by a shape deformation. Systematic additions of A1 were found to change both the temperature of transformation, and also the morphology of the resultant lath structure, from fine platelets to broad facets (Wolff and Cortie, 1994; Wolff and Pretorius, 1994). More recently, studies of alloys based on Au - 18 wt.% Cu - 6 wt.% A1 have found that the Spangold structure and phenomenology are consistent with a beta compound (e/a ratio of 3/2), placing Spangold in the family of classic shape memory alloys. Shape deformation proceeds via an ordering reaction of the hightemperature phase from B2 to L2, positions, whereafter a displacive transformation between an incompletely ordered L2, parent phase and a tetragonal product of as yet indeterminate structure occurs (Levey et al., 2000). In addition to its advantageous composition, the 18carat Spangold alloys occupy the yellow to red wavelengths in the colour spectrum, joining the class of intermetallicspossessing intrinsiccolour. Spangold is also relatively malleable (it can be hot-worked), and has excellent castability, all of which make it attractive for jewellery applications. Of equal technological interest is its potential application as a shape memory alloy.
77
A recent addition to the ‘Spangold‘ concept is the 23-carat Au-AI alloy, which exhibits a p-phase akin to the body-centred p-electron compounds (Levey et al., 1998). Light chemical etching of the surface lamellae produces colourful interference effects, that have likewise been proposed for decorative purposes. The narrow range of stability, and brittleness of the alloy unfortunately renders it difficult to work with. For some stunning examples of the jewellers art in coloured gold intermetallics, the reader is referred to Cretu and Van der Lingen,1999.
8. Other Applications
A comprehensive survey of the precious metal compounds in other fields lies beyond the scope of this chapter. However, some unusual areas for further study have emerged in the recent literature, and some examples are given in Table 3. The reader is referred to the chapters in Volume 2 on ‘Shape Memory Alloys’, and ‘Dental IMCs’, and in this volume ‘Color in IMCs’ for a more detailed description of the precious metals in this field.
9. Supply and Availability of the Precious Metals Somewhat different dynamics govern the global gold and PGM markets, and what follows is an overview of the principal features. In recent years, physical demand for gold has comfortably outstripped mining production, but shortfalls have been easily met from the estimated 34000 tons in reserves held by central banks. There is therefore little fear that gold supplies and worldwide resources could not sustain demand from significant new applications. Gold usage is heavily concentrated - by far the largest application is in jewellery (3145.2 tons, or 85% (Gold 1999, Gold Fields Mineral Services Ltd), although industrial demand is substantial. However, gold’s (albeit diminished) monetary role ensures a high degree of elasticity in supply and demand. World PGM markets are considerably less flexible, for a number of reasons, some of which have already been mentioned: 0
World PGM production is geographically highly concentrated, most supplies originating from just two centres (Southern Africa, and the former Soviet Union).
Structure and Composition
78
Table 3 Some novel developmentsin the precious metal ICs
Application
Compound
System
Description
References
Icosahedral
Quasipriodic structures (or quasicrystals) show unusual thermophysical properties.
Anlage, S. M., Fultz, B., and Krishnan, K. M. (1988). J. Mater Res., 3(3), 421. Shield, J. E., Chumbley,L. S., McCallum, R. W., and Goldman, A. I. (1993). J. Mater. Res., 8(1), 44.
Hydrogen precursors
Pd,Mn CeRu,
Ag&k L1, MgCu,
Intermediateprecursors formed by N takeup. In Pd,Mn, an L1, structure can be stabilized with vastly different mechanical properties at RT. CeRu, undergoes amorphization. The systems allow the contributions from order-disorder, and phase transformation to be studied, depending on theintal history.
Nesbit, S., Craft, A., and Foley, R. (1991). Scripta Metall. et Mater., 25, 1183. Kim, Y -G., and Lee, J.-Y. (1990). Scrlpta Metall. et Mater., 24, 2123.
Hydrogen storage
FeTi +Pt.Pd
B2 (CSCI)
The hydrogen-storage capability of FeTi is improved by Swt.O/o additions of Pd w.r.t. activation, reduction in absorption-desorption pressures and reduced hysteresis.
Mulshreshtha. S. K., Jayakumar, 0. D., and Bhatt, K. B. (1993). J. Mater. Sci.. 28, 4229.
Extnnsic toughening
IrNb, Ir,,Nb, RuTa
Ll,, oPI2 Ll,, 3 2
Twm-Iike lamellae provide a mechanism for extrinsic toughenmg, yelding good toughness and plasticity at room temperature.
Ffeischer, R. L. (1992). Platintun Metals Rev., 36(3), 138. Fleischet, R. L., Field, R. D., and Briant, C . L. (1991). Metal]. Trans. A, 22A, 129.
Transfonnaaon induced plasticity (TRIP)
B2
A twinning or martensite-like deformation mode allows unusual room-temperature ductility in this high melting-point alloy (Tm- 1800aC). Sufscient plasticity and wear resistance allows it to be evaluated as a dental alloy.
Watezstrat, R. M. (1993) Platinum Metals Rev., 37(4), 194.
Electrocatalysis Ti,RuFe
cP2-Cscl
The electrocatalytic activity for the hydrogen evolution reaction in chlorate electrolysis can be considerably enhanced by nanocrystalline TiiRuFe application in dimensionally stable anodes (DSA).
Blouin, M., Guay, D., and Schulz, R. (1998) NanoStructured Materials, 10(4), 523.
Raney Ni catalysts find an equivalent in the PGMs. Raney Ru, Jr and Ir-Ru catalysts can be prepared by leaching their aluminides in KOH solution.
Semenova, A. D., Yankovskii, Kh. I., Kropotova, N. V., and Vovchenko, G. D. (1981). Russian Journal of Physical Cizemi.wy, §§(2),271.
Catalysis
RuAI, IrAl, RuIrAl
Not specified
Permanent magnets
FePt, Col?
Tetragonal tP4
Superplasticity
AI-PdAI,
Eutectic fcc/hcp
Darling, A. S . (1973). Znf. Metallurgical Renews, 18, 91. Fully superplastic behawour up to -400% at T>723 K up to strain rates of 3 x 10-2s-'
ThermoeIectncs
Tetragonal Below 1000K, the tetragonally distorted fcc compound delivers the maximum positive thermopower for any metal (see Vol. 2, Chapter 20).
Conducting films
D8b
Thin films of Mo,Ru3 and W,Ru, exhibit high hardness, electncal conductivity, good wear resistance and corrosion resistance.
Piatti, G., and Bardy, M. (1981). Metal Science, Feb., 55. Vdernikov, M. V., Terekhov, G. I., Sinyakova, S . I., and Ivanov, 0. S. (1969) Metals, 4, 191 (in Russian) Testardi. L. K., Royer, W.A., Bacon, D. D., Storm, A. R., and Wernick, J. W. (1973). Metallurgical Transactions34, 2195.
I9
Precious Metal Compounds Palladium
I
I
lrtdium
Ruthenium
I The minor PGMs are produced as by-products (in South Africa, of platinum mining, in Russia of nickel (or, arguably, palladium) mining) - production is thus governed by primary demand in the Pt/Pd markets, and strategic and economic considerations may dictate against increased production. Diversification is poor, placing heavy reliance on a few major applications. This is particularly true of platinum. Emission legislation creates a derived demand for Pt in catalytic converters; the other major application is in the field of jewellery. o Some of the minor PGMs have finite resources. Although, in theory, the PGM deposits in
Year
140,
I
Figure 14 The PGM markets, showing annual price range (dark shaded band) and total demand (bars) (data after Johnson Matthey, Platinum 1990-1999)
concentrations such as the Bushveld Complex are inexhaustible, some occur in concentrations and at depths that make them largely unexploitable. This puts practical limitations on the amount of metal available, irrespective of demand or price. Trends in PGM markets are reflected in Figure 14. The price ranges arc shown on an annualized basis (Platinum 1990-1999, Johnson Matthey). The salient features to note are: Production has steadily risen over the past decade, with producers carefully matching supply with
80
a
e
e
Structure and Composition
demand as far as it is possible to do so. Thus the primary products platinum and palladium have enjoyed a long period of price stability. Uncharacteristic price surges in the period 19991 2000 were largely attributable to legislative export constraints in the former Soviet bloc, leading to temporary shortfalls in the market. Among the minor PGMs, stability can be jeopardized by relatively minor fluctuations in demand. A case in point is the Rh market. In the early 199Os, a refinery bottleneck resulted in a smalt shortfall. The immediate consequence was a price spike, followed by a sustained drop in the Rh price as the effects of substitution took hold. Short-term profits bad to pay for the irrecoverable loss of applications and demand in the longer term. Market volatility is therefore carefully guarded against to ensure the longer-term viability of the market. Current and projected supplies of Ir are limited. Demand arising from significant new applications in recent years have on occasion seen the Ir price overtaking that of Pt. (This may in turn result in the loss of applications ID which its low price made it attractive initially.) Ru is acknowledged to exist in a state of slight oversupply, which has kept it the cheapest of all the PGMs. This situation is likely to be accentuated by increasingly greater recourse to the deeper UG2 reef, the world’s largest PGM resource, and which contains a relatively greater proportion of Ru than the Merensky Reef which has dominated production to date (Edwards, 1988). However, a growing number of new applications for Ru are set to increase demand in future (Wolff, 1999).
An assessment of estimated resources can be made from the data in Table 4.Pt and Pd reserves are vast
and capable of meeting sustained growth of up to 10% per annnm well into the next century. The Pt and Pd markets at the time of publication were balanced on the one hand by a declining trend in sales from Russian stockpiles, that characterized the immediate post-Cold War era, and by expansioii of the South African mines. In 1999, Pt was experiencing a boom, with the three largest producers in the world set to increase produo tion to raise projected Pt availability to 4.7 million tr. oz (-200 tons) per annum by 2004 (Financial Mail (South Afrrca), Oct. 1, 2999). The biggest growth areas are anticipated to be the commercialization of fuel-cell technology, as well as growth in Asian, particularly Chinese, jewellery and investment markets (Johnson Matthey, Platinum 1999). The case for Pd can be made in a similar vein, although the vagaries surrounding Russian production make this less deterministic. A final consideration relates to the PGM cycle, Owing to their nobility and intrinsic worth, the PGMs not only endure, but also provide a built-in incentive to recover as much metal as possible. Significant quantities of secondary metal have started entering the market, notably from recycled autocatalysts, and this trend is set to grow. In this respect, it is worth bearing in mind that the real cost of precious-metal based components is the interest cost plus the cost of refininglrecyclingat the end of the service life. Precious metals in any bulk engineering form are invariably recovered. In conclusion, the point that emerges is that, depending on the application, Au, Pt and Pd are precious metal commodities that can be discussed on a ‘tonnage’ basis for new applications. Use of the other PGMs is more narrowly constrained by physical production. However, the PGMs continue to occupy niche applications on the basis of their properties, and in balance with their price structures.
Table 4 Total PGM reserves of the major deposits (to 1200m depth) (after Viljoen, 1998; V m a a k and Van der Merwe, 2000) ~
PGM (in situ tons)
Pt
Pd
Ru
Rh
Ir
OS
South Africa Merensky Reef UG2 Reef Platreef Great Dyke, Zimbabwe Stillwater, Montana Noril’sk, Siberia Sudbury/Lacdes Isles, Canada
83 235.6 39 544.2 37 110.1 6581.3 10091.0 6627.91 20 255.0 878.9
42 104.6 21 750.4 17 534.1 2820.1 5656.0 1403.8 3985.3 343.1
27 102.9 13204.7 10 693.7 3204.5 3377.0 5058.5 14009.6 442.3
6898.9 2349.1 4287.2 262.6 590.0 33.8 776.4 31.2
4548.6 1419.6 2947.4 181.6 266.0 86.8 614.2 41.0
1618.8 454.1 1104.8 59.9 107.0 31.1 518.2 14.4
961.8 366.3 542.9 52.6 95.0 13.9 351.3 6.9
Total
121 088.4
53 492.8
49 990.3
8330.3
5556.6
2289.5
1428.9
Site ~~
~~~
Precious Metal Compounds
10. References Akhtar, A., and Teghtsoonian, E. (1971). Metall. Trans., 2, 2757. Anderson, S. A., and Lang, C, I. (1998). Scripfa Mater., 38(3), 493. Asada, T., Hoshino, T., and Kataoko, M, (19S5). J. Phys. F Met. Phys., 15, 1497. Baranov, N. V., and Barabanova, E. A. (1995). J Alloys and Compds, 219, 139. Blavette, D., Caron, P., and Khan, T. (1986) Scripta MetaZZ., 20(10), 1395. Brook. G. B. (1973). Gold Bulletm, 6(1), 8. Brook, G. B. and Iles, R. F. (1975). Gold BuNerin, 8(1), 16. Boniface, T. D. and Cornish, L. A. (1996). J. Alloys and Compds, 234,275. Bronger, W., Wrzesien, R., and Muller, P. (1997). Solid State Ionics, 101-103, 633. Bruemmer, S. M., Brimhall, J. L., and Henager, H. (1990). Mat. Res. Soc. Symp. Proc., 194,257. Cahn, R. W. (1998). Nature, 396, 523. Calderon, H. A., Fine, M. E., and Weertman, J. R. (1988). Metall. T r a m A., 19A, 1135. Chakravorty, S., Hashun, H., and West, D. R. F. (1985). J. Muter. Sea., 20, 2313. Chakravorty, S., and West D. R. F. (1985). Scp-ipta Metall., 19, 1355. Chang, L. C., and Read, T. A. (1951). Trans. AIME, 189,47. Chaston, J. C., and Sloboda, M. H. (1951). UK Patent No: 681, 4S4,May 8. Chaston, J. C. (1971) Gold Bulletin, 4(4), 70. Chen, B., and Franzen,H. (1990). J. Less-Corn. Metals, 159,343. Chen, Z., and Wang, Y. (1997). Thermec '97, International Conference on Thermornechanical Processing of Steels & Ofher Materials (eds T. chandrd, and T. Sakai). The Minerals, Metab & Matenals Society, p. 1449. Chiba, A., Ono, T., k,X. G., and TakahaslG, S. (1998). Intermetatlics, 6(1), 35. Chiba, A., Ono, T., Li, X. G., Kim, M. S., Hanada, S.. and Sugawara, T. (2000). Proc. Iridium (eds E. K. Ohriner, R. D. Lanam, P. Panifilov, and H. Harada). The Minerals, Metals and Matenals Society, 5 1. Chou, T. c. (1990). J. Mat. Res., 5(2), 378. Compton, D. N., Cornish, L. A., and Witcomb, M. J. (in press). J. Alloys and Compounds. Corti, C. W., Coupland, D. R., and Selman, G. L. (1980). Platinum Metals Review, 24,2. Cortie, M. B., Nana, S., and Wolff, I. M. (1995). South African Patent No: 9512676 (March 31). Coupland, D. R., McGill. I. R., Corti, C. W., and Selman, G. L. (1980). E n ~ i r o ~ ~ e Degradatzon nta~ of High Temperature Materials, Vol. 2, Douglas, Isle of Man, Apr.. 26. Coupland, D. R., Hall, C. W., and McGill, I. R. (1982a). Platinum Metals Rev., 26{4), 146. Coupland, U. R., McGill, I. R., and Corti, C. W. (1982b). Production and AppBcations of Less Common Metals,
81
Nov., Hangzhou, China, Metals SocietyIChineseSociety of Metals, 1. Coupland, D. R., and Pearson, D. P A. (1982) European Patent No: 0 065 812, Mar. 25. Cretu. C., and Van der Lingen, E. (1999). Gold Bulletin, 32(4), 115. Dahlgren, S. D. (1967). Trans. Metall. Soc. AIME, 239, 1867 Darling, A. S. (1973). International ~ e t a ~ l u r gReviews, ~al 18, 91. Das, B. K., and Lieberman. D. S. (1975). Acta Metall., 23,579. Das, B. K., Stern, E. A., and Liebeman, D. S. (1976). Acta Metall., 24* 37. Donkersloot, H. C., and Van Vucht, J. H. N. (1970). J. Less Corn. Met., 20, 83. Drews, J., Eberz, U., and Schuster, H.-W. (1986). J Less Common ~ e t u i 116,271. ~, Duhl, D. N., and Cetel, A. U. (1986). European Patent Application 0 208 645, June 3. DuPreez, G. and Wolff, I. M. (1 999). Unpublished research. Eberz, U. (1983). Farbtge Terniire tmd @aternure I n r e r ~ t a ~ l ~ c hPhasen e rnif P ~ a ~ m ~ e ~-aZiatli~en Phmen, Ph.D Dissertation, Umversity of Cologne. Edwards, D. (1988). S. A . Mining, Coal, Gold and Base Minerals, August, 5. Enami, K., Yoshida, T., and Nenno, S. (1986). Proc. Int. Con$ on Martensitic Transfor~at~ons, Japan Inst. of Metals, 103. Eow, K., Lu, D., and Pollock, T. M. (1998). Scripa Mater., 38(7), 1065. Fairbank, G. B., Humphreys, C. J., Kelly, A., and Jones, C. N. (1999). Infermetallics for the Third Millenium, Cincinnati, Nov. 1-4, ASM. Fisher, G.,Chan, W. Y., Datta, P. R.,and Burnell-Gray, J. S. (1999). Platinum Metals Rev., 43(2), 59. Flewher, R. L. (1991). US Patent No. 5,011,554. April 30. Fleischer, R. L.(1992%).Platinum Metals Rev., =3), 138. Fleischer, R. L. (1992b). US Patent No. 5,152.853, October 6. Fleischer, R. L. (1993a). Acta Mefall. Mater., 41(3), 863. Fleischer, R. L. (19936). Acta Metall. Marer., 41(4), 1197. Fleischer, R. L. (1993~).Metall. Trans. A, 24A, 227. Fleischer, R. L. and McKee, D. W. (1993). Metall. Trans.A, 24A. 759 Fleischer, R. L., Field, R. D., Denike, K. K., and Zabala, R. J. (1990). Metall. T r a m A, 21A, 3063. Fleischer, R. L., Field, R. D., and Briant, C. L. (1991). Metall. Trans. A, 22A, 403. Fogg, C. T., and Cornellisson, J. L. (1993). Ava~la&il~fy of Plaiinrdm and Platinum Group Meaak, Bureau of Mines, IC9338, U S Department of the Interior. Fonda, R. W.. Jones, H. N., and Vandemeer, R. A. (1998). Scripta Mater., 39(8), 1031. Giarnei, A. F., Pearson, D. D., and Anton, D. L. (1985). Mat. Res. Soc. Symp. Proc., 39, 293. Giessen, B. C., lbach, H., and Grant, N, J. (€964). Tram. Metall. Soc. AIME, 230, 113. Golberg, D., Xu, Y., Murakami, Y., Otsuka, K., Ueki, T., and Horikawa, H. (1995). Mater. Lett., 22(5-6), 241.
82
Structure and Composiiion
Gu, Y., Yamabe-Mitarai, Y., Ro, Y., Yokokawa, T., and Harada, H. (1998). Scripta Mater., 39(6), 723. Gu, Y., Yamabe-Mitarai, Y., Yu, X., and Harada, H. (1999). Mater. Lett., 41, 45. Gu, Y., Ydmabe-Mitarai, Y., and Harada, H. (2000). Proc. Iridiuwr (eds E. K. Ohriner, R. D. Lanam, P. Panfilov and H. Harada). The Minerals, Metals and Matenals Society, 73. Gtiard, R. W., and Westbrook, J. H. (1959). Trans. Metall. Soc. AZME, 215, 807. Gyurko, A. M., VignouI, G. E., Tien, J. K., and Sanchez, J. M. (1992). Metall. Trans. A, 23A, 3073. Gyurko, A. M., and Sanchez, J. M. (1993). Mater. Sci, and Eng., A170, 169. Harte, A. S., Hung, P. M., Horner, I. J., Hall, N., Cornish, L, A., and Witcambe, M. J. (1997). Advances m X-Ray Analysts, Vol. 39 (eds J. V. Gilfrich et al.).Plenum Press, New York. He, C., Ma, G., Wang, W., Wang, Y., and Zaho, H. (1983). Phase Diagrams of Precious Metal Alloys, The Metallurgical Industry Press, China. He. C., Zbou, Y., and Wang, W. (1993). PJfaseDiagrams of Precious Metal Alloys, First Supplement, The Metallurgical Industry Press, Ctuna. Hellstern. E., Feclit, H. J., Fu, Z., and Johnson, W. L. (1989). J. Mater Res.., 4(6), 1292. Higgmson, A. (1989). Br. Corros. J.,24(4), 297. Hill, A. D., Ford, D. A., Goulette, M. J., Arthey, R. P. (1984). US Patent No. 4488915. Dec. 18. Hill, P. J., Witcomb, M. J., and Cornish, L. A. (1997). Proc. Micro. Sac. of Southern Afrzca, 27, 18. Hill, P. J., Cornish, L. A., and Witcomb. M. J. (1998a). J. Alloys and Compds, 280, 240. Hill, P. J., Cornish, L. A., and Witcomb, M. J. (1998b). ICEMI4 Ekctron Microscopy 1998, Symposium U, Vol. 11, Cancun, Mexico, 31 Aug.4 Sept., 167. Hill,P. J. (1998). The Microstructures and Mechanical Properties of A6Ru-Ir Alloys, MSc Dissertation, University of the Witwatersrand,Johannesburg, South Afnca. Hill, P J., Cormsh, L. A., and Witcomb, M. J. (1999a). J. Alloys and Compds, 291, 130. 13i11, P. J., Cornish, L.A., and Witcomb, M. J. (1999b). 14th International Corrosion Congress, Cape Town, Sept. 27Oct. I, International Corrosion Council. Hill, P. J., WoHs I. M., Cornish, L. A., and Witcomb, M, J. (2000). Prof.. Iridium (eds E. K. Ohriner, R. D. Lanam, P. Panfilov, and H. Harada). The Minerals, Metals and Materials Society, 279. Hill, P. J., and Wolff, I. M. (1999). Mintek, unpublished research. Hill, P J., Biggs, T., Ellis, P., Hohls, J., Taylor, S., and WO& I. M. (ZOOla), Mater. Sci. and Eng. A., 301(2), 167. Hill, P. J., YamabpcMitarai, Y., and Wolff, I. M. (2001b). Scripta Mater., W(1),43. Hill, P. J., Adams, N., Biggs, T., Ellis, P., Hohls, J., Taylor, S., and Wolff, I. M. (in press@)). Presented at 5th Znt. Con$ Structural and Functional Intermetallics, Vancouver, July 2000.
Hill, P. J., Cornish, L. A., Witcomb, M. J., Ellis, P., and Wolff. I. M. (in press@)). Presented at High Temperature Corroston, Japan, September 2000. Hohls, J., Cornish, L. A., and Witcomb, M. J. (2000). f.ANo7v.s and Compds., 308, 205. Horner, I. J., Hall, N., Cornish, L. A., Witcomb, M. J., Cortie, M. B., and Booiface, T. D. (1998). J. Alloys and CompA. 264, 173. Hosoda, H., Miyazaki, S., Watanabe, S., and Hanada. S . (20004. Proc. Iridium (eds E. K. Ohriner, R.D. Lanam, P. Panfilov, and H. Harada). The Minerals, Metals and Matenals Society, 271. Hosoda, H., Yoshimi, K., Miyazaki, S., Watanabe, S., and Hanada, S . (2000b). Proc. Iridium (eds E. K. Ohrmer, R. D., Lanam. P., Panfilov, and N. Harada). The Minerals, Metals and Materials Society. 295. Hurly, J., and Wedepohl, P T. (1993). J. Mater. Sci., 28, 5648-5653. Ishyama, S., Hammelman, K. H., Buchkremer, H. P., and Stower, D. (1994). PM’94 Powder Metallurgy World Congress. Paris, France, Vol. 2, 1211. Jackson, M. R. (1990) United States Patent No: 4,980,244, Dec. 25. Jackson. M. R. (1991). United State.9 Patent No: 4,983,356, Jan. 8. Jackson, M. R. (1998). United States Patent No: 5,759,380, Jun. 2. Johnson, R.W., and Garland, C. M. (1990). J . Mater. Re.% 5(4), 746. Jorda, J. L., @a€, T., Schellenberg, L., Muller, J., Cennial, K., Gachon, J. C., and Hertz, J. (1988). J. Less-Comp1 Met., 136, 313. Jung, W.-G., and Kleppa, 0.J. (1992). Metall. Trans.B, 23B, 53. Kamm, J. L., and Milligan, W. W. (1994). Scripta Metall. et Mater., 31(1 l), 1461. Khataee, A., Flower, H. M., and West, D. R. F. (1989). Mater. SCI.and Tech., 5, 873. Km, M. J.. and Flanagan, W. F. (1967). Acta Metall., 15, 735. Kim, S., and Smith, G. D.W. (1997). Mat. Scz. and Eng., A239-240.229. Kim, S., Larson, D. J., and Smith, G . D. W. (1999). Intermetallics, 7 , 1283. Kudryavtsev, Y. V., and Semenova, E. L. (2000). Proc. Iridimn [eds E. K. Ohriner, R. D.hna m , P. Panfilov, and W. Harada). The Minerals, Metals and Matenals Society, 147. Lang, C . I. (1999). Proceedings 1st Znternatbnal Conference on ThermophysicalProperties of Matepiafs,ed. P. Hing, AMRC, Nanyung Technological Umvemty, Singapore,441. Lang, C. I., Towle, N., Nmla, M., and Miller, D. (2000). University of Cape Town, unpublished research. Lang, C . I., and Doyle, R. A. (1996). J. Appl. Phys., 79(12), 9382. Lee, K. N., and Worrel, W. L. (1989). Oxidation of Metals, 32, 357. Lee, K. N., and Worrel, W. L. (1994). Oxidation of Metals, 41, 37.
Precious Metal Compounds Levey, F. C., Cortie, M. B., and Cornish, L. A. (1998). Gold Bulletin, 31(3), 75. Levey, F. C., Cortie, M. B., and Cornish, L. A. (2000) Mefall. and Mater. Trans. A., 31A, 1917. Lin, W., Xu, J., and Freeman, A. J. (1992). J. Mater. Res., 7(3), 592. Lu, D. C.. and Pollock, T. M. (1999). Acta Mater., 47(3), 1035. Makita, T., and Nagasawa, A. (1984). Scrlpta Metall., 18, 1275. MalufF, E. I., and Cahn,R. W. (1985). J. Less Com. Me.tols, 114, 65. Manh, D. N., Paxton, A. T., Pettifor. D. G., and Pasturel, A. (1995). Infermetallics. 3, 9. Marezio, M., Dernier, P. D., and Chu, C. W (1971). Physical Review E, 4(8), 2825. Massalski, T. B. (ed.)(1990). Binary Alloy Phase Diagrams, ASM International, Materials Park, OH. McEwan, J. J., and Biggs, T. (1996). 13th International Corrosion Congress, Vol. 1., Melbourne, Australia, Nov. 25-29, Australia Corrosion Associatton, Victona, Australia. McGill, I. R., and Selman, G. L. (1978). UK Patent Application GB 2005649 A , Sept. 14. McKee, D. W., and Fleischer, R. L. (1991). Mat. Res. Soc. Symp. Proc., 213,969. Mitchell, R., Pans, H. G.,and Le Fevre, B. 0.(1973). Metall. Trans., 4, 833. Muldawer, L., and Feder, R. (1961). US Patent No. 3,012,882, 12 December 1961. Munroe, P. R. (1996). Intermetallics, 4, 5. Munroe, P. R., and Kong, C. H. (1996). IntermeraNic;r,4,403. Murakam. A., and Tsunoda, Y. (1999). J. Phys. and Chem. Soli&, 60, 1507. Murakami, H., Koimmi, Y.. Yokokawa, T., Yamabe Mitarai, Y., Yamagata, T., and Harada, H. (1998). Mater. Sci. and Eng., A250. 109. Murakami, H., H o m a , T., Yokokawa, T., Koizumi, Y., and Harada, H. (2000). Proc. Iridium (eds E. K. Ohnner, R. D. Lanam, P. Panfilov, and H. Marada). The Minerals, Metals and Materials Society, 121. Muto, S., Oshima, R., and Fujita, F. E. (1988). Metall. Trans. A , 19A, 2723. National Research Council (1996). Coatings for HighTemperature Stmctural Materials, National Academy Press, Washington D.C. Ngwenya, K. P., and WoKC I. M. (1999). Proc. Micros. Soc. Southern dfrica. 29, 19 Nordham, R., and Grant, N. J. (1952-54). J. institute of Metals, 82, 440. Obrowski, W. (1963). Metall., 17(2), 108. Ochiai, S., Oya, Y., and Suzdu, T. (1984). ActaMetaN.. 32(2), 289. Oota, A., Kusunoki, H., Tsuchya, M., Takei, H., and Noguchi, S. (1985). J. Phy.s. F: Met Phys., 15, 1103. Oshima, R., Muto, S., and Hamada, T. (1988). Platinum Metals Rev., 32(3), 110.
83
Otsuka, K., Oda, K., Ueno, Y., and Rao, M. (1993). Scripfa Metall. er Mater., 29, 1355. Otsuka, K., and Wayman, C. M. (1998). Shape Memory Materials, Cambridge University Press, New York. Otsuka, K., and Ren, X. (1999). Intermetallics, 7, 511. Oya, Y., Mishima, Y., and Sumki,T. (1987). Z . Metalucd>. 78(7), 485. Paxton, A. T., and Pettifor, D. G. (1992). Scripta Metall. et Mater., 26, 529. Pecora, L. M., and Ficalora, P. J. (1977). J. Elee. Mater., 6(5), 531. Petersen, A. (1989). Das System L i ~ h i ~ ~ - P l a t i ~ e t a l Z - 3 . Hauptgruppc~PhD Dissertation, University of Cologne. Piatti, G., and Bardy, M. (1981). Metal Science. 15, February, 55. Piatti, G., and Pellegrim, G. (1976). J. Mater. Sci., 11. 913. Putatunda, S. K., Naik, R., and Wager, L. E. (1998). Marer and Manuf. Proc.. 13(5), 671. Randin, J.-P. (1994). 5Gme Congress Europien de Chmnome'trie. Oct. 18-19, BesanFon. Session: Materiaux Cornmumcation 3, 5 pages. Raynor, G. V. (1976). Gold Bulletin. 9(2), 50. Rapson, W. (in press). The Science and Teehnology of Gold, Met-Chem Research Inc., Alberquerque. Raw, P. M. (1992). Precious Metals Horuon 2000, October 14-17, Nice, France, Eurornetaux, 75. Ridley, M., and Pops. H. (1970). Metall. Trrans., 1, 2867. Ritter, D. L., Giessen, B. C., and Grant, N. J. (1964). Trans. Metall. Soc. AIME, 230, 1250. Sabanz, A. L. R., and Taylor, G. (1997). Mat. Res. Soc. Symp. Proc,, 460,611. Sabariz, A. L. R., and Taylor, G. (1998). Poster paper KK5.30 at MRS, Boston, USA. Sabariz, A. L. R., and Taylor, G. (1999). Euromni 99, Munich, September. Savitsky, E., Polyakova, V., Gorina, N,, and Roshan, N. (1978). Physical Metallurgy of Platinum Metals, MIR Publishers, Moscow. Schmerling, M. A., Das, B. K., and Lieberman, D. S. (1970). Metall. Tram., 1, 3273. SchBnfeld, B. (1999). Prog. Mater. Sci., 44, 435. Schwartz, A. J., and Tanner, L. E. (1992). Proc. Int. Conf. on Martensitic Transformations, Monterey Institute for Advanced Studies, Monterey, California, USA, 95. Selman, G. L., and Midgley, R. J. (1975). UK Patent No. 1,520,630, Jul. 8. Semenova, E. L., and Kudryavtsev, Y. V (1994). J. Alloys and Compds, 203, 165. Semenova, E. L., Rusetskaya, N. Yu., and Petyukh, V. M. (1995). J. Phase Equilibria, 16(4), 297. Semenova, E. L., Rusetskaya, N. Yu., Petyukh, V. M., and L~~tovnichiy, V.Ye. (1995).PlatimMetalsRev., 39(4), 174. Smith, E. G., and Lang, C. I. (1995). Scripta Metall. et Mater., 33(8), 1225. Steinemann, S. G. (1994). 52me Congres Eztrope'en de Chronomitrte, Oct. 18-19, Besanpn, Session: Materiaux Cornmumcation 2, 7 pages.
84
Structure and Composition
Steyn, H., DeV., Wolff, I. M., and Goetzee, R. (1994). South Africm Patent No. 94/10180 (Dec. 22). Stoloff, N. S., and Davies, R. G. (1966). Progr. Mater. Scr., 13(1), 3. Stoloff, N. S., and Sims, C. T. (1987). Tn Superalloys II (eds C. T. Sims, N.S. Stoloff, and W. C. Hagel). Wiley. Suss, R., and Cretu, C . (1998). Proc. Micros. Soc. Southern dfrtca, 28, 20. Tadaki, T., Okazaki, H., Nakata, Y., and Shimini, Y. (1990). Mater. Trans. JIM, 31(11), 935. Taillard, R., and Pineau, A. (1982). Mater. Sci. and Eng., 54, 209. Taylor, S, S., and Biggs, T. (1996). South African Patent No: 9618932, Oct. 24. Towle. N. R. (1999). Order ffardeningofPlatinm Al1oy.q MSc Dissertation, University of Cape Town, South Afnca. Towle, N. R., Lang, C. I., and Miller, D. (1998). Proe. Micras. Soc. Southern Africa, 28, 15. Tsurikov, V. F., Sokolovskaya, E. M., and Kwkova, E. F (1980). Vestnik Mosliovskogo Universibeta, 35(5), 512. Vermaak, C. F., and van der Menve, M. J. (2000). The Platinum Mines and Deposits of the Bushveld Complex South Africa. Mintek, Randburg, South Afnca. p. 118. Viefhaus, H.. Row, J. P., and Grabke, H. J. (1993). Steel Research, 64(7), 369. Viljoen, M. (1998). World Platinum Congress, Nov. 18-19, Sandton, South Africa, Centre for Mining and Energy, 42 pages. Villars, P.,and Calvert, L. D. (1985). Pearson’s Handbook of CrystallographieDatafor Intermetallic Phases, ASM, Ohio. Vitek, V., Sodani, Y., and Cserti, J. (1991). Mut. Res. Soc. Proc., 213, 195. Walston, W. S., Ross, E. W., Pollock, T. M., O’Hatd, K. S., Murphy, W. €3. (1995). US Patent No. .5,455,120, Oct. 3. Wang, S.-P,, LI, Y.-J., Chang, X.-M., Yang, Y.-H. (1966). Precious Meials 1996, June, Newport Beach, California, USA, International Precious Metals Institute, 245. Warlimont, H., and Delaey, L. (1974). Prog. Matls Sci., 18. Warren, B. E. (1965). Trans. Metall. Soc. AIME, 233, 1802. Wayman, C. M., and Harrison, J. D. (1989). JOM, September 26. Wee, D.-M., Noguch, O., Oya, Y., and Suzuki, T. (1980). Tranx JIM, 21, 237. Wee, D.-M., Pope, D. P., and Vitek, V. (1984). Acta Metall., 32(6), 829. Wee, D.-M., and Suzuki, T. (1979). Trans. JIM, 20, 634. Westbrook, J. H. (1977). Metall. Trans. A , SA, 1327. Wolff, I. M. (1994). South African Patent No. 9413626 (25 May). Wolff, I. M. (1995). Endeavour, 19(1), 20. WOE, I. M. (1996). Metall. and Mater. Trans. A., 27A, 3688. Wolff, I. M. (1997). JOM, January, 34. Wolff,1. M. (1999). South African Journal of Science, 95, 539. Wolff, I. M., Cornish, L. A., Sauthoff, G., Steyn, H. DeV, and Coetzee, R. (1997). Structurat Intermetallics 1997, Proc. 2nd International Symposium OR Structural Intermetallics, Seven Springs, Champion, 21-26 Sept.
(eds M. V. Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. Yamaguchi). Tfhe Minerals, Metals tk Matertals Society, 815. WOW, I. M., Iorio, L, E., Rumpf, T., Scheers, P, V. T., and Potgieter, H. (1998). Mater. Sci. and Eng. A , 241 (1-2), 264. Wolff, I. M., and Cortie, M. B. (1993). South $ric.an Patent No: 9312674. Wolff, I. M., and Cortie, M. B. (1994). Gold BuIieiin, 27(2), 44. Wolff, I. M., and Hill, P. J. (2000). Proc. Iradiwn (eds E. K. Ohnner, R. D. Lanam, P Panfilov, and H, Harada). The Minerals, Metals and Matertals Society, 259. Wolff, I. M., and Pretonus, V. R. (1994). Gold Technology, 12, 7. Wolff, I. M., and Sauthoff, G . (1996a). Metail. and Muter. Trans. A., 27A, 1395. Wolff, I. M., and Sauthoff, G. (1996b). MetalZ. and Mater. Trans. A., 27A, 2642. Wolff, I. M., and Sauthoff, G. (1997). Acra Mater., 45(7), 2949. Wolff, I. M.,and Sauthoff, G. (1999). Unpublished research, Mintek, South Africa. Wong-Kian, M., Cortie, M. B., and Cornish, L. A. (1998). Meter. Sci. and Tech., 14, 1242. Wopersnow,W., and Raub, Ch. J. (1979). Metall., 33(7), 736. Wu, S. K., and Wayman, C. M. (1987). Meiallography, 20, 359. Xu, Y., Makhlouf, S. A., Ivanov, E., Wakoh, K., Sumiyama, K., and S U M ,K. (1994). Nanostructured Materials, 4(4), -437. Yamabe, Y., Koizurm, Y., M u r a k m , H., Ro, Y.. Maruko, T., and Harada, H. (1996). Scripza Mater., 35(2), 211. Yamabe-Mitarai, Y., Ro, Y., Maruko, T., Yokokawa. T., and Harada, H. (1997). Structural Intermetalks 1997, Proc. 2nd International Symposium on Structural Intermetallics, Seven Springs, Champion, 21-26 Sept. (eds M. V. Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M, Yamaguchi). The Minerals, Metals & Materials Society, 805. Yamabe-Mitarai, Y., Hong, M.-H., Ro, Y.. and Harada, H. (1999a). Phil. Mug. Leits, 79(9), 673. Yamabe-Mitarai, Y., Ro, Y., Maruko, T,, and Harada, H. (1999b). Scrpta Mater., 40(1), 109. Yamabe-Mitarai, Y., Gu, Y., Ro, Y., Nakazawa, T., Maruko, T., and Harada, H. (1999~).Scriptu Mater., 41(3), 305. Yi, S. S., Chen, B. H., and Franzen, H. F. (1988). J. Less Common Metals, 143, 243. Yu, X. H., Yamabe-Mitarai, Y., Ro, Y., Gu,Y., and Harada. H. (1999). Scripta Mater., 41(6), 651. Yu, X. H., Yamabe-Mitarai, Y., Ro, Y., and Harada, H. (2000). Key Engineering Materfals, 171-174, 677. Yuasa, S., Miyajima, H., and Otani, Y. (1994). J. Physical Soc. Japan, 63(8), 3129. Zhu, S . M., and Tjong, S . C. (1997). Metall. m d M a t . Trans. A., B A , 1095.
are-Earth
mpounds
Marfa L. Pornasini, Franco Merlo and Department of Chemistry and Industrial Chemistry, University of Genova, Genova, Italy
1. I~tro~uction Among intermetallicphases, the compounds containing rare-earth atoms form a numerous class, well studied for many years in both fundamentd and applied research. Nearly all known phases are binary or ternary compounds, with some general features, Binaries, formed by rare-earths with elements from manganese to the nitrogen group, are in most cases strictly stoichiometric, solid solutions being usually rare. Ternaries occur in most cases in R-T-X systems, where R is a rare-earth, T is a transition metal of the titanium to copper groups, and X is a post-transition element. In these phases, homogeneity ranges are frequently observed,coveringin some cases large fields in the compositional phase diagram. The term rare-earth refers to the Lanthanide family (lanthanum to lutetium), and for reasons of chemical similarity usually includes the other two elements of the third group, scandium and yttrium.* For these 17 elements the alloying behavior observed is not uniform, as shown by the experimental data. According to a simple analysis of composition and crystal structure of the existing phases, the rare earths can be divided into three groups: 1. The normally trivalent lanthanides, namely from
lanthanum to Iutetium, save europium and ytterbium, but including yttrium, 2. Europium and ytterbium, which in most systems behave as divalent a t o m , in a similar way to the alkaline-earths (calcium, strontium, barium), 3. Scandium. *See p.108 for a useful and amusing mnemonic for recalling the sequence of the rare-earth elements.
The three different behaviors are due both to the valence situation and for dimensional reasons. Moreover, in some systems cerium and ytterbium (and rarely europium, samarium and thulium) display valence instabilities,giving rise to anomalous physicd properties which are described by several theoretical models (intermediate valence, Kondo systems, heavy fermions).
2. Stability and Crystal Structure 2.1 Thermal Stability
A large number of rare-earth metal phase diagrams, especially of binary systems, has been determined, and the considerable volume of data on melting (or peritectic) temperatures gives information about the thermal stability of the intermediate phases, Figure l(a) reports all the phases found in the binary systems formed by gadolinium with the X elements from 7A to 15B groups of the periodic table. The phase diagrams are known for all Gd-X systems save for X = OS,Ir, Zn, Hg, and As. Considering the proposed Pr-0s diagram (Massalski et al., 1990) and the available thermal data on the Ce-Ir, Sm-Zn, La-Hg and Nd-As diagrams, it is possible to locate the solid phase with the highest melting temperature in each system. In nearly all cases this phase melts congruently, while GdMn,,, Gd,Fe,, and GdMg show a peritectic formation: the lines join these phases within every group, and the corresponding temperatures, indicated as T,,, are plotted in Figure I@). When X is a transition element, the composition of the phase with the maximum thermal stability is rich in X (at least GdX& and the corresponding T,, values are close to the melting temperatures of the X elements, which
Intermetallic Compounds: Vol. 3. Prmczples and Practice. Edited 1)y J. H. Westbrook and R. L. Fleischer. 02002 John Wiley & Sons, Ltd.
Ill I/ I //I I
0 Mn Fe
Rh
RU
OS
eu
Ni
CO
Pd
Ir
Pt
Mg
As
Al
Zn Au
Si
Ga Cd
Hg
As
Ge Ill
n
Sn
sb
Pb
Bi
Figore 1 (a) Schematx representation of the binary phase diagrams, For each system generally gadolinium IS chosen as representative and full circles indicate the existing intermediate phases. The lines join the phases with the highest melting or pentectic temperature (TmM) in each system; (b) Penodic trend of T,,
show a similar periodic trend. These two features indicate a prevailing influence of the chemical bond interactions o f the X sublattice. From the copper group onwards the composition of the most stable compound changes to GdX (or Gd,X, for X = Ge, Sn, or Pb), and the T,, values are considerably higher than the X melting points, reaching a difference of 1740°C for GdBi. This can be imputed to the strengthening of the R-X interactions, indicating a lower metallic character and an increase of the bond ionicity: in other words, a transition from purely metallic phases to true chemical compounds.
2.2 Volume ERects The formation of an intermetallic phase is accompanied by a volume effect. The average volume per atom Vexp, experimentally obtained from the ratio ‘elementary cell volume’/’number of atoms per cell’, is usually different from the ideal value VdCr calculated from the
proper combination of the elemental atomic volumes of the components. In most rare-earth compounds a volume contraction occurs, namely the quantity A V/V = ( Vexp- V&)/ Vcalcis negative, though slightly positive or zero values are sometimes observed. The phases formed by the divalent rare-earths (Eu, Yb) show the maximum AVlV values, and the heavy rareearths (from Gd to Lu) give volume contractions higher than light rare-earths. Within a binary R-X system, the AV/V values are different for each compound, but show a regdar trend versus composition, reaching the maximum for a given phase. Figure 2(a) reports these maximum AV/Vvalues in the binary Gd-X compounds, the observed contractions ranging from 1 to 15%. The 25% contraction shown by silicon and germanium phases (in particular GdSi, and GdGe,) is mainly a geometrical effect, due to the strongly different packing efficiency in the compounds compared to the diamond-type structure of elemental Si and Ge. The other factors influencing the volume
87
O
25-
x
atomic percent
loo
0
100
A V N ["h]
20 15
-
10 -
c-c--J
5-
Figure 2 (a) Periodic trend of the maximum volume contraction values (AV/V) observed in the binary Gd-X compounds; (b) Experimental average atomic volumes (V,,) versus composition in some binary gadolinium systems. Dotted lines show ideal trend; full curves are calculated according to Merlo and Fornasini (1993). The Vex,,scale is indicated only for the Gd-Ni system
deviations from the ideal linear trend between the elemental volumes are the elastic effects (energy necessary to change the volume of the Wigner-Seitz cells up to the equilibrium density in the compound) and the ionic effects (charge transfer due to the electronegativity difference between the component atoms). In Figure 2(b) some examples of average atomic volumes versus composition are plotted, for gadolinium systems with a large number of intermetallic compounds. The dotted lines indicate the ideal trend, while the full curves are calculated following an empirical model (Merlo, 1988; Merlo and Fornasini, 1993), which suggests the main role of charge transfer effects when X is a transition metal, and the greater
influence of elastic effects when X is a post-transition element.
2.3 Binary Compounds From a statistical point of view, the distribution of the intermediate phases in the R-X systems is not much different from that shown by Iandelli and Palenzona many years ago (1979). Taking into account the intennetallic data compilation (Villars, 1997) and the last five years of scientific literature, the current total of about 2500 binary rare-earth phases covers 65 different compositions, going from Sc,,Os,to YB,,. As shown in Figure 3, where the number of phases is reported as a
88
Structure w d Composition
1:2
1:l
400 QD
300 1:3
1 5 217
0
20
60
100
Figure 3 Statistical distributionof the R& phases as a function of the X content. The most frequent compositionsare indicated
function of composition, the most frequent formulae are RX, ,RX ,RX3and R,X,. The structurally known R,X, compounds crystallize in about 200 different structure types, with a large variety of bond-length values and atomic-coordination geometries. This situation i s the main reason behind the unusual chemical behaviors and exceptional physial properties which gwe rise to the numerous practical applications of rare-earth intermetallics. The importance i s clear of any theoretical or empirical method able both to rationalize the existing crystaltographic data and to obtain predictions on the existence, structure and properties of new phases (see chapter by Kiselyova). The method of structural maps, widely used for many years, and extensively applied to intermetallic compounds by Villars (1995), points out the physical factors which influence phase stability. Figure 4 shows the occurrence and the crystal structure of the binary phases with the four most frequent compositions formed by the trivalent rare-earths and yttrium. Only the phases stable at normal temperature and pressure conditions axe rqorted. The X elements are arranged as in the periodic table, and this sequence appears to be
one of the most important rationalizing factors; clustering of most structure types in certain regions is observed. Concerning the RX, phases, a great variety of structural types is observed. Save for the AuCu, (Pearson code cP4) type, which is spread over nearly the whole periodic table, the other types seem stable only in particular regions: from left to right the sequence is NbBe, (hR36) and CeNi, 01p24) first, then TiCu, (oP8), BiF, (cF16). and SnNi, (hP8). It is known that NbBe, and CeNi, are closely related, while the types AuCu,, SnNi,, TiCu,, ErCd, (oC16), HoA13 (hR60). TiNi, (hP16), BaPb, (hR36) and PuGa, (hR48) are all based on different stackings of close. packed layers. Stability domains of the different structural types can be recognized even for the RX2 compounds: for example, the Laves phases (MgCu, (cF24) and MgZn, (hP12) structure types) form mainly with transition elements, then the tetragonal MoSi,! (tI6) is encountered, followed by the AIB2 (W3), CeCu, (0112) and CeCd, (hP3) types, all three strictiy correlated and based on the trigonal prismatic coordination of the X
Rare-Earth Metal Compounds
Rx
Ra
0s
Ir
.I Pt
0
OCsCI
OCrB
mCaAt
oTlrNi(r)
O A
Mg
AI Si I.r901no
1%
Au
Hg
Tl
Bi
0 A O
#
* U
AFeB
1NaCI
Pb
vDyM
89
1
A
oAuCul
(e)
0
MnJSib
OQg&
Aw&$
APU&
t PlWbJsb,
Ph
Sb
fa)
Figure 4 Distribution of the structural types occurring for the known phases with composition RX, (a), RX, (b), RX (c) and R,X, (d), where R =trivalent rate-earths and yttrium. The symbols are drawn if at least one compound within the series exists
atoms. The types CaC, (tI6), a-ThSi, (t112), ZrSi, (oC12), TbGe, (oC24), Lap, (mC48), NdAsz (mP12) and SmSb, (oC24) occur only with the 4B and 5B group elements, and are mainly governed by the bond geometry of the X sublattice. A rather complex situation can be observed in the RX table of Figure 4, where the three most frequent types (CsC1 (cP2), CrB (oC8), FeB (oP8)) appear in partially overlapping domains. As already pointed out (Schob and Parthk, 1965; Hohnke and Parthk, 1966), the CrB- and FeB-type compounds, dominated by the trigonal prismatic coordination of the X atoms, are separated in two classes, and elongated or compressed prisms are found when X is a p-element or a transition metal, respectively. The phases crystallizing in the NaCl (cF8) type are restricted to the 5B group elements, and can be classified as normal valence compounds.
The hexagonal Mn,Si, type (hP16) frequently appears among the R,X3 phases. For many years it has been known that, by filling the octahedral interstices of this structure, a ternary ordered HfsSn3Cu (or MoSSi3C,hP18) type is formed, and several filled-up rare-earth stannides, plumbides and bismuthides have been reported. In more recent years this filled structure has been found for compounds such as LasGe3Z and La5Pb,Z, with a wide range of the Z element going from transition metals to halogens (Kwon et al., 1990; Guloy and Corbett, 1992, 1993, 1994). The corresponding binary compounds exist with the same unfilled MnsSi, structure, but in La&$ with Z = C,,O, this structure type is stabilized only by the insertion of the third element, whilst the parent compound LasSn3 crystallizes with the tetragonal WsSi3 (tI32) structure. Moreover, the filling with some particular elements may cause a structure
Structure and Composition
90
change, as in La,Pb,2 with Z = N , 0,where an interstitial derivative of the tetragonal Cr,B, (tI32) type is formed. Examples of contamination by hydrogen, probably present in the starting, commercial purity, rare-earth metals, were found in R5X3 phases with R = Eu, Yb and X = Sb, Bi, previously reported in the literature to have the orthorhombic P-YbsSb3 (oP32) structure type. They are really R,X3H, compounds, with hydrogen atoms in the tetrahedral interstitial sites, but after heating these products under high vacuum, high temperature conditions, hexagonal Mn,Si,-type phases can be obtained (Leon-Escamilla and Corbett, 1998). So, the capability of inserting a variety of elements in the host lattice in a controlled way can be useful either to stabilize compounds with that structure, or to obtain compounds with interesting physical and mechanical properties (Corbett et al., 1998). 2.4 T e m q Conrpoinndr
A compilation of crystal data and a discussion of the crystal chemistry of ternary rare-earth intermetallics were made by Parthe and Chabot (1984). Their proposed division of these phases into three classes is still valid: 1. Truly ternary phases, where the different elements
occupy different crystallographic sites. In some cases the structures are ternary ordered derivatives of binary types. 2. Truly ternary phases, where certain crystallographic sites are randomly occupied by different atoms. Extended homogeneity ranges are often observed. 3. Ternary solid solutions formed by binary phases with the third element. An updating of ternary phase equilibria and crystal chemistry in ternary rare-earth systems was reported by Gladyshevskii et al. (1990). The total number of known ternary phases increases every year, and nowadays about 5500 compounds are known, covering 21 1 different compositions. Figure 5 shows the distribution and the number of the 20 more common phases versus stoichiometry: RTX, RT,X, and RTX, are the most frequent formulae, which go from ScgFeGa (01142) to CeCr,Al, (cF184). The structurally known ternary compounds crystallize in at least 360 different structure types. Figures 6, 7 and 8 show the Qccurrence and crystal structure of the RTX, RT2X2and RTX, phases, respectively, considering only the trivalent rare-earths and yttrium. The T
X
Figure 5 Distribuhon of the 20 niore frequent RT rareearth intermetallic compounds versus composition base of each bar). The corresponding number of phases is proportional to the bars length, considering that about 1000 RTX phases are known
&
and X elements are placed on both axes and arranged according to the chemical scale proposed by Pettifor (1986a,b); the crystal structure symbols in each T-X box indicate the occurrence of one or more out of the 14 possible rare-earth phases. The triangular graph of the RTX phases IS complete on both axes, while for the other two compositions only the rectangular region containing all known phases is reported. The simplest crystal-chemical situation is shown by the RT,X, phases, which mostly belong to the very similar ThCr2Si2(tIl0) and CaBe,Ge, (tP10) types, two ordered derivatives of the B A 4 (t110) structure. The X atom is always an element of the 3B, 4B or 5B groups. A little more complex distribution is found for the RTX, phases, owing to the occurrence of 14 different structure types. If we regard the X atoms in phases crystallizingwith the three most frequent types, a general glance at the graph shows that the CeNiSi, type (oC16) occurs only with the 4B group, HfCuSi, (tP8) with the 5B group and YNiA1, (or MgCuAl,, oC16) with the 3B group elements. Most of the CeNiSi,-type phases are non-stoichiometric, with wide homogeneity ranges, and the true formula is
Rare-Earth Metal Compounds
91
Figure 6 Distribution of the crystal structure types for the RTX compounds (R=tnvalent rare-earths and yttnum; T, X =elements of the BA to 5B groups and Li, Ti, Mn). Both axes report the T and X elements arranged according to Pettifor's chemical scale. Each T-X box includes the 14 RTX systems, and the symbols refer to the structure type of the existing phases. The outlined region corresponds to four electron concentration values: 25/3, 26/3, 2713, 28/3 electrons/atom
usually R T a 2 with 0.1<x<1. An interpretation of such a situation was given in terms of a variable interstitial solution of the T atoms in a ZrSi,-type (oC12) RX, sublattice (Francois et al., 1990; Venturini et al., 1990). The maximum number of structure types (24), of observed T-X combinations (105) and compounds formed (more than 1000) is shown by the equiatomic RTX class. A four-dimensional structure map can be applied to these compounds, using the same parameters employed to analyze a part of these phases (Fornasini and Merlo, 1995): the size parameter Ar) = r ~ / ( ~ + c yQ), considering the elemental radii for coordination 12 (Teatum et aZ., 1968); the electrochemical parameter A@*)= -@ - @ff, where @* is the Miedema electronegativity (de Boer et al., 1988); the periodic parameter Px,namely the period number of the X element; and the valence electron concentration VEC. The problem in evaluating
+
this last quantity can be solved by taking the corresponding group number (e.g. 3 for La, 8 for Fe, or 14 for Si). These values reproduce the sequence of the elements in the periodic table, whose influence on crystal structure distribution is recognizable in Figure 6: as can be seen, most phases occur in the systems with four VEC values, from 25/3 electrons/atom (e.g. for the WeSi phases) to 28/3 electrons/atom (e.g. for RNiP). The corresponding region, outlined in the figure, contains 181 different T-X pairs, with 2534 possible RTX compounds. About 980 phases are structurally known, crystallizing in 2 1 structure types. Therefore, the structural map contains the eight sections corresponding to VEC = 2513, 261'3, 2713, 28/ 3 electrons/atom; Px = 3;4 and Px = 5;6 (Figure 9). The graph reports only some border points within each series of 14 possible phases for a given T-X pair. Half of the structure types (PrPdSi (mP24), ScAuSi (hP6), MgZn, (hP12), LaIrSi (cP12), CeRhAl COP%),
92
Structure and Composition
Figure 7 Distribution of the crystal structure types for the RT,X, compounds (R=trivaIent rare-earths and yttrium; T, X = elements of the SA to SB groups and Cr, Mn, Re)
V
Figure 8 Distribuhon of the crystal structure types for the RTX, compounds (R= trivalent rare-earths and yttrium; T, X = elements of the 8A to 5B groups and Li, Mn, Re)
93
Rare-Earth Metal Compounds
....
f(4j*l 7
6 5 c
.5
.... ....
.6 @) .7
7
6
5
Fignre 9 Structural map for the RTX phases (R = trivalent rare-earth and yttrium; T, X =elements of the 8A to 5B groups). The coordinates are the combination of the elemental radii,Ar) = r ~ / (+r px), ~ the combination of the electronegativitieson the Miedema scale*)= @ + @$the valence electron concentration ( Y E Q and the period number of the X element (Px). The symbols indicate the structure types
e,
CeGoAl (mC12), LuNiGe (oc48), EuNiGe (mP12), WdSi (oP24), gYbAuGe (oP36), TiFeSi (0136)) occur rarely, and orrly in limited regions of the map. Some types are frequent and well localized, such as the LiGaGe (hP6) and MgAgAs (cF12) types, whose occurrence appears mainly related to the VEC value of 28/3 electrons/atom, and the GeFeSi (tP6) phases, clustered in a region defined by Px=3;4 and VEC= 2513; 26/3 electrons/atom. On the other hand, the very frequent TiNiSi (oP12) and ZrNiAl (hP9) types spread over a large tridimensional region, and nearly everywhere the correspondingdomains overlap. Though the spatial separation of the different structural domains appears incomplete, this type of data representation can be considered useful, as some predictions of the structure of the unknown phases are possible, The imperfect structural definition of the map i s mainly due to the small energy differences among structure types which are geometrically related.
The best example is the structural family derived from the AlB, type. As pointed out by many authors (e.g. see Gladyshevskii et al., 1992; Kussmann et a[., 1998), all these types are based on the trigonal prismatic coordination of the non-rare-earth atoms, and can be obtained from the parent type by different ordering on the T and X sublattices, and/or deformation of the elementary cell. Figure 10 reports some examples of 1:1:l types, which maintain the hexagonal symmetry, while Figure 11 shows some orthorhombic and monoclinic variants, usually derived from the simplest form, the CeCua type. 2.4.1 Crystal Structure qf the
Nd, (Fe, -xTiJ29- Type Phases awing to their interesting magnetic properties, a particular note on the crystal chemistry of the novel R,(Fe, --xMx)29phases is given. The magnetic ternary
94
Structure and Composition n
2
YbAgPb
LiGaGe
ScAuSi
nisa
Figure 10 Hexagonal structures denved from the AIB, type. The AIB, cell is viewed along c; all others are projected along the [I 101 direction. R, X and T atoms are represented by large, medium and small circles, respectively. Open and full circles are at heights 0 and ;in the case of YLiSn the double circles are at heights $ and $. The lines connecting the R atoms show the trigonal prisms. For YbAgPb, see Merlo et nl. (1996); for all other structures, see W a r s (1997)
4
phase Nd,(Fe, -xTix)29 was reported with monoclinic symmetry (Li et al., 1994). This phase is not formed in binary alloys of rare-earths with iron, but a third element, such as Ti, V, Cr, Mn, Nb, MO, or Re, is necessary to stabilize it. The quantity of additional element can bc as small as x=0.022-0.067 for Ti, larger as x = 0.14-0.17 for Cr and x = 0.39 for Mn, or a mixture of additive elements can be used. Owing to the potential application of these materials for permanent magnets, a lot of work has been done in recent years on their characterization and magnetic properties determination. Several phases were studied R,(Fe,Ti),, with R = Ce, Pr, Nd, Sm, Gd, Tb, Dy, or Y (Fuerst et al., 1994a; Ibarra et al., 1994; Margarian et al., 1996; Li et aZ., 1996; Yang et al., 1997); R,(Fe,M),,
-
with R = Ce, Nd, Sm, Gd, Tb, Dy, or Y and M V, Cr, or MO (Han et al., 1997; Hu et al., 1998); Nd,(Fe,Mn), (Fuerst et aZ., 1994b); Sm3(Fe,Re)29 (Ivanova et al., 1996) and Gd,(Fe,Nb),, (Huang et al., 1999). By powder neutron diffraction the site occupation of some phases was determined (Yelon and Hu, 1996), while recently success in single-crystal growth has been obtained (Courtois et al., 1998; Mendoza and Shaheen, 1999; Yang et al., 1999a,b). The structure of Nd,(Fe,Ti)29(mS64) was described as belonging to the monoclimc A2/m space group (Kalogirou et al., 1995) and to contain 2 Nd and 11 Fe/ Ti sites. It belongs to the series of structures derived from RT5 (CaCurtype), where R is a me-earth (or actinide) element and T is a transition element, by
95
Rare-Earth Metal Compounds
a
b
EuAuln {Ti~iSi)
EuAuGe
a
a-U FeGe a
b YPdSi
b
EuAuSn
CaAuSn Figure 11 RTX structures derived from the CeCu, type by different ordering of the T, X atoms and/or deformation. Large, medium and small circles: R, T and X atoms. Open and full circles represent atoms on layers separated by half the period along the projection axis. To facilitate the comparison,the R distorted trigonal prrsms are connected by lines. The references are: oc- and y-YbAuGe (Merlo et al., 1998), YPdSi (Prots’ ef al., 19981, EuAuGe (Psttgen, 1995), EuAuSn (Pottgen er al., 1997), EuAuIn (Pottgen, 1996), oc-UFeGe (Canepa eb al., 1996), CaAuSn (Kussmann et al., 1998)
96
Structure and Composition
a43 Th22n17
a
ThMn, Rgure 12 Sectionsat zero height of the Nd,(Fe,Ti),,, Th,Zn,, and ThMn,, cells. Large circles: Nd or Th; small circles: Fe/Ti, Zn or Mn in the section plane; double circles: Fe/Ti, Zn or Mn atoms above and below the section plane. The segments of the '€h,Zn,-, and ThMn,, structures forming the Nd,(Fe,TiX, structure are outlined
replacing some R a t o m with T-atom pairs (dumbbeus) coaxial with the hexagonal c axis of the CaCu, cell. A classification of these structures was made by Stadelmaier (1984), who proposed the formula Rm-nT3mc2n, where m denotes the number of units RT, making up the new structure, and n is the total number of R atoms being replaced. For the 3:29 phase we have m = 5 and n = 2. This series also includes the well-known 2:17 stnictures (hexagonal Th2Ni,7 and rhombohedral ThZZnl7) and the tetragonal ThMn,, structure. So, it is not surprising that the structure of the 329 phase can be regarded as formed by an intergrowth of the Th2Zn,, and ThMn,, types, as already mentioned by several authors. The structure of Nd,(Fe,Ti), is represented in Figure 12 together with the Th2Zn17and T W n l 2structures, the axis perpendicular to the sheet being b for the monoclinic 329 cell, a
for the orthohexagonal 237 cell and a2 for the tetragonal 1: 12 cell. Along !his common direction the lattice translation is -&SA. Only the basal plane of each ?I1 is drawn, but the entire outlined blocks, 8.5 A thick, are identically reproduced in the 329 structure, maintaining the same composition of the component 2:17 and 1:12 structures.
-
3. PhyslcaI Properties
Many technological applications are based upon those special thermodynamic, magnetic, electronic, optical and purely chemical effects of solid materials, that are mostly found in intermetallic systems. Table 1 reports a fist of such effects, their main applications and one example of a useful material for each effect. Some
97
Rare-Earth Metal Compounh Table 1 Examples of technological applications based on physico-chemicalproperties of IMCs
Physico-chemicaleffect
Application
Material
Magnetic ordering Kerr effect Magnetostriction
Hard and soft magnets Magneto-optical data storage Actuators Read heads Cryogenerators Energy storage Refrigerators
Nd2Fe,,B BiMn, amorphous Tb-Fe-Co Tb, ,Dy, ,Fe, (“Terfenol-D’) InSb-NiSb, La,,,Ca, ,,MnQ, Gd, @,Gel -J6
Magnetoresistance Magnetocalonc effect Hydrogen absorption Peltier effect Chemical reactivity Thermionic effect Secondary emission Semiconductivity
Superconductivity
Getters Electron guns
Photomultipliers Electromc devices
Charge transport
substances have been in use for many years, others are not yet commercially employed, but the role played by rare-earths is very important. The 1995 edition of Intermetallic Compounds and the present one consider in detail nearly all the listed applications, and some other important topics such as structural superalloys uses. In the following, a sampling of literature data of the last five years will be given for five different applications of rare-earth intermetallics.
3.1 Magnetic Properties An overview of magnetic properties and applications of intermetallics was given in Vol. 2 of this treatise by Stadelmaier and Reinsch (1995). The properties of ordered magnetic systems have been described by theoretical models, showing the role of the different types of interactions in determining the material performance. The following general reviews may be cited: Gignoux and Schmitt (1995), Duc (1997), Richter (1998). Cadogan, in another chapter of this work, extensively reviews magnetic interactions of IMC constituents and in particular magnetic phase diagrams. Here we summarize the structures and properties of rare-earth intermetallics of magnetic interest. 3.1.1 Nd2Fe,,B Phase
Nd2Fe14Bis the well-known basis of high performing hard magnets and for several years its manufacture has entered into commercial production. Recent developments have increased coercivity and corrosion resistance and decreased material and production costs. Other efforts have been directed to overcome the poor thermal stability of this compound, due to its
LaNi,
Bi,Te, Ba,A1,3 LaB, Cs,Sb GaAs
Nb,Sn, Nb-Ti solid solution
low Curie temperature and the strong temperature coefficient of remanence and of coercive field. Several types of substituent or additive elements were studied, but the problem is that each of them, while improving some properties, is unfavourable for others. So, a proper control of the amounts of the added elements is essential. Typical substitutions are made by replacing the Nd atoms with Dy or Tb, and Fe atoms with CO,Ni or Cr, thereby achieving a considerable change of intrinsic properties, such as the magnetization, the Curie temperature, and the magnetocrystalline anisotropy. Additive refractory elements such as Nb, MO, V, W, Zr. or Ti both suppress to some extent the formation of iron particles, which could act as nucleation centers for Bloch walls, and inhibit grain growth during sintering, which would reduce the coercivity (Buschow et al., 1995). In order to obtain a uniform, fine grain size, stable during thermal processing, a dispersion of a stable second phase may be used. The addition of Ti + C, which produces particles of T i c in the form of a fine precipitate, improves the microstructure of the Nd,Fe,,B phase without changing the properties of the base alloy (Branagan and McCallum, 1995). With an addition of 0.25 at.% Cu to a NdI6Fe,,B8 alloy during a powder metallurgy processing route a coercivity HC> 1000kA m-l was obtained while maintaining a remanence of 1.3 T (Ragg and Harris, 1987); a further improvement with H,= 3290 kA m- was reached by substituting Dy for 30 at.% Nd in melt-spun Nd-rich alloys (Harland and Davies, 1998). The Dy snbstjtution is effective also for high-temperature properties. For better corrosion resistance, it is necessary to change the Nd-rich grain boundary from a reactive to a non-reactive phase. It was found that a combined addition of very small quantities of Cu,CO and 0 in
98
Structure and Composition
(Nd,Dy)-Fe-B alloys substantially improves the coercivity, high-temperature properties and corrosion resistance, without a reduction of the remanence (Rim and Camp, 1996). Furthermore, the addition of V and Nb decreases the oxidation rate in sintered NdFe-B phases (Steyaert et af., 1998). Herbst (1991) has provided a detailed review of R2Fe14Bcompounds. A precise crystal and magnetic structure determination of Nd2Fe14Bwas made 011 a single crystal by neutron diffraction, in order to better understand the nature of the spin reorientation transition, which occurs below 14OK. It was shown that this transition, besides changing the anisotropy from an easy c axis to a canted structure, causes a lowering of the crystal symmetry, which changes to the monoclinic Cm space group (Wolfers et al., 1996). This monoclinic distortion was also confirmed by a single-crystal X-ray detennination (Obbade et al., 1996). 3.1.2 R2(Fe,M)17 Phases
During the last five years, much work has been carried out on the R2Fel7-,M, compounds with M =AI, Ga or Si, studying also the effects of interstitial H, C, and W atoms. Some regularities are observed. The maximum numher of iron atoms being replaced by A1 or Ga is x = 8-9, and generally this substitution, even in sniall quantity, stabilizes the Th,Zn,, structure (space group Rgm). This is also the case for the heavy rareearths. Since the atomic size of AI or Ga is larger than that of Fe, the guest elements firstly replace iron in the I8h site, and then in sites 6c and 18f,always avoiding Pd, the site with the smallest WignerSeitz cell. However, a limited quantity of Si enters into the host structure (x = 3-44), and only the f8h site is occupied. The introduction of Al or Ga atoms increases the cell volume, while the Si introduction causes a decrease. In any case, a significant effect on the magnetic properties of this family of ferromagneticphases is observed. An x increase leads to an increase of the Curie temperature Tc and in some cases to a change of the anisotropy from an easy plane to the easy c axis. As T, often passes through a maximum for M =AI or Ga, and the magnetization always decreases with x,a compromise has to be chosen. Though several rare-earths were examined, the most promising is Sm, for which a substitution of 2-3 A1 or Ga atoms for Fe reaches the maximum T, value and is more than enough to get uniaxial anisotropy (Cheng et aZ., 1995; Li and Morrish, 1995). The combined effect of the iron replacement and the insertion of interstitial atoms (€3, C, or N) is
particularly interesting. It is known that the interstitial compounds Sm2Fel,N, and Sm2Fe17Cy fy = 2-3) have excellent properties relative to the parent compound, such as a noticeable increase of Tc and spontaneous polarization, due to the lattice expansion, and a strong uniaxial magnetic anisotropy. Another example is the spectacular effect given by nitrogen insertion in CeJ?e17:a transition from an antiferromagnetic state with T, =225K in the starting compound to a ferromagnetic state with Tc = 728 K in Ce,Fe17N, (Tsnard et al., 1996). The problem with these materials, that they decompose when heated above 900 K, can be overcome by a substitution of iron with AI, Ga or Si, which stabilizes the phases and allows the use of high-temperature powder metallurgy to form dense magnets ( M U k et al., 1996). For Sm2Fel,66Gao.,4Ny an anisotropic field MA= 18,4T, a rernanence B,.= 12.3kG, a coercive field H, = 10kOe, a saturation magnetization M, = 137 emu g-l, with T, = 751 K (Yu et al., 1995) are realized; while for the Sm-rich srn2Fe,4si2N,,6 phase, HA= 157kOe, M, = 117emu g- I, and Tc=602K (Hadjipanayis et al., 1995) were obtained. Substitution of iron with other d-elements (Ti, V, Cr, Nb, Zr, etc.) was also examined by neutron diffraction (Luo et al., 1997), since the diversity in site preference of the substituents could give different modificaQons of the Fe-Fe exchange in the Fe sublattice. Moreover, it was found that the insertion of carbon in Nd2Fe,7-xTi, induces a change in the Ti site occupation, passing from the Fe 6c site to the 18f and 18h sites. Recently, new routes were explored in preparing these materials. A carbonitride Sm,Fe,,(C,N),,, obtained by the economical method of mechanically grinding Sm,f;e17 with pyrazine, without crystallization treatment, showed a coercivity of 0.7 T (Jakubowicz and Jurczyk, 1998). An HDDR-processing (hydrogenationdecomposition-desorption-recombination)applied to Sm,Fe,,M (M = AI, Ga and Si) and their nitrides and carbides achieved high coereivities up to 3 T and 2.3 T for nitrogenated and carburized samples, respectively (Kubis et a€., 1999). For special applications at temperatures higher than it is necessary to improve the long-term thermal stability of the material, and this can be obtained in cobalt-rich compounds. In sintered Sm(Coo92_xFe,C~.06Zr~.",7.~ magnets, with TCmIlOOK, an HC=8.3kOe at 400 "C is found when x = 7 wt,% Fe (Chen et al., 1998). More recently, the compound Sm2Co,,A12 shows Tc=904K, M,=61.3em~g-~at 1.5K and &HA=8.4T at room temperature (Shen et al., 1999).
Rare-Earth h4etaE Compounds
3.1.3 R(Fe,M)IzPhases The RFe,,-xMx compounds with ThMn,, structure form another family with interesting ferromagnetic properties. Here the presence of a metal M = Ti, V, Cr, Mn, N13, MO, Ta, W, Re, or Si partially replacing iron is necessary to stabilize the phase. Since the introduction of the above elements always has the negative effect of decreasing both the Curie temperature and the magnetization, a minimum substitution is called for, typical values being x = 1 for Ti, x= 1-2 for V, Cr, MO,W, or Si and x = 0.5-0.7 for Nb and Ta. Out of the three Fe sites present in the ThMn,, structure (space group 14/mmm), the guest element (except Si) always occupies the 8i site, since it has the largest Wigner-Seitz cell and only one rare-earth atom neighbor. This behavior is in agreement with the greater size of the guest element and with its low affinity for the electropositive rare-earths. On the contrary, Si generally prefers the other two Fe sites, Sj and SA allowing contacts with two rare-earth atoms (Lin et al., 1995; Moze and Buschow, 1996; Ayres de Campos er al., 1998; Tang et al., 1999). As for the R2Fet7compounds, the introduction of interstitial H, C, or N atoms improves the Curie temperature and magnetization. By neutron diffraction on RFe,2-xM,Hy phases, hydrogen was found to occupy firstly the octahedral 2b site (2R-4Fe) up to y = 1 and then partially the triangular bipyramidal 8h site (lR-2Fe/M-2Fe) (Obbade et al., 1997), while carbon and nitrogen, absorbed in a smaller amount, enter the 2b sitc too. These materials have the advantage of a good corrosion resistance, but it is difficult to reach high coercivities for application as permanent magnets. Among the rare-earths, Pr, Nd, Cd, Dy, or Y or a mixture of them seem to be the most promising. Good values were obtained for NdFe,Co3TiN, with T, = 860 K, M , = 150emu g-I and &,HA=7.2T (Fujii et al., 1995) and for (Pro.8Dyo.z)Fei,5V1.5Nywith Tc = 815K, M, = 146emug-I and hHA=13.2T (Yang et al., 1996). The new tantalum-containing series and their hydrides and carbides form stable phases only with the heavy rare-earths p e r t et al., 1999). A comprehensive review on the magnetic properties of the ThMn,,-type compounds was made by Suski (1996).
,
3.1.4 R,(Fe,M), Phases
The physical properties of the R3(Fe.M),, phases reflect the characteristics of their crystal structure,
99
composed of blocks of the rhombohedral Th,Zn17 and tetragonal ThMn,, structures in the ratio 1:1. So, in examining the ferromagnetic properties of this family, it was pointed out that the saturation magnetization of compounds containing a low concentration of stabilizing M element (M = Ti, V, or Cr) can be roughly calculated from a combination of the saturation magnetization of the component 217 and 1:12 units (Han et al., 1997; Fang et al., 1997). As already found for the ferromagnetic R2Fe,, and R(Fe,M)lz compounds, the introduction of interstitial H, C, or N atoms here too strongly enhances both the Curie temperature and the magnetization, in relation with a cell volume increase of 2.3-2.8% for H, 4 6 % for C and 5-8% for N insertion. The two positions occupied by nitrogen in the unit cell were determined by neutron diffraction and found to be the same as in 2:17 and 1:12 structures, midway between two neighboring rare-earth atoms, reaching octahedral 2R-4Fe environments (Hu et al., 1996). According to the available nitrogen sites, the maximum number of N atoms per formula unit should be four, but several authors reported also 5-8. A farmation of rare-earth nitrides was proposed as a possible explanation. Another important point concerns the easy magnetization direction in these phases. It is not unique, as in the 2:17 and 1:12 phases, where, when uniaxial anisotropy is reached, it is always found along the c axis. In the monoclinic 3:29 structure the directions [ 1021 and [ZOl] correspond to the c axis of the 2: 17 and 1:12 units, respectively (see Figure 12). Thus we have two directions in competition, the choice depending on several factors, such as the type of rare-earth, the Mstabilizing elements and the presence of interstitial atoms. For instance, for Srn,(Fel-,M$, phases with M=Ti, V, or Cr it was found that the parent compound and the hydrogenated compound show the easy magnetization direction along [ZOl], while nitrogenated compounds show [102] as the easy axis (Koyama et al., 1996). As for the 2:17 phases, Sm seems to be the most promising rare-earth. For Sm~(Fe~,~3~Tio.o~,)29N~ a Curie temperature T, = 750 K, a saturation magnetization M , = 140emu g-l, an anisotropy field p0HA= 12.8 T and a coercivity of 0.83 T were reported (Yang et al., 1994). For S r n 3 ~ e o , ~ ~ * C r oa. , ~ ~ ~ C coercive field of 0.80 T was obtained, with a good corrosion resistance in comparison with a powder sample of commercial Nd-Fe-B (Wang et al., 1997).
100
Structure and Composition
3.1.5 Tetragonal Phases Derived from the NaZn,,
M-doped R-Fe-B permanent magnets, improving the corrosion resistance and coercivity of the alloys. The crystal structure was firstly determined for La6Co,,Ga3 The body-centered-tetragonal CeNi, sSL,5 (tI56) struc(t180) with space group I4/mcm (Sichevich et al., 1985), ture (Bodak, 1979) represents a ternary derivative of where the Ga atoms completeiy fill the 4a site and the well-known cubic NaZnI3 (cF112) type, with partially substitute for CO in a 162 site. When the M partial ordering of the Ni and Si atoms. Several atoms occupy only the 4a site, the structure becomes RT13-xMxcompounds were studied with T = Fe, CO, perfectly ordered and corresponds to the Nd6Fe13Si or Ni and M=Al, Ga, Si, or Ge (Villars, 1997). In type (Allemand et d.,1990). This relatively complex some cases a small amount of p-element stabilizes the structure contains two rare-earth and four iron sites. cubic structure, with a transition to the tetragonal Regarding magnetic properties, in some cases structure with a greater amount. For instance, in the magnetic order near room temperature was observed, PrCo,,-,Si, system for x = 0, no 1:13 phase is formed, but with very low magnetization values. Antiferrowhile for 1 Cx$2 a NaZn13-type phase and for magnetic ordering of the four Fe and two R sublattices 2.5GxG4.5 a CeNi8,,Si45-type phase are found was determined by neutron diffraction, for example in (Huang et al., 1995). A perfectly ordered structure Pr6Fe,,Si and Nd6Fe,,Au (Schobinger-Papamantellos was also reported for compounds with stoichiometry et a).,1999). However, for some NdFe13Mcompounds 1:9:4, for example, LaFe9Si4 (Tang et al., 1994), conflicting magnetic results were reported, probably PrCo,Si, and NdCo,Si4 (Tang et al., 1995). The due to traces of ferromagnetic NdzFe17impurity. transition from cubic to tetragonal seems also to be Even though these materials appear to be of no induced by a prolonged annealing, as was observed in interest for commercial permanent magnets, a number the system LaCo,,-,Si, for as-cast samples with x = 2 of interesting properties has been pointed out. and 2.5 (Rao et al., 1994). Different magnetic anisotropy directions were found In the binary rare-earth-transition element systems by changing the M atom in Nd6FeI3Mphases: for a 1:13 phase exists only for LaCo13(NaZn13-type),with M=Cu, Ag, Au, Sb, or Bi a basal anisotropy, for interesting ferromagnetic properties: a magnetization M=In, T1, Sn, or Pb an axial anisotropy, and for of 126emug-I and a Curie temperature of 1290K M = Si, or Ge a basal anisotropy at low temperatures, (Velge and Buschow, 1968). The combined occurrence but an axial anisotropy at room temperature were of this transition-metal-rich compound and the reported (Hautot et al., 1998). The influence of the possibility of changing to a tetragonal structure upon non-magnetic element on the direction of the magnetic appropriate CO replacement, thus improving the anisotropy may be related to the particular features of magnetocrystalline anisotropy, have raised the interest this structure, where slabs of iron atoms alternate with of many researchers. Several types of element subslabs containing rare-earth and M atoms only. So, it is stitution have been tried, extending the study also to proposed by the same authors that the magnetic other rare-earths and to Fe-containing phases, and anisotropy may be controlled by the s-p hybridization nitrogenation of the alloys has also been explored, in of the Nd-M bonds. analogy with what was successfully attained for the Another property seemingly related to the crystal 2:17 phases (Huang et al., 1996; Liu et al., 1995). structure is the noticeable hydrogen uptake of these Unfortunately, in this case the introduction of nonphases, where 13-20 H-atoms per formula unit are magnetic elements was found to strongly deteriorate absorbed, with a strong anisotropy in the lattice the magnetic properties necessary for a good permaexpansion, an order of magnitude greater along c nent magnet. than along a (Leithe-Jasper et al., 1996). Hydrogenated compounds such as Nd6Fe&UH2*,@(Coey et al., 1994) and Nd,Fe,,GaW,, (Yartys et al., 1997) were reported 3.1.6 R&,-,M, Phases in which a cell volume expansion of about 15% was reached. Remembering that hydrogen is preferentially Phases with the composition R6Fe14-,M, = La, Pr, bound to rare-earth atoms, a favourabie situation for Nd, or Sm; M = 1B to 5B group element; x = 1-32 its absorption i s given by the several available have been studied by several groups over the I'dst few tetrahedral and octahedral interstitial sites inside the years (Weitzer et al., 1993; Weitzer et at., 1994; rare-earth slabs. Kuncser et al., 1997; Hautot et al., 1998). The interest In Figure 13 the structures of LaFe9S& and comes both from their structural and magnetic properNd6Fe13Siare represented together with the related ties and from their occurrence as secondary phases in
Rare-Earth Metal Compounds Ce(Mn,Ni),, (Kalychak et al., 1975) and the ferromagnetic Th4Fe13Sn5(Manfrinetti et al., 1997). These are four tetragonal structures characterized by the same slabs, formed by iron atoms in Nd6Fe13Siand Th,Fe,,Sn,, and by Fe/Si and MnlNi atoms in LaFe9Si4 and Ce(Mn,Ni), respectively. These slabs, which contain an extended network of bonds, highlighted in the figure by icosahedra around the most symmetric atom, alternate with slabs formed by the other atoms of the structure: Nd bicapped square antiprisms centered by Si in Nd,Fe13Si, Th octahedra centered by Sn and layers of Sn in Th4Fel,Sn5. In LaFe9Si4the Fe/% slabs are simply separated by a La layer, while in Ce(Mn,Ni),, the Mn/Ni icosahedra are joined by an edge giving rise to chains parallel to the c axis.
101
3.1.7 Permanent Magnet Preparation
A brief survey of the main manufacturing routes follows, mostly referring to Nd2Fe,,B production. The starting alloys are usually prepared by induction melting of the constituent elements in a protective atmosphere. The second step is magnetic hardening, to obtain fine-grained powders with high coercivity. The main processes, namely metallurgical sintering, melt spinning, HDDR process (Takeshita, 1993) and mechanical alloying are schematized in Table 2. In powder metallurgical sintering the cast alloy is milled, and the powder aligned, pressed and vacuum sintered. After a heat treatment the material is again machined and finally magnetized. The first milling step can be replaced by the so-called hydrogen
Figure 13 Four tetragonal structures characterized by the same slabs of centered icosahedra (dark stippling). The c axis is vertical. In Ce(Mn,Ni),, and LaFe,Si, the atoms belonging to the icosahedra are Mn/Ni or Fe and Si and circles represent Ce or La atoms. In Nd$e,,Si and Th,Fe,,Sn, the slabs of iron icosahedra alternate with slabs of Nd bicapped square antiprisms centered by Si (light stippling) or with slabs containing Th octahedra centered by Sn (light stippling) and layers of Sn (full circles)
Structure and Composition
102
decrepitation @ID) technique, where the coarsegrained alloy is broken up into a fine powder by exposure to hydrogen at around 1 bar pressure and room temperature. The metallurgical route is fairly expensive, but anisotropic magnets are achievcd (Buschow, 1998). Melt spinning i s a rapid-quenchingtechnique. A fine jet of molten alloy, directed onto the outer surface of a rapidly spinning wheel, produces a thin solidified ribbon. The ribbons are then typically ground down into a fine powder. This process produces isotropic powders, due to the random orientation of the grains (Croat, 1997). Another fairly economical way is the HDDR technique. proposed by Harris and McGuiness (1991) and Nakayama and Takeshita (1993). This process i s applied to the Nd-Fe-B magnets and consists essentially of four steps: hydrogenation of Nd2Fe14Bat low temperature, decomposition of Nd2Fe14BH, into NdH, +8 -i- Fe -i- Fe2B, desorption of H, gas from NdH,+I, and, finally, recombination of Nd + Fe + FezB into NdzFe14B.Because the formation of Nd2Fe,J3 grains in the last step is a solid-state reaction, it proceeds at a much lower rate than during normal solidification from the melt. The average grain size remains small, about 0.3pm, and leads to large coercivities.This process was also adapted for magnetic hardening of interstitial Sm2Fe,7N,compounds applying the RDDR treatment to Sm2Fe17N, before nitrogenation (Takeshita, 1993; Muller et al., 1996). In the mechanical alloying technique, the initial alloy preparation is avoided. Powders of the constituent elements are mixed and directly cold worked in a ball mill under an inert atmosphere to yield amorphous
or layered nanostructures, which are then heat treated to induce crystallization. This method, initially used for making highly coercive Sm,Fe,,N, magnets, was successfully applied also to Nd-Fe-B (Harada and Kuji, 1996). Melt-spun Nd-Fe-B may be further treated to produce dense, well-aligned magnets. Full density is achieved by hot pressing, typicaUy at I kbar and 973K under an inert atmosphere. A second hot pressing in a larger die cavity to allow deformation transverse to the press direction (die upsetting) introduces a preferred magnetization direction parallel to the press direction (Lee, 1985). An increase in remanence and fairly large energy products are thus obtained. The die upsetting technique can also be applied to mechanically alloyed material to produce dense anisotropic magnets. Considering the final manufacturing of these materials, bonded magnets constitute a significant fraction of large-scale production. They are prepared by encapsulating the magnetic powder in a resin or polymer and then compacting or molding the material to the final shape. The principal advantages are their good mechanical properties and the possibility of having shaped parts practically ready for use. Normally, the magnetic powder i s obtained by melt spinning, but more recently the HDDR technique was applied and anisotropic bonded magnets produced. Indeed, the grains formed by the HDDR process are aligned to some extent, in the same direction as the parent grains in the original precursor alloy. A proposed explanation for this ‘memory’ i s the presence of fine, undissociated Nd2Fel,B particles which serve as nucleation centres for the recrystallized HDDR grains (Coey and O’Donnell, 1997). A
Table 2 Scheme of the main processes used in permanent magnet preparation
SINTERING
RAPID QUENCHING
ALLOY PREPARATION
ALLOY PREPARATION
i
MILLING
I
I
MELT SPINNING
I
MECHANICAL ALLOYlNG ALLOY PREPARATION
1
HOMOGENIZATION
I
CRUSHING
CRUSHING
1
-1
SINTERING
HEAT TREATMENT
RDDR
i
1
PARTICLE ALIGNMENT AND PRESSXNG
1
HEAT TREATMENT
1
MAGNET
MAGNET POWDER
i
CRUSHING
1
MAGNET POWDER
MILLING OF ELEMENTS
1
HEAT TREATMENT
1
MAGNET POWDER
Rare-Earth Metal Compomds
comprehensive review on processing and applications of rare-earth iron permanent magnets can be found in the book edited by Coey (1996).
3.2 S ~ p e r c ~ ~ ~ c ~ ~ ) ~ t y Stekly and Gregory (1995) presented an overview of superconducting behavior and applications of intermetallics as appeared in the early 1990s. However, the discovery of superconductivity in rare-earth transition metal borocarbides (Nagarajan et al., 1994; Cava et al., 1994d) has opened up a new area of research. These compounds have the formula RNi2B2Cwrth R =: rareearth and have a crystal structure of the tetragonal LuNi,B,C (tI12) type (Siegrist et al., 1994b). The highest critical temperatures in this family were found for LuNi,B,C with T,= 16.6K and YNi,B,C with T, = 15.6K, both containing non-magnetic rareearths; but magnetic rare-earths such as Dy, Ro, Er, Tm also form the same phase, showing interesting phenomena associated with the interplay between superconductivity and magnetic order, never observed before for any type of superconducting material. Moreover, in a multiphase sample of composition YFd,B3Co superconductivity at 23 K was found. which is the highest T, reported for a bulk rare-earth intermetallic alloy (Cava et aZ., 1994c) and close to the record T, of Nb,Ge, 23.2K. A review of the vanety of effects shown by the RNi,B2C phases was made by Gupta (1997). Indeed, DyNi,B,C orders antiferromagnetically at TN= I1 K and becomes a superconductor at 6K, whereas ErNi2B2Cand TmNi2B2Cbecome superconductors at T, = 10.5 K and 11 K, respectively, but order antiferromagnetically at 6.5K and l S K , this long-range order coexisting with superconductivity. HoNi2B2C shows first a magnetic transition and then goes into the superconducting state at almost the same temperature, T C a8 K. So, three different behaviors are encountered with T,< TN,T,> TNand T,x TN.DyNi2B2Cpresents the rather rare transition to superconductivity in an already magnetically ordered lattice. Its magnetic structure is formed by layers of iso-oriented Dy moments stacked antiferromagnetically along the c axis. In ErNizBzC the Er moments undergo an incommensurate antiferromagnetic ordering, being again confined in the ab plane, whereas in TmNi,B2C they are oriented along the c axis. A remarkable feature of HoNi,B,C is its double re-entrant superconducting transition. This compound undergoes a first magnetic transition at 8.5 K and then goes into the superconducting state at almost the same N
103
temperature ( T c z8K). Next it re-enters the normal state at.., 6 K with a change in magnetic structure, and finally regains superconductivity below 5 K, with a commensurate antiferromagnetic order, similar to that of the Dy phase. This behavior, which is very sensitive to the chemical composition and heat treatment of the sample, has been observed for the first time under the condition of zero applied field. Modifications in the nature of the magnetic ordering, in particular the occurrence of incommensurate structures, seem to be responsible for the disappearance of superconductivity. Superconducting RPt2B,C phases, crystallizing with the same structure type, were reported for R = La, Pr and Y, with T, approximately 10K, 6K, and 10K, respectively (Cava et al., 1994b). Moreover, it was observed that, in light rare-earth phases. substitution of Au for Pt and appropriate annealing improve phase purity, while maintaining the same properties (Cava et al., 1994a). Superconductivity is found also in metastable, as-melted ScNi,B2C samples, with T, = 15-16 K (Ku et aZ., 1994), and in an annealed sample of YRu,B,C with T, = 9.7 K (Hsu et al., 1998). With reference to the multiphase U-Pd-B-C alloy showing Tc=23K, much work has been done to identify the superconducting phase. Some authors reached the common conclusion that the phase has a composiQon close to YPd2R2C,with a body-centered tetragonal structure, identical or closely related to the LuNi2B2Ctype, but, owing to the probable metastable nature of this phase, attempts to prepare single-phase specimens have not yet been successful (Jia et al., 1994; Sun et al., 1994; Tominez et aE., 1998). Other superconducting phases, but with different compositions, were found in borocarbide and boronitride systems. In RNiBC compounds, crystallizingwith the LuNiBC (tP8) structure (Siegrist et al., 1994b), upon substitution of Cu for Ni two phases were obtained, for example YNi,-,CqBC with TCz8.9K and LuNi,_,Cu,BC with T,w6.4 K (Gangopadhyay and Schilling, 1996). Superconductivity wa5 also found in the boronitride La,Ni,B,N, with T c z13K (Cava et al., 1994e). Physical properties measurements have been made on both polycrystalline samples and single crystals. Within two months after the discovery of superconductivity in these materials a method of singlecrystal growth based on a Ni,B ffux technique was developed at the Ames Laboratory and extensively applied in the following years to the study of all heavy rare-earth RNi2B2C phases (Canfield et al., 1997). Crystals of YNi2B2C and HoNi,B,C with centimeter dimensions were also grown using a floating-zone
104
Structure and Composition
method (Takeya et al., 1996). In recent years, alternative preparation routes were explored, in view of possible practical applications. Rapidly quenched YC(NiB), alloys with x = 2,3,4 were prepared by both splat-cooling and melt-spinning techniques, obtaining a critical temperature of 16K in a rapidly quenched YNi3B3C ribbon sample (Kim et al., 1998). In nanocrystalline or amorphous YNi,B,C, produced under different conditions of ball milling, it was observed that superconducting properties deteriorate upon milling and disappear when the amorphous phase is achieved. However, upon annealing the amorphous phase recovers its superconductivity (Ledig et al., 1999). Once again the superconducting state appears closely connected with the crystalline nature of the material. The LLIN~~B,C structure (Siegrist et al., 1994b) can be considered a filled-up variant of the well-known ThCr2Si2type, where the insertion of carbon into the rare-earth plane causes an expansion of the cell along the c axis. NaCl-type layers with composition LuC alternate with Ni2B, layers, where nickel is tetrahedrally coordinated by four boron atoms. Even though the structural framework is three-dimensional, the layered nature is pointed out by the results of Mossbauer studies on the isotypic DyNi,B2C (Cupta, 3999, suggesting that the Dy-C plane is insulating and electric conduction takes place within the Ni2B2layers. A drawing of the LuNi2B2C structure and related ThCr,Si,-type LuFe2B, (Villars, 1997) structure is reported in Figure 14. A detailed analysis of this structure in the RNi2B2C series and in other isotypic borocarbide compounds was made by Siegrist et al. (1994a). Strong anisotropy throughout the RNi,B2C series is observed: while the a constant of the tetragonal cell decreases regularly from La to Lu, following the normal lanthanide contraction, the c constant, on the contrary, increases. The proposed explanation for this anomalous trend is closely related to the structural features of these phases. The chain element B-C-B. aligned along the c axis, is very rigid, the B-C distanceo maintaining practically the same value of 1.47-1.48 A through the entire series. Thus, the radius variation of the rareearth will affect only the R-C distance within the NaCltype plane, regulating the a-constant value. without changing the B-C distance. On the other hand, since the Ni-B distance in the Ni2B, layers is also constant through the series, this constancy can be obtained only by an increase of the B-Ni-B tetrahedral angle on going from La to Lu, with a consequent thickening of the Ni2B2layer and c-constant expansion.
The structures of LuNiBC and La,Ni,B,N, (tI20) (Zandbergen et al., 1994b) are also given in Figure 14. It is easy to see that all these structures are related and the quaternary phases in particular can be described by a general formula (RX),(Ni,B,) with X = C or N. The first members of this family, LuNi2B2C,LuNiBC and La,Ni2B2N,have n = 1, 2 and 3, respectively, where n represents the number of NaC1-type layers inserted between two Ni,B, layers. Other quaternary nonsuperconducting phases found in these systems are LaNiBN, isotypic with LuNiBC (Zandbergen et al., 1994b) and Lu2NiBC2, identified by high resolution electron microscopy, and containing also a 6, layer between two lutetium layers (Zandbergen et al., 1994a).
3.3 Magmtoresistmce Magnetoresistance (MR) is the change in the electrical resistance that occurs under application of magnetic fields. The materials with giant magnetoresistance (GMR) and colossal magnetoresistance (CMR) show giant values (Aplp = 1040%) or colossal values ( A p [ p = 100% or more) of MR, usually in the vicinity of an order-disorder magnetic transition temperature. The applications of these materials can be as magnetic field sensors, read heads for hard drives, magnetic random access memories, or galvanic isolators. An explosion of research work has occurred on the perovskite-like manganites derived from LaMnO,. Partial substitution of lanthanum by an alkalineearth element produces spectacular effects, such as the appearance of a mixed valence state (Mn+3/Mn+4), the shifting of the magnetic ordering temperature, the competition between ferro- and antiferromagnetic interactions, the occurrence of a metal-insulator transition near the magnetic transition temperature, and correspondingly huge MR values. CMR materials are found also in rare-earth intermetallics, but their application is difficult because of the operating temperature, which is usually well below room temperature. Some examples are Tb,Ni,Si, with magnetoresistance change Aplp = 85% in an applied field B= 5T at T = 9 K (Mazumdar et al., 1997); CeAg,Sb, with Aplp = 40%, 3 = 5 T, T = 9 K (Guzik and Pierre, 1998); DyNiSb with A p / p = 32%, B = 4T, T = 2 K (Karla et al., 1998); Er0,4T~,&Oz with Ap/p = 300%, B = 10T, T = 0.5K (Hauser et al., 1999). The only rare-earth materials which show high MR values at slightly higher temperatures are Eu,,MnSb,, and Eu14MnBill,with magnetic ordering points at 92K and 32K, respectively (Chan et al., 1998).
105
Rare-Earth Metal Compounds
LaN LaN Ni2B2 LaN
LUC
LaN
LUC
Ni2B2 Y1
Ni262
Y I
LaN LaN
La3Ni2B2N3 Figure 14 Tetragonal structure types in superconducting borocarbides. The LuFe,B, structure (ThCr,Si,-type) is also reported for comparison. The c axis is vertical. Large open circles: Lu or La; medium full circles: Ni: small open circles: B; small full circles: C or N. The characteristic tetrahedral connection of Ni with four boron atoms and the 15°C or 8-N bonds are drawn
properties were favourab~e,the lattice t h e r ~ a lconductivity was too high and could not be s~~fficieiitly For many years compounds with the~oelectric lowered by part~alreplace~ent of the constituent as electric power g e n e ~ ~ t o r s ~lemeiits.Best results were o b t ~ ~ n ewith d a closely The chapter by V e ~ e r ~ i k o ~ ~ related class of compounds called ‘filled s k ~ t t e r u ~ i t e s ~ , 111 the preceding edition of ~ ~ ~ where t ~ ~ atoms ~ ~ up ~ the voids ~ lin the l ~ rare-earth fill provides a basis for both pr~ncip~es and s~utterLid~te s t r ~ ~ t u rThese e. phases have the ~ o r ~ u I a practical applications of the classical ther~oelectric RT4X12,where R = rare-earth r generally La-Nd, Sm, yed in a certain range of EL$, or Yb), T = Fe,Ru, or OS and X = P, As, or Sb, ?Se3alloys below 500 K, The series RCO~P,, is also known, while several ~~~~~~~~, and Si-Ge alloys above a n t i ~ o n i ~ eare s formed by the al~aline~earths (Ca, roblem of ther~oelectricdevices is and a r s ~ i i i d eby~ Sr, or Ba) and a few ~hosp~iides asured by the dimensionless figure aiid U (Villars, 1997; Even et al., 1995; Kaiser and ere T is the t e m ~ e r a t ~and ~ r eZ = ~~~x ~eitschko,1998), eebeck coefficient, ~i aiid x A p e c ~ I i a r iof t ~ several of these phases is that the R are the elect~icaland thermal ~ o ~ d u ~ t i respecy ~ t y ~ atoms ~~~~n~ the voids are loosely bound and show actually used, the m a x i ~ ~ m unusually large thermal d~spla~enient para~eters,For 1 Since good ther~noelectric an enhanced effect the R atom should find an oversized ~ ~ t e r shou~d ~ a ~ sbe n 1 i ~ ~ abetween y metals (high cage, and this is realized in some a ~ i t ~ ~ ~ n iIn des. elec~rical~ o ~ i d u ~ t ibut v i tlow ~ see be^^ c ~ e ~ c iand e~t thermal d ~ s p ~ a c e m parameter e~t in ermal co~ductivity~ and insula4Sb,2 and ~ a ~ e is4about ~ b four ~ ~ coeEcicnt arid low thermal times, and in ~ d ~mores than ~ six~times~ that~of 2 ~ o i i ~ ~ ~ c t i vbut i t yalso , low electrical c o n ~ ~ c t i v i tity ~ ~ the other Fe or Sb atoms. Figure 15 is a d ~ a w ~ of n gthe is clear why s e i ~ i c o i ~ d ~have ~ ~ treceived o r ~ most of the LaFe4Sb12structure. The structure is c o ~ p o s e dof a t r i d i n ~ e ~ s i oarray n ~ l of distorted FeSb, ~ o ~ i e r - ~ h a ~ i n ~ the search for more e ~ c i e materials, ~t comoctahedra. This arrangement gives rise to rectangu~ar ~ O L ~ such r ~ ~as S CoSb,, rSb3, c ~ s t a l l i ~ i ning planar Sb4 units which co~nectadjacent o c t ~ ~ e d r a , ucture (CoAs,-type, cf32) and form nearly icosahedral voids where the large rareet al., 1995; Tritt et al., 1996; earth atoms are located, being ~ ~ r r o ~ ibyi d12eSb. ~ A Caillst et al., 199 ever, although the electronic partial occupancy of the R site is often observed; for instatice, it is 9496 for LaFe,Sb,, arid 83% for NdFe,Sb, ?. The ~ n ~t s~ ~e vibr~tion ~~ l~ aa i ~~ ~ ~ l i tof u ~the e sR atoms in the ant~nionidec o ~ ~ o u n dconfer s a very favourable efTect in enhancing thernioelectric properties, since this ‘rattling’ motion gives rise to strong phonon scattering and, as a consequence, to a remarkable reduction of the lattice thermal ~ ~ n d L ~ c ~ ~ ity. On the other hand, these materials maintain relatively good electronic c~nductiont ~ o u g hthe framework formed by t ~ d n s i t i o n ~ ~ eand t a i anti~11ony andb ~ ~ atoms. The c o ~ ~ studied, o ~ ~ ~n ~~ e ~4 S =La - ~ F e 4 - ~ ~ 0with ~Sb~ ~ or Ce seen1 to be the most ~ ~ o m~ ~~ so i rand ~~ ~l eisner, ~ i 1995; Sales et al., 1996, 199’7; Chen et al., 1997; Sales, 1998), as theimal c ~ n d u c t i v is ~strongly ~~ reduced coinp~redto unfilled ~kutterudite§,being 6-8 times lower at room t e m p e ~ ~ t u with r e ~ values a p p r ~ a c ~ i nthose g of amorphous SiO,, The partial replacei~entof CO for Fe has two main effects: it inc~ease~ the ~ e e b e cc o~e ~ c i ~up nt Crystal ~ t r u ~of ~ the u ~ f ~i l l e d - ~ ~ ~ ~ t t ~ r u d i ~ ~ to a ~ ~ ~ b s t level i t not ~ ~ exceeding ~ o ~ x=2.5, and it (~134).Sb, ~ ~ ~ l i around ~ d r athe Fe atoms are decreases the rare-earth o c c ~ p ~ n c factor. y For . Large circles: La; small circles: Sb ~~~~~~~~~~~
I
Rare-Earth Metal Compounds
Lao,9Fe,CoSb,, and Ceo,9Fe,CoSb,, a Seebeck coefficient Sx 100,uV/IC, an electrical conductivlty o x 1000R-l cm-I, a lattice thermal conductivity ~ ~ x 0 . 0W 1 5K-l cm-' are reported at room temperature, and for the Ce compound ET is greater than one at temperatures above 700K. A similar behavior was recently observed in the nickel-substituted compounds Ce,Fe,-,Ni,Sb,,, where, upon substitution, the Seebeck coefficient increases up to 180pV/K at x = 1.1, but for x = 1,5 becomes small and negative, while the Ce content regularly decreases (Chapon et aE., 1999). Filled skutterudites MxCo4Sb,2with partial occupation of the voids by M = Sn or Pb were also prepared by a high pressure-high temperature treatment (Takizawa et al., 1999). An improvement of the electric and thermal properties of these materials should be achieved by varying the carrier concentration througb an accurate electronic balance, of course preserving the low lattice thermal conductivity, which is a peculiarity of the filled skutterudites. This should be possible, because the lattice contribution to the thermal conductivity is higher than the electronic term and does not depend strongly on carrier concentration.
-
3.5 Energy Storage The commercial production of Ni-MH batteries started nearly ten years ago as described by SchEapbach et al. (1995) and quickly increased, reaching 570 million cells in 1997, representing 40% of the market for small rechargeable batteries. As is well known, this system is based on the noticeable property of the intermetallic compound LaNi, to absorb, rapidly and reversibly, more than six hydrogen atoms per formula unit. The electrochernical reactions involved are: Negative Positive Overall
charge M + H 2 0+ e- 2 MH + OHdischarge
Ni(OH),
+OH- 2 NiOOH +H 2 0+e-
M + Ni(OH), 2 MH + NiOOH
The charge-discharge mechanism is simply a movement of hydrogen between a metal hydride (MH) electrode and a nickel hydroxide (Ni) electrode in an alkaline electrolyte, typically 6M KOH. The MH electrode is formed by an alloy powder pasted onto a nickel-plated steel sheet. The alloy
107
composition is MmNi, & O ~ , ~ M ~ ~ ,37, A where ~ , La is substituted by the more economical mischmetal (Mm) and Ni is partly replaced by elements chosen for improving the performance of the cell. In particular, the combined substitution of cobalt and aluminum for Ni results both in corrosion inhibition and in considerable reduction of the volume expansion from hydrogen absorption, which is about 24% for pure LaNi,, and is responsible for a pulverization of the alloy and a consequent loss o f capacity during repeated charge-discharge cycles (Sakai et al., 1995). Moreover, cobalt lowers the hardness of the alloy, another important point for a high-cycling stability (Chartouni et al., 1996). The presence of Mn is useful from several aspects. During the activation cycles the alloy is cracked by the hydridingldehydridingvolume changes, allowing diffusion of the electrolyte solution inside the particles. This operation is required prior to the alloy cycling process between the hydride and metal phase. The addition of Mn to the alloy produces small equiaxed grains in the ingot, facilitating cracking and improving rate capability. Manganese, together with aluminum, contributes also to a decrease in the hydrogen pressure in the alloy, favouring hydrogen diffusion inside the cell and maintaining charge efficiency. Other essential requirements for an MH electrode are a homogeneous microstructure, obtained by a suitable heat treatment of the MmNi,-based alloy produced by rapid solidification, and a large, wellconducting surface area obtained on forming a nickelrich layer. Practical capacities per volume are in the range 9001200mA-h/cm3, with an energy density per unit volume of 360 W-h/l and an energy density per unit weight of 70 W-h/kg. These cells are used in various portable appliances such as cellular phones, video cameras, and lap-top computers, where compactness is very important; and recently, a high-power Ni-MH battery has been placed on the power-tool market. Other applications are in the field of the electric vehicles, where high power Ni-MH batteries either are directly employed, or in fuel cell systems assist a methanol reformer in hydrogen production (Sakai et al., 1999). Basic research and technology for applications have been closely connected in this field, more so thm in others. Research and development efforts are now directed on the one hand, to improving technology and producing light and compact batteries with high power, and on the other hand toward reducing costs. The cost per cell has greatly decreased to about onethird of its initial value, but still remains high,
108
Structure and Composition
primarily because of the incorporation of cobalt in the alloy. A partial substitution of CO with Fe, which maintains good cycle stability and rate capability, was proposed by Ziittetel et al. (1997), while an alioy with the cobalt completeiy replaced by a mixture of Cu, Fe, and Cr, and a small fraction of the rare-earths replaced by Ti showed good cycling stability although with a lower capacity (Hu, 1999). Chemical treatments of the electrode surface were also studied. By dipping and stirring the alloy in an aqueous HCl solution, several beneficial effects are observed: a partial solution of rare-earihs and Mn generates micropares, which increase the specific surface area and activate the discharge reaction; the outer surface area is enriched in Ni and CO metal, enhancing electric conductivity and catalyzing hydrogen absorption and desorption reactions (Imoto et at., 1999). Treatment of the allay powder with KOH solution at 80 "Cproduces a surface containing Mm hydroxides, Mn oxides and metallic Ni, and prevents the alloy bulk from further corrosion, enhancing the cycle life (Ikoma e t al., 1999). Moreover, it was found that surface deposition of trace amounts of platinum-group metals on the MH electrode considerably improves absorptioafdesorption rates and increases resistance to deactivation ('Willey e t at., 1999). The more than 50 papers on the MH electrodes and Ni-MH batteries, presented in the 1998 International Symposium on Metal-Hydrogen Systems - Fundamentals and Applications, held in Hangzhou, China, and published in J. A l b y ~Compd., 293-295 (1999), form a basis for updating the research and bibliography, and demonstrate the continuing interest in this subject. The search for new energy-storage alloys of high performance. and IOW costs is still an open field.
Ac~nQwled~e~e~t Dedicated to Professor Erwin ParthC.
Note: The sequence of the rare-earths (Sc, Y, La, Ce, Pr, Nd, Pm,Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu) is readily recalled with the following mnemonic:
-Some young Ladies gan't
e t Nickels groperly into
-Slot-machines: Every S r l Tries Daily However Every Time u o u Look.
J. C. Chaston
References Allemand, J., Letant, A., Moreau, J. M., Nozieres, J. P., and Perrier de la Bathie, R. (1990). J. Less-Common Mer., 166, 73. Ayres de Campos. J., Ferreira, L. P., Godinho, M., Gil, J. M., Mendes, P. J., Ayes de Campos, N., Ferreira, I. G., Bououdina, M., Bacmann, M., Soubeyroux, J. L., Fruchart, D., and Collornb, A. (1998). J. Ph-vs.: Condensed Matter, 10,4101. Bodak, 0. I. (1979). Sov. Phys. CrystaZZogr.,24, 732. Branagan, D. J., and McCallum, R. W. (1995). J. Alloys Compd., 218, 143, Buschow, K. H. J. (1998). Permanent-Magnet Materials and their Applications. Trans Tech Publications, Switzerland, p. 24 and references cited therein. Buschow, K. H. J., Feijen, F. H., and de Kort, K. (1995). J. M e n . Magn. Mater., 140-144,9. Caillat, T.,Borshchevsky, A.. and Fleurial, J.-P. (1996). J. Appl. Phys., 80, 4442. Canepa, F., Manfrinetti, P., Pani, M., and Palenzona, A. (1996). J. Alloys Compd., 234,225. Canfield, P. C., Bud'ko, S. L., Cho, B. K., Beyermann, W. P., and Yatskar, A. (1997). J. Alloys Compd., 250, 596 and references cited therein. Cava, R. J., Batlogg, B., Krajewslu, J. J., Peck Jr., W. F., Siegrist, T., Fleming, R. M., Carter, S., Takagi, H., Felder, R. J., and van Dover, R. B. (1994a). Physica C, 226. 170. Cava, R. J., Batlogg, B., Siegrist,T., Krajewski, J. J., Peck Jr., W. F., Carter, S., Felder, R. J., Takagi, H., and van Dover, R. B. (1994b). Phys. Rev. B, 49, 12384. Cava, R. J., Takagi, H., BatIogg, B., Zandbergen, H. W., Krajewski, J. J., Peck Jr., W- F., van Dover, R. B., Felder, R. J., Siegnst, T,, Mizuhashi, K., tee, J. O.,Eisaki, H., Carter, S. A., and Uchida, S . (1994~).Nature, 367, 146. Cava, R. J., Takag3H., Zandbergen, H, W., Krajewski, J. J., Peck Jr., W. F., Siegrist, T., Batlogg, B., van Dover, R. B., Felder, R. J., Mizuhashi, K., Lee, J, O., Eisakt, H., and Uchida, S. (1994d). Nature, 367, 252. Cava, R. J., Zandbergen, H. W., Batiogg, B., Eisaki, H., Takagi, H., Krajewski, J. J., Peck Jr., W. F., Gyorgy, E, M., and Uchida, S . (1994e). Nature, 372, 245. Chan, J. Y., Kauzlarich. S . M., Klavms, P., Sheiton, R. N., and Webb, D. J. (1998). Phys. Rev. B, 57, R8103. Chapon, L., Ravot, D., and Tedenac, J. C. (1999). J . Alloys Compd.,282, 58. Chartouni, D., Meli, F., Zilttel, A., Gross, K., and Schlapbach, L,(1996). J. Alloys Compd., 241, 160. Chen, B., Xu, J.-H., Uher, C., Morelli, D.T., Meisner, G. P., Fleurial, J.-P., Caillat, T., and Borshchevsky, A. (1997). Phys. Rev. B, 55, 1476. Chen, C . H., Walmer, M. S., Walmer, M. H., Liu, S., Kuhl, E., and Simon, G. (1998). J. Appl. Phys., 83, 6706. Cheng, Z. H., Shen, B. G., Wang, F. W., Zhang, J. X., Gong, H. Y., and Zhan, W. S. (1995). J. Phys.: Condensed Matter, 7, 3195.
Rare-Earth Me fa[ Compounds
Coey J. M. D. (ed.) (1996). Rare-Earth Iron Permanent Magnets. Oxford University Press, New York. Coey, J. M. D., Qi, Q,, Knoch, K. G., Leithe-Jasper, A., and Rogl, P. (1994). J . Magn. Magn. Mater., 129, 87. Coey, J. M. D., and O'Donnell, K. (1997). J. Appl. Phys., 81, 4810. Corbett, J. D., Garcia, E,, Guloy, A. M., Hurng, W.-M., Kwon, Y.-U,, and Leon-Escamilla, E. A. (1998). Chem. Mater., 10, 2824. Courtois, D., Givord, D., Lambert-Andron, B., BourgeatLami, E,, Amako, Y., Li, H. S., and Cadogan, J. M. (1998). J . Magn. M a p . Mater., 189, 173. Croat, J. J. (1997). J. Appl. Phys., 81, 4804. de Boer, F. R., Boom. R., Mattens, W. C . M., Miedema, A. R., and Niessen, A. K. (1988). Cohesion in Metals. Transition Metal Alloys, North-Holland, Amsterdam. Due, N. H. (1997). In Handbook on the Physics and Chemistrv of Rare Earths, Vol. 24 (eds K. A. Gschneidner Jr. and L. Eyring). Elsevier, Amsterdam, p. 339. Evers, C. B. H,, Jeitscbko, W,, Boonk, L., Braun, D. J., Ebel, T., and Scholz, U. D. (1995). J. Alloys Compd., 224, 184. Fornasini, M. L., and Merlo, F. (1995). J. Alloys Compd., 219, 63. Francois, M., Venturini, G., Malarnan, B., and Roques, B. (1990). J. Less-Common Met., 160, 197. Fuerst, C. D., Pinkerton, F. E., and Herbst, J. F. (1994a). J. Appl. Phys., 76, 6144, Fuerst, C. D., Pinkerton, F. E., and Herbst, J. F. (1994b). J. Magn. Magn, Mater., 129, L115. Fujii, H., Miyazaki, Y., Tatami, K., Sun, H., Morii, Y., Akayama, M, and Funahashi, S. (1995). J. Magn. Magn. Mater., 140-144, 1089. Gangopadhyay, A. K., and Scbilliing, J. S. (1996). Ph,ps. Rev. B, 54, 10107. Gignoux, D., and Schmitt. D. (1995). In Handbook on the Physics and Chemistry of Rare Earfhs, Vol. 20 (eds K. A. Gschneidner Jr. and L. Eyring). Elsevier, Amsterdam, p293. Gladyshevskii E. I., Bodak, 0. I., and Pecharsky, V. K. (1990). fn Handbook on the Physics and Chemisfyof Rare Earths, Vol. 13 (4s. K. A. Gschneidner Jr. and L. Eynng), Elsevier, Amsterdam, p. 1. Gladyshevskii, R. E., Cenzual, K., and Parthi, E. (1992). J. Alloys Compd., 189, 221. Guloy, A. M., and Corbett, J. D. (1992). Z . Anorg. Allg. Chem., 616, 61. Guloy, A, M., and Corbett, J. D . (1993). Jnorg. Chem., 32, 3532. Guloy, A. M., and Corbett, J. D. (1994). J. Solid State Chem., 109,352. Gupta, L. C. (1997). J . Alloys Compd., 262-263, 22 and references nted thereia. Guzik, A., and Pierre, J. (1998). J. Alloys Compd., 264, 8. Hadjipanayis, G. C., Zheng, Y. H., Murthy, A. S., Gong, W., and Yang, F. M. (1995). J. Alloys Compd., 222,49.
109
Han, X. F., Yang, F. M., Pan, H. G., Wang, Y. G., Wang, J. L., Liu, H. L., Tang, N., Zhao, R. W., and Li, H. S . (1997). J. Appl. Phys., 81, 7450. Harada, T., and Kuji, T, (1996). J. Alloys Compd., 232,238. HarIand, C.L., and Davies, H. A. (1998). J. Alloys Compd., 282, 37. Harris, I. R., and McGuiness, P. J. (1991). J . Less-Common Met., 172-174, 1273. Hauser, R., Bauer, E., Gratz, E., Michor, H., Hilscher, G.. and Markosyan, A. S . (1999). J. Magn. Magn. Mater., 196197,647. Hautot, D., Long, G. J., Grandjean, F., de Groot, C. H., and Buschow K. H. J. (1998). J. Appl. Phys., 83, 1554. Herbst, J. F. (1991). Rev. Mod. Phys., 63, 819. Hohnke, D. and Partbe, E. (1966). Acta Crystallogr., 20, 572. Hsu, Y. Y., Chang, H. C., and Ku, €3. C. (1998). J. Appl. Phys., 83, 6789. Hu, J., Mei, L., Luan, K., Yang, F., Pan, H., Han, X., and Wang, J. (1998). J. Alloys Compd., 278, 17. Hu, W.-K. (1999). J. A l h y ~Compd., 289, 299. Hu, Z., Yelon, W. B., Kalogrrou, O., and Psycharis, V. (1996). J. Appl. Phys., 80, 2955. Huang, F., Liang, J. K., Liu, Q. L., Chen, X. L., and Hou, G. Y. (1999). J. Appl. Phys., 85, 6686. Huang, M. Q., Wallace, W. E., Obermyer, R. T., Sirnizu, S., and Sankar, S . G. (1995). J. Magn. Magn. Mater., 151, 150. Huang, M. Q., Wallace, W. E., Obermyer, R. T., Simizu, S., McHenry, M., and Sankar. S. G. (1996). J. Appl. Phys., 79, 5949. landelli, A., and Palenzona, A. (1979). In Handbook on the Physics and Chemistry of Rare Earfhs, Vol. 2 (eds. K. A. Gschneidner Jr. and L. Eyring), Elsevier, Amsterdam,p. 1. Ibarra, M. R., Morellon, L., Blasco, J.. Pareti, L., Algarabel, P. A., Garcia, J., Albertini, F., Paoluzi, A., and Turilli, G. (1994). J. Phys.. Condensed Matter, 6, L717. Ikoma, M., Kornori, K., Kaida, S., and Iwakura, C. (1999). J. Alloys Compd., 284, 92. Imoto. T., Kato, K., Higashiyama, N., Kimoto, M., Itoh, Y., and Nishio, K. (1999). J. AZloys Compd., 282, 274. Isnard, 0.. Miraglia, S., Fruchart, D., Giorgetti, C., Dartyge, E., and Ktill, G. (1996). J. Phys.: Condknsed Mutier, 8, 2437. Ivanova, G. V., Makarova, G. M., Shcherbakova, Ye. V., Belozerov, Ye. V., and Yermolenko, A. S . (1996). J . Alloys Compd., 243, 95. Jakubowicz, J, and Jurczyk, M. (1998). J. Alloys Compd., 266, 318. Jia, C. L., Xu, Y.-H., Beyss, M., Peter, F., and Urban, IC. (1994). Physica C, 229, 325. Kaiser, J. W., and Jeitschko, W, (1999). J. Alloys Compd., 291, 66. KaloBrou, O., Psycharis, V,,Saettas, L., and Niarchos, D. (1995). J. Magn. Magn. Mater., 146, 335. Kalychak, Ya. M., Akselrud, L. G., Yarmolyuk, Ya. P., Bodak, 0. I., and Gladyshevskii, E. 1. (1975). Sov. Phys. Crystallogr., 20, 639.
110
Structure und Composition
Karla, I., Pierre, J., and Skolozdra, R. V. (1998). J. Alloys Compd., 26542. Kim, A. S., and Camp, F. E. (1996). J. Appl. Phys., 79, 5035. Kim, K.-S., Yu, S.-C., Strom, V., and Rao, K. V. (1998). J. Magn. Magn. Mater., 177-181, 553. Koyama, K., Fujii, H., and Suzuh, S. (1996). J. Magn. Magn. Mater., 161, 118. Ku, H. C., Lai, C. C., You, Y. B., Shieh, J. H., and Guan, W. Y. (1994). Phys. Rev. B, 50, 351. Kubn, M., GutAeisch, O., Gebel, B., Miiller, K. H., Harris, I. R., and Schultz, L. (1999). J. Alloys Compd., 283, 296. Kmcser, V., Rosenberg, M., Buschow, K. H. J., and Filoti, G. (1997). J Alloys Compd., 255, 60. Kussmann, D., Hoffmann, R.-D. and Pottgen, R. (1998). 2. Anorg. Allg. Chem., 624, 1727. Kwon, Y -U., Rzemik, M. A,, Guloy, A., and Corbett, J. D. (1990). Chem. Mater., 2, 546. Ledig, L., Hough, D., Oertel, C.-G., Eckert, J., and Skrotzki, W. (1999). J. Alloys Compd., 285, 27. Lee, R. W. (1985). Appl. Phys. Left.,46, 790. Leithe-Jasper, A., Skomski, R., Qi, Q., Coey, J. M. D., Weitzer, F., and Rogl, P. (1996). J. Phys.: Condensed Matter, 8, 3453. Leon-Eseamilla, E. A., and Corbett, J. D. (1998). J. Alloys Compd., 265, 104. Li, H. S., Cadogan, J. M., Davis, R. L., Margarian, A., and Dunlop, J. B. (1994). Solid. State Commun., 90, 487. Li?H. S., Xu, J. M., Courtois, D., Cadogan, J. M., Liu, H. K., and Dou, S. X. (1996). J. Phys.: Condensed Matter, 8, 2881. Li, Z . W., and Mornsh, A. H. (1995). J. Phys.: Condensed Mutter, 7, 6727. Lin, C., Xu, X. F., Sun,Y. X., Liu, Z. X.. Yang, J. L., Zhang, B. S., Du, H. L., and Ye, C. T. (1995). J. Phys.. Condensed Matter, 7, 7293. Liu, J. P., Tang, N., de Boer, F. R., de Chatel, P. F., and Buschow, I<. H. J. (1995). J. Magn. Magn. Mater., 140144, 1035. Luo, H., Hu, Z., Chen, M.. Yelon, W. B., Marasinghe, G. K., Ezekwenna, P. C.,James, W. J., Chang, W C., and Tsai, S, H. (1997). J. AppL Phys., 81, 4542. Manfiinetti, P., Canepa, F., Palenzona, A., Fornasmi, M. L., and Gianniru, E. (1997). J. Alfoys Compd., 247, 109. Margarian, A., Li, H. S., Dunlop, J. B., and Cadogan, J. M. (1996). J. Alloys Compd., 239, 27. Massalski, T. B., Okamoto, H., Subramanian, P R., and Kacpnak, L. (eds) (1990). Binary Alloy Phase Diagram, 2nd Ed., ASM International, Matenals Park, OH. Mazumdar, C., Nigam, A. K., Nagarajan, R., Gupta, L. C., Chandra, G., Padalia, B. D., Godart, C . , and Vijaynraghaven, R. (1997). J. Appl. Phys., 81,5781. Mendoza, W. A., and Shaheen, S. A. (1999). J. Magn. Magn. Mater., 195, 136. Merlo, F. (1988). J. Phys. I;: Met. Phys., 18, 1905. Merlo, F., and Fornasini, M. L. (1993). J. Alloys Compd., 197. 213.
Merlo, F., Pani, M., and Fornasini, M. L. (1996). J.Alloys Compd., 232, 289. Merlo, F., Pani, M., Canepa, F., and Fornasini, M. L. (1998). J. Altoys Compd., 264, 82. Morelli, D. T., and Meisner, G. P. (1995). J. Appl. Phys., 77, 3777. Moze, O., and Buschow, K. H. J. (1996). J. Alloys Compd., 233, 165. Miiller, K.-H., Cao, L., Dempsey, N. M., and Wendhausen, P. A. P. (1996). J . Appl. Phys., 79, 5045. Nagarajan, R., Mammdar, C., Hossain, Z., Dhar, S . K., Gopalttkrishnan, K. V,,Gupta, L. C., Godart, C., Padalia, B. D., and Vijayaraghavan, R. (1994). Phys. Rev. Lett., 72, 274. Nakayama, R., and Takeshlta, T. (1993). J . Alloys Compd., 193, 259. Obbade, S., Wolfers, P., Fruchart, D., Argoud, R., Muller, J., and Palacios, E. (1996). J. AOoys Compd., 242, 85. Obbade, S., Fruchart, D., Bououdina, M., Miraglia, S., Soubeyroux, J. L., ,and Isnard, 0. (1997). J. Alloys Compd., 253-254,298. Parthk, E., and Chabot, B. (1984). In Handbook on the Physics and Chemistry of Rare Earths, Vol. 6 (eds K. A. Gschneidner Jr. and L. Eyring), Elsevier, Amsterdam, p, 113, Pettifor, D. G. (1986a). J. Phys., C19, 285. Pettifor, D. G . (1986b). J Phys., C19, 315. Pottgen, R. (1995). J. Mater. Chem., 5, 505. Pottgen, R. (1996). J. Mater. Chem., 6,63. Pottgen, R., Hoffmann, R.-D., Miillmann, R., Mosel B. D., and Kotzyba, G . (1997). Chem. Eur. J., 3, 1852. Prots’, Y . M., Pottgen, R., and Jeitschko, W. (1998). 2. Anorg. Allg. Chem., 624, 425. Ragg, 0. M., and Hams, I. R. (1997). J. Alloys Compd., 256, 252. Rao. G. H., Liang, J. K., Zhang, Y . L., Cheng, X. R., and Tang, W. H. (1994). Appl. PhyLq.Lett., 64, 1650. Richter, M. (1998). J. Phys. Dr Appl. Phys., 31, 1017. Sakai, T., Matsuoka, M., and Iwakura, C. (1995). In Handbook on &hePhysics and Chemistry of Rare Earths, Vol. 21 (eds K. A. Gschneidner Jr. and L. Eyrmg). Elsemer, Amsterdam, p. 133. Sakai, T., Uehara, I. and Ishikawa, H. (1999). J. Alloys Compd., 293-295, 762. Sales. B. C. (1998). MRS BuEleih, January, p. 15. Sales, B. C., Mandms, D., and Williams, R. K. (1996). Science, 272, 1325. Sales, B. C., Mandrus, D., Chakoumakos,B. C., Keppens, V., and Thompson, J. R. (1997). Phys. Rev. B, 56, 15081. Schlapbach, L., Meli, F., and Ziittel, A. (1995). In Intermetallic Compounds: Principles and Practice, Vol. 2 (eds J. H. Westbrook and R. L. Fleiseher). Wiley, Chtchester, ch. 21, p.475. Schob, 0. and Parthi-, E. (1965). Acta Crystallogr., 19, 214. Schobinger-Papamantelfos,P., Buschow, K. H. J., de Groot, C. H., de Boer, F. R., Biittger, G., and Ritter, C. (1999). J . Phys.: Condensed Matter, 11,4469.
Rare-Earth Metal Compounds Sharp, J. W., Jones, E. C., Williams, R. K., Martin, P. M., and Sales, B. C. (1995). J. Appl. Phys., 78, 1013. Shen. B. G., Cheng, 2. H., Zhang, S. Y., Wang, J. Y., Liang, B.. Zhang, H. W., and Zhan, W S. (1999). J. Appl. Phys., 85. 2787. Sichevich, 0. M., Lapunova, R. V., Sobolev, A. N., Gnn, Yu. N., and Yarmolyuk, Ya. P (1985). Sov. Phys. Crystallogr., 30, 627. Siegrist, T., Cava, R. J., Krajewski, J. J., and Peck Jr., W. F. (1994a). J. Alloys Compd., 216, 135. Siegrist, T., Zandbergen, H. W., Cava, R. J., Krajewski, J. J.. and Peck Jr., W. F. (1994b). Nature, 367, 254. Stadelmaier, H. H. (1984). Z . Metallkd., 75, 227. Stadelmaier, H. H., and Reinsch, B. (1995). In Zntermetallic CompoundAy:Principles and Practice. Vol. 2 (eds J. H. Westbrook and R. L. Fleischer). Wiley, Chichester, ch. 14, p. 303. Stekly, Z. J. J., and Gregory, E. (1995). In ZntermetaNic Compounds: Principles and Practice, Vol. 2 (eds J. H. Westbrook and R. L. Fleischer). Wiley, Chichester, ch. 16, p.351. Steyaert, S., Le Breton, J.-M.. and Teillet, J. (1998). Solid State Commun., 108, 33. Sun, Y. Y., Rusakova, I., Meng, R. L., Cao, Y., GautierPicard, P., and Chu, C . W. (1994). Physica C, 230,435. Suski, W (1996). In Handbook on the Physics and Chemistry of Rare Earths, Vol. 22 (eds K. A. Gschneidner Jr. and L. Eyring). Elsevier, Amsterdam. p. 143. Takeshita, T. (1993). J. Alloys Compd., 193, 231. Takeya, H., Kadowaki, K., Hirata, K., and Hirano, T. (1996). J. Alloys Compd., 245, 94. Takizawa, H., Miura, K., Ito, M., Suzuki, T., and Endo, T. (1999). J. Alloys Compd., 282, 79. Tang, N.. Zhao, 2. G., Briick, E., de Boer, F. R., Buschow, K. H. J., Wang, J. L., and Yang, F. M. (1999). J. Phys.. Condensed Matter, 11. 5855. Tang, W., Liang, J., Chen, X., and Rao, G. (1994). J . Appl. Phys., 76, 4095. Tang, W., Liang, J., Zhao, Y., Guo, Y., Rao, G., and Yang, F. (1995). J. Alfoys Compd., 226, 139. Teatum, E., Cschneidner Jr., K. A., and Waber, J. (1968). Rep. LA-4003, NTIS, Springfield, VA. Tominez, E., Berger, P., Alleno, E., DCcamps. B., Schiffmacher, G., Bohn, M., and Godart, C. (1998). J. Alloys Compd., 275-277, 123. Tntt, T. M., Nolas, G. S., Slack, G. A., Ehrlich, A. C., Gillespie, D. J., and Cohn, J. L. (1996). J . Appl. Phys., 79, 8412. Vedernikov, M. V. (1995). In Zntermetallic Compounds: Principles and Practice, Vol. 2 (eds J. H. Westbrook and R. L. Fleischer). John Wiley & Sons Ltd, Chichester, Chapter 20, p. 453.
111
Velge, W. A. J. J., and Buschow, K. H. J. (1968). J. Appl. Ph-vs.. 39, 1717. Ventunni, G., Francois, M., Malaman. B., and Roques, B. (1990). J . Less-Common Met., 160, 215. Vert, R., Bououdina, M., Fruchart, D., Gignoux, D., Kalychak, Y M.. and Skolozdra, R. V. (1999). J. Alloys Compd., 287, 38. Villars. P. (1 995). In Znlermetallic Compounds: Principfes and Practice, Vol. 1 (eds J. H. Westbrook and R. L. Fleischer). Wiley, New York, p. 227 and references cited therein. Villars, P. (1997). Pearson’s Handbook Desk Edition. Crystallographrc Data for Zntermetallic Phases, Vols 1-2, ASM International, Matenals Park, OH. Wang, Y. Z., Hu, B. P., Liu, G. C., Li, H. S., Han, X. F.. and Yang, C. P. (1997). J , Phys.. Condensed Matter, 9, 2793. Weitzer, F., Leithe-Jasper, A., Rogl, P., Hiebl, K., NoEl, H., Wiesinger. G.. and Stemer, W. (1993). J. Solid State Chem., 104, 368. Weitzer, F., Leithe-Jasper, A., Rogl. P.. Hiebl, K., Rainbacher, A., Wiesmger, G., Steiner, W., Fnedl. J., and Wagner, F. E. (1994). J. Appl. Phys., 75, 7745. Willey, D. B., Harris, I. R., and Pratt, A. S. (1999). J. Alloys Compd., 293-295, 613. Wolfers, P., Obbade, S., Fruchart, D., and Verhoef, R. (1996). J. Alloys Compd., 242. 74. Yang, C. P., Wang, Y. Z., Wu, G . H., Ran. X. F., Hu. B. P., Ma, C. L., Chen X. Y., and Zhu, J. (1999a). J. Phys.: Condensed Matter. 11, 6843. Yang, C. P., Wang, Y. Z., Wu, G. H., Hu. B. P., Du, J., Han, X. F., Wang, J. L., Jiang, 2. L., and Zhu, J. (1999b). Solid State Commun., 111, 247. Yang, F. M., Nasunjilegal, B., Wang, J. L., Pan, H. Y., Qing, W. D., Zhao, R. W.. Hu. B. P., Wang, Y. Z., Liu, G. C., Li. H. S., and Cadogan, J. M. (1994). J. Appl. Phys., 76, 1971. Yang, F. RI., Han, X. F., de Boer, F R., and Li, H. S. (1997). J . Phys.. Condensed Matter, 9, 1339. Yang, J., Dong, S., Mao, W., and Yang, Y (1996). J. Magn. Magn. Mater., 152, 396. Yartys, V. A., Gutfleisch, O., Panasyuk, V. V., and Harris, I. R. (1997). J. Alloys Compd., 253-254, 128. Yelon, W. R.,and Hu, 2 . (1996). J. Appl. Phys., 79. 1330. Yu, S., Zhou, S., Zhang, M., Li, F., Wang, R., Lin, G., Chai, L., Wu, X., Yin, Y., and Lin, Y. (1995). J. Alloys Compd., 222. 96. Zandbergen, H. W., Cava, R. J., Krajewski, J. J., and Peck Jr., W. F. (1994a). J. Solid State Chem., 110, 196. Zandbergen, H. W., Jansen, J., Cava, R. J., Krajewski, J. J.. and Peck Jr., W. F. (1994b). Nature, 372, 759. Zuttel, A., Chartouni, D., Gross, K.. Spatz, P., Bachler, M., Lichtenberg, F., Folzer, A., and Adkins, N. J. E. (1997). J. Alloy3 Compd., 253-254, 626.
This Page Intentionally Left Blank
Slavi G,Sevov Department of Chemistry and Biochemistry, University of Notre Dame,IN, USA
1. Intxodnction The subject of this chapter is a special class of intermetallics called ‘Zintl phases’ named after Edward ZintS who pioneered their exploration and the rationale for their structures (Zintl and Dullmkopf, 1932; Zintl and Brouer, 1933; Zintl, 1939). Zintl’s view applied to the classical Zintl phase NaTl assumes compSete electron transfer from the more electropositive sodium to the more electronegative thallium (Zintl and Woltersdorf, 1935}, just as in genuine ionic salts such as NaCI. The difference is that the resulting anions do not necessarily achieve an electronic octet as isoErated species but may rather bond to each other in order to do so. Thus, 3-in Na+TY behaves like an element of the group to the right, group 24, and forms a diamond network, typical for the elements of that group, that is stuffed by the Na+ cations. Such phases are the link between the metallic, alloy-type intermetallics on the one hand and the typical valence cornpounds on the other. The number of Zintl phases has increased many-fold since Zintl’s time, and a number of reviews on the subject have been published in recent years (Corbett, 1996, 1997, 2000a; Miller, 1996; Eisemann and Cordier, 1996; van der Lugt, 199% Belin and Tillard-Charbonnel, 1993). Since the definition of a ‘Zintl phase’ has never been very exact, often compounds that include nonmetals have been considered in this class. The intent here is to concentrate more on clearly intermetallic ZintI phases, i.e. phases that contain main-group metals, semimetals, or semiconductorsonly. Thus, only compounds of the alkali metals with groups 13 (without B and AI), 14 (without C), and 15 (without N and P) will be discussed here.
1.1 Def&ion and Synthesis
For the purposes of this chapter ‘classical’ or true Zintl phases are defined as follows. They are compounds that: 1. contain an alkali or alkaline-earth metal and a p-
element (or elements) that is a metal, semimetal, or small-gap semiconductor; 2. are electronically balanced or closed-shell compounds, i.e. the number of electrons provided by the constituting elements equals the number of electrons needed for covalent bonding in the structure; 3. have very narrow or no homogeneity width, i.e. they are line compounds; 4. are semiconductors or poor conductors; 5. are diamagnetic or show very weak, temperatureindependent paramagnetism; 6. are brittle. Criteria (2) through (6) are the same as for normal valence compounds, while (1) provides the connection to the general class of intermetallics. Requirement (2) needs to be explained in some more detail, especially the meaning of ‘number of electrons needed for covalent bonding’. For structures with ‘normal’ 2center-2-electron localized bonds this number is simply the sum, 2 x (number of bonds) t 2 x (number of lone pairs), where all atoms follow the octet rule. For compounds with delocalized bonding, on the other hand, the valence rules are different. The bonding electrons in compounds with deltahedral clusters are calculated by Wade’s rules (Wade, 1972, 1976) that have been deveIoped for the borane c a p but also work equally well for naked main-group clusters. Furthermore, there are combinations of the two, i.e.
Interwetalfic compound^: Vol. 3, Principles and Practice. Edited by J. R. Westbrook and R. L. Fleischer. 02002 John Wiley & Sons, Ltd.
114
Structure md Composition
compounds that contain networks of deitahedral clusters bonded to each other via localized bonds, and the sum of bonding electrons for such systems will include the electrons for cluster bonding, those for intercluster bonding, and the lone pairs (usually on cluster atoms not involved in intercluster bonding). Since the Zintl phases usually contain alkali or alkaline-earth metals and are also air- and moisturesensitive themselves, their synthesis and handling is carried out in vacuum or inert atmosphere. Typically, they are synthesized by heating (anywhere between 300 and 9OO0C) mixtures of the constituting elements sealed in containers made of ductile and high-melting niobium or tantalum. Since the latter are easily oxidized at these temperatures they are additionally 'jacketed' in evacuated ampules of fused silica.
1.2 Histopicat Ovepvicw For alloys and intermetallic compounds made of similar elements, the correlation between structure and valence electron concentration (VEC) had been established before Zintl's time. The rules developed by Hume-Rothery and Westgren were already stated, and explained why different structures form for different VECs (Hume-Rothery, 1926; Westgren and Phragmen, 1926). Also prior to ZintI the GrimmSomnierfeld rule for tetrahedral frameworks was suggested (Grimm and Sommerfield, 1926). Zintl also studied intermetallic compounds but only those made of elements with very different electronegativities, i.e. combinations of the very electropositive alkali or alkaline-earthmetals with the post-transition elements. It W B Sclear that the phases in these systems behaved quite differently than Hume-Rothery's metallic phases and were rather saltlike since they had melting points higher than the elements, were brittle, and conducted poorly, It was also noticed that all compounds that were not Hume-Rothery phases exhibited atomic volume contraction, presumably due to contraction of the more electropositive atoms. This suggested that the 'latter were cations in the structure, or in other words, they have transferred their electrons onto the more electronegative atoms. This was realized by Zintl and became the basis for rationalizing the electronic structure and bonding in the new class of compounds, now called Zintl phases. Thus, instead of using VEC for the prediction of the structure, it was proposed to look at the electronic state of the more electronegative element after assuming complete transfer of the valence electrons from the more electropositive element (Zintl and Woltersdorf, 1935). Later, Klemm
developed the idea further into what is known as the 'pseudo-atom' concept, the negatively charged atom would behave structurally similar to the corresponding isoelectronic neutral element of a group to its right (Klemm, 1956, 1959). This concept, combined nowadays with the electronic requirements for delocalized bonding in closed-sheil clusters, is a very powerful tool correlating the structures of the Zintl phases with their stoichiometnes.
2. Zintf Phases with Ddocaiized Bonding 2.1 Isolated Clusters
Usually deltahedral clusters are associated with the borane cages and their derivatives of the type B,,H&-, B,&?;, etc. The bonding electrons in such species are delocalized, i.e. they are shared by all atoms of the cluster. Therefore, the electronic requirements in such species can not be described by simple 2-center-2electron bonding but rather by rules developed by Wade (Wade, 1972, 1976) and known as Wade's rules. The latter postulate that closo- (no vertices missing), nidn- (one vertex missing), and arachrzo(two vertices missing) species require 2n+2, 2n44, and 2n -I-6 ebctrons for bonding, respectively, where n is the number of vertices of the cluster. The same rules can be applied for main-group 'naked' deltahedral clusters, i.e. clusters without ligands, where one has to consider the fact that each vertex will carry a lone pair of electrons. With this requirement in mind we can review the suitability and potential problems for forming such clusters from different main-group elements. The isolated borane-type deltahedral clusters of the closo-type of groups 13, 14 and 15 will have formulas E$"2)-, E;-, and respectively. The nido-species will have charges of (n+4)-, 4-, and (n-4)+, respectively, and the charges for the corresponding arachno-species will be (n f 6) -, 6 -, and (n-6)+, respectively. One can clearly see that formation of negatively charged clusters of group 15 is basically impossible. The only two possibilities for elements of this group are arachno Ej- (square) and E,- (pentagon). None of them has been observed in neat solids (solids made directly by solid-state reactions from the elements and that do not contain organic or similar moieties), although the former anion can be crystallized from ethylenediamine solutions of K,Bi, in the compound E-cryptJ2B4 where crypt = 4,7,13,16,21,24-hexaoxa-l, 10-diazabicyclo[8.8.8]hexacosane (Cisar and Corbett, 1977). For
Zintl Phases clusters of group 13, on the other hand, the negative charges are going to be too high. For example, a closo-cluster of 11 atoms will require the enormous charge of 13- . Its 10 and 9 atom nzdo- and avachnoversions will have even higher charges of 14- and 15-, respectively. Therefore, it is very unlikely that such clusters will fonn for this group either. The charges in group 14 are independent of the cluster size, they are always 2 -, 4 - , and 6 - , for closo, nido, and arachno species. The charge of 2- is perhaps too small for a relatively large cluster since it will require only two countercations in a compound such as A,[E,,]. The two cations will be too few and too small for packing eficiently with the large clusters and for separating and screening them effectively from each other. No such clusters have been found in neat solids, although they exist in compounds precipitated from ethylenediamine solutions with huge countercations such as [A-crypt]. Known in such compounds are E?- trigonal bipyramids (Edwards and Corbett, 1977; Campbell and Schrobilgen, 1997), and E:tricapped trigonal prisms (Belin et al., 1977). Until recently, the nido and arachno clusters of group 14 were also thought to have charges too low for formation in neat solids. However, in recent years this concept has been proven wrong with the synthesis of a number of Zintl phases with stoichiometry A,E, that contained the nid0-E;- clusters (Queneau and Sevov, 1997a, 1998a,b; Todorov and Sevov, 1998; von Schnering et al., 1997). 2.1.1 Group 14
Small tetrahedral species El- in the equimolar Zintl compounds AE (= A,E4) are known for all elements of this group (Schafer and Eisenmann, 1985). Counting electrons in the tetrahedron is somewhat debatable, since there are two different, but equally satisfactory, ways to do that. One way to look at it is as a nido deltahedron derived from the closo trigonal bipyramid E:- by removing a vertex, despite the fact that all of its faces are triangular and that it does not really look like a geometrical body with a missing vertex. Nevertheless, when taken as such, i.e. assuming delocalized bonding, its charge is going to be 4- as for any nido species of this group. The second rationalization of the electron count is to assume simple 2-center-2-electron localized bonding between all the vertices. Thus, the tetrahedron has 6 edges which require a total of 6 x 2 = 12 bonding electrons. When the 8 electrons for the 4 lone pairs are included, the total for the cluster becomes 20 electrons, or the available 4 x 4 = 16 electrons provided by the 4
115
atoms are again 4 electrons short from the 20 needed by the cluster. Yet a third and perhaps the best way to view these compounds is to use the pseudo-atom concept where E- in AE is a pseudo-element of group 15 such as phosphorus, and as such it forms tetrahedra that are typical for one of the allotropic forms of that element. Tt should also be mentioned here that pseudophosphorus can be made not only from atoms of the neighboring group, but also from atoms of group 13, i.e. two groups to the left of phosphorus. Thus the compounds A2E (= A,E,) where E = In or T1 contain the same tetrahedral species although with a doubled charge, E:- (Smith and Hansen, 1967; Sevov and Corbett, 1993a). Larger deltahedral clusters of this group have been known for quite some time but all in compounds crystallized from ethylenediamine solutions with cryptated alkali metal countercations (Corbett, 1997, 2000a. and references therein). The synthetic niethodology starts usually with the preparation of a precursor AXE, by melting the two elements. It was always thought that these precursors were simple featureless melts or alloys with no particular structure, or even less likely, structures with clusters. The precursor is dissolved in ethylenediamine which usually also contains 2,2,2-crypt as an additive designed to capture the alkali metal. It was thought that the cluster formation always occurred at this step during the dissolution of the precursor. Nevertheless, our and others’ studies have shown that many of the precursors in these systems are Zintl phases with already pre-existing clusters, and that the dissolution process often is reduced to simply extracting them into the solution (Queneau and Sevov, 1997a. 1998a,b: Todorov and Sevov, 1998; von Schnering et al., 1997; Somer et al., 1998; Xu and Sevov, 1999). The first structurally characterized compound with isolated 9-atom deltahedral clusters was Cs,Ce9 (tP208) shown in Figure 1 (Queneau and Sevov, 1997a). Choice of this particular combination of elements for study was not accidental; it combines the largest cation, Cs, and the smallest element of this group, Ge, for which a 9-atom cluster has been characterized from solution. As already discussed, it was thought that E$- clusters cannot exist in neat solids due to the mismatch of large clusters and small cations, and therefore the best chance for formation of such species had the smallest clusters combined with the largest cations. With the knowledge that at least Ge$- exist in neat solids it was more than logical to test thc possibilities for similar clusters of the heavier analogs Sn and Pb.
116
Structure and Composition
Figure 1 The structure of Cs,Ge, (left), the first Zintl phase with isolated %atom clusters of group 14.The clusters Ge;- (right) are nzdo deltahedra with the shape of monocapped square antipnsms. There are 208 atoms in the unit cell
Thus, the next step was to investigate the combination of the largest cation and the largest cluster, Cs4Pb, (mP52) (Todorov and Sevov, 1998). Again, it was somewhat surprising to discover that the corresponding Zintl phase can be synthesized. Although the atomic ratio is the same as in Cs4Ge, the packing and therefore the structure of Cs4Pb9is different. Lastly, the other extreme combination of cation and anion, K4Pb9(mP52), with a very small cation and the largest cluster was tried (Queneau and Sevov, 1998a). The compound forms, the structure is different from the above two (Figure 2), and contains two different types of Pbg4- clusters, labeled A and B (Figure 3). The clusters of type A have the expected geometry for this charge, i.e. monocdpped square antiprisms, but those of type B are rather like tricapped trigonal prisms. This latter geometry indicates, of course, closo species of %atoms, and those therefore should rather
carry a charge of 2 - . Nevertheless, this particular cluster is elongated along the three-fold axis, and, as a result of this, one otherwise antibonding and empty orbital becomes bonding and filled, and the charge is again 4-. It should be mentioned here that compounds with Sn$- clusters were also made. Furthermore, in addition to the structural characterization most compounds were also studied by Raman spectroscopy (von Schnering et al., 3997). It was shown that the 9-atom clusters exhibit a very characteristic set of vibrations that can be easily recognized in these spectra. Thus, Raman spectroscopy can be a very useful tool for identification of compounds with isoIated ionic clusters of different sizes by simply matching experimental spectra with spectra from a database. AI1 the 9-atom clusters of group 14 mentioned so far, Gey, Sn,, and Pb,, were known from solution work
Zintl Phases
117
Figure 2 The structure of K,Ph, w t h two different types of Phi- clusters, monocapped square anhprisms (darker polyhedra) and elongated tricapped trigonal prisms (lighter polyhedra). The potassium cations are shown as isolated spheres
prior to finding them in Zintl phases. The element missing from this collection is obviously silicon since no silicon clusters (larger than the tetrahedra found in A,Si4) had ever been made either from solution or in the solid state. Nevertheless, with the knowledge that clusters of the heavier analogs can be made in neat solids, it was natural to explore the possibility for similar silicon clusters. Thus, reactions with stoichiometry A&, were carried out, and the resulting phase indeed contained Si$- deltahedral clusters although the stoichiometry of the compound was somewhat different (Queneau and Sevov, 1998b). The new compound, AI2Sil7where A=K, Rb, Cs (mP464), contains not only 9-atom clusters but also the tetrahedral species Sii- (Figure 4). The formula can therefore be rewritten as 12A++ 2(Si$-) + Si$-.
2.1.2 Group I3 Although this group was initially thought to be not very promising for deltahedral clusters (see above), it turns out that it actually provides the largest and most diverse collection of clusters with delocalized bonding. The first cluster of this group was found in a phase which, ironically, is not exactly a true Zintl phase
Figure 3 Closer view of the two types of Ph$- clusters found in K4PhP.A) monocapped square antipnsm; and B) elongated tncapped trigonal prism
according to the definition given in the Introduction (above), and the cluster itself is not of borane-type geometry. It is an 11-atom cluster of indium, Inyc, found in the compound &In,, (hR114) shown in Figure 5 (Sevov and Corbett, 1991). The discrepancy between charge and number of cations means that there is an extra electron per formula. Therefore, it is not an electronically balanced phase and cannot be called a true Zintl phase according to our rules. It better fits a group of phases which can be called ‘almost Zintl phases’ or ‘metallic Zintl phases’, discussed later on. The extra electron is apparently
118
Structure and Composition
Figure 5 The structure of K,In,, (rhombohedral. R3c) is of close-packed layers of In:; clusters (polyhedra) stacked tn a cubic fashion along the c =is ( v e r t d ) and separated by double layers of potassium cations (spheres)
Figure 4 Shown is the structure of Rb,,Si,, which contains 9and ‘$-atom clusters of silicon, Sii- and §ij- in ratio 1:2 shown as lighter and darker polyhedra, respectively. The rubidium canons are shown as isolated spheres. This structure and other Zintl structures may be viewed in color at www.nd.edu/ sbobev/interm,htrrd
n
N
delocalized over the alkali-metal atoms and perhaps the cluster antibonding orbitals, and the compound is Paulj-paramagnetic and metallic. Nevertheless, the cluster is deltahedral and exhibits delocalized bonding. Yet, its geometry and electron count differ from that expected for a borane cage of 11 atoms. The cluster in this case is a pentacapped trigonal prism that has been compressed along the three-fold axis (Figure 6), and the number of electrons needed for skeletal bonding within the cluster is not 2n+2 but rather 2 n - 4 = 18, Thus, instead of carrying the enormous charge of 13 corresponding to (n + 2) - for closo species of group 13 (see above), its charge i s the more reasonable 7 - .
Figure 6 A closer view of the indium cluster in K&,. Its shape can be considered as a pentacapped trigonal prism that has been compressed along the threefold axis {vertical). Three of the capping atoms cap rectangular faces at the waist of the prism, and the remaining two cap the two triangular bases, top and bottom of the figure
Zintt Phases
Molecular orbital calculations clearly show that a good size HOMO-LUMO (Highest Occupied Molecular Orbital-Lowest Unoccupied Molecular Orbital) gap of 1.2eV opens at 18 bonding electrons for this geometry, i.e. the cluster IS hypo-electronic with respect to the classical chic, geometry. It can be rationalized in the following way. Starting from a classical tricapped trigonal prism with 2n + 2 = 20 electrons, i.e. In;1-, we add two more indium vertices to cap the triangular bases. This addition does not bring new orbitals, i.e. the number of required electrons remains 20, but only two additional electrons, the p-electrons of the indium atoms, and the cluster becomes In?,-. Next, the cluster is compressed along the three-fold axis, and this distortion makes a formerly bonding a; orbital into an antibonding and empty one (Figure 7). This translates into the lower charge for the cluster, In:, . Besides K81nilthe clusters 1n:T also exist in RbsInil and Cs81nll. Clusters with the same geometry and charge were later found for thallium in A8Tlll where A = K, Rb, or Cs (Dong and Corbett, 1995a), A1,Tl,, (hP42) where A = R b or Cs (Dong and Corbett, 1996a), and K,8T12&u3 (hP443) (Dong and Corbett, 1995b), and for gallium in Cs,Gal, (Henning and Corbett, 1997). Furthermore, it should be pointed out that these AsEl, phases can be made electronically balanced by ‘controlled’ one-electron oxidation without changing their structures. Thus, the chlorineintercalated derivatives Cs,GallC1 and Cs81nl,CI (hR120) contain the same 11-atom clusters and the same overall structure but are semiconducting and diamagnetic (Nenning and Corbett, 1997). The chiorine occupies an available site that is empty in the AsEl, structure. These two phases satisfy all but one requirement in our definition of Zintl phases (see above). Criterion (1) is violated since chlorine is obviously a nonmetal. Nevertheless, CI is one atom out of 20, and furthermore the phase can exist without it. This clearly shows how complicated defining Zintl phases can be! In addition to these compounds, there is also a substitutional phase that is electronically balanced. This is K&n,,,Hg] (hP38) with the same 11-atom cluster as in K,Inll but with one indium atom substituted by Hg, an element from the group to the left (Sevov et al., 1993). Due to the substitution the cluster assumes charge 8-, i.e. [In,oHg]s-, and the compound is a true Zintl phase. Although here we have a ternary system, the heteroatom does not cause changes in the cluster shape. It rather acts only as an electron gatherer and causes localization on the cluster of the otherwise delocalized extra electron.
119
Similarly, there are two denvatives of substitution in the Tl&- clusters. The isoelectronic [TI,CC~~]~~is the result of three thallium atoms substituted by an element to the left, and the cluster keeps the same overall geometry (Huang and Corbett, 1999). Nevertheless, the compound in which it is found, Na9KldTllBCd3 (hP92), is not a classical Zintl phase since it has an extra electron which makes the compound paramagnetic. Another derivative is fr19Au21g- which has even fewer electrons, 2n-6 rather than 2n-4 (Dong and Corbett, 1995b). The pentacapped tngonal prism in this case IS compressed even ftrther, so much so that a bond is formed
-5
-6
/ e’ -7
-8 ISXI (
~
~
H
~
Figure 7 The molecular orbital levels for p orbitals in In,, clusters with D,, symmetry: left, the classical pentacapped tngonal prism with 20 electrons; and right, the observed axially compressed cluster with 18 electrons. The HOMOLUMO separations are shown with broken lines
120
Structure and Composition
et al., 1998). Two of these, Tl:- and Tli-, follow Wade’s ruies and carry the prescribed charges of (n + 2) - . The geometries, as one might expect, are also borane-like, a trigonal bipyramid and an octahedron for the two, respectively. The other three species are again hypoelectronic with nonclassical shapes: T1:and T17- are compressed square and pentagonal bipyramids, respectively, with bonds between the two axial vertices, and TI8- has a strange geometry that is best described as the imaginary product of the removal of four neighboring vertices from a Tl-centered icosahedron (Figwe 9). There are a number of observations that can be Figure 8 The cluster Tl&u?_ XI K,8T120Au3.It can be made at this stage for both groups 13 and 14. One is considered as an extremely compressed pentacapped that clearly nido and arachno species of group 14 have trigonal pnsm (see also Figures 6 and 7) with the gold atoms (light) capping the two triangular bases. The the ‘right’ charges and can form in the solid state. Two, compression leads to the formation of a bond between the the problem of high negative charges of group 13 has two gold atoms been solved by simply forming deltahedral clusters with nonclassical shapes that require fewer electrons between the gold atoms capping the bases (Figure 8). for skeletal bonding, i.e. hypoelectronic clusters. This, Alternatively, the shape can be considered as made of however, is only one of the solutions of the problem. thrce face-fused octahedra, all sharing an edge, the Another solution is to form centered clusters where Au-Au bond. Again, compression and condensation the centerkg atom brings additional electrons and/or lead to a decrease of the number of electrons needed causes geometrical distortions that lead to hypoelecfor bonding and a lower charge. tronic species. This is the case in K,E,Jn (tP38) for For indium and gallium it seems the 11-atom cluster E = In or T1 (Sevov and Corbett, 1993b; Dong et al., is the only isolated species. Thallium, on the other 1997), KloEloM(op252) for E = In or Tl and M = Ni, hand, forms five additional isolated clusters: Tli- in Pd, or Pt (Sevov and Corbett, 1993~;Corbett, 2000a). Na2KzlTlI9 (oCl68) (Dong and Corbett, 1994), and NaloGa,,,Ni (Henning and Corbett, 1999). The ZnNa23K9Tl1533* (hP96) (Dong and Corbett, 1996~)and centered clusters [El&nIs- are close to the classical Na9K,6TllxCd3(Huang and Corbett, 1999), Tli- in shape expected for 10-atom species (Figure lO), a AT1 where A = K or Cs (Dong and Corbett, 1993, bicapped square antiprism, but again compressed 1996b), Tlg- in Na,,K,Tl,,M (hP39) for M = Mg, Zn, Cd, or Hg @ong and Corbett, 1996d), T17- in Kl0Tl7 along the four-fold axis. The compression is clearly a result of the equidistant positioning of all cluster atoms (mP68)(Kaskel and Corbett, 2000) and Na,2K38T14xA~2 from the central zinc. The effect of the compression is (hP100) (Huang et al., 1998), and T18- in Na2K2,TlI9 the same as in all compressed clusters described so far, (Dong and Corbett, 1994) and Na,2K38TbxAu2 (Huang i.e. molecular orbitals that are otherwise bonding and filled become antibonding and empty, and the charge *Non-integral proportions indicate presence of vacancies in on the cluster is lowered. Thus, instead of 2 n + 2 the the position.
Figure 9 Shown are three thallium clusters: left TY- with the shape of a Compressed square bipyramrd; center, T1:- with the shape of a compressed pentagonal bipyramid; and nght, T1;- with the shape of a self-centered icosahedronof thallium where four vertices have been removed
Zintl Phases
121
Figure 10 Two centered In,, clusters: top, In,,,Zns- with the shape of a compressed bicapped square antiptism; and bottom, In,$Ji'o- with the shape of a compressed
tetracapped tngonal prism (capped are the three rectangular faces and the upper tnangular base) Zn-centered cluster requires just 2n = 20 electrons for bonding. The central Zn and the 10 In atoms provide 2 + 10 = 12 electrons (lone pairs are not counted), and the resulting charge is 20 - 12 = 8 - . Exactly the same is observed in the [E,,M]'O- clusters. The clusters are tetracapped trigonal prisms that have been compressed along the three-fold axis (Figure 10). The cause for the distortion i s again the central Ni, Pd, or Pt atom being equidistant from the cluster atoms. The distortion makes the cluster again hypoelectronic with 2n = 20 electrons. In this case the central atoms do not provide bonding electrons, and consequently the charge is 20 - 10 = 10- . Isolated and centered clusters of thallium are again more numerous than those of the lighter analogs. In addition to the above-mentioned centered species it also forms self-centered icosahedra (Tl,,Tl)lo- and (T112T1)*1-(Dong and Corbett, 1995c), Na- and Hcentered icosahedra (T11,Na)13- and an octahedron (T16H)?- (Dong and Corbett, 1995d), and icosahedra
Figure 11 The dimer of corner-sharing, the zinc atom, trigonal hipyramids of Ge,Zn6- found in Cs,Ge,Zn. It can also be considered as two germanium tetrahedra linked by a
zinc atom. The two Zn-capped faces are eclipsed making the zinc in a tngonal prismatic environment centered by two-electron donors (Tl12M)12- where M = Mg,Zn, Cd, or Hg (Dong and Corbett, 1996d). Notice that the self-centered icosahedra can exist with different charges, 10 - and 11 - , and therefore with both even and odd numbers of electrons. More importantly, the cluster with the odd number of electrons (49), (Tli2Tl)10-,is one electron short of the required number for a closed shell. This has apparently been explained with an electron hole localized on the cluster, since the compound is paramagnetic. 2.1.3 Heteroatomic The substituted 11-atom clusters of group 13, &n,,HgI8-, [ v u 2 l 9 - and [Tl,Cd,]'O- have been
122
Structure and Composition
discussed already. For group 14 there are no such substituted species. Rather, all attempted similar substitutions with elements of group 12 lead to species made of tetrahedra of group 14 that are connected by the heteroatom. Thus, Cs,[Ge,Zn] ( o n 50) (Figure 11) contains isolated [GeXZnl6- clusters that can be considered as corner shanng (the Zn atom) trigonal bipyramids or as two Ge4-tetrahedrajoined together via a zinc atom (Queneau and Sevov, 1997b). For the latter, the zinc atoms cap faces of tetrahedra that are eclipsed with rcspect to each other so that the zinc is in trigonal prismatic coordination. Coincidentally, had the zinc substituted for a germanium atom in the G$clusters, the resulting specles would have had the same stoichiometry and charge, [Ge8Znlh-.The explanation of why zinc does not substitute in Geg- and In;, clusters is perhaps the fact that it carries only two electrons and its s-orbital is not inert (stable) enough. Mercury, for example, can and does substitute for In in Inll. The zinc atom prefers a position with rather a sphericalenvironment, a geometry that does not require a lone pair. Cadmium and mercury also do not substitute in the 9-atom clusters of group 14. Instead, they form compounds that are structurally different from Cs,[Ge8Zn] but stoichiometrically equivalent with it. The cadmium analog, K6[Pb8Cd](mC90) (Todorov and Sevov, 1999), contans isolated PbI--tetrahedra as well as tetramers of Pb4-tetrahedra connected via cadmium atoms, [(Pb4)Cd(Pb4)Cd(Pb4)Cd(Pb4)]10(Figure 12). The latter can be also viewed as made of a pair of [PbXCdl6-species analogous to [Ge8ZnI6- that have been further linked by the third cadmium. All tetrahedra are linked via faces, but the faces are
staggered wrth respect to each other, so that the cadmium atoms are in a trigonal antiprismatic environment. Lastly, Hg-connected dimers of tin tetrahedra exist in Cs,[Sn8Hg], but the bridging is between edges rather than faces. The mercury atoms are in elongated tetrahedral coordination. 2.2 Networks of Clusters Besides Zintl phases with isolated clusters there are a number of compounds with networks of interconnected clusters, predominantly in the systems group 13 - alkali metal. Such compounds combine two different types of bonding, delocalized within the clusters and localized between them. Naturally, the intercluster bond distances (exo-bonds) are shorter than those within the clusters (endo-bonds) since the former are ‘normal’ 2-center-2-electron bonds while the latter is delocalized bonding. It turns out that many of the network compounds are not electronically balanced but very close to being such. Many of them have a few extra electrons (or holes) relative to those needed for the bonding. Such compounds can be referred to as ‘metallic Zintl phases’, and in order to consider them in this chapter we have ,to somehow broaden the definition given before. We can extend and paraphrase Hughbanks’ interpretation of Zintl phases as phases with ‘structures for which we can at least think‘ of as valence compounds (Hughbanks, 1992). We will call ‘metallic Zintl phases’ those compounds that conform to all rules outlined in the Introduction (see above) except rule (2). We modify rule (2) to state that the phases
Figure 12 The tetramer of four lead tetrahedra joined by three cadmium atoms, (Pb,)Cd(Pb4)Cd(F‘b4)Cd(Pb4),found in K6Pb,Cd
123
Zintl Phases ‘look like’ they were nearly electronically balanced according to their structures but have a few extra electrons or holes.
2.2.1 Group 13
Gallium is perhaps the element with the greatest tendency to form networks of interconnected clusters, while it has only one compound with isolated clusters. The tendency for network formation diminishes drastically on going down the group, with thallium exhibiting only two such compounds, A,5T127 for A=Rb, Cs (Dong and Corbett, 1996a) and K6Tlll (oC184) (Corbett, 2000b), but a large number of isolated clusters. Tndium is clearly in between with many examples of both networks and isolated species. The structural chemistry of the systems alkali metal gallium has been extensively reviewed by Belin (Belin and Tillard-Charbonnel, 1993). A good example of a gallium network is the structure of KGa, (tZ24) (Belin and Ling, 1982) which contains 8-atom clusters of gallium, each with 8 exo-bonds, and isolated 4-bonded gallium atoms, the two species in equimolar ratio (Figure 13). The clusters are of the closo-type and therefore require 2n+2 = 18 electrons for skeletal bonding. Also, since each exo-bond is a normal 2center-2-electron bond, half of the electrons needed for them, i.e. (8 x 2)/2 = 8, are counted for the cluster as well. The total number of electrons required for the endo- and em-bonding of the cluster is therefore l8+8=26. The eight gallium atoms of the cluster bring 8 x 3 = 24 electrons and the charge of the cluster is 2-, i.e. Gag-. The 4-bonded isolated gaslium is in tetrahedral coordination and therefore has a formal charge of 1 -, i.e. Gal-, in order to achieve an octet. Thus, the equimolar combination of Gai- and Galwill require 3 potassium atoms to balance the charge, and therefore the formula K3Ga, (=KGa,) can be represented as 3K* iGai- +Gal-. The compound is electronically balanced and a true Zintl phase. Other examples o f electronically balanced gallium networks of clusters are: &Gal, (oC128) with 12and 1I-atom clusters, LizGa7(hR18) with icosahedra, Na,Ga13 (kR360) with icosahedra and 15-atom ‘spacers’, RbGa, (hR32) with icosahedra connected via 3-center-2-electron bonds, etc. More complex structures with fused deltahedra and/or partially occupied gallium sites also exist, and these are often ‘metallic Zintl phases’. Such are Li5Gag (oC56), Na,,Ga39 (oZ244), Li,Nn,Ga19 57 (oE896), Li,K3Ga2, (oC328), Na13K4Ga4,,57 (hR138), etc.
Figure 13 The structure of KGa, made of interconnected gallium (dark spheres) 8-atom cfoso deltahedra and isolated atoms
(Belin and Tillard-Charbonnel, 1993, and references therein). There are also many indium networks made up of interconnected deltahedral clusters. Perhaps the simplest such network is found in Rb2fn3(= Rb41n6)(tI20) (Sevov and Corbett, 1993d), layers of 4-bonded octahedra made of indium (Figure 14). The electron count is straightforward: 2n 2 = 14 for skeletal bonding, 2 x 2 = 4 €or lone pairs on the two non-exobonded vertices, and (4 x 2)/2 = 4 for the 4 exo-bonds. This totals 22 electrons required for bonding while the 6 indium atoms of the cluster provide only 6 x 3 = 18 electrons. ?‘lie additional 4 electxons come from the 4 rubidium atoms, and the compound is electronically balanced. This is not the case, however, in K,Naz,In4,
+
Structure and Composition a
Figure 14 The structure of Rb&, made of layers of 4-bonded closo octahedra of indium and separated by the rubidium cations (isolated spheres)
figure 15 The structure of K,N%J%,. The anionic network IS made of intercoiinected icosahedra (darker polyhedra) and avachno 12atom deltahedra (drum shaped, lighter polyhedra) of mdium. The c a ~ o n s of potassiuni and sodium are shown as larger and smaller isolated spheres, respectively
125
Zintl Phases
U Figure 17 A closer view of the 12-bonded closo-In,, found in Na,In,,,. The shape can be considered as a tetracapped truncated tetrabedron. The 4 capping atoms are 6-bonded within the cluster and do not have exo-bonds. The truncated tetrahedron is shown with thicker bonds
Figure 14 The structure of Na,In,,, which contains 12bonded closo-In,, clusters and nido icosahedra
(cP154) (Sevov and Corbett, 1993e), also with a network of clusters (Figure 15). The network is made of 12-bonded icosahedra (closo-Tn,J and 6-bonded hexagonal antiprisms (~rachno-In,~) in ratio 1:3. Each icosahedron requires 2n f 2 = 26 and (12 x 2)/2 = 12 electrons for endo- and exo-bonding, respectively, while these numbers for the aruchno-In12 are 2n I-6 = 30 and (12 x 2)/2 = 12 electrons, respectively. Combined with the available 12 x 3 = 36 electrons from the 12 indium atoms forming the clusters, the charges are 2- and 6- for the closo and aruchno species, respectively, i.e. clom-InfF and urachno-Infy in ratio 1:3. Thus the number of needed extra electrons to balance the charges of 1 x 12 + 3 x 12 = 48 indium atoms is 1x 2 + 3 x 6 = 20. However, available are 3 potassium and 26 sodium atoms according to the formula, and therefore the compound is a ‘metallic Zintl phase’ with 9 extra electrons per formula. Twelve-atom clusters, icosahedra, are typical for this group which is sometimes cdled the group of the icosagens. Nevertheless, indium has shown capabilities to form even larger closo-species. Sixteen-atom clusters of indium havc been found in two different nonstoichiometric compounds, NaJn,, (tP228) and
Na151n27,4 (oc344) (Sevov and Corbett, 1992, 1993a). The network of the former (Figure 16) is made of three different species, 12-bonded closo-in,,, 11-bonded nido-In,,, and 4-bonded isolated indium atoms. The numbers of required and available electrons per cell are 504 and 507, respectively, or only 3 extra electrons (or 0.6%) available per 504 electrons and the compound is metallic. an ‘almost Zintl phase’. The 16-atom deltahedral cluster is a new species found for the first time in this compound (Figure 17). Its shape can be understood starting from a tetrahedron, then truncating its 4 corners to form a 12-atom cluster species with 4 hexagonal faces, and finally capping those hexagonal faces with 4 additional indium atoms. It is highly spherical with the ideal geometry of point group T,. As a result of this particular tetrahedral symmetry, the bonding within the cluster needs rather 2n + 4 electrons instead of the ‘normal’ 2n+2 electrons for closospecies. The cluster is hyperelectronic. 2.2.2 Heteroatomic
In addition to the described networks of homoatomic clusters there are also many group 13-based networks of heteroatomic ones. Such are Nalo2Cu36Ga279 (hR417) and Na,02Zn,2Ga243(hR417) with species of 3 face-fused icosahedra, Na&d2&+aS6 (cF460) and Li3RZn34Ga67 (hP834) with 16-atom tetracapped truncated tetrahedra, and Li,,Cu,Ga2, (~1160)with interconnected icosahedra (Belin and Tilfard-Charbonnel, 1993, and references therein). Indium-based ternary and quaternary compounds with heteroatomic clusters are also numerous, K1n2-,Cd, (hP90) and
126
Structure and Composition
Figure 18 T h e structure of Kh,-,Cd, with arackno-In,, deltahedra (lighter polyhedra) in the shape of double drums and icosahedra (darker polyhedra)
K3,1n,9-xCdx (hR636) (Sevov, 1993) with 12-bonded araclano 18-atom clusters (Figure 18), Na2,In3,,,Zn, and Na,31n,9BAu3,4(hP132) (Sevov, 1993) with 15bonded closo 15-atom clusters that are truncated trigonal prisms (Figure 19), and Naglnl6,,Zn, (cF448) with 16-bonded closo 16-atom clusters (Sevov, 1993). All these are either electronically balanced or very close to being such. Once more, the thallium-based heteroatomic network compounds are only a very few. Known are RI,Cd,T121 (hP88) with interconnected pentagonal bipyramids (TillardCharbonnel et al., 1997) and CssT1,1Cd2with columns of fused Cd-centered pentagonal antiprisms (Corbett, 2000b). Perhaps the most spectacular ternary Zintl phases based on group 13 elements are those with fused ‘fLillerene’4ike cages made of indium (Figure 20), Na&&, (kP195) (Sevov and Corbett, 19930 and Na1721nt92Z2 (oP712) for Z=Ni, Pd, Pt (Sevov and Corbett, 1996). The fullerene cages in these phases are made of pentagonal and hexagonal faces and contain
concentric spheres of sodium where the number of sodium atoms equals the number of fullerene faces (Figure 21). Nickel-centered clusters of Inlo, just like those in K,o[In,oNi](see above) are found inside the sodium spheres. The formula of one fullerene cage can be written as Ni@In,o@Na,,@fn,, where the symbol @ is used for ‘within’ (Figure 21). We should point out here that the stoichiometries of these compounds are extremely close to another Zintl phase, NaIn (=Na,In,) with a diamond framework of indium (Zintl and Dullenkopf, 1932). The difference i s just 2 additional nickel atoms per nearly 200 atoms of sodium and indium! Although there is such a negligible difference in stoichiometry,the two structures, those of Na&r~~~Ni, and Na961np6,have nothing in common. Furthermore, although structurally extremely complicated, the Nag61n9,Z,compounds are diamagnetic, and therefore true Zintl phases. Another group of compounds containing networks of heteroatomic clusters are based on mixtures of groups 12 and 14. Two compounds, Na,,Cd,E,
Zintl Phases
Figure 19 A view of the 15-bonded closo-In,, with the shape of a truncated trigonal prism (the three-fold axis is vertical) found in Na.&,,,,Zn4,, and Na,,In,,Au,,.,
made of Figure 20 The indium framework in NaJn,Ni, face-fused fidlerene-like cages of 74 and 60 atoms. The spheres are ‘stuffed’ with sodium cations and Ni-centered In,, clusters (see Figure 21). The layers of cages are stacked in double hexagonal fashion, ABACAB. along the c-axis (vertical)
127
Figure 21 A closer view of a fullerene-like cage of indium made of 74 atoms. A nickel atom centers the sphere, a cluster of 10 lndium atoms surrounds the nickel, and a sphere of 36 sodium atoms surrounds the indium cluster. The formation can be written as Ni@In,,@Na,,@In,
(cII60) for E = P b , Sn (Todorov and Sevov, 1997a) and Na49Cd58.sSn37,(hR580) (Todorov and Sevov, 1997b), were synthesized recently. Both compounds contain empty icosahedra, the only examples of such species that do not involve group 13 elements. One way to rationalize that is, of course, to consider the isoelectronic relationship between an atom of group 13 and the average of an atom of group 12 and another one of group 14. In addition to the icosahedra, Na,9Cd58,Sn,7,, also contains 12-bonded closo deltahedra of 18 atoms, the largest closo deltahedra known so far {Figure 22). This cluster is also interesting from an electronic point of view since, as in the case of closoIn,,, it is hyperelectroaic and requires 2n + 4 bonding electrons. However, the violation o f Wade’s rules in this case is not due to the particular symmetry as in the case of In,, but is rather due to the size of the cluster. Since the cluster is rather elongated, an ‘extra’ orbital that is bonding withm the two hemispheres and antibonding between them becomes net-bonding and filled. Also of interest here is that the non-exobonded atoms of the cluster do not carry lone pairs of electrons, apparently due to the flatness of the coordination of these atoms. Tlre planarity is a direct result of the size of the cluster, and therefore this cluster is beyond or at some critical size, above which atoms without exo bonds will have empty porbitals.
128
Structure and Composition
Figure 22 The structure of Na,,Cd,,,,Sn,, with the largest closo deltahedron of 18 atoms. The cluster IS 12-bonded to other cluster species in the structure. The potassium cations are shown as isolttted spheres
3. Zintl Phases with Localized Bonding
The compounds in this category contain only regular 2-center-2-electron bonds between the p-elements. Most of them involve the relatively electron-richer pelements, ie. elements from the right-hand side of the p-block. The majority of compounds involve at least two p-elements of different groups and alkali metals, and are therefore mostly ternaries. An excellent review of these compounds by Eisenmann et al. has appeared recently (Eisenmann and Cordier, 1996). Many of the compounds contain tetrahedral units made of one of the p-elements and centered by the other, and often share corners or edges with each other. For example, isolated Sn-centered tetrahedra of antimony, [SnSb,]*-, are found in Na8SnSb4(cF104) (Eisenmann and Ktein, 1988). Chains and higher dimensional motifs of corner-sharing tetrahedra are found in many compounds. Chains characterize the structure of Na,SnSb,, while a 3-0 network of tetrahedra is found in Cs,h4Bi, (aP17) (Bobev and Sevov, 1999a}, for example (Figure 23). There are also structures
Figure 23 The structure of Cs,In,Bi, with chains of edgesharing In-centered tetrahedra of bismuth. The chains are interconnected via In-In bonds (shown) to form a 3dimensional network. The cesium cations are shown as isolated circles
where the tetrahedra are both corner- and edge-shared such as Ga-centered As4-tetrahedra in K,Ga3As, (OR@), In-centered As4-tetrahedra in K,In2As3 (oC64), etc. The number of such heteroatomic compounds and their structural types is very large. In addition to the number of different ways of interconnecting tetrahedra, there are structures with isolated or connected flat AB, units, or mixtures of interconnected tetrahedra and such AB3 units, etc. (Eisenmann and Cordier, 1996). Nevertheless, despite the structural variety, all these compounds have simple bonding, are easily understood by simple octet rule consideration, and since they have been extensively reviewed already (Eisenmann and Cordier, 1996; von Schnering and Hiinle, 1988; Schafer and Eisenmann, 1985; von Schnering, 1981; Schlfer and Eisemann, 1981) they will not be discussed any further in this
Zintl Phases
Figure 24 The structure of Cs3Bi2with isolated dimers of Bif. An extra electron is apparently delocalized on the cesium cations and antibonding Bi-orbitals
chapter. More emphasis will be given to new homoatomic compounds, i.e. one p-element combined with alkali metals.
3.2 Group 15 It is quite surprising that until recently there were blank spots of structurally unknown compounds in the simple binary systems alkali-metatantimony and
129
alkali-metal-bismuth, and even in alkali-metalarsenic. Studies of the corresponding liquid systems (van der Aart et al., 2000) have suggested that structural units such as dimers, tetramers, squares, etc., exist in melts with stoichiometries A5E4and A3E2 Recently, structural studies in these systems revealed that similar units exist in the corresponding solid-state compounds (Gascoin and Sevov, 2000). Thus, compounds A,Bi2 (oC15) were found for A = R, Rb, or Cs. They Fontain Bi2-dimers with a Bi-Bi distance of 2.976A in Cs3Bi2 (Figure 24). Compounds with the stoiclxiometry A5E4 were characterized for both bismuth and antimony with A = K, Rb,or Cs. They contain planar zig-zag tetrameTs of bismuth (Figure 25) wi!h end distances of 2.972 A and a middle oiie of 3,036A. Magnetic and two-probe conductivity measurements suggest that the compounds are metallic. This combined with the stoichiometries and the structures of the compounds (the possible presence of hydrogen has been discarded based on neutron diffraction studies and indirect synthetic evidence) leads to only one possible explanation of their electronic structures. Similar to K&,, which can be written as 8K+ + InTr + e- and other similar ‘metallic Zintl phases’ with extra dectrons (see above), the two compounds A,& (oCl5) and A5E4(oCl8) contain the species E:- and E f - , respectively, and an extra
Figure 25 The structure of Cs,Bi, ulth isolated tetramers of Bi:-. electron that makes the compound metallic
Similarly to Cs,Bi, (see Figure 24) there IS a delocalized
Structure and Composition
I30
electron each. Their formulas can be written as 3A+ + E:- + e- and 5A+ +E!- + e-, respectively. Furthermore, both bismuth compounds dissolve in ethylenediamine, and the compounds that can be crystallized from the solutions contain double-bonded dimeric fBi=Bi12- molecules (Xu et al., 2000). This indicates that the dimer in A3Bi2 also has a double bond, and that the planar tetramer in A,Bi4 has one delocalized double bond with all bismuth atoms in sp2 hybridization, The extra electrons in both compounds are delocalized over the alkali-metal cations and antibonding Bi-Bi orbitals (just as in K81nl,, see above). The latter makes the bismuth distances somewhat longer than for the correspondingmultiple bonding. 3.2 Group 14 Species with multiple bonding have been found for the elements of this group as well. Thus planar rings of silicon and germanium, Sif- and Gej-, isoelectronic with cyclopentene, have been found in LilZE7(oP152) (von Schnering et al., 1980). Also, planar triangular star-shaped Silt- and Ge:- are found in the same compounds. These species are isoelectronic with CO$and similarly have a delocalized double bond. Recently, a class of compounds called clathrates based on group-14 elements have attracted much attention due to their potential for thermoelectric applications. These compounds are AsE46(cP54) and A,E1,6 (cF160) (ideal stoichiometries) where A = alkali metal and E = group-14 element, identified as clathrate-I and clathrate-11, respectively (Cros et al., 1965; Kasper et al., 1965). The structures of both are built of fused cages of 4-bonded tetrahedral E-atoms, and the alkali metals occupy the centers of the cages. C I a ~ a t e - Itype phases (analogous to the zeolite melanophlogite) crystallize in a primitive cubic lattice, space group P m h (Figure 26). The 46 clathrand atoms form two cages of diEerent sizes: one is a 20-atom pentagonal dodecahedron [512], and the other is a 24atom tetrakaidecahedron [512fi2] (the symbol [51262] denotes a polyhedron with 12 pentagonal and 2 hexagonal faces). There are 8 cages per formula unit, 2 smaller and 6 larger, and therefore the clathrate-I formula can be written as A2AiE4;. Clathrate-I1 (analogous to the zeolite ZSM-39) crystallizes in the face-centeredcubic F&rn space group (Figure 27). The framework-building atoms also form two different polyhedra. The difference, however, is that here in addition to the pentagonal dodecahedra [512] the second type of polyhedra are built of 28 atoms, hexakaidecahedra [51264], that have 12 pentagonal
Figure 26 The structure of clathrate-I with cages made of 4bonded elements of group 14,The clathrate has cages of 20 and 24 atoms shown as gray and white polyhedra, respectively. Alkali-metalcations (not shown) center the cages
and 4 hexagonal faces. The ratio of smaller to larger cavities in clathrate-If is 16:8 and therefore the formula can be written as A\6A$E,36. Four-bonded atoms of group 14 should have zero formal charge, and therefore both clathrates should be *metaIlicZintl phases’ considering a full transfer of electrons from the alkali metal atoms to the framework atoms. Nevertheless, there are ways to achieve electronically balanced clathrates, i.e. to ‘stabilize’ true Zintl
Figure 27 The structure of clathrate-I1 with cages of 20 and 28 atoms shown as gray and white polyhedra, respectively. Two types of cations with very different sizes such as Cs+ and Na+ stabilize this structure by occupying the centers of the two different cages
131
Zintl Phases
phases. One way is to form vacancies at the network sites as in A,Sn4u2 where is a vacancy at a tin site. Each missing tin atom leaves the 4 neighboring atoms with incomplete valence shells, i.e. 4 extra electrons will be needed to complete their octet configurations wth lone pairs. Another way to balance the ‘extra’ electrons from the alkali metals is to substitute group14 atoms with an electron-poorer element, from groups 13 or 12 for example. This idea has been widely utilized in the last 10-15 years for making new ternary phases with the clathrate-1-structure. The ideal stoichiometries will be A8EiE3*and &&E42 where E and E are group-13 and -12 elements, respectively. The existence of many of these compositions and their compliance with the Zintl concept have been proven. Even more complicated and difficult for interpretation are the compounds with the clathrate-11-type structure, Until recently, only the non-stoicbiometric NaxSi136 and NaxGe136phases were known in the binary A-E systems (Cros et al., 1965, 1970). The structure of the former was recently refined from X-ray powder diffraction in order to determine the distribution of the sodium atoms within the two available alkali-metal sites, i.e. the centers of the pentagonal dodecahedra and the hexakaidecahedra (Ramachandran et al., 1999; Reny et al., 1998). It was concluded that for xG8, the Na-atoms occupy preferably the larger [512fi2]cage. Until very recently, there was no unequivocal structure determination for compounds with the clathrate-I1 type structure. The first wellrefined structures from single crystal data and reproducible direct syntheses were reported for the stoichiometricand completely-filled silicon and germanium clathrate-I1compounds with mixed alkali metals: C S ~ N ~ ~ , S ~ ,CssNa,6Ge136, ~,, Rb8Na16Si136, and Rb8Na16Ge136 (Bobev and Sevov, 1999b, 2000). These compounds are stoichiometric and due to the extra 24 electrons per 136 group-1.8 atoms, they are also metallic Zintl phases.
a
4. Conclusion
Both the true Zintl phases and the ‘metallic Zintl phases’ are classes of fascinating compounds, some with clusters presenting delocalized bonding and some with ‘normal’ localized bonds. The variety and structural richness is quite obvious, and this is only a small percent of all possible combinations to be explored. The importance of these intermetallic compounds is primarily in the understanding of the chemistry of the p-elements in their negative oxidation
states. When in such unusual oxidation states many of these elements form bonds between themselves, form clusters and networks of clusters. These phases are potential starting materials for deposition of thin films or formation of nanoparticles of the corresponding pelement. Some of them may have good thermoelectric and semiconducting properties, and even may show superconductivity. There are definitely many new and more exciting compounds to be found.
5. References Belin, C., Corbett, J. D., and Cisar, A. (1977). J . Am. Chem. Soc., 99, 7163. Belin, C., and Ling, R. G. (1982). Acad. Sci. Paris, 294, 1083. Belin, C., and Tillard-Charbonnel, M. (1993). Prog. Sotid St. Chem., 22, 59. Bobev, S.,and Sevov. S. C. (1999a). Inorg. Chem.. 38,2672. Bobev, S., and Sevov, S. C. (1999b). J. Am. Chem. Soc., 121, 3795. Bobev, S.,and Sevov, S . C. (2000). J. Solid State Chem., 153, 92. Campbell, J., and Schrobilgen, C.(1997). Inorg. Chem., 36, 4078. Cisar, A., and Corbett, 3. D. (1977). Inorg. Chem., 16, 2482. Corbett, J. D, (1996). In Chemistry, Structure, and Bonding of Zintl Phases and Ions (ed. S. M. ICauzkdrich). VCH, New York, p. 139. Corbett, J. D. (1997). StructureLBonding, 87, 157. Corbett. J. D. (2000a). Angew. Chem. Int. Ed., 39, 670. Corbett, J. D. (2000b). private communicalon. Cros, C., Poucliard, M., and Hagenmuller, P. (1965). Compt. Rend., 260, 4164. Cros, C., Pouchard, M., and NagenmuIIer, P (1970). J. Solid State Chem., 2, 570. Dong, Z.-C., and Corbett. J. D. (1993). J. Am. Chem. Soc., 115, 11299. Dong, L C . , and Corbett, J. D. (1994). J. Am. Chem. Soc.. 116, 3429. Dong, Z.-C., and Corbett, J. D. (1995a). J. Cluster Sci., 6, 187. Dong, Z.-C., and Corbett, J. D. (1995b). Inorg. Chem., 34, 5042. Dong, Z.-C., and Corbett, J. D. (1995~).J. Am. Chem. Soc., 117, 6447. Dong, Z.-C., and Corbett, J. D. (1995d). Inorg. Chem., 34, 5709. Dong, Z.-C., and Corbett, J. D. (1996a). Inorg. Chem., 35, 1444. Dong, Z.-C., and Corbett, .T. D. (1996b). Inorg. Chem., 35, 2301. Dong, 2.-C., and Corbett, J. D. (1996c). Inorg. Chem., 35, 3 107.
132
Structure and Composition
Dong, Z.-C., and Corbett, J. D. (1996d). Angew. Chem. Int. Ed. Engl., 35, 1006. Dong, Z.-C., Henning, R. W., and Corbett, J. D. (1997). Inorg. Chem., 36, 3559 Edwards, P. A., and Corbett, J. D. (1977). Inorg. Chem., 16, 903. fisemam, B., and Cordier, G. (1996). IR Chemistry, Structure, m d Bonding of Zintl Phases and Ions (d. S. M. Kadarich). VCH Pubtishers, Inc., New York. NY,p. 61. Eisenmann, B., and Klein, J. (1988). 2. Naturforschung, 43B, 69. Gascom, F., and Sevov, S. C. (2000). J. Am. Cltern. Soc., 122, 10251. Orimm, H.O., and Sommerfeld, A. (1926). Z . Phys., 36, 36, Henning, R. W., and Corbett, J. D. (1997). Inorg. Chem., 36, 6045. Hennmg, R. W., and Corbett, J. D. (1999). Inorg. Chem., 38, 3883. Huang, D.-P., Dong, 2.-C., and Corbett, J. D. (1998). Inorg. Chem., 37, 5881. Huang, D.-P., and Corbett, J. D. (1999). Inorg. Chem., 38, 316. Hughbanks, T. (I 992). In Inorganometullic Chemistry (ed. T. P. Fehlner). Plenum Press, New York, p. 289. Hume-Rothery, W. (1926). J. Inst. Metals, 35, 295. Kaskel, S., and Corbett, J. D. (2000). Inorg. Chem., 39, 778. Kasper, J. S., Hagenmuller. P., Pouchard, M., and Cros, C. (1965). Science, 150, 1713. Klemm, W. (1956). Trab. Hem, Int. React. Solidos3rd, 1,447. Klemm, W, (1959). Proc. Chem. Soc., 329. Miller, G. J. (1996). In Chemistry, Structure, and Bonding of Zintl Phaseasand Ions (ed. S. M. Kauzlarich). VCH, New York, p. 1. Queneau, V., and Sevov, S. C. (1997a). Angew. Chem. Int. Ed. Engl., 36, t 754. Queneau, V., and Sevov, S. C. (1997b), J. Am. Chem. Soc., 119,8109. Qimeau, V., and Sevov, S. C. (1998a). Inorg. Chem., 37, 1358. Queneau, V., and Sevov, S. C. (1998b). J. Am. Chem. Soc., 120, 3263. Ramachandran, G. K., Dong, J., Diefenbacher,J., Gryko, J., Mamke, R. F., Sankey, 0.F., and McMiIlan, P. F. (1999). J. Solid State Chem., 145, 716. Reny, E., Gravereau, P., Cros, C., and Pouchard, M. (1998). J . Mater. Chenz., 8, 2839. Scliafer, H., and Eisenrnann, B. (1981). Rev. Inorg. Chem., 3, 29. Schiifer, W., and Eisenmann. B. (1985). Rev. Mat. Sci., 15, 1. Sevov, S. C . (1993). Ph.D. Thesis, Iowa State University, Ames, IA. Sevov. S. C., and Corbett, . I . D. (1991). Inorg. Chem., 30, 4875. Sevov, S. C., and Corbett, J. D. (1992). Inorg. Chem., 31, 1895.
Sevov, S. C., and Corbett, J. D. (1993a). J . Solid State Chem., 103, 114. Sevov, S. C., and Corbett, J. D. (1993b). Inorg. Chem., 32, 1059. Sevov, S. C., and Corbett, J. D. (1493~).J. Am. Chcm. Soc., 115, 9089. Sevov, S. C., and Corbett, J. D. (1993d). Z. Anorg. AElg. Chem., 619, 128. Sevov, S. C., and Corbett, J. D. (1993e). Inorg. Chem., 32, 1612, Sevov, S. C., and Corbetl. J. D. (1993f). Science, 262, 880. Sevov, S. C., and Corbett, J. D. (1996). J. Solid State Chem., 123, 344. Sevov, S. C., Corbett, J. D., and Ostenson, J. E. (1993). J. Alloys Comp., 202, 289. Smith J. F., and Hansen, D. A. (1967). Acta Crystallogr., 22, 836. Somer, M., Carrillo-Cabrera, W., Peters, E. M., Peters, I<., and van Schensing, H.-G. (1998). 2.Anorg. dlfg. Chew?., 624, 1915. Tillard-Charbonnel, M., Chahme, A., Belin, C., Rousseau, R.,and Canadell, E. (1997). Chem. Eur. J., 3, 799. Todorov, E., and Sevov, S. C. (1997a). Inorg. Chem., % . , 4298. Todorov, E., and Sevov, S. C . (1997b). J. Am. Chem. Soc., 119,2869. Todorov, E., and Sevov, S . C. (1998). Inorg. Chem., 37,3889. Todorov, E., and Sevov, S. C. (1999). Angew. Chem. Int. Ed., 38, 1775. van der Aart, S. A., Verhoeven, V. W. J., Verkerk. P., and van der Lugt, W. (2000). J. Chem. Phys., 112, 857. van der Lugt, W. (1996). In Chemistry, Structure, and Bonding of Zintl Phases and Ions (ed. S. M. I(aWkmCh). VCH, New York, p. 183. von Schnering, H A . , Nesper, R., Curda, J., and Tebbe, K.-F. (1980). Angew. Chem. Int. Ed. Engl., 19, 1033. von Schnering, H.-G. (198 1). Angew. Chem. Int. Ed. Engl., 20, 33. von Schnemg, H.-G.. and HBnle, W. (1988). Chem. Rev., 88, 243. von Schnermg, H.-G,, Bartinger, M., Bolle, U., CarrelloCabrera, W., Curda, J., Gm, Y., Heinemann, F., Llanos, J., Peters, IS.,Schmeding, A., and Somer, M. (1997). 2. Anorg. Allg. Chem., 623, 1037. Wade, K. (1972). Inorg. Nucl. Chem. Leit., 8, 559. Wade, K. (1976). Adv. Inorg. Chem. Radiochem., 18, 1. Westgren, A. F., and Phragmen, G. (1926). 2. Metaifkunde, 18, 279. Xu, L., and Sevov, S. C. (1999). J. Am. Chem. Soc., 121,9245. Xu, L., Bobev, S., El-Bahraoui, J., and Sevov, S, C. (2000). J. Am. Chem. SOC.,122, 1838. Zintl, E. (1939). Angew. Chem., 52, 1 . Zintl, E., and Brouer, G. (1933). 2.Phys. Chem,, BU1, 245. Zitl, E., and Dullenkopf, W. (1932). Z . Phys. Chem., B16,183. Zintl, E., and Woltersdorf, G. (1935). Z. Elecrrochrm., 41, 876.
Books must follow science, not science books. Francis Bacon, 15
The verb “to theorize” is now conjugated as follows: ‘I built a model; you formulated a hypothesis; he made a conjecture.’ ohn Ziman, 1925-
It is a different sort of genius that audaciously posits the subtIe interactions among a complicated ensemble of atoms. Marry A. A ~ ~ t e 1921r,
Esther Belin-Ferr6 Universitd Pierre et Marie Curie - Centre National de Za Recherche Scientifique, Paris, France
1. Introduction The properties of intermetallics may be better understood through insight into their electronic as well as their atomic structures. Knowledge of the electronic structure and the nature of bonding may also give general trends, helpful when looking either for new intermetallics or for improving physical characteristics along definite lines such as enhancing resistance to corrosion, modifying mechanical properties, etc. Therefore, numerous theoretical as well as experimentalmeans have been developed to analyse the band structure and densities of electronic states (DOS) of these solids. For further background the reader may wish to consult the following chapters from Vol. 1 of this treatise: Turchi, Ch.2, on electronic theories of phase stability; Singh, Ch. 6, on band structures and their interpretation: and pintschovius, Ch.7, on phonon dispersion curves and their interpretation. In this chapter a few experimental spectroscopic techniques will be presented. These are interesting as far as electronic structure is concerned because they investigate transitions between two different quantum states of the solid, each one characterized by its energy and lifetime. Accordingly they provide direct relevant information on the nature and energy distribution of the electronic states and their interactions, therefore on the chemical bonds in the material. Total overlap of partial electronic bands in the valence band of a specimen implies a complete mixing of the corresponding states and suggests metallic or covalent bonding which is generally found for hard materials. Changes of the electronic and physical properties from one sample to another involve modifications in the electronic structure, for example, deviation of the electronic
energy distributions for a given intennetallic with respect to free-electron or d-like metals will underline that the specimen departs from a metallic-like behaviour. Also, shift of the valence and conduction band edges from the Fermi level will end up in the formation of a pseudo-gap or a real gap which clearly will imply deviation from metallic behaviour. We are not going to consider here all of the numerous spectroscopic techniques useful for investigating the electronic structure of solids (visible, UV and infra-red optical spectroscopies, Raman, NMR, ellipsometry . . .) but will restrict ourselves to only a few which we have chosen because they give direct information on the electromc distributions and can be compared directly with DOS calculations with no further assumptions and do not require sophisticated specimen preparation. These are photoemission, soft X-ray emission, X-ray photoabsorption, electron energy loss, and isochromat Bremsstrahlung spectroscopies. The first two techniques allow one to analyse electronic occupied states whereas the last three are well adapted to a description of the electronic distnbutions of normally unoccupied states. For each of these techniques, the underlying general principles will be presented. Their usefulness and main characteristics will be illustrated with only a few selected recent examples of their application to intermetallics presented from the large volume of the experimental information published so far.
2. Investigation of Occupied States
As mentioned,the occupied electronicstates of a solid can be probed by means of photoemission and soft X-ray
Intermetallic Compounuk Vol. 3, Principles and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 02002 John Wiley L Sons, Ltd.
136
Bonding and Stability
photoemission process involves three successive steps: first, the excitation of the photoelectron, second, its travel through the solid and third, its ejection into the vacuum from the very surface. This last step is governed by the photoelectron transmission probability. Accordingly, the binding energy EB o f the photoelectron is related to its kinetic energy by:
Figure E Scheme of the energy levels involved in binding energy measurements.EFis the Fermi level. EK(v)is the kinetic energy with respect to the vacuum energy level of the sample, and EB IS the binding energy of the photoelectron created by the incoming photons. It is measured wth a speGtrometer having the work function @'spech.om
emission spectroscopies. For a few decades now, since the original work of K. Siegbahn and his group in Uppsala (Siegbahn cf al., 1967, 1969), so-called ESCA (electron spectroscopy for chemical analysis) or photoemission spectroscopy (PES) has become quite a popular tcchniquc, so much so that very efficient and powerful spectrometers are produced and marketed by various companies. On the contrary, for soft X-ray emission spectroscopy, appropriate spectrometers are not commercially available; they mostly have to be designed and constructed in the laboratories that intend to use this technique. However, this situation will not long prevail for nowadays several groups, all around the world, operate such home-made equipments using the very intense light beams of new generation synchrotron facilities.
2.1 ~~otoerniss~oB Spectpnseollies Photoemission spectroscopies (PES) probe the kinetic energies EK of photoelectrons emitted from inner levels of a solid as well as from the outermost occupied band (OB) as a consequence of the interaction of the solid with an incoming X-ray (XPS) or ultra-violet (UPS) radiation of energy hv. Therefore, the initial state i s the ground state of the system whereas in the final state there is a hole created by the incoming energetic beam. Within the framework of the frozen orbitals (assuming that the presence of a hole in the final state does not affect significantly the remaining orbitals), the
where Q, denotes the difference between the work functions of the spectrometer and of the sample (Figure I). Usually, it is not necessary to know Q, exactly provided the EB scale is properly calibrated, e.g. by referring to a well-known level such as the 1s level of contaminating carbon, always present at the surface of a sample not prepared in a clean, high vacuum, or the Au 4f7j2 levels if Au can be evaporated onto the sample itself, It i s to be noted that the Fermi levels of the spectrometer and metallic specimens adjust to each other, whereas this does not hold true for non-conducting substances for which charging effects induce shifts that can be as large as several eV. As to the final energy resolution of PES measurements, it depends upon the natural energy width of the incident photon beam and the spectrometer experimental fwction. For XPS it is generally about 0.4-O.XeV and for U P S it may be as good as a few tenth of an eV. The binding energies of the inner levels of a specimen are obtained SeparateIy. On the contrary, for the OB, the contributions of all the electrons are summed9 modulated by the photoelectron cross-sections o:
where \Irf and yii denote the wave functions of the final and initial states, respectively. During its travel through the solid, the photoelectron is subjected to inelastic interactions with the electrons of the medium; therefore, the sampled depth is governed by the mean free path of the photoelectrons, denoted A. Thickness up to 3h are generally probed, which makes PES basically a surface-sensitive technique. Consequently, it is necessary to use very careful preparation and characterization of the surface under study. In addition, because of the inelastic interactions, the shape of the background on which the photoelectron peaks are superposed is not a simple function, and it is necessary to model it. In narrow energy ranges, say for instance for inner level peaks, a linear background can be subtracted, but this cannot be done for OB states that usually extend over more than 10-12eV.
137
Speetroscopic Approaches
The number n(c) of the emitted photoelectrons is a fmction of several parameters and can be written as
n(&)cx n ( ~ v ) . N ( E ~ i . N ( E ) ~ ~ ~ . a ( E ) . ~(3) (E) where n(hv) is the number of incident photons, N(E), and IV(E)~stand for the DOS at initial and final states, respectively, f& is the Fermi-Dirac temperature function, a(E) is the photoemission cross-section and A(E) is the detector transmission factor. The contribution to the total intensity due to an element E at a depth x in a direction referred to by its angle 9 with the surface of the sample is connected to its concentration and to the electron's mean free path. Therefore, accurate nieasurements of the intensity of a chosen core level of a given element in a sample make it possible to determine its concentration by comparing the same photoelectron peak in the specimen and the pure element. Also, profile variations in concentration can be achieved by tuning the angle 0. There are two kinds of PES techniques by which final states may be explored. In integrated photoernission all possible final states contribute to the spectra. As photoemission cross-sections depend on hv, tuning the energy of the incident photons allows one to achieve some site selectivity and to distinguish between the features due to individual contributions to the OB of an alloy, provided their respective EB values are different enough to be resolved experimentally. Yet, the photoemission cross-sections significantly favour d andf states relative to p or s states (Yeh and Lindau 1985; Yeh, 1993). Accordingly, no features arising from sp constituents are observed in the OB spectra of alloys which contain sp as well as d and f elements. This integrated photoemission technique is mainly applied to polycrystalline specimens. The experimental data are usually compared to total DOS calculations when available. Such comparison is made more meaningful by convoluting the theoretical DOS with appropriate functions to account for the photoemission cross-sections and experimental functions. One example is shown in Figure 2 (Brown et aZ., 1997). Here, the complete overlap between Pt and Mn d-like sub-bands is indicative of a metallic bond in this material. Xn angle resolved photoemission, the photoelectrons are collected over a small solid angle, and their damping is weak enough that only transitions with k momentum conservation are involved. Thus, energy dispersion curves can be probed, which allows one to investigate band structure topology as we11 as Fermi surfaces. The data are often treated using the free-
8
I
.
I
I
1
n
c 8
6
2
0
Binding Energy (e Figure 2 Panel (a) PES spectra for Pt,Mn with partial contributions of Pt and Mn. In panel CO) the calculated DOS (Tohyama et al., 1989) (thin solid lines) are properly broadened for a meaningful comparison with the experimental data (thick solid line)
electron final state model; and, when possible, they are compared to band structure calculations. This technique is labelled ARUPS or ARXPS according to whether incident ultraviolet or X-ray photons are used; it is applied mainly to single crystals. In addition to the inner level peaks and to the OB band, the photoelectron spectra show many satellite
Bonding and Stability
radiation emitted by the X-ray tube. It is often difficult to account exactly for this background in the OB range, and models are necessary; however, for core levels, a linear background can generally be subtracted from the raw data. The intrinsic shape of the energy distribution of a core-level peak is a Lorentzian curve whose full width depends at half maximum intensity (FWHM), .4&,1,,, on the life time z of the inner hole through: AE,,,,, =hz-'
7
l , , , . l ~ . . . l . * , , . l . * ~ . *
$IQ
BOO
890
BEO
am
IN^^^ ENERGY (ev) Figure 3 Ce 3d PES spectra in a senes of Ce(Pd,,Cux), IntermeVduics taken at 1486.6 eV incident energy (Ogawa et al., 1993)
contributions that may carry important information about the electronic structure and bonding. They arise from characteristic energy losses, Auger peaks, shakeup and shake-off processes, plasmon excitations, surface states, etc. as well as final-state effects. Piasmon excitations are found towards high I& of the peaks, whereas surface states give a contribution to the low EB side. Final-state effects usually do not dominate in the spectra except for elements with unfilled inner shells, like the 4f levels of rare earths. Then, the photoemission spectra show complex features andlor multiplets involving f", f n + l and fn+2 final state configurations (Figure 3) (Ogawa et al., 1993). In addition to the photoemission peaks that are superimposed on a background which arises from the inelastic diffusion of the electrons; for X P S there is an additional contribution from the Bremsstrahlung
(4)
Therefore, core level spectra are usually decomposed into Lorentzian curves and compared to standards. Actually, for meaningful decompositions, Voigt functions can also be used, since it is commonly admitted that the experimental function of the PES spectrometer is a Gaussian distribution of energy. The PES core-level spectra bring information from both qualitative and quantitative st'andpoints. We already mentioned that their study may be advantageous for measuring concentration. In addition, they are very sensitive to the chermcal environment; hence shape modifications (for example, broadenings in amorphous systems as compared to crystals) as well as shifts up to a few eV (usually less than 5 eV) with respect to pure elements can be observed in different saniples. Indeed, the energy of an electron in a corclevel depends on the attractive potential of the nucleus of the atom and the repulsive interaction due to the other electrons of the surrounding cloud. As during the photoemission process, all the orbitals of the system contribute to the screening of the hole; the screening energy involves two terms, one is dtte to the intraatomic relaxation of the electronic cloud of the perturbed atom itself and the other term comes from the relaxation of neighbour atoms. Therefore, corelevels are affected by any change of the potentials to which they are subjected. To account exactly for such core-level shifts requires reference to models. These involve relaxation energy terms, electronegativities, formation enthalpies, etc. Accordingly, investigation of the chemical shifts in an intermetallic compound is of great interest, as it brings information 011 the chemical state of the probed element. The example of Figure 4 represents the A1 2p3p inner level in fcc Al, an amorphous intermetallic AI-Cu-Fe and the oxide present at the surface of the samples. Note that the feature towards low kinetic energies (high binding energies) at about 3 eV from the maximum is due to the contributioii of the Cu 3p level to the spectrum of the intermetallic sample.
139
Spectroscopic Approaches
fee AI
1400.00
1404.00
1408.00
netic energy (eV) Figure 4 Core-level peak in amorphous AI-Cu-Fe (open rhombs) and fcc Al (solid line) xn the kinetic energy range around 1404 eV (binding energy around 75 ev). The vertical bars show the maxlma of the AI 2pSizpeaks in the oxide (maximum intensity) and the samples and so illustrate chemical shifts from fcc A1 to the amorphous AI-Cu-Fe alloy (Belin-Fe&, unpublished)
3.0
0.0
BINDING ENERGY (eV) Another interesting piece of information is provided by examination of the shape of the Fermi edge in relation to the metallic character of a sample. This comparison requires measurements with a high energy resolution; and, accordingly, UPS experiments are necessary. Many studies have been dedicated to the investigation of the Fermi edge of intermetallics; in recent years the presence of a pseudo-gap at EF in quasicrystalline alloys was verified for icosahedral as well as decagonal phases which gives indication that the electronic properties of such alloys depart from being metal-like. Figure 5 shows the Fermi edges of fcc Al and Au and icosahedrai Al-Cu-Fe and points out a clear deviation of the OB edge of the icosahedral intermetallic with respect to what is expected for a freeelectron-like system (Mori et al., 1991). As mentioned, ARUPS helps to map out the energy bands of a solid. As an example, we show in Figure 6 the binding energies versus k l for the r X direction of €-Fe% (100) measured between the third and fourth Brillouio zones and those calculated using a direct transition model with a free-electron final-state band
Figure 5 From bottom to top, expenmental Fermi edges (dots) and calculated Fermi edges (solid lines) for pure gold, fcc AI and an icosahedral AI-Cu-Fe quasicrystal. For this Iatter, the F e r n edge i s calculated assuming a freeelectronlike behaviour. The discrepancies between calculation and experiment demonstrate that the quasicrystal is by no means a freselectron-like system and emphasizes the low DOS at EF, therefore establishing the presence of a pseudo-gap
(Castro et al., 1997). The data are compared to band structure calculations by Mattheis and Hammann (1993). Topological investigations are based on the fact that the photoelectrons are diracted by the crystal lattice and therefore carry structural information. Fermi surfaces are analysed in such a way (Aebi et al., 1996) as exemplified in Figure 7, which shows the Fermi surface mappings of Pt (111) and its comparison with theoretical calculations (Pilo et al., 1999a and b). In conciuding this section, we note that only in very few cases of ordered alloys, have core-level as well as
140
Bonding and Stability
OB PES photo-diffraction patterns been used to decompose the OB into the partial densities of states of its components. We present in Figure 8 the total and partial DOS of AuCus (001) (Stuck et al., 1990). More information on PES techniques can he found in many textbooks, such as Barr, 1993; Brundle and Baker, 1978; Cardona and Ley, 1978 and 1979, or in Hiifner, 1995, etc. 2.2 X-ruy Emissiorr Spectroscopy
Figure 6 Experimental binding energies (solid dots) versus kL for the rX direction of t--FeSi (100). The solid lines are the calculated bands taken from Mattheis and Hammann (1993). These are shifted by 0.1 eV towards low binding energes for a better adjustment to the experimental results
An ionized or excited system is not stable and evolves spontaneously. Therefore, a hole in a core-level of a solid recombines, radiatively or not, from either a more external level or the OB. The energy implied in this recombination is the difference between the energies of the initial and final states. The initial state presents a hole in an inner level, and there remains a more external hole in the final state. The radiative process is the X-ray emission investigated by X-ray emission spectroscopy (XES). The non-radiative counterpart is the Auger effect, not treated here, which ends up emitting electrons and leaving two holes in the final state. Within the one-electron approximation, thus neglecting relaxation energy terms, X-ray emission involves two steps: first, creation of the inner hole at level L(n, 1, j ) and second, reorganization of the electronic cloud from another level L(d,Z',j') or from the OB. This is schematized in Figures 9a and 9b. The energy distribution of the emitted radiation is given by
Figare 7 Mapping of the Fermi surface of Pt(l1X) (left side) from PES measurements at room temperature with an incident radiation of energy 21.2 eV and its comparison with theoretical calculations (right side). Black dots of the calcuiation are to be compared with hgh intensiaes in the experiment
141
Spectroscopic Approaches
Figure 8 Partial and total DOS of AuC%(001) measured in the (1 11) direction. Calibration on the DOS scale was made assuming 10 and 30 electrons/ceIl for Au and Cu, respectively
or
for two levels. C(n, 2, j ) and L(n‘, I’, j’), equation (3, or for one level and the OB (equation (6)) participating in the transition, respectively. In equation (G), N ( c ) denotes the probed OB densities of states. Mf-, is the matrix element of the transition probability; it is constant or varies slowly with energy depending on l(PflrlfPi)lwhere P,and Pfare the wave functions of the initial and final states. Therefore, in equation (3, the initial and final states are highly localized; in contrast, in equation (6), Pi is a localized state whereas Pfdescribes a hole in the 03.3.Spatial overlapping between the initial and final wave functions is required for a transition to take place, and this requirement implies that only the amplitude of Pfin the core hole region will give a significant contribution to the spectrum. Accordingly, XES has a local character. In addition, it is necessary that the radiative transition
Figure 9 Sketch of the pnnaple of X-ray emission. Panel a represents an X-ray transition between two core-levels. Panel b corresponds to an X-ray emission involving the occupied band
probability differs from 0, which is achieved provided the transitions fulfil dipolar selection rules, namely: AI=fl,Aj=0, f l
(7)
As a consequence, if a core hole with symmetry 1 is created in an element A, alone or in a compound, the local density-of-states with I f 1 symmetry around this element, denoted by N(k)’*’ will be probed by the XES technique, thus giving also a partial character to XES. Note that (i) transitions with j’ = 0 --j = 0 are not allowed; (ii) transitions to 1 f 1 states are favoured with respect to those to l - 1, therefore, from a p hole, states with d character will be probed with a higher
142
Bonding and Stability
probability than states with s character. Consequently, comparison of the experimental XES spectra with partial densities of states calculations is possible. More details can be found in many textbooks, as for instance Agarwal (19791, or in Bonnelle (1987). The shape of the distribution for a transition involving two inner levels is a Lorentzian curve whose FWHM is the sum of the FWHMs corresponding to the two levels involved in the transition. The shape of the emitted intensity distribution to which the OB contributes results from the convolution of the probed partial local DOS and the Lorentzian distribution L(n, 5 j ) . The initial hole can be created either with impinging electrons or with photons. Irradiation by electrons easily produces multiple interactions; hence, satellite emissions due to the X-ray transition in the multipleionized or excited system may also be observed in the spectra. The XES technique is straightforward, directly applicable to any kind of material: crystalline, amorphous, insulator, bulk, powder, etc. In addition, according to the conditions of the experimental setup it can inform the investigator either with reference to the bulk sample or to a thin layer just below its surface. OB of deep interfaces in solids may also be investigated with XES induced by electrons by tuning the energy of the incident beam so that it reaches the depth of the interface in question (Fargues et al.? 1985; Szasz and Kojnok, 1985). The total energy resolution in an experiment results from the natural width A q n , I, j ) of the inner level L(n, 1, j ) and the instrumental function. AE(n, I, j j varies from a few meV up to several eV according to the atomic number and considered level (Krause and Oliver, 1979); and it is generally the most important parameter in choosing the appropriate transitions to investigate. Despite the fact that no absolute DOS values are obtained, it makes sense to compare thc DOS of a given element in various materials; because, according to equation (6), any change in the shape and energy of the spectral distribution will give direct insight into modifications of the corresponding N(f>. An example is given in Figure 10, which shows the Mg 3p distribution in the pure metal and in Mg2CoHs (Belin et al., 1987). The feature in the low transition energy range, not present in the Mg 3p band in the lower curve, corresponds to additional states due to Mg-H bonding in the hydrogenated intermetalk. A relevant characteristic of XES is that the contributions of the various components to the OB of a solid can be investigated separately, each one obtained in its own transition energy scale. TOachieve
I
, 1292.00
1296.00
1300.00
1304.00
X-ray transition energy (eV) Figure 10 Mg 3p distribubon of states in the pure metal (lower curve) and in NiMg2H, (upper curve) on the X-ray transition energy scale
a picture of the OB and have insight into the electronic interactions that are present, further adjustment of the various partial contributions on the energy scale is necessary. This can be realized on the binding-energy scale but requires knowing the Fermi energy EF for each individual X-ray transition energy scale. To do so, it i s necessary to determine the binding energy with respect to EF of the inner level participating in each Xray transition; this can be achieved thanks to complementary X-ray and (or) inner-level PES measurements. An example is shown in Figure I1 for Al~Cu2Fe.It presents the curves corresponding to the A1 3p, A1 3s-d, Cu 3d-4s and Fe 3d-4s contributions to the OB, respectively, as adjusted to the binding energy scale from the measurements of AI 2p3!2, Cu 2 p 3 p Fe 2p3p inner levels and the X-ray transikon AI Zp3p -+A1 Is. Each curve is normalized to its own maximum intensity. Note that as transition probabilities strongly favour d states, no actual view of the Cu and Fe 4s states is achieved. The overlap between the peaks for Fe 3d and AI 3p curves denotes a mixing of the corresponding states and therefore illustrates some hybridized, covalent character of the A1-Fe bond. The maximum of the Cu 3d peak is almost at a relative minimum of the A1 3s-d distribution curve. Here, this can be interpreted as a Fano-like interaction (Terakura, 1977) between the localized d states and the extended AI states which splits them into two bonding and non-bonding parts, located on each side of the Cu
C..,,n+r,,n*rrnirr
I 15
10
I
I
t
5
0
5
Binding energy (eV)
Figure 11 Al3s-d (dots), Al3p (solid tine), Cu 3d (stars) and Fe 3d (triangles) electronic distribution curves in OB of crystalline Al,Cu,Fe. The full vertical line at the ongm represents the F e r n level, which IS located within fO.1eV for A1 and f0.3eV for Cu and Fe respectively
12.00
8.00
4.00
,
143
Approaches
3dmaximum (Belin et al.,1992). There is a decrease of intensity at EF of the various counterparts as compared to pure metallic elements, which indicates the opening of a small pseudo-gap at EF in agreement with the fact that this alloy is a Hume-Rothery intermetallic. Comparison between experimental spectra and partial DOSS is shown in Figure 12 (Trambly de Laissardikre ef al.,1995). The agreement is very good for the Cu and Fe states, but for Al 3p and 3d states there is a slight discrepancy in the energy range 4 eV below EF which is still a matter of investigation by theoreticians (Papaconstaiitopoulos, 1997). Note that the features towards low binding energies from the maximum in the experimental curves of panels a and b are satellite contributions to the main peak. The situation is less simple when the sample contains elements with open shells. The initial state of XES is then the same as the final state in PES. Therefore many features are observed on the emission spectral distribution curves, which arise from final state effects due to both ionized and excited configurations. Figure 13 displays the Ce 4f-3d emission spectrum of CeNiz. Structures due to transitions involving excited and ionized 3d94f1, 3dI04f0, 3d945-' configurations can be recognized. In this example, the occurrence in the spectrum of peaks and bumps (arising from peaks of smaller intensity) characteristic of both Ce3+ and Ce4+ established that the CeNi2 intermetallic is actually a mixed valence alloy (Sonder, 1990).
0.00
Binding energy (eV) Figure 12 Experimental electronic distribution curves (symbols) and calculated partial DOS'S (solid lines) for AI,Cu,Fe. From bottom to top, the panels correspond to Cu d, Fed, Alp and AI s-d states, respectively. The thick line of the upper panel IS for the A1 s states whereas the thin one is for d states
0.00
10.00
20.00
30.00
40.00
Nmber of Steps Figure 13 Ce 3d4f emission spectrum of CeNi,. The two sets of peaks correspond to transitions involving 3dJI2and 3dsi2 inner levels, respectively. The bars show the features characteristic of Ce3+ whereas the arrows show those due to Ce4-i-
Bonding and Stability
144
2.3 C o r n ~ ~ e ~ n tofa PES r i ~ ~and XES Techniques
whereas Ru 5d states (triangles in XES curves) are near the EF.PES shows that the Cu states are domiiiant and also that the total DOS at EF is very low. XES establishes the low intensity at the EF, especially for the N states, and shows the interactions between the A1 and the Cu states on the one hand and between the AI and Ru states on the other hand. For Ru the energy resolution of XES spectroscopy is not very good, which makes the Ru spectrum broad. The high resistivity of quasicrystalline Al-Cu-Ru (around 1800 Q cm at room temperature (Berger et aZ., 1993)), is in line with the reduced density of the extended-like states, namely N 3 p , in the vicinity of EF which is about 20% of the intensity in fcc Al, and the marked pseudo-gap at EF as assessed from the energy distance of the AI VB edges from Ep. Another example is shown in Figure 15 for FeSi where the X P S and XES experimental results are
As PES provides a view of the total OB and XES gives the partial contributions, it may be interesting to combine both techniques for a better understanding of the electronic structure of a solid. This is exemplified in Figure 14, which presents the PES and XES OB spectra of icosahedral quasicrystalline Al-Cu-Ru. For such an intermetallic, no DOS calculations are available due to the lack of translational properties of the quasicrystalline atomic structure. Therefore, it is advantageous to investigate the electronic structure experimentally, Note that it is of particular interest to have a view of the AI (or Mg) states and of their interaction with the other components in such intermetallics, since the sp states are less localized in character than the d states and therefore are more sensitive to changes in chemical and topological environments. In addition, they can be principally involved in many physical and electronic properties; yet, due to cross-sections that noticeably favour d and f states, they cannot be recognized in PES spectra. Figure 14 shows as an example PES curves for A~~SCUZORUIS (top curve) and A&~~.sCUZSRU~~.~ (lower curve) icosahedral quasicrystals; taken from Nakamura and Mizutani (1994) and the XES curves for A16sCuzoRu15 from Belin-Ferri. et al. (1996). From both sets of measurements it appears that Cu 3d states (crosses in XES spectra) are in the middle of the OB
2o
i
15
10 5
0 n
>m
3
>
2
2 CJ
t
,
' O
0.4
$ -&
*
c
0.3
0.2 0.1
y-
0.0
.^*
0.3
0.4
X c
c a)
Q
12.00
8.00
4.00
0.W
-4.00
-8.00
Binding energy (eV)
Figure 14 PE5 (XPS) (two upper curves) and XES (four lower curves) for icosahedral Al-Cu-Ru plotted on the same energy scale with the vertical line at the origin showing the Fermi level. All measurements were earned out at room temperature
0.2 0.1
0.0
Energy (eV) Figure 15 Electronic structure of FeSi. Top panel: the XPS spectrum is the dotted line and the total DOS calculated with the LMTO method is the continuous line. Lower paiels: from top to bottom, the experimental Fe 3 d 4 , Fe 4p, Si 3s-dand Si 3p curves are the dotted and broken lies. The partial calculated DOS are solid lines for Fe 3d, Fe 4p, Si 3s and Si 3p, respectively, and the broken line is for Si 3d (Galakov et al., 1995)
Spectroscopic Approaches
compared to total and partial DOS calculations; the good agreement between the data validates the theoretical model (Galakov et al., 1995).
3. Investigation of Unoccupied States 3.1 X-ray Photoab~orpt~on Speetroseopy An incoming photon can deliver its energy to a core electron (n,l , j ) of a solid and promote it to a normally unoccupied level, provided the energy of the incident photon is equal or higher than the ionization threshold. This is schematized in Figure 16. Such a process leaves an inner hole in the core state. Therefore, photoabsorption spectroscopy (XAS) and XES can be considered to be complementary in that, whereas one spectroscopy refers to occupied states, the other concerns the unoccupied ones. Similar to XES, in the one-electron approximation, the energy distribution of a photoabsorption transition involving the unoccupied band (UB) results from the convolution between the energy distributions of the core-level and the conduction DOS respectively, weighted by the corresponding transition probabilities. The spectral energy distribution associated with the photoabsorption process is
.i)
~ ~ ~ - t l ! 2 ~ ( & R * 1, ~ ( ~ ,
(8)
Here also, the photoabsorption transition is expected to follow the dipole selection rules as expressed in
145
equation (7), and its probability depends upon the overlap between a final state, which may be of somewhat extended character, and a more localized initial one. This makes XAS both site and symmetry selective. Accordingly, the photoabsorption spectrum describes the sum of all the transitions from the core level towards continuum states of appropriate angular momentum. Since the matrix element of the transition probability is constant or varies very slowly with energy, XAS spectra are directly related to the unoccupied DOSS of the solid; but, similar to XES, no absolute values can be derived from the experiments. Here also, as for XES. the total energy resolution in an experiment results from the natural width AE(n, I, j ) of the inner level L(n, I, j ) and the instrumental function. If 10 is the intensity of the incident photon beam, and I, the intensity after transmission through a thickness x of the sample, I and 10are connected via the Beer-Lambert law:
I = Ioe-ltx
fi can be taken as the linear photoabsorption coefficient since this effect is largely dominant as compared to others such as the Compton and Thomson diffusions, Auger emission, etc. U , is a function of the photoabsorption cross-sections, CT, of the different constituents of the solid. Again, the variation of CT is proportional to both @i and @f, the initial and final wave functions. ‘@idescribes the initial state, here the ground state, and @f is the final state, namely the electron ejected into the UB. Measurements of I and 10 allow one to determine the variation of 1.1. As a result of the photoabsorption process, the core-level vacancy can be filled owing to the Auger process. Consequently a photocurrent, the so-called photoyield current Y, is associated with the electrons escaping from the sample. This current can be measured and gives rise to the so-called yield technique. The fraction of the photon intensity absorbed in the sample is (10- I ) / & where Z is the photon beam intensity at the depth x. Therefore, the photoyield cwrent Y can be written
Y = (Io - l ) / I o= y(1 - e-”)
(10)
where y is a constant. Because the penetration depth is much larger than the electron-escaping depth, Y can be considered to be ypx. Hence, Y depends on the mean-free-path of the escaping electrons, and the yield technique is basically surface-sensitive in contrast to the transmission technique, which is mainly related to the bulk material. According to the particular sample,
-
Figure 16 Sketch of the principle of X-ray photoabsorption
(9)
146
Bonding and Stability
the incident photon beam is totally absorbed through a thickness x that can vary from tens of nm to pm, However, the yield technique must be selected when films a few tens of nm thick, which are often required €or a good experimental accuracy in the transmission mode, are not available. TOsummarize, XAS allows one to investigate partial local densities of unoccupied states; and, therefore, comparison with theoretical partial DOSSmakes sense. As €or XES, further adjustment of an absolute energy scale is necessary to reveal a picture of the UB distribution; and again, measurements of inner-level binding energies allow one to adjust the various contributions on the binding energy scale. To exemplify, Figure 17a displays the unoccupied electronic distributions of Al p , Fe d-s and p as well as Cu d-s and p in Al7CuzFe (Sadoc et al., 1993); and Figure 17b shows the partid DOS calculations by Trambly de Laissardiere et al. (1995), broadened in order to account for the core-level life-time as well as instrumental broadenings. Similar to XES, qualitative changes in spectral shapes from pure elements to intermetallics or from one intermetallic to another one reflect differences in bonding due to modifications in charge transfer or hybridization. For instance, advantage has been taken of such a peculiarity to investigate mixed valence interrnetaifics and evaluate the valency rates. By combining both X A S and XES spectra, it is then possible to have a full view of the electronic distributions in a solid and a complete derivation of the electronic interactions. This is exemplified in Figure 18 which shows the experimental OB and UB in icosahedral quasicrystaliine AI-Pd-Mn. As the shape of the Mn UB d-like counterpart is similar to that in the pure metal (not shown in the figure here), this adjustment points out that localized states, Mn OB and UB states, essentially, are found on each side of EF.Accordingly, there is no significant charge transfer from Al to Mn states and only a small part of the A1 3p states are hybridized to Mn 3d states (actually 4s states are not seen due to transition probabilities that favour d states). The maxima of Mn and Pd UB states correspond to relative minima of UB AI p ; hence, this confirms the states are not mixed. Pure Al s OB states are found at about 6 eV below EF.Assuming the transition probabilities are constant in the investigated energy range, the AI p distribution is adjusted to be the same intensity at EF as its OB counterpart. The OB and U3 Al p-like sub-bands show a marked asymmetry with a minimum at EF which points out the formation of a pseudo-gap and a low intensity of the
0 -5 Binding energy eV
5
-10
Figure 17a Experimental empty states distributions in AI,Cu,Fe. Solid line: Fe d-s, crosses: Fe p , triangles: Cu d-s. diamonds: C u p , dots: AI p . All sub-bands are normalized to their own maxima. In this alloy, Fe d-s states are predominantly present at the Fernu level, mainly overlapping A1 p states, whereas Cu d-s states are found about 2eV above EF
-4.00
0.00
4.00
8.00
Binding energy (eV)
Figure 1% Theoretlcal partial DOS in A1,aFe frrambly de Laissardihre et al., 1995) broadened to account for the core-level lifk-.hmes involved in the XAS expenments and the instrumental functions. Fe d, Fe s, Cu d, Cu s, AI p , Fe p , and Cu p curves are shown. Note that in the calculations, the energies are counted opposite in sign to those in the measurements. Good agreement is found in the energy positions between the expenmental curves of Figure 17a and the calculations. Analysis of the shapes shows that empty Fe and Cu states are of mixed of d and s characters
147
Spectroscopic Approaches
12
4
8
0
-4
-8
Binding mergy (eV) Figure 18 OB and UB contributions in an icosahedral AI-Pd-Mn quasicrystal as obtained from XES and XAS expenments, respectively. Small dots: A1 3s-d, full line: A1 3p, diamonds: Pd dd-Ss, triangles: Mn 3 d 4 , large dots: Pd d-s, open dots: Mn d-s, broken line: AI p. Ail the curves are normalized to their maximum intensity except for the AI p distribution which. assuming the transiaon probabilities are constant, is adjusted to be the same intensity at EF as its OB counterpart
rather extended UB Al p counterpart above EF is observed. This is in line with the high resistivity of this intermetallic alloy (about 1000 f'l cm at room temperature at which the spectroscopic measurements are made) (Belin et aZ., 1994; Belin-Fed and Dubois, 1996). Note that XAS which refers to an energy range of a few tens of eV above the absorption edges is often labelled as XANES, namely X-ray absorption nearedge structure in contrast to EXAFS, extended X-ray absorption fine structures, which is concerned with energy ranges of a few hundreds of eV above the edges and mainly carries structural information. 3.2 Electron Energy Loss Spectroscopy A mono-energetic incident electron beam directed into a solid is slowed down because of collisions, either elastic or inelastic, with the core electrons. When the incident energy is high enough, ionizations and excitations of the electrons of the solid can be produced, and inner shell electrons can be promoted to unoccupied states of appropriate symmetry. Thus, electron energy loss spectroscopy (EELS) allows one to explore the distribution of unoccupied states (Figure
14). Note that as in XAS, there is a remaining core hole in the final state, therefore this spectroscopy is also site selective. The energy loss is measured with respect to the elastic peak in the reflection or the transmission modes for small scattering angles. The measurements give the fraction of the incident electron beam with energy E which is scattered into a solid angle dR with the energy from E1 to El+. If we denote by q the momentum transferred from the incident to the scattered beam, then the cross-section of EELS is given by: d2a/dQdE cx (l/q2)Im(--l/c)
(11)
1It) is the so-called energy loss function and E is the dielectric constant which is connected to the photoabsorption coefficient: For energies around lOeV, the loss function is mainly determined by the collective excitations of the valence electrons, whereas for higher energies core excitations are taken into account. Finally, in the independent particle approximation, and for small momentum transfers,
fin(-
148
Bonding and Stability
stie peak ero toss)
f ~
Energy Loss (ev)
Figure 19 Principle of electron energy loss spectroscopy. The energy losses are measured with respect to the elastic peak which is taken as the origin
Figure 20 EELS spectra describing AI s empty states in fcc A1 (upper curve) and in icosahedral quasicrystalline Al,,Cu,Ru,, (lower curve), respectively (Terauchi et al., 1996)
3.3 Bremsstrahlung Isochvomat Spectvoscopy
Consequently, EELS is subject to fulfilment of dipole selection rules. In the transmission mode, namely using particles of high incident energy, the spectra are equivalent to those measured using transmission photoabsorption, whereas for low energy excitations, the spectra obtained in the reflection mode are dominated by electron mean-free-path effects and therefore are essentially surface-sensitive, analogous to yield photoabsorption spectra. Note that when the incident electron energy is close to that of an absorption edge, extra features may appear on the spectra due to non-dipolar transitions. Note also that as for XAS, no absolute DOS values are achieved with EELS measurements, but comparisons between the same element in various samples is possible as the shapes of the spectral curves are related to the unoccupied DOSS (Figure 20). Features due to such energy losses are commonly observed in electron scanning transmission microscopy and are also found in XPS spectra where they generally give rise to broad peaks corresponding to the high binding energies of the main lines, so-called plasmon peaks. In UPS spectra, both surface and volume plasmons can be detected easily. Investigation of such plasmon features may provide information about the nature of the solid, in particular about the freeelectron-like character and number of oscillating electrons (Osterwalder et al., 1990).
An electron beam which penetrates a solid loses energy due to its interaction with the strong fields of the atoms. As a result, photons are emitted continuously up to a limit which corresponds to the very energy of the incident electron beam. This is the so-called Bremsstrahlung radiation (from the German word for braking radiation) which carries information on the solid itself. Its detection is therefore of high interest and gives rise to the Bremsstrahlung spectroscopies. There are two principal ways to detect the Bremsstrahlung radiation, as schematized in Figures 21a and b. In all cases, the incident electron is high in the continuum and is thus considered as a free electron. In
I
a
e-varies
b
Figure 21 Principle of Bremsstrahlung spectroscopy a: in the spectral mode and b: in the isochromate mode
Spectroscopic Approaches
process a, incoming electrons of constant energy are dumped into the unoccupied states and the photons emitted are scanned according to their energy. Because of energy conservation, their variation depicts the variation of the empty states. This is the 'spectral' mode. In process b, the detection is performed at constant photon energy by varying the energy of the incident electron beam; here also, the incident electrons populate the initially empty band, and the description of the variation of empty states is made in the 'isochromate' mode. The latter process is also often denoted as 'inverse photoemission' (IPES); hence, the spectral mode is generally labelled inappropriately as Bremsstrahlung isochromate spectroscopy (BIS). In any case, whatever the detection procedure is, these BIS andlor IPES spectroscopies describe the totality of the empty states and do not provide chemical selectivity. However, comparison with total DOS calculations is meaningful. In addition, Bremsstrahlung spectroscopies are surface-sensitive, since the mean-free-path of the electrons governs the depth sampled. Many studies have been dedicated to analysis of the role of transition matrix
149
elements and intensity distributions as well as manybody effects (Speier et al., 1985; Hoekstra et al., 1986; Sobczack and Auleytner, 1988; Speir et al., 1988). Experimental investigations have been carried out so far mainly at synchrotron facilities but also using home-made equipments. Two examples are shown, Figure 22 (Czyzyk et al., 1992) and Figure 23 (Qgawa et al., 3 993). In the latter, final-state effects due to Ce 4 . configurations can be recognized. r
.*r. P
-*
*.it
1
Figure 22 Bremsstrahlung isochromate spectrum of MoNi,. The upper curve is the raw experimental data, the dotted curve below is the experimental curve with the inelastic contribution to the background subtracted. The lowest solid curve is the calculated DOS. Peaks labelled A, C, D, E and features F, G, H correspond to both MOand Ni d states, M O and Ni p , Ni s, Ni p and to MO s states and MO and Ni p states, respectively
0
.
1
2
.
I -
1
4
.
1
.
1
.
6
I
ENERGY ABOVE EF {sv) Figure 23 Bremsstrahlung isochromate spectra of Ce(Pd,,Cu,.),. Multiplet structure arising from Ce f I and f configurations is observable whatever x . The second peak above EF IS due to both Ce and Pd X states
150
Bonding and Stability
1
-8
-6
-4
-2
0
2
4
Energy relative to E, (eV) c Figure 24 Principle of Bremsstrahluiig spectroscopy in the constant initial state inode
Another type of spectroscopy that involves Bremsstrahlung radiation is the constant initial state spectroscopy (CIS), in which the initial state is the one electron in the conduction fundamental state iband, and the final state is a core-level hole two electrons in the conduction band, as shown in Figure 24. Therefore, the observed spectra result from the autoconvolution of the empty DOS. In contrast to BIS, CIS can involve a deep inner level and so can be bulk-sensitive.
+
4. Conclusion
The various spectroscopic techniques introduced in this paper are only a few among those that can investigate DOSs of intermetallics. They have been chosen because they are straightforward. thus giving direct insight into the DOS, although not providing absolute DOS values. They are all complementary, each one revealing a special aspect of DOS. As far as occupied states are concerned, PES gives a total view of OB states modulated by cross-sections, but it is surface-sensitive and charging effects may occur for insulators and semiconductors. On the other hand, XES can investigate all partial local occupied electronic distributions in a solid whatever they may be, but XES is mainly related to the bulk (Figure 14). If one is interested in empty states descriptions, XAS and EELS provide partial local electronic distributions that may
Figure 25 Occupied and unoccupied densities of states in Y(Co,_&l,), investigatedwth X P S (left side of the figure) and BIS (right side of the figure), respectively. for different x concentrations. The vertical line at the origin is set at the Fermi level. and the curves are each normalized to the total area
be either bulk or surface-sensitive, whereas BIS techniques refer to total empty DOSs and are more likely applicable to the study of the surface. Note that as XAS is both site- and symmetry-selective, the combination of BIS and XAS techniques may give insight, for instance, into the role of the core-hole effects due to strongly correlated empty levels or to levels close to the Fermi level. It is also possible to combine PES and BIS techniques using the same instrument, therefore within the same accuracy, and have a view of total occupied and empty DOSs (Figwe 25) (Son et al., 1999). Similarly, in the same spectrometer, under the same experimental conditions, and hence with the same broadening, it is possible to analyse selected partial local occupied and empty DOSs that involve the same inner level from both XES and XAS measurements, provided films of suitable thickness are available for the photoabsorption experiments or the yield technique is used (Figure 18). All parameters that influence DOS, such as changes due to modifications in atomic structure or chemical environmentscan be investigated and characterized with all these spectroscopic techniques. Note that other spectroscopic techniques, not presented here, may also be complementary to one or more of the techniques described above. For instance, Nh4R or MBssbauer spectroscopies can be related to PES or XES measurements as they both give some view of the participation of states at EF. See, for
Spectroscopic Approaches
example, a recent study of the electronic structure of ion-beamprepared A31-,Fe, samples which combined Rutherford back-scattering, conversion electron Mossbauer and X-ray emission spectroscopies from which the metallic character of the compounds was ascertained (Traverse et al., 1996).
5. References Aebi, P., Kreutz, T. J., Ostenvalder, J., Fasel, R., Schwaller, P., and Sclilapbach, L. (1996). Phys. Rev. Lett., 76, 1150. Aganval, B. K. (1979). X-Ray Spectrosropy (Optical Series, S987, Springer Verlag, Berlin). Barr, T. L. (1993). Modem ESCA. The Prinnples and Practice of X-Ray Photoelectron Spectroscopy(CRC Press, Boca Raton). Belin, E., Gupta, M., Zoiliker, P., and Yvon, K. (1987). J. of Less Comm. Metats. 130, 267. Belin, E., Dmkhazi, Z., Sadoc, A., Calvayrac, Y., Klein, T., and Dubois, J. M. (1992). J. of P h y x Cond. Matter, 4,4459. Belin, E., Dankhazi, Z., Sadoc, A., and Dubois, J. M. (1994). J. of Phys.: Cond, Matter, 6. 8771. Belin-FerrB, E., Dankhazi, Z., Sadoc, A., Berger, C., Muller, H., and Kirchmayr, A. (1996). J. of Php.. Cond, Matter, 8, 3513.
Belin-Fed, E., and Dubois, J. M. (1996). .J. ofPhyS.. Cond. Matter, 8, L-717. Berger, C., Belin, E., and Mayou. D. (1993). Annales de Chimie, Fr., 18, 485. Bonnelle, C . (1987). Annual Report C, R. Soc. Chemistry of London, 201. Brown. D., Crapper, M. D., Bedwell, K. H., Butterfield M. T., Guilfoyle, S. J., Malins, A. E. R.. and Petty, M. (1997)., J. of Phys.: Cond. Matter, 9,9435. Brundle, C., R., and Baker, A. D. (1978). Electron Spectroscopy. Theory, Techniques and Applications (Academic Press, London). Cardona, M., and Ley, L. (1978 and 1979). “Photoemissionin Solids” in Topics tn Applied Physrcs, 26 and 27 (Spnnger Verlag, Berlin). Castro, G. R., Alvarez, J., Davila, M. E., Asencio, M. C., and Michel E. G. (1997). J. gf Phys.: Cond. Matter, 9, 1871, Czyzyk, M. T., Lawmczak-Jablonska. K., and Mobilio, S. (1992). Phys. Rev. B, 45, 1581. Fargues. D., Vergand, F., and Bonnelle, C. (1985). Surface Science, 163, 489. Galakov, V. R., Kunnaev, E. Z., Cherkaslicnko, V. M., Yarmoshenko, Yu. M., Shamin, S . N., Postikov, A. V., Uhlenbrock, St., Neumann, M., Lu, Z. W., Klein, B. M., and Shi Zhu-Pei (1995). J. ofPhys.. Cond. Matter, 7,5529. Hoekstra, H. J. W.M., Sperer, W., ZelIer, R., and Fuggle. J. C. (1986). Phys. Rev. B, 34,5177 Hiifner, S. (1995). PhotoeZectron Spectroscopy. Second Edition (Springer Verlag, Berlin). Krause, M. O., and Oliver, J. H. (1979). J Phys. Chern. Re5 Data, 8, 329.
151
Mattheiss, L. F., and Hammann. D. R. (1993). Phys. Rev. 3. 47, 13114.
Mon, M., Matsuo, S., Ishmasa, T., Matsuura, T., Kamiyd, K., Inokuchi, H., and Matsukawa, T. (1991). J. ofPhys.. Cond. Matter, 3, 767. Nakamura, Y., and Mizutant, U. (1994). Mater. Scz. and Eng., A181-182, 790. Ogawa, S., Suga, S., Bocquet, A. E., Iga, F., Kasaya, M., Kasnya, T., and Fujimori, A. (1993). J. of Phys. Soc. of Japan, 62, 3575. Ostenvalder, J., Greber, T., Hiifner, S.. and Schlapbach, L. (1990). Phys. Rev. Lett.. 64,2883. Papaconstantopoulos, D. A. (1997). Private communication. Pilo, Th., Hayoz, J., Berger, H., Gnom, M.. Schlapbach, L., and Aebi, P. (1999a). Phys. Rev. k i t . , 83, 3494. Pilo, Th., Hayoz, J.. Aebi, P., and Schlapbach, L. (1999b). Physica B, 259-261, 1118. Sadoc, A., Belin. E., Dankhazl, Z., and Flank, A.-M. (1993). J. of Non-Cr”vst.Sal, 153 & 154, 338. Siegbahn, K., Nordling, C., Fahlman, A., Nordberg, R., Hamrin, K., Hedman, J., Johansson, G., Bergmark, T., Karlsson, S. E., Lindgren, I., and Lmdberg, B. (1967). Nova Acta Regiae Soc Sci. Ups, 4, 20. Siegbahn, K.. Nordling, C., Johansson. G., Nordberg, R,, Hedman, J., Hedh, P-F., Hamrm, K., Gblius, U., Bergmark, T., Werne, L. O., Manne. R., and Baer, Y. (1969). ESCA. Applied to Free Molerules (North Holland, Amsterdam). Sobczak, E., and Auleytner, J. (1988). Phys. Rev. B, 39.6251. Son, J.-Y., Konoshi, T., Mizokawa, T., Fujimorr, A.. Koui, K., and Goto, T. (1999). Phys. Rev. B, 60. 538. Sonder, A. (1990). Thbe de I‘tTniversitt5 Paris Vi, Paris, unpublished. Speier, W., Fuggle, J. C., Dwham, P., Zeller, R.$ BIake, R. J., and Sterne. P. (1988). J. Ph-vs. C. 21, 2621. Speier. W.. Zeller, R., and Fuggle, J. C. (1985). Phys. Rev. B. 32,3597
Stuck, A., Osterwalder, J., Greber, T., Hiifner, S., and Schlapbach. L. (1990). Phys. Rev. Letr., 65, 3029. S~asz,A., and Kojnok, J. (1985). Applied Surface Science, 24, 34,
Terauchi, M., Tanaka. M., Teal, A. P., Inoue, A., and Masumoto, T. (1996). Phil. Mag. Let., 74. 107 Terakura, K. (1977). J Phys F, 7, 1773. Tohyama, T., Ohta, Y ., and Shimizu, M. (1989). J of Phys.: Cond. Matter, 1, 1789. Trambly & Lassardi&re,G.. Dankhan, Z., Belin, E., Sadoc, A., Nguyen Manh, D., Mayou, D., Papaconstantopoulos, D. A., and Keegan, M. (1995). Phys. Rev. B, 51,4035. Traverse, A., Belin-FerrB, E., Dankhan, Z., Mendoza-ZBlis. L., Laborde. O., and Portier, R. (1996). J. of Phys.. Cond. Matter, 8, 3843. Yeh, J. J. (1993). Atonair Calculation of Photoionisation CrossSections and Asymmetry Parameters (Gordon and Breacli, New York). Yeh, J. J., and Lindau, 1. (1985). Al. Data Nucl. Data Tables, 32, 1.
This Page Intentionally Left Blank
Chapter 8 Effects of Pressure on Stability and Properties Y. V. Levinsky Moscow Institute for Fine Chemical Technology, Moscow, Russia
During recent decades studies of effects of pressure on solids have been undertaken with exponential proliferation. There are at least four reasons promoting the specific interest. First of all, new high-vacuum or highpressure technologies have been developed; secondly, there is the opportunity for new materials with promising physical, chemical and mechanical properties; thirdly, investigations of processes under high pressure make a valuable contribution to further understanding of the quantum mechanical nature of the interactions in solid materials; and the last reason is a persistent improvement of experimental techniques in high vacuum and in particular under high pressure. Special attention is given to solid materials under pressures above 108 Pa. Surveying the thousands of entries in this area, the subject of intermetallic compounds has been given particular attention. Intermetallic studies, as a part of all inorganic investigations (pure elements, simple and complex oxides, halides, chalcogenides, phosphides, hydrides, borides, nitrides, carbides, etc.), have been increasing constantly. Thus, until 1990 the number of published intermetallic entries was about 2% (Tonkov, 1992) of all highpressure publications, whereas nowadays it has increased to 10-15%. And the reason is not a limit of simple substances to study but the constant discovery of new high-pressure phenomena in intermetallic compounds. Several aspects of the effects of high pressure on intermetallic compounds can be pointed out: Q
A thermodynamic one, i.e. phase transitions, shifts in phase-equilibrium compositions, different types of graphical representation of phase equilibrium.
Q
A kinetic one that focuses on the rate of transition to equilibrium, the stability of high-pressure phases and their decomposition under standard conditions. A structural one related to the mechanism of deformation of the lattice and the transformation of one crystal structure to another. A physical one that helps understanding of the quantum mechanical nature of intermetallic bonds, metal-superconductor transition, semiconductormetal transition, etc., changes of electrical, magnetic, optical, elastic and other properties. An applied one that outlines prospects for highpressure intermetallic phases useful in industry and technology; application of the obtained results in geology, astronomy, physics of the Earth; etc.
Thermodynamically it is convenient to examine a pressure/intermetallic (composition, structure) correlation, by analyzing an equation o f state that is of the general form: Q i d ~ l =0
(1)
where Q is the extensive parameter and pi the intensive parameter. If all the intensive parameters but temperature, chemical potentials and pressure are constant or their variations can be neglected, equation (1) yields SdT - Vdp -t
n,dpc,= 0
(2)
where S is the entropy, T, the temperature; V, the volume; p, the pressure; n, and p,, the mole fraction and chemical potential of the i-component. Thus, the analysis of the pressure effect on phase transitions and
Intermetallic Compounds: Vol. 3, Princ+les and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 02002 John Wiley & Sons, Ltd.
154
Bonding and StahiZity
equilibrium compositions comes to a comparison of thermal (SdT), mechanical (Vdp) and chemical (Xn,dp.,) energy contributions in the total energy balance of the system. Experimental methods allow examination of equilibria in systems over a wide pressure range (from 10-* to 10l2 Pa) that by convention can be divided into high- and low-pressure intervals. Low pressure is that commensurable with the partial pressure of one of the elements of the intermetallic compound, whereas high pressure can be defined as the pressure under which the dT/dp value for the first type of phase transition can be found in experiments.The condition assumed above has a conditional character because, for different intermetallic compounds, boundaries between the assigned areas vary greatly. At low pressure it’s practically impossible to imagine an experiment where the equilibrium gas pressure above the intermetallic compound would be higher than the external one. Practically, for all observed intermetallics, the partial pressure of the most volatile component above the intermetallic compound is the main contribution to the total cquilibrlum pressure (ptotl), i.e. it can be assumed that pi =ptoti. Since = - RT lnp, the external pressure variations result in changing both the mechanical (ptotldv)and chemical (bqdni) energy components in the total energy balance. Calculationsshow that under low pressures the value of the mechanical component of the free energy can be disregarded, Thus the effect of external pressure on equilibrium will be observed only through the changing activity of the more volatile component of the intermetallic compound. The effect of external pressure is equivalent to changing the activity of tbe volatile component. The case can be simulated in the q u a s i ~ e q u i l i b r i ~ scheme when inert gases, non-reactive with the intermetallic compound, are introduced into the system. These conditions can be realized, for example, using a mixture of inert gases with a very low partial pressure of the volatile component of the intermetallic compound or by making use of a gas mixture that contains a small amount of the vapour of the element which is a component of the equilibrium of condensed phases (for example, oxygen in the mixtures Hz/H20, CO/C02; sulphur in the mixtures H2/H2SI etc.) or in isopiestic and electrolytic schemes. The examples stated above permit observation of intermetallic compounds under pressures much lower than those that can be produced by rarefaction of gas. Thus composition and properties of some chalcogenides have been examined under conditions that
simulate a pressure of IO-” Pa. Gas rarefaction of this magnitude has not been produced in experiments and on the whole makes no physical sense. The effect of high pressure on the equilibrium of condensed phases in intermetallic systems becomes perceptible typically at pressures above 108 Pa (Tonkov, 1992). In experiments the effect of pressure on equilibrium has been investigated at static pressures up to 5 x 10” Pa (Sung, 1997; Jayirnian, 1986; Hemley and Mao, 1997), whereas under dynamic conditions up to 10” Pa (Altshuler et al., 1999; Neinz and Yealoz, 1984; Kanel et al., 1996). Despite there being a wide range of pressures under which phase equilibria in binary and multicomponent systems have been observed, the principles of their graphical plotting are the same. It has been shown (Pelton and Schrnalzried, 1973; Pelton and Tliompson, 1975) that a graphical representation of equilibrium in a two-coordinate system coinpleteiy depends on whether intensive (qi)or extensive (Q,) parameters are the fixed parameters. Substitution of one parameter for another within the same group does not affect the pattern of the phase diagram. - Q2/& Since only three combinations q1- q2. 9% and Q1/&3 - Q2/Q3 (where Q d Q 3 and Q2/Q3 are the specific extensive properties conjugate to the intensive parameters (qland q2)are possible), there are three basic types of two-coordinate representation of phase diagrams. Figure 1 illustrates their salient features and conjugation. On the diagram of the first type (Figure la) the intensive properties ql and q2 are plotted along the axes, the value 40, is determined by the values of ql and q2, and the other (i- 3) intensive parameters must have constant values. For a binary system the constant value must belong to an intensive property. In Figure lathe two-phase equilibrium a 8 exists along a curve (line), whereas the three-phase equilibrium a + 8 y i s represented by a triple point. The second type of phase diagram is represented by Figure l b and Figure lc. At constant q4,qs . . ., the parameters q 1 and cp2 are plotted along the ordinate and the ratios Qz/Q3 and Ql/Qs are plotted along the abscissae, respectively. Each line representing a twophase equilibrium in Figure l a corresponds to a gap on the phase boundary outlining fields of two-phase equilibrium in Figure l b and Figure lc, and this is the fundamental difference in the topology of diagrams of the first and second types. Figure Id represents the third type of phase diagram with the coordinates Q I / Q ~and QZ/@. Each triple point in the diagram of the first type corresponds to a
+
+
Efects of Pressure on Stability and Properties
6-
w 4 7
-
Figure 1 Fundamental types of phase diagram representation (Pelton and Schmalmed, 1973; Pelton and Thompson, 1975)
triangle in the diagram of third type representing equilibrium of the three phases with constant extensive parameters; the latter are defined by the triangle angles. All the above-mentioned features of the three types of phase diagrams pertain equally to unary, binary, ternary and n-component systems as well as to the unlimited number of meaningful types of energy and, consequently, the same number of intensive parameters. If we examne the effect of pressure on equilibrium in systems with intermetallic compounds (with no external variable field of force), the intensive parameters are commonly the temperature, chemical potential and pressure. So three-dimensional, not two-dimensional, diagrams represent equilibrium in a binary system. Pelton and Thompson (1975) found that temperature, pressure and composition are the most convenient coordinates (i.e. P-T-X phase diagrams). Isobaric and isothermal cross-sections of
155
these diagrams belong to phase diagrams of the second type. Three-dimensional P-T-X phase diagrams of systems with intermetallic compounds are not common because of their rather complex representation. P-T diagrams, being projections of the three-phase equilibrium curves (lines) of the P-T-X diagrams on the P-T plane, are the most common and convenient representations, although they do not contain the compositions of the equilibrium phases. Extensive information on P-T diagrams, their construction methods and usage can be found in monographs by Vogel, 1959; Rhines, 1956; and Levinsky, 1990, 1997. Crucial points in these diagrams are coordinates of points of non-variant four-phase equilibria. The parameters of four-phase equilibria and phase compositions for those binary systems with intermetallic compounds that have been examined in detail are listed in Table 1* (Levinsky, 1997). It follows from Table 1 that thermodynamic characteristics of equilibrium transitions with intermetallic compounds have been investigated in more detail under low pressures than high ones. Cd-Sb (Aptecar et al., 1977) and Zn-Sb (Ponyatovsky and Belash, 1977) are examples of systems studied at high pressure. We will examine the effect of pressure. within the limits of 10-2-1010 Pa on the graphical representation of equilibria in the Cd-Sb system. Figures 2a-c give the isobaric cross-sections of the P-T-X Cd-Sb phase diagram (Aptecar et al., 1977) constructed on the basis of the experimentally obtained results. These results, as well as the data of pressure and equilibrium vapour content under low pressure (Borg, 1961), permitted construction of the P-T phase diagram of the system (Figure 3) (Levinsky, 1997). One can easily show that diagrams Figures 2a-c and 3 are not in contradiction and that the diagram of Figure 3 allows us to construct the isobaric and isothermal cross-sections with CdSb intermetallic compound (diagrams of the second type) within the limits of 10-2-1010 Pa and 200-700 "C correspondingly. Monographs (Levinsky, 1982, 1990, and 1997) report methods of graphical representation of this type of section for the Cd-Sb system and others.
*References to the orignal equilibnum studies are given in Levinsky, 1997. It is recognized that the phases shown in the table as taking part at equilibrium under the stated conditions of T and P, according to the referenced literature, are not always consistent with the accepted phases under standard '7 and P conditions as shown in the phase diagram literature. These are issues that must eventually be resolved.
156
Bonding and Stability
Table 1 The parameters of non-vanant four-phase equilibrium in binary systems with intermetallic compounds (Levinsky, 1997)
System 1
AI-Bd AI-Ba Al-Ba AI-Ba AI-Mg AI-Mg AI-Mg AI-Mg AI-Mg AS-CO AS-CO AS-CO AS-CO As-Cd As-Cd As-Cd As-Cd As-Cd As-Cd As-Cd ASK AS-K AS-K
AS-K AS-K AS-Pd As-Pd As-Pd As-Si As-% As-Si As-Zn As-Zn As-Zn As-Zn Au-Zn Au-Zn Au-Zn Au-Zn Au-Zn AU-ZII Au-Zn Ca-Hg Ca-Hg Ca-Hg Ca-Hg Ca-Hg Ca-Hg Ca-Hg Ca-Mg Ca-Mg Ca-Mg Ca-Pb Ca-Pb Ca-Pb Ca-Pb
Temperature (“C)
Phases, taking part at equilibrium (composition in parentheses in atomic %; L=liqmd, G=gas)
2
538 730 750 914 320 370 450 450 437 918 923 958 1014 465 465 595 595 610 62 1 620 386 386 530 584 53 1 780 610 728 797 977 1097 651 65 1 750 723 403 683 438 490 520 585 654 84 486 513 530 587 660 N39 517 445 370 638 968 1127 750
4 1 15 4 3x 10-2 5x10-‘ 0.9 -1 1.2 -0.3 -2 20 103 9x 1 . 3 ~104 1.2~104 1.7~10~ 5x104 1.3~10~ ~ x I O - ~ 3x 1 103 (?) 3x (?)
2X1O6 7.5 2 3x106 3x106 4x105 104 103 9x103 103 10 30 102 1.2~103 4 x 102 4.5x 102 1 7 x 10-* 16 6 2 20 4 0 - 3
1 0.2~5 3~10-~ 8x10-3 40 2 x 102 40
Ba, A1,Ba4, L, G Al,Ba4, AI5Ba3,L, G AI,Ba,, Al,,Ba,, L, G Al,,Ba,, Al,Ba, L, G P(38 Mg), ~(50 Mg), R (42 Mg), G P(38 Mg), $55 Mg), R (42 Mg), G a (18 Mg), P(38 Mg), L, G P(38 Mg), $45 Mg), L, G ~ ( 6 0Mg), 490 Mg), L, G CO, Co,As,, L, G Co,As,, Co,As, L, G Co3As,, Co,As, L, G CoAs, Co3As,, L, G d-Cd3AS2,d-Cd,As,, L, G d-Cd3As2, d-Cd,As,, CdAss, G P-Cd,As,, dr-Cd3AS2,CdAs,, G aN-Cd3As,, ,B-Cd3AS2,L, G P-Cd3AS2,CdAs,, L, G CdAs,, L, G CdAs,, As, L, G a-KAs,, P-KAs,, KAs, G a-KA%,,P-KAs,, P - K ~ A s G ~~, KAs, P-KAs, L, G K,As4, KAs, L, G K3As, K3As,, L, G As, As,Pd, L, G As,Pd, AsPd,, L, G AsPd,, A%Pd,, L, G As, SiAs,, L, G SiAs,, SiAs, L, G SiAs,, Si, L, G cu-Zn,As,, D-Zn,As,, L, G a-Zn,As,, P-Zn,As,, ZnAs,, G P-Zn,As,, ZnAs, L, G ZnAs,, As. L, G a(30 Zn), o1(28 Zn), ,B (37 Zn), G a( 42 Zn), P (28 Zn), L (45 Zn), G Zn, ~ ( 8 9Zn), L, G ~ ( 8 5Zn), y3 (83,5 Zn), L, G y(76 Zn), y2 (75 Zn), -r3 (78 Zn), G ~ ( 6 2 3zn), y3 (78 Zn), L. G P (57 zn), r(62,5 Zn), L, G CaHg, Ca,,Hg,,, L, G Ca,Hg, Ca, L, G C%W, Ca,Hg, L, G Ca,Hg,, Ca3Hg, L, G CaHg,, CaPg,, L, G CaHg, Ca,Hg,, L, G Hg, CaHg, L, G Mg, CaMg, L, G y-Ca, CaMh. L, G B-Ca, y-Ca, L, G CaPb, CaPb,, L, G Ca,Pb3 CaPb, L, G Ca,Pb, Ca,Pb3, L, G Ca, Ca,Pb, L, G continued
Effects of Pressure on Stability and Properties
157
Table 1 continued System 1
Ca-Si Ca-Si Ca-Si Ca-Si Ca-Zn Ca-Zn Ca-Zn Ca-Zn Ca-Zn Ca-Zn Ca-Zn Ca-Zn Ca-Zn Cd-Mg Cd-Pd Cd-Pd Cd-Pd Cd-Pt Cd-Pt Cd-Pt Cd-Pt Cd-Pt Cd-Sb Cd-Sb Cd-Sb CO-Mg CO-Mg Co-Zn Co-Zn Co-Zn Co-Zn CO-Zn CO-Zn Co-Zn CO-Zn Co-Zn Cr-Si Cr-Si Cr-Si Cs-Sb Cs-Sb Cs-Sb Cs-Sb Cs-Sb Cs-Sb Cu-Mg Cu-Mg Cu-Mg Fe-Zn Fe-Zn Fe-Zn Fe-Zn Ge-Mg Ge-Mg Ge-U Ge-U
Temperature (“C) 2 1030 1040 792 1268 394 391 414 439 420 638 642 690 669 650 320 ~490 660 314 670 710 730 770 290 445 3 10 635 970 -419 566 675 690 746 895 924 925 966 1305 1413 1380 473 417 404 553 459 487 485 552 725 425 530 782 560 636 696 -93 1 1071
Pressure Pal 3
0.2 0.4 10, -102-103 9x1W3 9 x 10-, 3~10-~ 0.1 10 9x102 8X1O2 2.5~ 102 4xld 15 10 ~ 3 0 30 15 2x104 - 2 . 5 ~ 104 - 2 . 5 ~ 104 - 2 . 7 ~104 1 1o2 3 x 102 2x 102 20 - 7 ~ - 5 ~103 -5~10~ 1 . 4 104 ~ -8X 104 N105 Nl05
-1.5 x 105 -5~10-* -0.3 -0.1 -3.5x 10-2
-0.2 -0.7 - 2 . 5 ~103 N2X 102 N8X 102 10 10 30 10 2 x loz 104 2 x 102 N2X1o2 3 -10-4 5 x IO-‘
Phases, taking part at equilibrium (composition in parentheses in atomc %; L=liquid, G=gas) 4 CaSi,, Si, L, G CaSi, CaSi,, L, G O-Ca, Ca,Si, L. G Ca,Si, CaSi, L, G Ca,Zn, Ca, L, G Ca,Zn,, Ca,Zn, L, G CaZn, Ca,Zn,, L, G CaZn, C a n , , L, G Zn, CaZn,,, L, G Ca,Zn,, CaZn,, L, G Ca,Zn,, CaZn,, L, G CaZn,, CaZn,,, L. G CaZn,,, CaZn,,, L, G 4 2 8 Mg), ,436 Mg),L, G Cd, $19 Pd), L, G ~ ( 1 7Pd), Y,(23 Pd), L, G P 4 4 Pd), 71 (20 Pd), L, G y(87 Cd), Cd, L, G 71 (76 Cd), ~ ( 8 Cd), 5 L, G y,(74 Cd), y2 (73 Cd), L. G F(69 Cd), 7 2 (73 Cd), L, G PI (51 Cd), E (65 Cd), L, G CdSb, Cd, L, G CdSb, Sb, L, G Cd, Sb, CdSb, L Mg, MgCo,, L, G MgCo,, CO, L, G Zn, y2 (92.8 Zn), L, G y2 (92 zn), y,(88.6 Zn), L, G yL(88 Zn), 6(89 Zn), L, G ~ ~ ( 8 7 Zn), . 4 6(88 Zn), $85.4 Zn), G 4 8 9 Zn), y(85 Zn), L, G PI (56.5 Zn), 1175.2 Zn), L, G Pi (50 Zn), PO0 Zn), L, G p,(47.9 Zn), p(49 Zn), a(36 Zn), G m(36 Zn), ,449 Zn), L, G Si, CrSi,, L, G CrSi, Cr,Si3, L, G CrSi,, CrSi, L, G Cs,Sb,, Cs, L, G CsSb,, Cs,Sb,, L, G CsSb, CsSb,, L, G Cs,Sb4, CsSb, L, G Cs2Sb. Cs,Sb,, L, G Cs,Sb, Cs,Sb, L, G Mg, CuMg, L, G Cu2Mg, CuMg, L, G Cu,Cy,Mg, L, G E (93 Zn), Zn, L, G 6 (96 Zn), E (93 Zn), L, G a(41 Zn), r(70 Zn), L, G r(75 Zn), 486 Zn), L, G Mg,Ge, Mg, L, G Mg,Ge. Ge. L, G Ge, UGe,, L, G U,Ge,, y-U, L, G continued
Bonding and Stability
15% Table 1 continued System 1
Temperature (“C) 2
Pressure @a) 3
Hg-Mg Hg-Mg Hg-Mg Hg-Mg Hg-Mg Hg-Mg Hg-Ni Hg-Pd Hg-Ti Hg-Ti Hg-U Hg-U Hg-U Hg-U Hg-U In-Sb K-Sb K-Sb K-Sb K-Sb K-Sb Mg-Ni Mg-Ni Mg-Ni Mg-Si Mg-Si Mn-Si Mn-Si Mn-Si Mn-Si Mn-Si Mn-Si Mn-Si Mn-Si Na-Sb Na-Sb Na-Sb Nb-Sn Ni-Zn Ni-Zn Ni-Zn Ni-Zn Ni-Zn Ni-Zn Pd-Zn Pd-Zn Pd-Zn Pd-Zn Pd-Zn Rb-Sb Rb-Sb Rb-Sb Rb-Sb Rb-Sb Rb-Sb Sb-Zn
170 558 560 519 508 453 232 238 300 300 455 669 725 773 735 492 -63 407 397 460 507 506 760 1097 637 950 1150 1155 1234 1070 1040 1060 1205 1155 -97 -435 400 2130 418.5 490 675 810 875 1040 425 565 780 845 530 37 398 418 456 515 439 400
- 3 ~102 - 3 ~10’
-
Phases, taking part at equilibrium (composition in parentheses in atomic %; L=liqmd, G=gas) 4
- 2 . 5 ~103 - 1 . 5 ~10, -6 6x 103 104 3x104 1.2~104 10’ - 5 ~ 2x 106 - 4 ~106 -107
0.1 -1.1 - 7 ~10-4 -8X w4 5 -10 16 1.3~10~
-103 1.5~10~ 0.12 0.5 2.5 3.5 2.5 4 60 30 -10-5
-1o-’ 3 x 10-4 20 80
-103 1.5~10~ 2.5~10~ -6104 -10 50 -103 (?) 4 0 3 (?) 30 w10-4 3~10-~ 4x10-3 -10-2 -10 50 - 7 ~109
-
InSb; Sb, L,-G K3Sb,K, L, G KSb,, Sb, L, G KSb,, KSb, L, G K,Sb, K,Sb4, L, G K5Sb4,KSb, L, G Mg, Mg,Ni, L, G Mg2Ni, MgNi,, L, G MgNi,, Ni, L, G Mg,Si. Mg, L, G Mg,Si, Si, L, G MnSi, 75-n, Si, L, G MnSi, Mn, 75 .Si, L, G Mn,Si,, MnSi, L, G Mn3Si,Mn&, L, G Mn,Si,, Mn,Si, L, G p-Mn, Mn,Si,, L, G 6-Mn, y-Mn, L, G p-Mn, y-Mn, L, G Na,Sb. Na, L, G Na,Sb, NaSb, L, G NaSb, Sb, L, G Nb,Sn, Nb, L, G fi(89 Zn}, Zn, L, G y(85 Zn), 6(89 Zn), L, G p, (51.8 Zn), p(50 Zn), L, G a(32 Zn), @,(45.3 Zn), p(47.3 Zn), G p(58.3 Zn), y(74 Zn), L, G n(38 Zn), p(50 Zn), L, G $85 Zn), Zn, L, G Pd46 zn), P’ (58 zn), Y (76 Zn), G (63 Zn), y(76 Zn), G (65 Zn), $76 Zn), L, G PI (56 Zn), y (76 Zn), PdZn,, G Rb,Sb, Rb, L, G RbSb, RbSb,, L, G RbSb,, Rb,Sb,, L, G Rb,Sb, Rb, L, G Rb,Sb,, RbSb, L, G Rb,Sb,, Rb,Sb4, L, G ZnSb, Zn, Sb, L continued
Eflects of Pressirre on Stability and Properties
159
Table 1 continued System
Temperature ("C>
1
2
Sb-Zn Sb-Zn Sb-Zn Sh-Zn Sb-Zn Sb-Zn Si-IJ Si-U Si- U Si-U Si-U Sn-U sn-U Sn-U Th-Zn Th-Zn Th-Zn Th-Zn Th-Zn Y -Zn Y -Zn Y-Zn Y-Zn Y-Zll Y-Zn Y-Zn Y-Zn Y -Zn Y-zn Y-Zn
--
460 430 -510
-530 -510 -540 1320 1570 1575 1510 -1610 1350 1380 1136 -419 1041 946 1045 996 420 630 750 685 1015 870 880
905 895 860 870
Phases, taking part at equilibrium (co~positionin parentheses in atoinic %; L =liquid, G =gas)
Pressure (Pa)
4
3
-2X 109 -lOW -5 x 10' -2x 108 2 x lox -2X 1oS N
~
10-3 X
1.5~10 4x10 1.5 x 10-' -0.3 -1 -1 -2X 10-? -10 -103 7.5x 103 1.1x 1OS 1 . 6 105 ~ -20 - 2 . 5 ~102 -3 x 10, - 6 ~10, 2x 104 ~ 5 x 1 0 ~ -4~10~ -3~10~ -4x 104 - 6 ~104 2 x 10,
ZiiSb, @-Zn,Sb,, Zn, L P-Zn,Sb,, J-Zn,Sb,, Zn, L ZnSb, P-Zn,Sb,, y-Zn,Sb,, L P-Zn,Sb,, y-Zn,Sb,, <-Zn,Sb,. L y-Zn,Sb,, q-Zn,Sb,, 5-Zn,Sb,, ,O-Zn,Sb, y-Zn,Sb,, q-Zn,Sb,, C-Zn,Sb,, L IJ,Si, L, Si, G U,Si,, USi, L, G U,Si,, USi, L, G USi,, US&,L, G USi,, U,Si,, L, G Sn,U, Sn,U3, L, C Sn,U,, Sn,IJ,, L, G Sn21J3,y - u , L, G Th,Zn,7, Zn, L, G Th,Zn, Q-Zn, L, G ThZn,, Th,Zn, L, G ThZn,, ThZn,, L, G Th2ZnI7,ThZn,, L, G YZn,,, Zn, L, G a-YZn,, P-YZn,, YZn,, G a-YZn,, P-YZn,, YZn, G YZn,,,Y,Zn,,, L, G P-YZn,, YZn, L, G YZn,, Y,?ZnSR, L, G Y13Zn58, Y,Zn,,, L, G VZn,, p-YZn,, L, G Y3Znll,YZn,, L, G Y2Zii17,YZn,, L, G YZn, Y, L, G
The sections of a P-T-X-binary phase diagram at constant concentration do not belong to any of the three types of phase diagrams discussed above, and do not contain information on equilibrium phases for reasons similar to those for polythermal sections of three-component systems at constant pressure. It follows from this that the widespread representations of the first type of phase transitions of intermetallic compounds in pressure-temperature coordinates generally assume that cornposition is variable and P-T binary diagrams with intermetallic compounds are diagrams of the f k s t type (Figure la), but with the requirement that the chemical potential of one of elements of the intermetallic compound is a constant. This condition has not been met in experiments, at least those the author has been acquainted with (an experiment obeying these require~entsis unlikely to be proposed). Consequently, all the published P-T diagrams of phase transitions for binary (and nioreover for multicomponent) intermetallic compounds
cannot be regarded in principle as the equilibrium ones. This is likely to be the reason for the frequently observed differences in interpretation of experimental P-T diagrams of intermetallic compounds. The variation of composition of intermetallic compounds in phase transitioiis of the first type under constant, and in particular under variable, pressure has been noted many times (Levinsky, 1974; Degtyareva el al., 1983; Degtyareva and Ponyatovsky, 1987). But most investigations of high pressure on structure and properties of intermetallic compounds are carried out in physical laboratories; and, as a rule, the phase compositions before and after applied pressure are not defined with the required accuracy of at least 0.01 at.%. The kinetics of phase transitions in such cases are quite complex, for they depend on both temperature and pressure. The rate of approach to equilibrium at rather high homologous temperatures, T/T,n,lt > 0.70.8, is high and depends weakly on pressure. Usually, it
Bonding and ~tability
160 Sb, mass. % 600 S0Q 400
300 200
6uu
5u0
400
300
Sb,at. %
takes several minutes to obtain equilibrium and the rate depends on the heating unit in the high pressure apparatus. A di~usion~ e c ~ a n i sismlikely under static pressures, whereas for dynamic pressure applications the rate of phase change corresponds to an explosive wave velocity and the velocity of sound (Altshuler et anel et al., 1994). In the latter case, the rnartensitic mechanism of transition is more likely (Alt$~uleret al., 1999; Bacanov, 1984). It will be noted that the structure and composition of the high-pressure phases produced at the same temperature and pressure, either in static or in dynamic pressure
Fi Isobaric section of the Cd-Sb System (Aptecar et al., 1977) (a) curve 1 - 1 ~ 1 0 ~2 ;- 1 0 ~ 1 0 Pa; ~ 3 - 1 9 ~ 1 0Pa; ~ 42 7 ~ 1 0Pa; ~ 5 - 3 5 ~ 1 0 Pa; ~ 6 - 3 6 ~ 1 0 ~Pa; (b) p< 108 Pa; (c) p = 10 Pa
application, are practicdly the same (Altshuler et al., 1999; Bacanov, 1986). At relatively low homolo~ouste~p$ratures,achieving equilibrium may take several hours. A hysteresis of the direct and reverse tran~i~ions is found quite often under these conditions; as the t e m p ~ ~ a t increases, ~re the hysteresis becomes smaller and fiiially disappears (Tonkov, 1992; Chupenko and Degtyareva, 1984). ~uenchingkinetics are c o ~ p l i ~ a t efor d in~~rmetallic cornpouiids that are in e~uilibriumat high temperature and high pressure and are then transferred to room temperat~re and standard pressure, as are the
Eflects of Pressure on Stability and Properticy
t--ra:L,G \
;taa
305
400
$05
600
70#
T,"C ure 3 P-T phase diagram of the Cd-Sb system in the range 200 to 700 "C and 10-2 to 6x109 Pa (Levinsky, 1997)
transitions of quenched and metastable phases into the equilibrium state. Quenching parameters quite often depend on both the cooling rate and time of pressure as well as on how both processes are correlated. Most experiments report that the high-pressure phases are preserved in the standard state. Metastable phases as a rule are slow to transform and can exist at room temperature and even lower temperatures and at standard pressure for a long time; heating to several hundred degrees Centigrade is required to bring the inetastable phases to equilibrium (Tonkov, 1992). The effect of high pressure on the crystal structures of intermetallic compounds becomes apparent during elastic compression (reduction of the specific volume) and fornnation of phases with new lattices. Elastic compression of intermetallic compounds is performed at high pressure and is a basis for calculating elastic constants and related phenomena. Lattice structures of intermetallic compounds stable only at the high pressure are different from those under normal conditions. Nevertheless, existing experimental results show some general trends (Degtyareva and Ponyatovs~y~ 1987). The basic effect is that as the pressure rises so does the density of the intermetallic phase, i.e. the number of atoms in a unit volume of the intermetallic compou~dincreases. Higher pressures increase the probability of crystal structures with larger numbers of external electrons per atorn. Seiniconductor-type compounds are transformed to
161
metallic ones. The above-stated phenomena have been found for phase transitions under high pressure in the following systems InAs, GaAs, InP (Bao et al., 1995/96); CoGe, ReGe (Larchev and Popova, 1982); Ti3Al (Sahu et al., 1997); AuGe (6-phase), AgGe (5phase) (Fujinaga et al., 1991); GaSb ( 1996; Malyushitskaya, 1994; Ilina, (Matsumara et al., 1997); BaSi2 (Imai et al., 1997; Imai and Hirano, 1995; Imai et al., et al., 1998); AlSb, CdsAs:!; CrAsz; InAs; MnAs; FeMnAs; Z113A.s~; CeCuzSi2; CoGe; VFeSb; GaSb; R SnSb (Tonkov, 1992). An assumption has been made ( Ponyatovsky, 1987; Tissen et al., 19 prediction of high-pressure phases in a homologous series of intermetalli~compounds. Application of high pressure causes the same changes in phase compositjon as does substitution of one element of the intermetallic compound by another heavier member of the same group. For example, there are three intermetallic phases ZnSb, Zn4Sb3, and Zn3Sb2 in the Zn-Sb system at atmos~heric pressure. When pressure i s increased, the isobaric cross-sections become less complicated, and at 3 CPa only the compound ZnSb is observed in the T-X phase d i a g r a ~ . of diagram but with a heavier element Cd fCd-Sb system) is characterized by the same pattern at atmospheric pressure (Pon.yatovsky and Belash, 1977). The systems of Pb-Sb and analogous examples. Under standard pressure the isobaric cross-section. of Pb-Sb eutectic, whereas in the system intermediate ephase. A similar phase is found in the Pb-Sb system under pressure 3 GPa. is similarly followed in the In-Sb and (Degtyareva and Ponyatovsky, 1987; De 1998). As the pressure increases, new phases near the eutectic composition at standard pressure can be fou in the Ag-Ge (Degtyareva et al., 1996), Fe-Mo(Mordovetz and Rachek, 19951, and In-Sb (Miiiomura et al., 1994; Asaumi et al., 1975; 1976) systems, Amorphous alloys that are affected by pressure are often crystallized under different conditions from those at st an~ardpressure. ~rystallizatio~ of amorphous alloys under pressure has been seen in the systems Nb-Si (Malyushitskaya, 1994); Nb-Ge (Iwasaki et a 1984); Al-Ge (De~tyarevaet al., 1~96);Fe-Mo(Mordovetz and Rachek, 1995); and In-Sb (Minomum et al., 1974; Asaumi et al., 1975; and Asaumi et al., 1976).
Bonding and Stability
162
Most examiimtions of high-pressure on the properties of intermetallic compounds have two main objectives: 1) expanding the theoretical knowledge of the quantum mechanical nature of the interaction forces and the electron structures of intermetallic compounds; 2) searching for new phases with extraordinary physical or mechanical properties. The first objective is achieved mostly through direct measurements of the properties of specimens under pressure; for the second purpose, the properties are examined under standard conditions on quenched, metastable phases. Basic research methods for specimens under pressure are X-ray-diffraction studies, measuring magnetic properties and measuring electrical resistance. X-ray-diffraction measurements convey information for crystal structure definition and also the basic data for calculation of the elastic constants, density, and energetic characteristics. Knowledge of magnetic (or electrical) properties vs. temperature (or pressure) extends our knowledge about the nature of, and allows prediction of, such phenomena as superconductivity, Kondo effect and Hall effect, shifting Curie or NCel points, etc. Tonkov's handbook (1992) contains a detailed survey of results in this field €or the period up to 1990; more recent results include those of Movshovich et al. (1996), Tomas et al. (1996), Tang et al. (1996), Watako et al. (1996), Gavriliuk et al. (1996), Lord et al. (1996), Cornelius et al. (19971, Thomasson et al. (1998), Kojama et al. (1998), Kobajashi et al. (1998), Thessicu et al. (1998), Hauser et al. (1998), and Koyama et al. (1999). No intermetallic phase generated by application of high pressure has yet been found to exhibit properties of such interest for industrial applications as to be comparable to the impact of diamond and cubic boron nitride. But more extensive investigation of new phases at high pressure, improving the experimental techniques, and constructing industrial apparatus for superhigh pressure and high temperature (Sung, 1997) allow us to predict that interesting new phases will be unveiled in the years to come.
1. References Altshuler, L. E.,Trunin, R. F., Urlin, V. D., Forto, V E., and Funtikov, A. I. (1999). Uspechi Fiz. Nauk., 169(3), 323 (Russ.). Aptecar I. Z., Belash, I. T., and Ponyatovsky, E. G. (1977). High Temp.-High Pressures, 9, 641.
Asaumi K., Shmomura, O., and Minomura, S. (1975). Rev. Phys. Chem. Japan, Spec. Issue, p. 3 1 1 . Asaumi K., Shimomura, O., and Minomura, S. (1976). J . Phys. Soc. Japan, 41, 1630. Bacanov, S. S. (1986). Uspechi Khzm.. 55(4), 579 (Russ.). Bao, Z . , Anderson, Y . R., and Schmidt, V. H. (1995/1996). High Temp.-High Pressures. 27/28, 383. Borg, R.J. (1961). Trans. AZME, 221, 242. Brazhkin, V V., Lyapin, A. G., Khvostantsev, L. G., Sidorov, V. A., Tsiok, 0. B., Baylits, S . C., Sapelkin, A. V., and Clark, S. M. (1996). Phys. Rev., 54(3), 1808. Chupenko, G. V., and Degtyareva, V. F. (1984). Fiz. Tverd. Tela (Leningrad), 26(4), 1210 (Russ.). Cornelius, A. L., Gangopadhyay, A. K., Schilling, J. S., and Assmus, W. (1997). Phys. Rev. B: Condensed Matter, 55(21), 14109. Degtyareva, V. F., Belash, 1. T., Chipenko, G. V., Ponyatovsky, E. G., and Raschupkin, V. I. (1983). Fiz. Tverd. Tela (Leningrad), 25( lO), 2968 (Russ.). Deglyareva, V. F., Porsch, F., Ponyatovskii, E. G., and Holzapfel, W. B. (1996). Phys. Rev. B: Condens. Matter, 53(13), 8337. Degtyareva, V. F., Winzenick, M., and Holzapfel, W. B. (1998). Phys. Rev. B: Condens. Matter, 57(9), 4975. Degtyareva, V. F., and Ponyatovskii, E. G. (1987). In VZijanze Vysokich Davlenii na veshtesrvo, A. N. Piljankevich, Kiev, Izd. Naukov Dumka, p. 20 (Russ.). Fujinaga, Y., Kusaba, K., Syono, Y., Iwasaki, H., and Kikegawa. T. (1991). J. Less-Common Met., 170(2), 277. Gavriliuk, A. G., Stepanov, G. N., Sidorov, V. A., and Irkaev, S. M. (1996). J. Appl. Phys., 79(5), 2609. Hauser, R., Bauer, E., and Gratz, E. (1998). Phys. Rev. B, 57(5), 2904. H e m , D. L., and Yealoz, R. (1984). J. Appl. Phys., 49, 3776. Hemley, R. J., and Mao, H. K. (1997). Encycl. Appl. Phys., 18, 555. Ilinn, M. A. (1987). Fiz. Nizk. Temp., 13, 201 (Russ.) (Sov. J. Low-Temp. Phys.), 13, 111. Imai, M., and Hir'mo, T. (1995). J. Alloys & Comp., 224, 111. Imai, M., Hirata, K., and Hirano. T. (1995). Physica C., 245, 12. Imai, M., Hirano, T.. Kikegawa, T., and Schmonura, 0. (1997). Phys. Rev. B, 55(1), 132. Iwasaki, H., Okajima, M., Kondo, S., Wang, W. K., and Toyota, N. (1984). Mater. Res. Soc. Symp. Proc., 22, 6770. Jayarman, A. (1986). Rev. Scr. Instrum., 57, 1013. Kanel, G. I., Rasorenov, S. V., Utkin, A. V., and Fortov, V. E. (1996). The Blast-Wave Processes in a Condensed Substance. Moscow, Yanus-k (Russ.). Kobayash, T., Miyazu, T., Takeshita, N., Shimizu, K., Amaya, K., Kitaoka, Y., and Onuki, Y. (1998). J Phys. Soc. Japan, 67(3), 996. Koyama, K., Goto, T., Fujii, H., Takeshta, N., Non, N., Fukuda, H., and Janssen, Y (1998). J. Phys. Soc Japan, 67(6), 1879
Efects of Pressure on Stability and Properties Koyama, K., Goto. T., Takeshi, K., and Ryunosuke, N. (1999). J. Ph-ys. Soc. Japan. 68(5), 1693. Larcliev, V. I., and Popova, S. V. (1982). J. Less-Common Metals, 87, 53. Levinsky, Yu. V. (1974). Zh. Fiz. Khim., 118, 1818 (Russ.). Levinsky, Yu. V. (1982). P-T-X Phase Diagrams of Binary Systems. Moscow, Izd. Metallurgija (Russ.). Levinsky, Yu. V. (1990). P-T-X Phase Diagrams of Binary Metallic Systems. Moscow, Izd. Metallurgija. vol. 1, vol. 2 (Russ.). Levmsky, Yu. V . (1997). Pressure Dependent Phase Diagrams of Binary Alloys, ASM-MSI, New York, vol. 1, vol. 2. Lord, J. S., Riedi, P. C., Tomka, G. J., Kapusta, Cz., and Buschov, K. H. (1996). Phys. Rev. B. Condens. Matter, 53(1), 283. Malyushitskaja, Z. V. (1994). Neorganicheskze Materiali, 30, 1360 (Russ.) (Znorg. Mat., 1994, 30, 1267). Matsumara, T., Kosaka, T., Tang, J., Matsumoto, T., Takahaslu, H., Mori, N., and Suzulu, T. (1997). Phys. Rev. Left., 78(6), 1138. Minomura, S., Shimomura, O., and Sakai, N. (1974). In Tetrahedr. Bond. Amorph. Semicond. Int. Conf: Yorktown Heights, New York, p. 234. Mordovetz, N. M., and Rachek. A. P. (1995). Ukr. Fiz. Zh, 40(3-4), 254 (Ukrain) 8241. Movshovich, R., Lawrence, J. M., Hundley, M. F., Neumeier, J., Thompson, J. D., Lacerda, A., and Fick, Z. (1996). Phys. Rev. B: Condens. Matter, 53(9), 5465: 1996, 53(3), 8241. Pelton, A. D., and Schmalzried. H. (1973). Metall. Trans., 3, 1393.
163
Pelton. A. D., and Thompson. W. T. (1975). Phase Diagrams. Progress zn Solid State Chemi,stry, 10, 119. Ponyatovsky, E. G. and Belash, I. T. (1977). High Temp.-High Pressures, 9(6), 645. Rhines, F. N. (1956). Phase Diagrams in Metallurgy: Their Development and Application. McGraw Hill, New York. Sahu, P. Ch., Candro Shekar, N.V.. Mohammed Yonsuf, and Govinda Rajan, K. (1997). Phys. Rev. Letr., 78, 1054. Sung, Chien-Moin (1997). High Temp.-High Pressures. 29(3), 253. Tang, J., Matsushita, A., Kitazawa, H., and Matsumoto, T. (1996). Physica E, 217(1-2), 97. Thessicu, C., Kamishima, K., Goto, T., and Lapertot, G. (1998). J. Phys. Soc. Japan, 67(10), 3605. Thornasson, Y., Okayama, V., Sheikon, I., Bnson, J -P., and Braithwaite, D. (1998). Solid State Conimun., 106(9), 631; 637. Tissen, V. G., Degtyareva, V. F., Nefedova. M. V.. Ponyatovskii, E. G., and Holzapfel, W. B. (1998). J Phys.: Condens. Matter, 10, 7303. Tomas, F.. Ayache, C., Fomine, I. A., Thornasson, J., and Geibei, C. (1996) J. Phys: Condens. Matter. 8(4), L51. Tonkov, E. Yn. (1992). High Pressure Phase Transformations - A Handbook, vols 1-2, Gordon and Breach, Philadelphia. Vogel, R. (1959). The Heterogeneous Equzlibria. Geest und Poztig, Leipug. Watako, Y., Ishii, T., Oomi, G., Takahashi. H., Mon. N., Thompson, J. D., Shero. J. S., Madru, D., and Fisk. Z. (1996). J. Phys. Soc. Japan, 65(1), 27.
This Page Intentionally Left Blank
Chapter Magnetic Phase J. M. Cadogan School of Physics, The University of New South Wales, Sydney, Australia
1. Introduction A magnetic phase diagram shows the type of magnetic structure or magnetic ordering of a material as a function of parameters such as temperature, magnetic field, composition, pressure, etc. Most people will be familiar with the basic magnetic structures formed by intermetallic compounds, e.g. ferromagnet, antiferromagnet and ferrimagnet, and the reader is referred to Chapters 19 (Roth) and 40 (Kouvel) in Volume 1 of this series for discussions of the basic interactions and magnetic structures found in intermetallics. The reader i s also referred to the excellent summary of magnetism, magnetic structures and magnetic interactions in the Solid State Physics Source Book (Parker, 1988) for descriptions and definitions of magnetic phenomena. Examples of magnetic phase diagrams begin with Figure I , which will be discussed later. Each structure discussed in this chapter will be identified at its first mention with its ‘Strukturbericht’ type (where available), Pearson structure symbol, space group and space group number (#), based on the 1983 International X-Ray Tables. At its most fundamental level, the magnetic structure adopted by a compound is the result of energy minimization, and in this chapter I will discuss the magnetic structures found in intermetallic compounds, with an emphasis on intermetallics formed between rare-earth (R) and transition-metal (T) elements, in particular R-Fe-based compounds. There are two reasons for this emphasis; (1) such compounds have been studied extensively over the past 15 years due to their application as permanent magnet materials, with the tetragonal compound NdzFe14B (tP68, P42/mnm, #136) forming the basis for the world‘s strongest
permanent magnets and (2) unusually complex magnetic structures can arise in R-T intermetallic compounds as a result of the interaction between the localized 4f electrons of the R atom and the itinerant, non-localized 3d electrons of the T atom. Intermetallic compounds find numerous applications besides as permanent magnets and the reader is referred to the chapter by Stadelmaier and Reinsch (Chapter 14, Volume 2) of this series for a discussion of the practical applications o f magnetic materials. We refer the reader to the review article on the RzFe14B series of compounds by Herbst (1991) for an excellent summary of these principal interactions and the treatments thereof. Fornasini et al. in another chapter of this work, extensively review various rare-earth structures. Before considering the fundamental interactions which determine the magnetic structure of a compound and the associated magnetic phase diagrams it is instructive to consider briefly how one determines the nidgnetic structure of a compound. This will be followed by a discussion of some of the fundamental interactions (exchange, crystal-field) responsible for determining the magnetic ordering mode. Finally, I will illustrate various types of magnetic phase diagram with examples from a number of different intermetallic structures, once again with an emphasis on R-Fe based compounds.
2. Determination of Magnetic Structure
2.1 Neutron Diflraction Without a doubt neutron diffraction has made an inestimable contribution to the determination of
Intermetallic Compounds: Vol. 3 , Principles and Practice. Edited by J. H. Westbrook and R.L.Fleischer. 02002 John Wiley & Sons, Ltd.
magnetic structures in intemetallic compounds and it remains the most powerful meaiis of determining the detailed magnetic structure of a compound. The neutron interacts with the atoms in an interinetallic compound in two ways: (i) the nuclear interaction which gives rise to the nuclear scattering, used to d ~ t e ~ m i nthe e crystal structure (cell type, lattice parameters, space group, atomic positions etc.) and (iij the magnetic interaction of the neutron’s dipole moment with any magnetic moments 011 the atoms iii the crystal. The scattering produces peaks in the diffraction pattern according to the well-known Bragg formula
and the crystal structure is deterinined by (i) noting the syste~aticabsences of certain reflections (hkl) in a diffraction pattern obtained at temperatures above the magnetic ordering temperature so that the magiietic scattering is absent and (ii) fitting the relative intensities of the observed reflections. For exaiiiple, a body-centred cubic structure gives rise to reflections iller iiidices obey the condition w the magnetic ordering temperature, ~ d d i t i omagnetic ~ ~ ~ scattering arises from atomic planes along which there is some component of the ~ t o m i cmagnetic moment. In other words, those crystallographic planes which are perpendicular to the atomic magnetic moment do not contribute to the magnetic s c ~ ~ ~ t ~ rthe i n gscattering : intensity varies as cos2a where rx is tlie angle between the atomic magnetic moment and the crystal plane. Thus, one may often deduce intuitively the magnetic ordering direction of a crystal from a consideration of the presence or absence of any nzagnetic contribution to the nuclear scattered peaks in the ~nagneticallyordered temperature range. hirane (1959), the ability of neutron diffraction to determiiie the magnetic orderiizg direc, however, if one is studying tation arises from the supersition of non-identical reflections having the same one must consider scattering from a family of planes, each of which has the same interpla~arspacing and hence angle. The consequence of this av~ragingprocess is that one cannot determine the magnetic ordering direction at all for powder saniples of cubic crystals. For uniaxial crystals (tetragonal, hexagonal, rhombohedral) one can only determine the angle ( U ) between the magnetic moment and the unique (‘c’j axis; the azimuthal orientation (4) is indeterminate. For lower symmetry, e.g. orthorlioinbic, one may deterin~neboth 8 and #.
The first magnetic neutron diffraction pattern was obtained by Shull aiid mart (1949) from Pace-centred cubic MiiO (B1, cF8, F m h , fit225) and is shown in Figure 1. Below the magnetic ordering temperature of 122 K, extra peaks appeared in the pattern which could not be explained by the face-centred cubic crystal cell. In fact, the extra peaks corresponded to half-integer Miller indices and the ma gnetic structure was explained in terms of a magnetic unit cell which is obtained from the crystallographic cell by doubling the crystallographic cell along the three orthogonal directions a, b, c. This doubling of the crystal periodicity implies an antiferroniagnetic arrangement of neighbouring spins since a ferrornagnetic arrangement of moments would have identical magnetic and crystal cells and hence integer Miller indices. More exotic magnetic structure$ such as spirals, i n c ~ ~ m e n s u r a t e with the crystal lattice, give rise to peaks which may be indexed with Miller indices which are neither integer nor half-integer. An excellent accoirnt of the use of neutron diffraction to determine magnetic structures is that by
ossbauer spectroscopy is a local technique in the sense of providing a ineasurement of the local atomic magiietic moment, indirectly via the hyperfine field. Thus, one may estimate an atomic magnetic moment even in an antiferroinagiietic structure which would give a zero net niagnetizatioii when studied with a technique such as ~agnetoiiie try or thermogravimetry (in an applied magnetic field). The atomic magnetic inomeiit is only an estimate since one is actually measuring the liyperfine magnetic field at the nucleus produced by the atomic magnetic moment and there is, at times, coiisiderable uncertaiiity iii the conversion factor betweeii hyperfine field and atomic magnetic moineiit. The difficulty associated with preparing bulk magnetically aligned powder samples severely limits the ability of M~ssbauerspectroscopy to determine the magnetic ordering direction, but it is possible to observe the effects of changes in the ordering direction (spin-reorientations) via the associated change in magnetic liyperfine field resulting from the change in the orbital magnetic moinent and hence the orbital contribution to the hyperfine field. It is possible to carry out Mossbauer spectroscopy on single crystals but tlie crystal must be extremely thin to allow sufficient ~ a m ~ a - r atransmission. y
~ a ~ n ~Phase i i cDiagram
Figure P Neutron diffraction patterns of MnO obtained above and below the Ntel temperature (12% K). The top pattern, obtained in the a ~ ~ t ~ € e r r ~ ~ regime, a ~ i i e thas i c been indexed in terms of a magnetic cell whrch is o b t ~from i ~the ~ crystdlograpbic ~ cell b y doubling (Shull aizd Smart, 1949)
of the study of a magnetic crystal is that by Koon et aE. (1986) who studied the spin-reorientatio~in NdzFel4 (Figure 2). The ~ l e b s c h - ~ o r d a ncoefiicients which give the probabilities for t~ansitionsbetween nuclear levels show that the relative areas of the six lines in a magnetically split 57 Fe sextet are in the ratio
3:
~ 1 sin2(, . ~
1
4 sin20
+ cos26 :1:1:-----1 + cos28 : 3
where 8 is the angle between the hyperfine field (atomic magnetic moment) and the incident garnmaray direction. For a powder sample, this ratio reduces to the well-known 3 : 2 : 1 : 1 : 2 : 3 area ratio after powder averaging. The single crystal spectra shown in Figure 2 clearly demonstrate the tipping of the Fe magnetic ~ o r n e ~away t from the crystallographic c-axis. At room temperature, the atomic ~ a ~ ~~ eo t~ i eare ~~ atl is~ ~ ealong d the tetra
axis, corr~spondi~lg to 6 L=I 0" and g i ~ i na ~line ratio 3 : 0: 1 : 1 : 0 : 3. As the moments tip away from the caxis with decreasing ternperatu~e8 increases and the relative intensities of lines 2 and 5 in the spectrum increase. This spin-reorientation is ~ e t r i m e n t ato~ the use o f Nd2Fef4B as a p e r m a n e n ~ - ~ a g ~material et below 140 K. Spin-reorientations of the magnetization direction can also be detected by i ~ i e ~ the s ~temperature ~ ~ n ~ d ~ p e ~ d e n cofe the h ~ p e r field. ~~e contribution to the hyperfine field the spin component (p,) of the atomic magnetic moment but there is also a c o n t r i ~ ~ from ~ i o ~the ~ orbital c o ~ p o ~( pe L~) of t the ~ a ~ n ~~ t i oc ~The~ orbital moment depends on the orientation of the moment within a crystal structure, and ;i spinreorientation consequently produces a significant change in the hyperfine field:
~
Bonding and S ~ a ~ i 1 i ~ " v
action, one may detect spin~reorientations from the interplay of the electric and magnetic interactions in the total nuclear hyperfine Hami~tonian.For a site symmetry higher than ort~orhombic, the nuclear hyperfine Hamiltonian yields energy levels given by
[3cos2fl- 1][3m2- I(I+ l)] where fl is the angle between the hyperfine magnetic field and the principal Z-axis of the electric field gradient (EFG) whose largest component is VZZ,The first term in this equation is the eeman interaction of the nuclear magnetic moment with the hyperfine magnetic field. The second term represents the interaction between the electric q~adrupolemoment of the excited I = 3/2 state of 57Fe and the EFC, which is determined by the ch~rgesof the s u r r o L ~ ~ dlattice, i~g
4.2 K, with permissmn from Elsevier Science
where 01 and are constants, with opposite signs. In igure 3a we show the temperature dependence of the field at the 57Fe sites in hexagonal , P63/mmc, #194), as measured by ure also shows the hyperfine field at the crystallographic 4f Fe sites - the so-called 'dumb-beIl> sites. At room temperature, the Er3+ sub~atticeanisotropy is determined almost exclusively by the second~ordercrystal-field interaction. As the t e m p ~ r a is t ~reduced, ~ ~ ~ the higher-order (fourth and sixth) crystal-field terms at the Er3+ ion become stron.ger, relative to the second-order tern, and the ~r~+ moment rotates away from the room-temperature easy direction, The Fe atomic moment is therefore forced to rotate away from its preferred easy direction due to its exchange coupling with the Er3+ sublattice change in the orbital component of oment leads to an increase of the average hyperfine field at the 57Fenucleus of about 510%. A more detailed discussion of crystal-field effects i s given in the following section. In those cases where there is an electric quadrupole hyperfine interaction due to the surrounding charges in. the ~ ~ t t i c in e , addition to the ~ a g n e t i cdipole inter-
-0.1 5
0 F i g ~ 3r ~ (Top) The temperature dependence of the average 57Fehyperfine field and that at the 4f crystallographic site in Er,Fe,,C, (redrawn from Zhou et al. 1992), with permission from Elsevier Science; ( ~ o t t The o ~ te~perature ~ dependence of the ',Fe quadrupole splitting at the 12k crystal log rap hi^ site in Tni,Fe,,N,, (redrawn from Hu et u2. 1990b), reproduced with p ~ ~ ~ s s ~ o ~
169
Magnetic Phase Diagrams in addition to the electrons o f the host atom. Now, if it can be assumed that the EFG principal axis frame does not change direction with temperature, then any e in the physical orientation of the atomic magnetic moment will result in a change in 8 of the liyperfine magnetic field and therefore a change in the measured quadrupole split tin^ of the spectrum. In the extreiiic case of ail axis +-+plane ~ e o ~ ~ ~ there i ~ tis~a t i o ~ change in the experin2eiitally measured quadrupole splitting o f -2, being the extreme values of the second13 cos? 8 - 11 for order ~egendre polynomial O = O o - 90". In Figure 3b we show the measured quadru~olespiitting at the 12k Fe site in hexagonal ~ ~ fliP38, ~ ~~ ~ / ~ iy194), l l ~l as ~ ,~ ~ ~ ~by s ~1 l Hu et al. (1990b). The effect of the axis plane reorientation is clear. The reader is referred to onser (1975) for a detailed ossbauer spectroscopy and to the review on Nossbauer spectroscopy as applied to rareearth ~ e ~ ~ a magnets ~ e n t by ~ a d o (1996) ~ a ~ for further details.
4
~
~~
d
~
2
~
f-,
One may determine the easy direction of magnetization from ~eas u ~emen ts of the m ~ ~ n e t i z a t ~of o na single crystal as a ~ u n c of ~ iapplied ~ ~ ~ ~ a ~ n efield. t i c In practice, one orients the crystal using X-rays, fixes this orientation, then measures the magnetization along the various crystal directions. For ferromagnets one simply finds the direction along which the saturation magnetization is largest. In Figure 4 we show the results of a ~ i ~ ~ l ~i i- ~~ ~~ ~~ i §~study t t ~~ by l~ ~Cadogzln t i o ~er ~al. (1988) of the pe~anent-magnetmaterial Nd2Fel4 y do these data show clearly the tipping of the zation away from the crystal c-axis (the magnealong the 'easy direction' is 3% larger than that measured along the r-axis) but it is clear that drastic N
These so-called FO magnetization processes) are invaluable in the determination of crystal-field parameters and are the result of the complex interplay between the various energy terms in the anisotropy, whereby the magnetic structure adopts the minimum energy configuration. A detailed analysis of such processes has been given by Asti and Bolzoni (1980). actia
ethod to deternine the easy ation of a ~ e r r ~ m a g n e tor i~
~agnetizationcurves of single-crystal Nd,Fe,,B (Cadogan ei al., 1988). The canted magnetic structure of Nd,Fe,,B below the spin-reorientation tempera turc (140 I<>ts also shown, ca produced with p c r ~ i s ~ i oalfl 1OF
~
176)
Bondirtg and Stability
netic powder material is to disperse a finely or Nkl). Details of the pheiiomenolo~icaltreatment of powdered sample in an epoxy resin and allow the the exchange interaction as applied to mixture to set in an applied magnetic field. One can tallics may be found in the paper by Cadogan et al. then determine the easy ~agnetizationdirection from ( 1988). the cliffraction pattern by noting which (hld) peaks are The exchange interaction is dependent on the enhanced and which are reduced, or even absent, iiiteratoinic separation and is generally negative for n. This method relies on the short distances, leading to antiparallel coupling, and diff'raction is, in effect, positive for larger distances, leading to parallel layers of the powder and is co~~pling. Tlierc are a number of ways in which the not a bulk t e ~ ~ i ~due i ~ to u ethe limited penetration of magnetic structure of an iiitermetallic ciln be changed s mentioned with regard to the use of by c h a n g ~ ~the g inte~atoniicseparation, and thus the sbauer spectroscopy, it is cxtremely exchange Interaction. Such means include (i) atomic in fu'ull a~ignmeiitof the entire sample. substitution, which relies on differences in atomic radii, sim~~icity of this eth hod^ some care is necessary when (ii) application of hydrostatic pressure and (iii) absorpstudying antif~rr~iiiagnetic materials since the magtion of i~iterstitialeleinents such as H, N, 6 ,which leads netic susceptibi~~~y of an antiferromagnet in small to volume expansion. An example applied fields Is largest in the direction ~ e ~ ~ to e ~ interatomic ~ i ~ ~ separation i l ~ on ~ the exchan the axis along which the m o ~ e n t lie, s leading ofteii to hence the magnetic structure can be seen in the pseudoan incorrect idcntification of the casy direction of ternary system LaXV1-,Mn2Siz which is discussed in detail in section 4.5. The La3+ion is larger than Y3+and both ~ o s s b a u e rand n ~ ~d~ffraction t r ~ experime~ts ~ have been used to show that there is a conconiitant change in magnetic structure from f e r ~ o ~ a ~ ~atethe tic La end of the series to antiferromagnetic at the Y end. In passing, I would like to note that the magnetism of Mn and its intermetallics is notoriously complex and the e behaviour of tlie La,Y~-, n2Siz system is more enls in R-T iiitermetallics are in coinplicated than this simplc intera toinjc s ~ p ~ r a t i o ~ h the possible exceptioiis of Ce, argument would suggest. amiltonian dcscribing the magnetic There are interesting examples of magneti~structures which arise from the com~etition between different exchange terms. For example, if we consider 7 - l ~ o m 1= zrnrra-uion-ltr + 'Jfsrpm-arbii + K@\LI~ +R c ~ a chain of r n o i ~ e in ~ ~at crystal ~ structure and allow JI- 7 - l , ~ ~ ~ tion n ~ t~ ~~ ~ s ~ ~ ~i ~ ~ ~ ~ r ~ ~ only nearest-neig~bour (J1)l and next-nearest-~eigh~ bour ( J z ) exchange interactions, with different signs, where the various tesins are respectively (i) the intrathen it is possible to obtain a spiral atomic Coulomb interaction, (ii) the spin-orbit cou(Figure 51, such as in the case of pling, (iii) i ~ t ~ r ~ t magnetic o ~ i ~ c exchange coupling, approximation. Writing the excha (iv) the crystal field acting on the R3+ ion, (v) Heisenberg form ~agnetostrictiveeffects and (vi) any external effects e e ~ a ni~teractionof the moment with an externally applied magnetic field. The exchange and crystal-field terms are the dominant tions responsible for the magnetic behaviour of + nionient in a crystal.
"mean' field), and thc strcngth of this coupling may bc ed from the i n a ~ ~ e tordering ic temp~rature(Curie
it is straightforward to show that the exchange energy of a particular spin is E: = -J1(2 cos 4)- J2(2 cos 2 4 ) and the turn angle between adjacetit spins is simply 4 = c o s 1(-J1/4J2) (see Enz, 1960).
n addition to the ~ a g n e t i cexchange interaction, one must also take into account the magnetocrystalline anisotropy o f the R3) ion in an R-T intermetallic, which in some cases may be considerable, as illustrat~d
Magnetic Phase Diagrams
171
for hexagonal symmetry. The polar angles of the magnetic moment in the crystal axis frame are (0, 4)” The anisotropy energy terms K1, K2 etc. (which can be positive or negative and have quite ~ ~ ~ e rternperaent ture dependences) can be determined by fitting inagnetization curves obtained in applied magnetic field along the various directions in the crystal (torque measurements, Sucks~ith-Thompson etc.), The use of the K, anisotropy e fitting single-crystal inagnetization data, together with the relationship between these terms and the underlying crystal-field parameters &,,, can be found in the paper by Cadogan et al. (1988) and the reviews by Herbst (1991) and Franse and The case of XFtexch < 7 - l ~usually ~ applies to ionic compounds rather than to intermetal~i~s but does occur in R intermetallics which have very low magnetic ordering temperatures. The more complex case, when 7-tFlexcjz 7-lcF,requires that the full Harniltonian be diagonali~edand, in this case, no simple apprsximation can be made. As stated above, the main contribution to the magnetocrystalline anisotro tallics is the crystal-field electrons with surrounding and conduction electrons. The most common description of this tion is in terms of the point charge model which treats the interacinteraction between the tion as a simple C localized 4f electrons and a surrou~dinglattice of point charges. The interaction is then expressed in terms of a crystal-field Hamiltonian 7 - i ~ ~with : (Stevens, 1952) sp~n-operator equivale~ts ~ ~ ~ ( ~ which are polynomials in the angular moment~m operators, and crystal-field parameters energy terms whose m a g n i t ~ ~and ~ ss strength of the magiietocrystalline anisotropy and the symmetry of the crystal lattice PCM is a crude approximation nevertheless one may use the fo to great advantage under the protection of the crystal symmetry which the PCM, with all its i~perfections, ultimately must reflect! To begin with, the electrostatic potential at the 4f electrons (F$is written as a simple Coulomb summation over the neighbo~ringpoint charges (qj) at lattice positions (Rj)and then expanded in tei-ms of Legendre polynomials as N
*
Helical magnetic structur~o f Dy (schematic)
by the use of Nd@el& as the basis for the world’s strongest perinanent-~a~net material. The magnetocrystalline a n i s o t r ~ yi s a consequence of the crystalfield interaction acting on the R3+ion and the reader is directed to the review article by Hutchings (1964) for a complete description of the crystal-field interaction and its mathematical development in terms of spin operators. The exchange and crystal-field terms play the ~ o m i ~ arole n t in determin~n~ the magnetic properties of the R3+ ions in the crystal. In the situation where XFlexch > X C F , which is the case for R-T p e ~ a n e n t ~ m a g n materials, et the crystal field can be treated as a perturbation on the exchange interaction, and the anisotropy energy can be written in the classical form
for tetragonal symmetry, and E = K1sin20
+ rC, sin48 + K3 sin‘ 8 + K; sin‘ 8 cos 6d,
172
Bonding and St~~i~it”v
where cog is the angle between r;(0i, #J and Ri(Oj, #J). This may be rendered more tractable by using the spherical harinonic addition theorem to re-express Y ( r z )in terns of Tesseral harmonics and ~ nal l yin lar momentum operators J z , J2, J+ and J- of the 4f electron configuration, via the ~ i g n e r - E ~ k a rtheorem t (see, for example, Edmonds, 1957)).Finally, we arrive at the familiar form of the crystal-field Hanil1t onian describing the magnetocrystalline anisotropy o f the 4f shell:
n=O
m=O
For example, the dominant second-order diagonal operator is 0 2 0 = 3J5 - J 2 and the off-diagonal term Q& is Of;, r=: i ( J $ J!). The 7 - t ~ cannot ~; contain ternis of odd YI due to the inherent inversion symmetry of the problem; the anisotropy energy must be an even function of 0. Furthermore, if the quantization z-axis of XCF is a c-fold crystallographic rotation axis then XCF must contain terms of the form B,,,. The most general expression for 7 - l contains ~ ~ 27 terms in this expansion, but the crystal~o~raphic point symmetry of the atomic site in question reduces this to more manageable sizes. For example, in the case of hexa~onal sym~etry,such as in the R2Fel7 cornpounds with heavy R e l em e n~(and ~ SmCos), %er; takes the simple form
+
which is evidently more complex than the simpler form for the ThIVfnlz structure in which the R3+ point ~ tetragonal. syinmetry # / m is ~indeed To determine the magnetic structure adopted by a compound we must add the exchange interaction to the crystal field and, using standard notation, we therefore have
is an externally applied magnetic field and of the subl at t ~~e producing the is the m ag~et i zat i o~ exchiniige interaction. The exchange or molecular field or mean-field constant is A. The R3+ magnetizatio~is ~ ~ n{is the ~ number ) , density of R = ~ g ~ where R3+ moments and gJ is the Land6 g-factor. The equilibrium density matrix p i = (1/Z>exp ( - PEi) are the eigenvalues of where 1= ( l / k ~ Tand ) the 3t = 7i.exc/2+ 7 - t ~ ~ . The partition function , The ensemble av which leads to a iiig direction and to study spin-reorientations in the magnetization direction we need to study the angulardependent behaviour of the free energy of the system F(0, p) = - / t ~ T l n Z and the final magnetic configuration corresponds to the m i n ~ value ~ u of ~ F(0, q). We can separate the crystal-field parameters BE, into a 4f part and a lattice part: 4,m
in which the ‘diagonal terms’, i.e. with m = 0, determine the overall anisotropy (i.e. easy c-axis, cplane or inter~ediate)and the ‘0~“diagona~’ term with rn = 4 determines the in-plane anisotropy. Another efYect of the off-diagonal term in the ~ Y ~ isc Fto reduce the R3+ magnetic moment by crystal-field quenching. In the case of tetragonal symmetry, such as found at )~~ which have the R3+ site in the ~ ( F e , M coinpounds the ThMnlz structure (2)2b, tI26, 14/mmm, ##l39),we have 7 - 1 fI= ~ B~2 0 0 2 0
+ B&oi4 -k
$- B40040
.b)60060
+ &o&
At this point it is worth noting that 7 - t is~ determined ~ by the point symmetry of the site in question and nat the overall space symmetry. As an example we may whose overall crystal structure is ace group P42/mnm; but the Nd3+ site has the orthorhombic point symmetry mm, and the app~opriateXcF for the Nd3+ ions is
xci; = &O020
i60
3.B40040
+ & 2 0 i 2 + Bi40i4
+ B& o&+ B24o&+ B&jo&
=;
6,(
~
~
~
4
~
~
~
,
E
The Stevens (1952) constants & depend on the 4f electron configuration of the R3+ ion, whereas the crystal-fie~d lattice coe~cientsARm depend on the arrangement of charges in the surroundin are, to a first approximation, usually assumed to be ~onstantacross a series of compou~ds.Thus, within this f r a m e ~ o rthe ~ information on the sign of the crystal-field terms is contained within the 0, conn . example. Nd3+ stants for the R3+ ion in ~ u ~ s t i oFor has Oz .c: 0 whereas Tm3+ has 02 > 0 and one would therefore expect these ions to have opposing anisotropies (e.g. easy-axis vs- e a s y ~ p ~ in ~ nthe e ~absence of higher-order terms, The role of the sign of BEm in determining the easy direction of ma~netization may be illustrated simply by cons~derin~ the second-order operator term B20020= Bzo(3J; - J2).The corresponding energy is then ECF= ~ 2 0 ( 0 2 0 )= B20(3J; - J ( J + 1)) and we can consider the energies of two states, as shown in Figure 6 . The lJZl = J state, which corresponds to easy-axis ordering, has E ~ =FB20J(2J - 1) whereas the ~ J z = ] 0 state, which corresponds to easy-plane
173
Mugnetic Phase Diagrams m
**II R U I
.............................. e', .......'~*~*.........*.....".&...",.*. ................ I
I
..
m
*#?
Figure 6 The role of the second-order crystal-fieldterm in determining the magnetic ordermg direction: (left) easy-axis or (riglit) easy-plane
ordering, has EcF = -&0J(J f 1). Thus, 2320 > 0 favours easy-plane order and B20 < 0 favours easyaxis order. The relatively weak magnetocrystalline anisotropy of the Fe sublattice is not amenable to the above approach, due to the delocalized nature of the 36 electrons, so it it usually treated phenomenologically; and a siniple expression E, = K F sill28 for the anisotropy energy of the Fe sublattice is generally -T i~ter~eta11~cs7 where 0 is the nd tlie c-axis. In Nd2Fel4B, the Fe sublattice has axial anisotropy with lill ,-., 17 K/f.u. at room temperature.
As stated earlier, one may use the point charge model to great advantage provided one keeps in mind the crudity of this approximation. The symnietry will be correct but the magnitudes of the crystal field par~metersdeduced by the PCN summation should never be treated as 'cast in stone' - the signs of the calculated second-order terms are usually correct and it may be useful to consider relative magnitudes of . cannot terms within a given order, e.g. ( ~ 2 / B 2 0 )One deduce 0 priori values for the crystal-field parameters and the det~rmination of a set of Brim values appropriate to describe a magnetic structure remains an empirical exercise. There are numerous experiinental techniques which one may use to determine a set of crystal-field paramete~s,including single-crystal magnetization hyperfine interactions as f ( T ) heat capacity inelastic neutron scattering single-crys~alac-susceptibilit~
and the reader is referred to the review by for a discussion of these options.
s in
The R2Fel 4~compounds crysta~~ize in the tetragona~ Nd2Fel4B structure and, as mentioned earlier, are important as p e ~ a n e n t ~ m a g nmaterials. et The anisotropy of the Fe sublatttice is an easy-c axis. Above 140 K, the anisotropy of the R3+ sublattice is dictated by the second-order crystal-field term since the fourth- and sixth-order terms have more rapid temperature dependences and are only significant at lower temperatures. Those R3+ ions with I320 < 0 (Nd, Pr, Tb, Dy, €30) have easy-c-axis anisotropy whereas those R ions with B20 > 0 (Sm, Er, Tm) have easyplane anisotropy. As we have already mentioned, when Nd2Fel4B is cooled below 140 K, the hig~e~-order (fourth and sixth) crystal-field terms acting on the Nd3+ ions become important and the net magnetization tips away from the c-axis since these higher-order terms have opposite signs to the second-order term. In the early days of the R2Fel4B series much work was done on pseudo-ternary compounds in which two R elements were used, in an attempt to optiniize the per~anent-magnet properties. For example, it is known that Tb3+ has a significantly stronger crystalfield interaction, and hence anisotropy, than Nd3+ and partial substitution of Tb3+ for Nd3+ can be used to increase the overall anisotropy of the rare-earth sublattice. However, the Tb magnetic moment, being a heavy rare-earth, couples antiferro~a~netically to the Fe sublattice thereby reducing the overall ~agnetization,and one must t h e r ~ f o ~find e a compro-
inise between the increasing anisotropy and the decreasing magnetization when investigating the permanent-magnet potential of complex phases. A mixed system such as (NdYEr)2Fel4B(Figure 7) shows the effect of the competition between the negative secondorder crystal-field term of the Nd3+ and the positive tern of the Er3+. Thus, one obtains a transition from easy-c axis (& < 0) to easy-c plane (B20 > 0) as the Er content iiicreases (Cadogan and Li, 1992). ( ~ 12~ ,
The R(Fe,T) 12 compounds crystallize in the tetragonal ThMn12 structure (D2b, tI26, 14/mmm, #139) which ed in some detail in Chapter With the possible exception o f 12 reported by Cadieu et a/. 991), it has not been possible to prepare a binary 2 phase; a third element is needed to stabilize this 1112 structure. To date, the structure has been formed with T = Al, Si, Ti, V, Cr, MO, Mn, W, Nb or Ta: a high Fe content of around RFel1.3T0.7 has been obtained with Ta or N b as the stabilizing element. The anisotropy of the Fe su~latticeis easy-c axis, whereas that of the COsublattice in the isomorphous R(Co,T)]* compounds is easy plane. In general, R-sublatticc anisotropies are signi~cantlylarger than that of the Fe sublattice but in the R(Fe,T)12 compounds the two anisotropies are c o m p ~ r ~ b lThis e . near equality leads to a large number o f spin-reorientations being observed in these compounds. In Figure 8 we show
The competition between the axial anisotropy of R(Fe,T)lz and the planar anisotropy of R(Co,T)l? is illustrated in the phase diagrams (Figure 10) of the mixed series R(Fe,Co)llTi with R = Y, Dy, Gd, Ho, Tb and Er studied by Cheng et a/. (1990). As we shall see later, in section 4.6, the R(Fe,T)12 compounds provide viable permanent-magnet materials alternative to NdzFel4B after interstitial modification of their ma~neticbehaviour by the absorption of nitrogen or carbon. 17
The R2(Fe,M)17 compounds crystallize in either (i) the rhombohedral ThzZn17 structure (hR19, R h , #I 66)
for light R elements or (ii) the hexagonal Th2Nil7 structure (hP38, P63/mmc, #194) for heavy The anisotropy of the Fe sublatttice is easy-c axis. As mentioned in section 3. I , the exchange interaction between two moments depends critically on the separation of tlie moments. The effect of the application of hydrostatic pressure on the interatomic separation, exchange interaction and hence the magnetic structure of an intermetallic has been demonstrated in hexagonal Y2Fei7 by Nikitin ei al. (1991), who were able to produce a non-collin~ar helimagnetic structure starting from a collinear ferromagnet with increasing pressure (Figure 11).
The La(T,M)13 compounds form with the cubic NaZli13 structure (D23, cF112, F d c , #226) aizd this system has the highest transition~metalcontent of Et-T intermetallics, although the reduced anisotropy associated with the cubic symmetry probably precludes their use as p e r n i a n e n t - ~ a ~ ~materials. iet The only R-T binary with this structure is LaCol3, which is ferrornagnetic below 1290 I<. Fe-rich cornpounds in the range Fe1o.s-ll.s can be formed with stabilizing elements M = Si and A1 (Palstra et al., 1985). In Figure 12 we show the magnetic phase diagram of La(Fe1-x,Al.x)13 in which one observes antiferromagnetism at A1 contents in the range 0.08-0.14, ferroinagnetism for x in the range 0.14-0.38 and finally, mictomagnetisin for x in the range 0.38-0.54. (A mictomagnet is similar to a spin glass in that small groups of moments, coupled by direct exchange, are embedded in a spin-glass matrix phase. Early (pre-spin glass) work on mictoma~neticsystems employed a model involving a mixture of f e r r ~ m a ~ n ~ t i and sm antiferromagnetism, hence the name.)
The RT2X2 compounds (T = transition metals and X = Si, C e ) crystallize in the tetragonal ThCrzSi;! structure (D13, tIIO, 14/mmm, ##139) and have been discussed in great detail in the book by Leciejewicz (1994). These compounds are interesting because the only transition-nietal element known to be magnetic in this series is Mn. In those compounds with ~ i ~ n - ~ a g n eT,t i the c sublattice orders magnetically at low temperatures ( < 6 0 due to the indirect (RKKY) exchange intcractio etween R iraoments. In Figure 13 we show some of the possible ma
~ a g ~Phase ~ t Diagrams i ~
175
a
t
ure 7 Magnetic phase diagram of (Nd,Er),Fc,,B (Cadogan and Li, 1992). The letters a,b,c,d correspond to the magnetic structures shown below the phase diagram, reproduced with permiss~o~ from Elsevier Science
176
Bonding and ~ t u ~ i l ~ t y
Magnetization curves of single-crystal DyFe,,Ti (Hu et nl., 1990a)
Figure 9 Schematic experi~entalmagnetic phase diagrams of the RFe,,Ti series (Rzrare-earth) (HLI et al., 1989), reproduced with ~ e r ~ i s sofi ~1OP n
ordering modes of the R sub~atticein the compounds. It is clear that this series warrants to itself given the variety of magnetic order found in this series (ferromagiietic (F), antiferrom and various modulated structures best . general, though, we may terms of spin waves ( S ~ ) )In summarize the magnetic ordering in terms of the crystal-field interactions at the R3" ion, as detailed earlier. Thus, those R"+ ions with Stevens constant 82 < 0 (e.g. Tb, Dy the c-axis, whereas those R3" ions wi Tm) order perpendicular to the c-axis. As an aside, we note that the compounds with = Ce exhibit a wealth of properties such as heavy~fermionbehaviour due to the valence instability of the Ce ion. Unlike the remaining R ions, the Ce ion can empty its 4f shell by adopting a 4+ state rather than the usual 3+ state. The corresponding magnetic moment fluctuates between 2.54 fiB (Ce3+)and 0 fiB (Ce4+). The ma~neticorder of the Mn §u~latticein the RMn2X2 compounds (X = Si, Ge) comprises planes of Mn which are usually ferromagiietic. The overall magnetic structure of the compound depends critically on the interplanar separation. As mentioned in section. 3.1, short interatomic distances are associated with negative (antiferromagnetic~coupling whereas longer distances are associated with positive (ferromagnetic) coupling. The same applies with regird to the interplanar separation in RMnzX2 compounds. In
Magrzetic Phase Diagrams
174
~ ~ g u 12 r e Magnetic phase diagram o f La(Fe,Al),, (Palstra et al. 1985)
, Tb, Dy, No and Er) (Chciig et al., 1990) as conipiled by Szytula and Leciejewicz, 1994, reprinted with permission from CRC Press, Boca Raton, Florida
c
F
ure I1 Magnetic phase diagram of Y2Fe,, as a function of lattice co~pressionfrom applied pressure. PM, FM and HM represent paramagnetic, ferroinagnetic and helimagnetic structures, respectivdy (Nikitm et d., t991), reproduced with permission from Elsevier Science
Figure 14 we show the magnetic phase diagram of SniMnzGe:! as a function of applied pressure (Dumj et al., 1988). The transition from ferromagiietic order (F) to antiferromagnetic (AF) with increasing pressure, corresponding to Ia ttice contraction and hence decreasing interplanar separation, is clear. A similar effect in the S ~ ~ - ~ ~ (K ~ =~ n z ~ e z shown in Figure 15 where one se transition being driven by the replacement of Sm with R elements of different radii, leading to a change in the lattice parameter and, once again, the interplanar separation (Szytula, 1992). As shown in Figure 13, the magnetic structures formed in the R ~ ~ compounds 2 ~ z can be quite complex, with numerous different antiferromagnetic phases being formed. As an example, in Figure 16 we show the temperature-dependent magnetic phase diagram of TbNiZSiz as a function of magnetic field applied along the tetragonal c-axis ( 1991). One obtains ferromagnetic, antiferromagnetic and amplitude-modulated magnetism depending on the field and temperature. The distance dependence of the exchange interactions, which are responsible for the W magnetic order in this series, is demonstrated by the study of series ( ~ a , Y ) ~ n : ! S(Campbell i~ et al., 1999). is ferromagnetic (larger Mn- n se~aration)whereas YMn2 Sil is antiferroinagnetic (smaller separation). In Figure 17 we show the magnetic
178
Bonding and Stability
i~~~~ 13 ~ u ~ ~ of m the a ~different y types of magnetic order found in the RT,X, system (taken from Saytula and Leaejewicz, 19941, reproduced with permission from Elsevier Science
phase diagram of the mixed ( ~ a , ~ ) M n 2series, ~2 deduced from neutron diffra~tion and Mossbauer measurements carried out on (La,Y)MnzSiz samples doped with 0.5 wt.% of isotopically enriched 57Fe(the ossbauer isotope comprises only about 2% of
In Figure 18 we show a ‘universal curve’ for the Mn2X2 compounds (Szytula, 1992) in which the magnetic ordering temperature of the Mn sublattice is plotted as a fun~tionof intralayer separation. The critical distance marking the tra~sition~from antiferromagnetism to f e r r o ~ a ~ n eist ~2.85 s ~A, which agrees with the model of localization/delocalizationo f the 3d electrons proposed by ~ o o d e n o u ~(1h963). Similar eRects are seen in the Fe-based compounds RFel+Al,, where the critical Fe-Fe distance is 2.60 A ( ~ z y t u ~and a Leciejewicz, 1394).
In 1990, Coey and Sun showed that iiiterstitial ~odificationof the Rz(Fe,T)l? cornpounds leads to a doubling of the Curie temperature (T& and, perhaps more importantly, it can produce a change in the sign of the magnetocrystalline anisotropy. Thus, a compound such as SmzFe17, which is effectively useless as a permanent magnet material due both to its Tc being around RT and to its having an easy-plane anisotropy, leading to minimal coercivity, becomes a contender for p e ~ a n e n t - ~ a g napplications et after interstitial modification to Sm2Fe17Nm3 with a STc 700 K and §ubstantial coercivity due to its easy-axis anisotropy. This effect of interstitial modification is also observed in the (Fe,T)12 compounds with MdFe1ITiN, being a possible ~ e ~ ~ a n e ~ t - ~ i a g ~ N
Magrzetic Phase Diagram
Magnetic phase diagram of SmMn,Ge, as a function of applied pressure (Duraj et al., 1988). F = ferromagnetic, AF = antiferromagnetic aiid P= paramagnetic. The line segment in the range T < 105 K represents a transition to re-entrant ferro~~gnetism? reproduced with permission from Elsevier Science
material. In both the 2: 17 and 1 : 12 structures the interstitial N or C atom is situated arouard the R3+sites and causes a dramatic change in the crystal-field at the R3+ atom. One approach to modelling this effect is the bonding-charge model, proposed by Li and Cadogan (1992) and based on the work of Pauling (19601, in which one calculates the degree of electron transfer between the R and the N or C using electronegativity difPerences. Despite the fact that this is another example of a point charge model calculation, it has been quite successful in interpreting such experimental data as the m a g ~ e t i c ~ o r d ebehaviour, ri~~ anisotropyfield enhan~ementand change in m~gnetic-ordering direction. In Figure 19 we show the magnetic phase diagram of the GdZFel4B series as a function of €3 content, hang et al. (1988). (The 2:14:1 structure does not accommodate any significant a ~ o u n t sof N or C as interstitial elements, but one
179
Figure 15 Magiietic phase diagram of (Siii,R)Ikan,Ge, as a function of applied pressure (Szytula, 1992) The temp~rature T, marks the first-order ferro- to antiferro~ag~etic trans~tion and T, represents the transition to re-entrant f e r r o n ~ ~ ~ n e t i s ~ , reproduced with perniission from Elsevier Science
Figure 16 Magnetic phase diagram of TbNi,Si, (Blanc0 et al., 1991). Phases I and IV are a m p l i t ~ d e - ~ o d u l a ~phase e d , I1 is a simple a n t i f e r r o ~ a ~ i i(correspond~ng et to AFIII in Figure 13), phase 111 1s long-period coinmensurate and phase V is a complex magnetic phase, reproduced with permission from Elsevier Science
Bonding m d Stability
1450
7 Magnetic phase diagram of (La,Y)Mn,Si, as a function of temperature. F = ferromagnetic and AF = antiferr~magnetic,respectively. (Campbell et al. 1999). The magnetic structure notation is that of Venturini et al. (1993, in which Fmc = canted ferromagnetic, AFI = planar ~~ntiferroma~~ietic, AFI = axial antiferrQ~agnetic,AFmc = c a ~ t ~ a ndt i f e r r o ~ ~ ~ nand ~ t iF,,,,, c = a conical magnetic a r r a n g ~ ~ e nwhich t coexists with the canted magnetic structure, reproduced with permission of IOP
Figure 19 Magnetic phase diagram of Gd,Fe,,BH, (Zhang et al., 1988), reproduced with permission from Elsevier Science
\
La
I
can form a ternary R2Fe14C phase with most R elements.) The interesting point to note is the presence: of a spin-reorientation from axis to plane with increasing H content. Given that Cd3+ is an s-state ion and therefore any crystal-field effects will be extremely small, it seems the introduction of hydrogen alters the Fe sublattice anisotropy, even though the H atoms are preferentially located around the Gd3+ site.
I
One of the most complex ~ a g ~ e t phase ic dia~rams mapped out to date (if not the most complex) i s that of CeSb which, strictly speaking, is semi-metallic. Ce crystallizes in the cubic NaCl structure (Bl, cF8, Fm?m, #225), and its magnetic behaviour as a function of temperature and applied magnetic field was first studied in 1977 by Rossat-Mi~nodet al. The Dependence of the magnetic ordering t e ~ p e r a t ~ r e magnetic phase diagram of CeSb i s sh and ordering type of the RMn2X, c o ~ p o u n d son ~ntralay~r 20. Below the Nhel t ~ ~ p of ~16 ~ ~ ~ ~ r separation (Szytula, 1992). F = ferromagnetic, AF = applied magnetic field, CeSb exhibits five distinct antiferron~agnetic and P = paramagnetic, reproduced with p ~ r ~ s s i from o n Elsevier Science ~ a g n e t i cstructures which are all ~ o d u l a t e dwith
e
Magmiic Phuse Diagrams
P
I
T re 20 Magnetic phase diagram of CeSb ( ~ o s s a t " ~ i ~ ieti oal., d 1977). The notations F, FP, AFF, using I;=ferrotiia~iiet~c, AF = aniiferromagnetic and P =paramagnetic; so FP, for example, t-cpresents a strwture wlth both ferromagnetic and paramagnetic moments, as explained in the text, reproduced with permission from Elscvler Science
wave vectors [OQq]along the r-axis. The value of q varies discontinuous~ywith t e ~ p e r a t u r range $ to The peculiarity of this order i s that there are nonmagnetic Ce planes interspersed between mag~etic Ce planes in a t 0 JT 0 J a ~ r a n g e ~ e n t . These structures change in a complicated m a n ~ e r when one applies an external magnetic field and there are no fewer than 15 different nnagnetic structures known for CeSb, as a function of T and field. The u n ~ e r l y i nbasis ~ for this rich magnetic behaviour seems to be a hyb~dizationbetween the Sb p-boles and the Ce3+ 4f states. Xn a recent theoretical study of CeSb by Martiiiez et al. (1999) a frustrated Kondo lattice model has been proposed iizcorporatiag inter~ l a n a rexchan~einter~ctions ~ e ~ ~ neareste e n and next~nearest-neig~bours. The presence of the paramagiietic planes is tlieii explained in terins of frustration.
3.
-Fe-
The random addition of antiferroniagnetic exckaiige interwtions to leads to a loss o f P frustration. In extreme cases, foriiied with randoin isotropis: spin frcezing an neither a iiet magnetizatioii nor lower levels of frustration the acteristics of both extremes as long-ranged FM order
temperature Txy, followed by the
182
Bonding and Stability
re 21 Predicted magnetic phase d i a g r a ~of a system with a €rustration fraction f. PM, FM and SG are paratnagnet, ferromtagnet and spin-glass, respectively. (Thomson et d.,1992), reproduced with perinissioii of American Physical Society
F direct exchange, can be tuned via the composition as a consequence of the amorphous nature of the material. The magnetic phase diagram predicted for such a system using a Heisenberg model of nearest-neighbour iiateractions with a fraction o f the FM ‘bonds’ randomly replaced by AF exchange, is shown in Figure 21 (Thomson et al., 1992). The material phase diagram of the Fe-Zr binary system exhibits two deep eutectics at 24at.% Fe and 90.2at.’?/a Fe, which make it possible to prepare amorphous samples by rapid-q~en~hing techniques such as melt spinning. This system shows a wide range of beliaviour from superconductor to ferromagnet, depending 011 composition. At tlae Fe-rich end of the system the magnetic structure below TC comprises longitudinal ferrornagnetisni with transverse spin-glass order. At lower temperatures Fe-Zr undergoes a further transition to a state in which the spin-glass order in the transverse direction is frozen, without the ferromagnetic order (often referred to, erroneously, as a re-entrant spin-glass by imperfect analogy with superconductivity transitions). The two ramagnet (PM) 4 FM and ave been ineasured by a variety of tecliniques including Mossbauer spectroscopy, magneization and muon spill 1995; Ryan et al., 1998,
Figure 22 Magnetic phase diagram of amorphous Fe-Zr (Ryan et al., 2000) showing data obtained by (i) muoii spin relaxation, (ii) ac-susceptibility and (iii) “Fe Mossbauer spectroscopy. T, aiid T,, correspond to the transitions shown m Figure 21, reproduced with permission of American Physical Society
0I
%w#
t
I
I
I
tt
~ ~ ~ 23 u rMagnetic e phase diagram of amorphous Fe-Ru-Zr (Ryan rt al., 1998) showing data obtained by (i) acsusceptibility, (ii) neutron depolarization, (iii) magnetometry (Arrott plots) and (iv) 57FeMossbauer spectroscopy. T,, is the temperature at which a break in the temperatme dependence of the hyperfine field measured by Mlissbauer spectroscopy was observed. Thus, T,, corresponds to Txy in Figures 21 and 22, reproduced with permission from Elsevier Science
Magnetic Phase Diagrams
Y
I (1
61001
I
I
li I
183
field. Gignoux and chmitt (1995) have reviewed the effects of such fundamental interactions as crystal-field effects, quadrupolar interactions and bilinear exchan frustration in causing these t ~ a n s i t i o ~ ~ . In Figure 24 we show the single-crystal magnetization curves obtained on tetragonal DyGozSi:! where two field-induced ~ e t a m a ~ n ~transition$ tic c-axis are observed (Iwata et al., 1990) behaviour has becn observed in TmGa3 by d.(1987) where q u ~ d r u ~ o l ainteraction^ r are responsible for the etam magnet ism. ignoux and Schniitt (1995) have also coiisidered the state o f theoretical work on these processes and the ~greementbetween e ~ e r - i m ~ r o v iexperimental n~ and theoret~ca~ techniques.
clusio Magnetic phase diagram of DyCo,Si, (lwata et al., 1990), reproduced with permission from Elsevier Science
2000). In Figure 22 we show the magnetic phase diagram of a-Fe-2.3 which includes all these data. As the Fe concentration increases the two transitions approach each other, as expected from the spin-glass mean-~eld treatments described earlier (Figure 20). Eventually, the two transitions merge and thcre is a single magnetic transition from paramagnet directly to spin-glass. ~nfortu~iately, it has not proven possible to prepare fully amorphous samples with Fe concentrations beyond the critical concentration. However, by partially substit~tingRu for Fe up to 5 at.% Ru, Ryan et al. (1998) were able to study the cross-over froin FM to SG since the addition of nonmagnetic Ru leads to a rapid increase in frustration. In Figure 23 we show the magnetic phase diagram (zerofield) of ainorplzous a-Feg+xRuxZrio showing susce~tibility,magnetometry, neutron depolarization and Mossbauer data, which clearly map out the PM 4SG and PM -+ FM -+ FM + SG regimes with a critical Ru ~o~centration of 2.5 at.%.
In rare-earth intermetallic compounds in which only the R sublattice is magnetic, numerous types of metam~gnetictransitions have been found. use the most general sense of the word metamagnetic to refer to the breaking of an antiferromagnetic structure usually due to the application of a magnetic
The determination of the magnetic phase diagrams of rare-earth, transition-metal intermetallic compounds provides invaluable data which enable one to study such fundament~linteractions as fields, quadrupolar ordering, etc. understanding of the magnetic behaviour of those inter~etallicswhere the exchange int~ractionis larger than the crystal-field interaction, such as is the case wi tli permanent-magizet nia terial s. tallic compounds in which this criterion does not hold continue to provide a challenge to our understanding. Besides providing informalion on the fundamentals of the magnet~smof such i n t e ~ e t a l l ithe ~ ~deter~ina~ tion of magnetic phase dia~ramsi s also related to t potential application o f compouiids such as NdzFe14 as permanent magnets.
s Asti, G., and Bolzoni, F. (1980). .I. Itir~gn.M L ~ ~Mater., E. 29 Blanco, J. A., Cignoux, D., Schniitt, D., and Vettier, G. (1991). J. Magn. Magn. Mat Cadieu, F. J., Wegde, H., Rani aivarathna, A., and Chen, K. (1991). Appl. Phys. Cadogan, J. M. (1996). J . Phys. Cadogan, J. M., Cavlgan, J. P., (1988). J . Phys. F: ~ ~ tPhys a l Cadogan, J. M., and Li, H. S . (1992). J. ~ ~Magn, g ~ ~~ t e .r . ~ 110, L20. Campbell, S. J., Cadogan, J. M., Zhao, and Li, H. S. (1999). J. Phys.. Con& ~
Cheng, S. F.. Deniczyk, R. G., Laughlin, D. E., and Wallace, W. E. (1990). J, ~ a g n Mcl~n. . iwclter.,
87, L251-4.
Nikitm, S. A., Tishin, A. M., Kuz'iin, M. D., and Spiclikm, Yu. I. (1991). P h j ~Letr, . A, Palstra, T. M. M., ~ i e u w e n h u y ~ G., J., Mydosh, J. A., and Buschow, K. H. J. (1985). Phys, Riev. B, 3
University Press, Ithacca, New York. Ren, H., and Ryari, D. H. (1995). Phys. Rev.B, Rossat"Mignod, J. (1986). ~ e t h o o~f s~ x ~ e r i ~ e Physzcs, n~al Vol. 23, Academic Press, New York. Rossat-Mignod, J., Burlet, P., Villain, .J,, Bartholin, H., Wang, T. S., Florence, D., and Vogt, 0..(1977). Pbtys, Rev. B, 16, 440. Rotli, W. L. (1995). I ~ ~ ~ r Compounds: ~ e ~ ~ Vol. l l 1, ~ ~ Principles. Chapter 19 (eds, J. El. Westbrook, and R. L. Fleischer). Wiley, New York. Ryan, D. I-l. (1992). Recmt Progress in an do^ Magnets, Chapter 1 (ed. D. H. Ryaii). World Scientific, Singapore, pp. 1-40. Ryan, L). H., Tun, Z., and Cadogan, J. M. (1998). J . M4xgn. Magn. Mater., 177-1 Ryaii, D. H., Cadogan, J. M.$and van Lierap, J. (2000~.Pbtys. Rev. B, 66, 6816. Shirane, G. (1959). Actcl Cryst., 1 Hu, B. P.. Li, H. S., Suii, H., tawler. J. F., and Coey, J. M. D. Shull, C. C., and Smart, J. S. (19 lid State ~ o ~ ~76,~587. ~ n . , 995). In t e r ~ e ~ ~ l l i c Stadel~aier~ H. H., and Reinsc T. (1964). Solid Stafe ~ ~ y s i c16, s , 227. ~ o ~ p o Vol. u n 2, ~ ~Practice Iwata, N., Honda, K., Shigeoka, T., Hashimoto, Y., and Westbrook, atid R. L. Fleischer). Wiley, New York. Magn. Magn, Muter., 90-93, 63. Stevens, E(. W. H. (1952). Proc. Phys. Soc., ., Gallen, E., Das, B. N., Liou, S. H., ith, W., and Thompson, J. E. (1954) Proc. Xoy. Soc., ~ ~ ~ i iR. a n(1986). , J ~ ~ g Magn. n . 362. ouvel, J. S. (1995). I n t e r ~ e ~ a I l i C c o ~ p Vol. ~ ~ 1,~ n ~ ~ ~A. (1992). J. Alloys and Compo Szytula, A., and Leciejewicz, J. (1994). P ~ ~ n ~ i p Chapter l e * ~ , 40 (eds. J. €3. Westbrook, and R, L. Structures and Magne fic Pro~ertie.~of Rare Eurth Fleischer). Wiley, New York. I n ~ e r ~ n e t u l lCRC i ~ ~ , Press Inc., Boca Raton. and Cadogan, J. M. (1 992). Solid State Cornmun., Thornson, J. R., Cuo, H., and Grant, M. (1992). Venturini, G., Welter, R., (1995). J . Magn. Magn. Zhang, L, V., Pourarian, Magn. Magn. Mater., 7B,203. Zhou, R. J., Kapusta, Cz., Rosenberg, M., a K. H. J. (1992). J . Alloys and C o ~ ~ o 1~ ? i ~ , A ~ s t e r d a pp. ~ , 493-624. Jersey. p. 37. Enz, U. (1960). P~yszca, Franse, J. J. M., and Radwanskr, R. J. (1996). Rare-E(~rtl~ Iron ~ ~ r ~ ~ c l~ ?c ligen ~e tts Chapter , 2 (ed. J. M. D. Coey). Oxford, pp. 58-158. Schmitt, D, (1995). J. Alloys and 423. Topics in Applied Physics Volume 5: ectros~opy, ~~ringer"Verlag, Berlin, Heidel~erg. Goodenough, J. B. (1963). Mclgriefismand the Chemical Bond, Wiley, New York. Herbsl, J, F. (1991). Rev. Mod. Phys., Hu, B. P., Li, H. S., Gavigan, J. P., ancl . C o ? ~ ~Matter, e n ~ 1, ~ 755. ~ . D., and Gavigan, J. P. ~
Chapter 10 Calculation of
ase Diagrams
Department of Materials Science, University of Culifornia, Berke€ey, CA, USA
1. Introduction
A phase diagram is a graphical object indicating phase relationships in thermodynamic space. Usually, one coordinate axis represents temperature; the others may represent pressure, volume, concentrations of various components, and so on. Here, we shall be concerned only with temperature-concentration diagrams, for the sake of illustration limited mainly to binary (twocomponent) systems. Since more than one component will be considered, the relevant thermodynamic systems will be alloys, by definition, of either metallic, ceramic, or semiconductor materials, the emphasis being placed primarily on metallic alloys. Phase diagrams can be classified broadly into two main categories: experimentally and theoretically determined. The object of the present chapter is the theoretical determination, i.e. the calculution of phase diagrams, ultimately, their prediction. But a calculation of phase diagrams can mean different things: prototype, fitted, and first principles. Prototype diagrams are calculated under the assumption that energy parameters are known a priori, or given arbitrarily. Fitted diagrams are those whose energy parameters are fitted to known experimentally determined diagrams or to given empirical thermodynamic data. First principles diagrams are calculated on the basis of energy parameters calculated from essentially only the knowledge of the atomic numbers of the constituents, hence by actually solving the relevant Schrodinger equation. Such is the ‘Holy Grail’ of alloy theory, an objective not yet fully attained, although great strides have been made in that direction recently. Actually, theory enters as well in the experimental determination of phase diagrams, as these diagrams
indicate not merely the location in thermodynamicspace of existing phases, but must conform to rigorous rules of thermodynamic equilibrium (stable or metastable). The fundamentd rule of equality of chemical potentials indeed imposes severe constraints on the graphical representation of phase diagrams, while also permitting an extraordinary variety of forms and shapes of phase diagrams to exist, even for binary systems. That is one of the attractions of the study of phase diagrams, experimental or theoretical: their great topological diversity subject to strict thermodynamic constraints. In addition, phase diagrams provide essential information for the understanding and designing of materials, and so are of vital importance to materials scientists. For the theoretician, firstprinciples (or ab initio) calculations of phase diagrams provide enormous challenges, requiring the use of advanced techniques of quantum and statistical mechanics. The main portion of this chapter covers basic principles (Section 2), and is followed by one example of application to the AI-Li phase diagram (Section 3). A discussion (Section 4) summarizes the principal results and compares qualitativelythe various methods described in the main portion of the text. The reader will profit by also consulting chapters of Volume 1 of this series on IntermeialEic Compounds, Principles (Westbrook and Fleischer, 1995). In particular, additional information on mean-field methods applied to phase diagram construction (Section 2.2 of the present chapter) will be found in Chapter 5 of Volume 1 (Ellner and Predel, 1995), and more detailed treatments of electronic structure calculations (Section 2.5) will be found in Chapter 2 (Turchi, 1995) and Chapter 3 (Carlsson and Meschter, 1995) of Volume 1.
Intermetallic Compounds: Vol. 3, PrincipEes and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 02002 John Wiley & Sons, Ltd.
I86
Bondiag and Sta~ility
he ther~odynamicsof phase equilibrium were laid ibbs over 100 years ago (Gibbs, 1875-1 878). s was strictly a ‘black box’ thermodynamics in the sense that each phase was considered as a uniform cont~nuumwhose internal structure did not have to be specified. If the %lack boxes’ were very small, then interfaces would have to be considered, which Gibbs 1 without having to describe their
provided in free energy~oncentrationspace by the coinrnon tangent rule, in binary systems, or cominon tangent hyperplane in multicomponent systems. A very complete account of multicomponent phase equilibrium and its graphical ~nterpretationcan be found iii the book by Palatnik and Landau (1964), and summarized in more readable form by Prince (1966). The reason that equation (1) leads to a common tangent construction rather than the simple search for the minima of free energy surfaces is that the equilibria represented by (1) are constrained minima, with coiistraint given by
lmholtz) free energy F (=I E - TA‘, perature and S is entropy), and
II
ZXI=:
I= 1
where 0 6 XI (of P, V, ir, say), even for the case of fluids or hydroally stressed solids, as considered here, are ally not known. Analytical functions have to be inv~nted’and their parameters obtained from experiment. ]If suitable free energy functions can be obtained, then correspon~ingphase diagrams can be constructed from the law of equality of chemical potentials. But what if the functions themselves are unreliable and/or the values of the parameters iiot determined over a region of thermodynamic space? Then to studying in detail the physics of the therinodyn~niicsystem under consideration, investigating its internal structure, and relating microscopic qua~titiesto ~acroscopicones by the techniques of statistical mechanics. This section will very briefly review some of the methods used, at various stages of the ‘~~lcu1ation’ of phase allow the elaboration of of which may be found in ~ ~ z review y ~ articles i ~ ~(de ontaine, 1979 and 1994), in a more extended version of the present article (de Fontaine, 1999), and in refe~encescited . A historical approach is given in the author’s Turnbull lecture (de Fontaine, 1996).
he conditioii of phase equilibrium is given by the set
(1) ~esignatiiigthe equality of chemical potentials (p) for c~~mpon1 e ~(== t 1 to n) in phases 01, p, . . . and A. ~ ~ n v e n i e ngra~hical t descrip~ionof equation (1) is
1
(2)
<1
designates the concentration of y the properties of the xjs, it follows that a convenient representation of concentration space for an YI-component system is that of a regular simplex in (n - 1)~dimensionalspace: a straight line segment of length 1 for binary systems, an equilateral triangle for ternaries (the so-called Gibbs triangle), a regular tetrahedron for ~ ~ a t e r n a r i e sand , so on (Prince, 1966; Paiatnik and Landau, 1964). The temperature axis is then constructed oithogoiial to the simplex. The phase ~ i a g r a m space thus has dimensions equal to the (non-indepeiident) number of components. Although heroic made in that direction (Cayron, 196 is clear that the full graphical representation of temperature-composition phase diagrams is practically impossible for anything beyond ternary systems, and awkward even beyond binaries. One then resorts to constructing two-dimensional sections, isotherins and isopliths (constant ratio of components). The variance f (or number of thermodynamic degrees of freedom) of a phase region where 4 ( 3 3) phases are in equilibrium is given by the fzimous Gibbs phase rule, derived from equations (1) and (Z), f = YI - @
+1
(at constant pressure)
(3)
It follows from this rule that when the number o f phases is equal to the number of components plus one, the equilibrium is invariant. Thus, at given pressure, there is only one, or a finite number of discrete temperatures at which y1 + 1 phases may coexist. Let us say that the set of y 1 + 1 phases in equilibrium is C == { E , p, . . ., A}. Then just above the invariant t e ~ p e r a t ~ rae ,particular subset of Z: will be found in the phase diagram, and just below, a different subset. lt is simpler to illustrate these concepts by a11 example. In isobaric binary systems, three coexisting
Calculatjon of Phase Diagrams phases constitute an invariant phase equilibrium, and only two topologically distinct cases are permitted: eutectic-type and peritectic-type, as illustrated in Figure 1. These diagrams represent schematically regions in teniper~ture~onceiitrationspace in a narrow temperature interval j list above and just below the three-phase coexistence line. Two-phase monovariant regions (.f= 1 in equation (3)) are represented by sets of horizontal (isothermal) lines, the tie-Zine‘~(or co-nodes), the extremities of which indicate the equilibrium concentrations of the coexisting phases at the temperature considered. In the eutectic case (Figure 1a), the high-temperature phase A (often the liquid phase) ceases to exist just below the invariant temperature, leaving only the (a+ /I) twophase equilibrium. In the peritectic case (Figure Ib), a new phase, p, makes its appearance along with one or the other type of two-phase material (a p) or (p A). It is seen from Figure 1 that, topologically speaking, these two cases are mirror images of each other, and there can be no other possibility. How the full phase diagram can be continued will be shown in Section 2.2 in the eutectic case calculated by means of ‘regular solution’ free energy curves. Ternary systems exhibit three topologically distinct possibilities: eutectic-type, peritectic-type and ‘intermediate’. In the ternary eutectic case, three 3-phase triangles come together to form a larger triangle at the invariant temperature, just below which the A phase disappears, leaving only the CI + p + y onov variant equilibrium. The peritectic case is, as in the binary case, its topological mirror image: the a + B + y triangle splits into three 3-phase triangles. In the ‘intermediate’case, the invariant is no longer a triangle but a quadrilateral; just above the invariant temperature, two triangles meet along one of the diagonals of the quadri~ateral,just below, it splits into two triangles along the other diagonal. For quaternary systems, the graphical representation is more difticult to describe, since the possible
+
+
187
invariant equilibria, of which there are four, eutectictype, peritectic-type, and two intermediate cases, involve the merging and splitting of tetrahedra along straight lines or planes of 4- or 5-vertex (invariant) figures (Prince, 1966; Cayron, 1960). For yet highercomponent systems, the full graphical r~presentationis just about hopeless (Cayron, 1960). The commercial firm Thermocalc (Jansson et al., 1993) does provide software which constructs two-dimensional sections through multidimensional phase diagrams, these diagrams themselves being constructed analytically by means of equation (1) applied to semi-empirical, multicomponent, free-energy functions. The reason for dwelling an the universal rules of invariant equilibria is that they play a fbdamental role in phase diagram determination. Regardless of the method employed, experimental or computational, prototype, fitted or ab initio, these rules must be respected, and often give a clue as to how certain features of the phase diagram must appear, rules which are not always respected even in published phase diagrams. They are, of course, respected in, for example, the three-volume compilation of evaluated experimentally determined binary phase diagrams, published by the American Society for Metals (Binary Alloy Phase D i a g r a ~ . ~1990). ,
The diagrams of Figure 1 are not even pro to types', they are schematic: the only requirements imposed is that they represent the basic types of phase equilibria in binary systems, and that the freeh~nd~drawn phase regions obey the phase rule. To be a little more quantitative, let us now look at the construction of a prototype binary eutectic diagram based on the simplest analytical free energy model, that of the regular solution. First, some useful general definitions. Consider some extensive thermodynamic quantity, such as energy, entropy, enthalpy, or free ener
F~~M 1 ~Binary e eutectic (a) and peritectic (b) topology near the three-phase invariant e q u i l i b r i ~ ~
188
Bonding and Stubility
For condensed matter, it is custoniary to equate the Gibbs and Helrnholtz free energies. That is not to say that equilibrium shall be calculated at constant volume, merely that the Gibbs free energy is evaluated at zero pressure; at reasonably small pressures, at or below the atmospheric, the internal state of condensed phases cbange very little with pressure. The total free energy of a binary solution AB for a given phase is then the sum of an ‘unmixed’ contribution, which is the concentration-weighted average of the free energy of the pure constituents A, I1, and a mixing term which takes into account the mutual interaction of the constituents. Therefore we have: (4)
the excess enthalpy depends quadra ticnlly on conceiitration. Then we have:
F*. = x(1 - X)W
(8)
where W is some generalized (concentration- and temperature-independent) interaction parameter. For the purpose of constructing a prototype binary phase diagram in the simplest way, based on the regular solution model for the solid (a) and liquid (A) phases, we write the pair of free energy expres$ions: J;f = x( 1 - X)W,
+ k B T [ xlog x + (1 - x) log( 1 - x)] (9a)
and
with F~lll X A F A
+ XBFB
(5)
The latter contribution, which we call the ‘linear term’ because suck is its dependence on concentration, contains thermodynaniic infor~ationpertaining only to the pure elements, The mixing term is the difficult one to calculate as it contains the important configurational energy and entropy of mixing contributions. The simplest mean field approximation regards Fn,,,as a slight deviation from the free energy of a completely random solution, i.e. one that would obtain for noninteracting particles (atoms, molecules). In the random case, the free energy is the ideal one, consisting oiily of the ideal free energy of mixing: F,d
(6)
-TSl,j
since the energy (or enthalpy) of mixing of noninteracting particles must vanish. The ideal configurational entropy of mixing per particle is given by the universal function
si, = - N k B ( X A
log X A f
xB log -“CB)
(7)
where N is the number of particles and k~ is Boltzinann’s constant. Expression (7) is easily geiieralized to multicomponent systems. What is left over in the general free energy is, by defi~ition,the excess term, in the mean-~eldempirical approximation taken to he a polynomial in concentration and temperature &(x, T ) . Here, since there is only one independent concentration because of equation (21, one writes x = xg, the concentration of the second constituent, and XA = 1 - x in the argument of F. In the regular solution model, the simplest formulation of the mean-field approximation, the simplifying assumption is also made that there is no t e ~ p e ~ a t udependence re of the excess entropy and that
These equations were arrived at by subtracting the linear term (equation (5)) of the solid from the free energy functions of both liquid and solid (hence the As in equation (gb)), and by dividing through by N, the number of particles. The lower-case letters symbols in equations (9a) and (9b) indicate that all extensive thermodynamic quantities have been normalized to a single particle. The (linear) A h term was considered to be temperature-inde~endent~and the As term was considered to be both temperature- and concentrationindependent, assuming that the entropies of melting of the two constituents were the same. Free energy curves for the solid only (equation (9b)) are shown at the top portion o f Figure 2 for a set of normalized temperatures (C= kB T/ ( w ( ) ,with wo 0. The locus of common tangency points gives the miscibility gap (MG) type of phase diagram shown at the bottom portion of Fi ure 2. Above the critical point ( t = O S ) , the two phases a and p are structurally indistinguishable. As and the constants a and b in A h = a 4bx were fixed so that the free energy curves of the soIid and liquid would intersect in a reasonable temperature interval. Within this interval, and above the eutectic (a -t- + A) temperature, common tangency involves both free energy curves, as shown in Figure 3: the full straight line segments are the equilibrium common tangents (filled circle contact points) between solid (flill curve) and liquid (dashed curve) free energies. For the liquid, we took I W RI< wo. The dashed common tangent (open circles) represents a metastable MG e ~ u i l i b r i uabove ~ the eutectic. Open square symbols in Figure 3 mark the intersections of the two free energy curves. The loci of these intersections on a phase diagram trace out the so-called To
Calculation of Phase Diagram
1813
0.0
-0.1
0% ion
0.b
i t
ility
kgT/lwl ~0.35
F ~ 3 ~Free ~energy ~ curves e for the solid (full line) and liquid (dashed line) according to equations (9a) and (9b)
0.6 0.
I--
0.1
indicated normalized t e ~ p e r a t u ~ and ~ s ) resulting miscibility gap
lines where the etast stable) d~sordered-state free energies are equal. Sets of both liquid and solid free energy curves are shown in Figure 4 for the same reduced temperatures ( t ) as were given in Figure 2. The phase diagram resulting from the locus of lowest common tangents (not shown) to curves such as those given in Figure 4, is shown in Figure 5. The melting points of pure A and B are respectively at reduced temperatures 0.47 and 0.43; the eutectic is at about 0.32.
This simple example suffices to demonstrate how phase diagrams can be generated by applying the equilibrium conditions given by equation (1) to model free energy functions. Of course, actual phase diagrams are mucl? more complicated, but the geiieral principles are always the same. Note that in these types of calculations the phase rule does not have to be imposed ‘externally’; it is built into the equilibrium conditions, indeed follows from. them. Thus, for example, the phase diagram o f Figure 5 clearly belongs to the eutecticclass illustrated schematicallyin Figure 1a. Peritectic systems are often encountered in cases where the difference of A and €3 melting temperatures is large with respect to the average normalized melting temperature t. The general procedure outlined above may, of course, be extended to mult~component systems. What if the interaction parameter W were negative? Physically, this means that unlike (A‘bonds’ are favored over the average of like (A-A, B-€3) ‘bonds’. The regular solution model cannot do justice to this case, so it i s necessary to introduce , ~ ~ ~ Z a ~ one t i c or ~ s more , of which will be preferentially occupied by a certain constituent. In a sense, this atomic ordering situation is not so unlike that of ~ ~ s ~ ~ f f ~ uillustrated t i ~ n , in Figure 3, where the two types of atoms separate out in two distinct regions of space: in the ordering case, the separation is between two i~iterpenetratingsublattices. Let there be two distinct
u
~
190
Bonding and Stability
Free energy curves, as in Figure 3, but at the iiidicated reduced temperatures
t,
generating the phase diagram of Figure 5
Calculation of Phase Diagrams
191
necessary to write Fx,as a pol~nomialof degree higher than the second in the relevant concentration variables, the temperature T also appearing, usually in low powers. The extension of the GBW method to include higher-degree polynomials in the excess free energy is the technique favored by the (see journal) group, a group devoted to the very useful task of collecting and assessing experimental t h e ~ o d y n a m i c data, and producing 0.25 calculated phase diagrams, agreeing as closely as possible with experimental evidence. In a sense, what 0.2 this group is doing is ‘ther~odynamicmodeling’, i.e. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 storing of thernio~yflami~data in the form of A ~~nc~n~ra~i~n B mathematical functions with known parameters. One Eutectic phase diagram generated from curves such may ask, if the phase diagram is already known, why as those of Figure 4 calculate it? The answer is that phase d i a g r a ~ shave often not been obtained in their entirety, even binary ones, and that calculated free energy functions allow sublattices, a and p (the same notation is purposely some e x t ~ ~ p o l ~ into ~ othe n u i i k i i o ~Also, ~~, ~ i i ~ w ~ i used as in the phase separation case). Tlie pertiiieiit free energies explicitly, allows one to extract many concentration variables are then (only the €3 concenother useful thermodynamic functions instead of only tration needs be specified) x“ and xp, wliich may be the phase diagram itself as a graphical object. Another each written as a linear combination of x, the overall useful, though uncertaiii aspect is the extrapolation of concentration of B, and q , an appropriate long-range known lower-dimensional diagrams into unknown order (LRO) parameter. In a binary ‘ordering’ system , o ~ ~ i with two s~iblattices~ there are now two i i i ~ e p ~ n d e n ~higher-dimensional ones. ~ i ~ ~ ~p l ~l ~~software can plot out two-dimensional sections through multicomposition variables: x and q. In general, there are one less order parameters than there are sublattices component systems. Such software is available introduced. When the ideal entropy, also containing commercially from Thermocalc (Jansson et al., 19931, linear combinations of x and q, multiplied by -7’is based in Stockholm. added on, we obtain so-called (generalized) GorskyIt is important to note, however, that the general~ p ~ g ~ (GSW) - ~ free i energies. ~ ~ ~ TQ ~ construct ~ ~ sa izatioiis just described - sublattices, Fouier ~ ~ a ~ s f ~ protQtype phase diagram, possibly featuring a variety polynomials - do not alter the degree of sophistication of ordered phases, it is now necessary to minimize of the approximation, which remains decidedly mean these free energy functions with respect to the LRQ field, that is, wliich replaces the averages of products parameter(s) at given x and T‘, then to construct (of concentrations) by products of averages. Such a common tangents. procedure can give rise to certain unacceptable Generalization to ordering is very important as it behavior of calculated diagrams, as explained in allows treatment of (ordered) compounds with extenSection 2.4. sive or narrow ranges of solubility. The CBW method is very convenient as it requires little in the way of computational machinery. When the GBW free energy is expanded to second order, then Fourier transThus far, the treatment has been restricted to a ‘black formed, it leads to the ‘method of Concentration box’ approach: each phase is coilsidered ns a themnowaves’, discussed in detail by Khachaturyan (1983), dynamic continuum, possibly in equilibrium with other with which it is convenient to study general properties phases. Even in the case of ordering, the approach is of ordering in solid solutions, particularly ordering still ~ f f c ~ o s c o peach i c , sublattice being also considered instabilities. However, to constructfitted phase diagrams as a thermodynamic continuum, possibly interacting which resemble those determined exper~ientally,it is with other sublattices. It is not often appreciated that, found that an excess free energy represented by a despite taking the crystal structure into ~ ~ n ~ i ~ e ~ ~ t quadratic form will not provide enough flexibility for in setting up the GBW model, the resulting free energy an adequate fit to theri~odynamicdata. It is then function still does not contain any geometrical 0.5
~~~~~~~
~~~~~~~
192
Bonding aPzd S ~ u ~ i l ~ ~ y
information: each sublattice could be located anyprocedure andlor using a ~ a m i l t o ~ ~based a n on where, and such things as coordination nuinbers can semi-enipirical potentials. To this we now turn: be readily incorporated in the effective iiiteractions W. deriving a variational method for the free energy, In fact, in the ‘mean-field ideal-entropy’ fomulawhich is at the heart of the cluster variation method tion, solids and liquids are treated in the same way: (CVM) (Kikuchi, 1951). termiiial solid solutions, compounds, liquids are The exact free energy F of a t~ermodynam~c system represented by similar ‘black-boxyfree energy curves, is given by and the lowest set of common tangents is constructed F = -kBTln at each temperature to generate the phase diagram. With such a simple technique available, which with partition function generally produces formally satisfactory results, why = e-E(state)/ksT (1 1) or several reasons: (I) the mean-field states overestimates the coiifigurational The sum over states in equation (1 1) may be partially entropy and enthalpy cont~b~itions, (2) the method decoupled (Ceder, 1993): often produces qualitatively incorrect results, particularly where ordering phenomena are concerned (de z + e-Ftat ( n ) l k ~ T (12) e (SRO) cannot Foutaine, 19’79), (3) s ~ o r t - r a n ~order {b) dyn be taken into account, (4) the CALPHAD method where cannot make the~odyiiamicpredictions based on ‘atornistics’. a much more elaborate ~ ~ c ~ o ~ c o ~ ~ ~ theory is re for the purpose of calculating ah irzitio phase diagrams. represents the sum of a replacive aiid a disp~u~ive To set up such a formalism, on the atomic scale, we energy. The former is the energy which results from the must define a reference lattice or structure, fcc, bcc, or rearrange~entsof types of atoms, still centered at their hcp, for example. The much more difficult case of associated lattice sites; this energy contributioii Iias liquids has not yet been worked out. Also, for also been called ‘ordering’, ‘ideal’, ‘~onfig~rational~, simplicity, we shall consider only binary alloys ‘chemical’. The displacive ener y is associated with static atomic displacernents. These displacements (A-B). At each lattice site (p), define a spin-like nfigurational operator bp equal to + l if atom of type themselves may produce volume effects (change in the volume of the unit cell), cel~~externaleffects is associated with it, -1 if B. At each site we also (change in shape of the unit cell at constant volume), attach a ( t ~ r e e - ~ ~ ~ e n s i ovector n a l ) up which describes and cell internal effects (relaxation of atoms inside the lacement (static or dynamic) of the atom from unit cell away from their ideal lattice positions) lattice site. For given configuration o (a boldfaced (r indicates a vector of No, = f l components, (Zunger, 1994). The sums over dynainical degrees of freedom may be replaced by Boltzmann factors o f the ven configuration in a supercell of N lattice sites), an appropriate ~amiltonian is then type Fdyn.(o) = ~ d ~ ~ ~ . (TSdyn.(~) b) to obtain finally a new partition function constructed, and the energy E is calculated at lir = P == 0 by quantum mechanical means. In grinci=;: e-*t4/ksT (14) ple, this program has to be carried out for a variety of In1 supercell configurations, lattice parameters and interwith ‘hybrid’ energy nal atomic displ~cements; then configurational averages are taken at given T and P to obtain exp~ctation values of appropriate t he~ odyna m i c functions, in particrtlar the free energy. Such calculaEven when considering the simplest case of leaving out tions must be repeated for all competing crystalline static displacive and dynai~icaleffects, and writing the phases, ordered or disordered, as a function of average replacive energy as a sum of nearest neighbor (nn) pair concentration x, and lowest tangents constructed at interactions, this lying ~ ~is i~possible ~ e tol ‘solve’ in various temperatures and (usually) at P = 0. As three dimensions, i.e. the suinmation in. the partition described, this task is an impossible one, so that rather function cannot be expressed in closed form. One must drastic a~~roximations must be introduced, such as then resort to Monte Carlo simulation or to variausing a variational principle to calculate the free tional solutions. To these several energies will correspond entropy contributious: particularly a energy instead of the correct partition function
+
z
Calc~lajionof Phase ~ ~ u g r a ~ s configurational entropy associated with the replacive energy, and a vibrational entropy associated with dynamical displacements. Let us consider the configuratio~alentropy. is to write the partition The idea of the C function, not as a sum o all possible configurations, but as a sum over selected cluster probabilities (x) for small groups of atoms occupying the sites of 1-, 2-, 3-, 4-, . . ., many-point clusters, such as pairs, triplets, q u a ~ r ~ p l .e .~ .~ of , lattice sites, for example. The transfoi~~~ion
z=
e-E(g)/kB’f’ + (4
cg( )e-E(x)/fcB = ceT
{XI
F(x)/!’
BT
{XI
(16 ) or number of ways the specified values There corresponds a configurational entropy
(17) hence a free energy F ( 5: E - TS),which is the one that appears in equation (16). The optimal variational free energy is then obtained by minimizing F ( x ) with respect to the selected set of cluster probabilities, and subject to c o ~ s i s t e n cconstraints. ~ The resulting F ( x * ) , with x* being the values of the cluster probabilities at the minimum of A?@), will be equal to or larger than the true free energy, hopefully close enough to the true free energy when an adequate set of clusters has been chosen. Kikuchi (1951) gave the first derivation of the g ( x ) factor for various cluster oices, followed by the more algebraic derivations of arker (1953) and Hijmans oer (1955). The simplest cluster approximation is of course that of the point, i.e. that of using only the lattice point as cluster. The symbolic formula for the complexion in this approximation is such that, inserted in the configurational entropy formula (equation (17)),it produces precisely the ideal entropy given, for binary systems, by the standard expression Thus the mean field entropy is identical to the C point a~prox~mation. To improve on the approximation, we need to go to higher clusters, pairs, triplets and so on. Of course, the result in^ entropy expressions will be quite a bit more complicated, but today, CVM entropy formulas can be derived ‘automatically’, for arbitrary lattices and cluster schemes, by computer codes based on group theory consi~erations.The choice of the proper cluster scheme to use is not a simple matter, although there now exist heuristic methods for selecting ‘good’ cluster
193
combinations (Finel, 1989; Vul and de Fontaine, 1993). The realization that config~rational entropy in disordered systems was really a many-body tbermodynamic expression led to the introduction of cluster es in free energy fLinctionals. variables led to the development of a very useful c l ~ s t e ~ algebra, to be described very briefly here (see the origiiial article (Sanchez et al., 1984), or various reviews (de Forrtaine, 1994, for e x a ~ p l e )for more details). The basic idea was to develop complete orthonormal sets of cluster ftinctions for expanding arbitrary functions of confi~urations,a sort of Fourier expansion for disordered systems. For s i ~ ~ ~ i cwe ~ty, shall look here only at binary systems, without explicitly considering sublattices. We shall see that the cluster algebra, applied to the replacive (or configurational) energy leads qujte ~ a t ~ r a l lto y a precise definition of effective cluster i ~ t e ~ ~ c t i o n s , similar to the pherromenol~gical w parameters of mean field equations (8) and (9). With such a rigorous definition available, it will be in principle possible to calculate those parameters from ‘first principles’, i.e. from the knowledge of only the atomic numbers of the constituents. Any function of configuration f ( o) can be expanded in a set of c l ~ s t ~ ~ . f ~ n cast ifollows o~s ( 1984): (18) CI
where the s u m ~ a t i o nis over all clusters of lattice points (‘a’ designating a cluster of fzcc points, of a certain geometrical type: tetrahedron, square, . . .), Fa are the expansion coefficients, and the “P, are suitably defined cluster functions. The cluster functions may be chosen in an infinity of ways but must form a complete set (Asta et al., 1991; Wolverton et al., 1991a; Sanchez, 1993). In the binary case, the cluster f u ~ c t ~ o may n s be chosen as products of o variables (= k l ) over the cluster sites. The sets may even be chosen ~ ~ t l ~ o n o r n i a l , in which case expansion (21) may be ‘inverted’ to yield the cluster coefficients
F; = PO
z=
J’(o)@CI(o)
(@as
f
(1%
(T
where the summation is now over all possible ZN configurations, and po is a suitable nor~alization factor. More general formulations have been given recently (Sanchez, 1993), but these simple formulas, (18) and (191, will suffice to illustrate the method. Note that the formal is^ i s exact and m o ~ e o v ~ r
194
Bonding and Stability
computationally useful, if series convergence is sufficiently rapid. In turn, this means that the burden and difficulty o f treating disordered systems can now be carried by the determination of the cluster coefficients Fa. An analogy may be useful here: in solving linear partial differential equations, for example, one generally does not ’solve’ directly for the unknown function, one instead describes it by its representation in a complete set of functions, and the burden of the work is then carried by the ~ e t e ~ i n a t i oof n the coefficients of the expansion. The most iniportant application of the clusterexpansion formalism is surely that of the calculation of the configurational energy of an alloy, though such theirnodynamic qtiantities as molar volume, vibrational entropy, elastic moduli and so on, as a function of configuration, can be obtained as well by this technique. In particular, application of equation (18) to the configurational energy E provides an exact expression for the expansion coefficients, called effective cluster interactions, or ECIs, V,. In the case of pair interactions (for lattice sites p and q, say), the effective pair i~teractionis given by
where ~!?IJ (I, 5 = A or E) IS the average energy of all configurations having atom of t.yp I at site p a i d J at y. ence it is seen that the If, parameters, those required by the Ising model thermodynamics, are by no means “pair potentials’, as they were often referred to in the past, but differences of energies, which can in fact be calculated. Generalizations of equation (20) can be obtained for multiplet interactions such as Vpqr as a function of AAA, AAB, etc., average energies, and so on. How does one then go about calculating the pair or cluster interactions in practice? One method makes use of the orthogonality property explicitly by calculating the average energies appearing in equation (20). To be sure, not all possible configurations (T are summed over, as required by equation (19), but a sampling is taken of say a few tens of configurations, selected at random around a given 13 pair, for instance, in a supercell of a few hundreds of atoms. Actually, there is a way to calculate directly the difference of average energies in equation (20) without having to take small differences of large numbers. Such a method i s called n~~ (DCA) the method of direct ~ o n ~ g z i r a t i oaveraging (Dreyssk et al., 1989; ~ o l v e r t o net al., 1991b, 1993; Wolverton and Zunger, 1994) It is unfortunately very computer intensive, and has been used thus far oiily in
the tight binding approximation of the ~amiltonian appearing in the Schrodinger equation. A convenient method of obtaining the ECls is that of the structure inversion method ( as the Connolly-Williams method from the names of those who originally proposed the idea (Connolly and Williams, 1983). It then turns out that the energy of a large-unit-cell structure can be obtained as a linear Combination of the energies of small-unit-cell structures, which are ~resum ab~y easy to compute, for instance by electronic structure calcu~ation.So what about electronic structure calculations? Section 2.5 gives a quick overview of standard techniques in the ‘alloy theory’ context, but before getting to that, let us already see the difference that the CVM makes, compared to the mean-field approximation, in the ~alculationof a prototype ‘ordering’ pkase diagram. 2.4
~
l
~
S
~
~
~
-
~
~
Thermodynamic quantities are macroscopic ones, actually expectation values of microscopic ones. Take the case of the energy; we need its expectation value ( E ) in cluster expanded form. For that we take the ensemble average of the general cluster expansion foimula (19) to obtain, per lattice point:
with correZatio~v a ~ ~ i adefined ~Z~~ by
the latter equality being valid for binary systems. The brackets indicate ensemble averages. The entropy, already an averaged quantity, is expressed in the CVM by means of cluster probabilities xor.These probabilities are related linearly to the correlation variables by means of the so-called configuration matrix (Sanchez and de Fontaine, 1978), the elements of which are most conveniently calculated by grouptheory computer codes. The required CVM free energy is obtained by combining energy (21) and entropy (17) contributions, and using Stirling’s approximation for the logarithm of the factorials appearing in the g(x) CVM expressions:
where the indices r and lr denote sequential ordering of the clusters used in the expansion, K being the order of
~
~
Calculation of’ Phase Diagrams
the largest cluster retained, and where the second summation is over all possible A/B configurations CQ of cluster k. The corre~at~ons c, are linearly independent variational parameters for the problem at hand, which means that the best choice for the correct free energy for tlie cluster appro~imation adopted is obtained by minimizing ( 3) with respect to the 5,. entropy expression are the socoefficients, which may be by a~propriategroup theory codes, as mentioned above. inimization of the free energy (23) leads to systems of simultaneous algebraic equations (sometimes as many as a few hundred) in the 5,. This task greatly limits the size of clusters that can he handled in practice. Note that, strictly speaking, the CVM is also a mean field formalisin but it does treat the ~ o r r e l a t i o ~(almost) s correctly for lattice distances included within the largest cluster considered in the ximation, and includes RO effects. Hence the is a ~ ~ l t i - s i mean te field model, whereas the MF is a siqgk-site model, the one we have referred to as the M F (ideal, regular solution, GBW, concentratioii wave, . . .) model. Of course the ‘point’ a p p r o x i ~ a t i o nor~ mean-field model, is very much simpler to handle than the ‘duster’ formalism. So how much is lost in terms of actual phase-diagram results in going from MF to CVN? That question was answered quite some time ago by exhibiting a very striking, yet siniple example, the case of ordering on the fcc lattice with only ~ r s t - n e i ~ h b o r eflective pair iiiteractions (de Fontaine, 1979). If one compares the calculated MF phase diagram constructed by the GBW mean field approxii~ation (Shockley, 1938), to that constructed by the CVM in tlie tetrahedron cluster approximation, it is seen that, even in a topological two diagrai~sare completely diflt‘erent: t diagram sliows a double second-order tra he central composition (XB = O S ) , whereas the CVM diagram shows three first-order transitions for the three ordered (L12-type ordered structure) and M phase diagram also shows two very small t~ree-phase regions which are actually peritectic-like, topologically, as in Figure 1b. There can be no doubt about which prototype diagram is correct, at least qualitatively: it is the CVM version, which in turn is very similar to the solid-state portion of the experimental Cu-Au phase diagram (Binary AZhy Phase ~ i a g ~ a ~1986; 7 z ~ in ~ ,this particular case, the first edition is to be preferred). The large discrepancy between the M F and CVM versions is mainly due to the fact that the basic ~eoinetricalfigure of the fcc
195
lattice is the nn equilateral triangle (and the associate ~ s t nn tetrahedron). Such a figure leads tca ~ where, as in this case, the first nn iiiteraction favors unlike atomic pairs. This reqL~ireme~t can be s a t ~ s ~ e d for two of the nil pairs making up the triangle, but the third pair must necessarily be a ‘like’ pair, hence the conflict. This frustration also lowers the transition temperature below what it would be in nonsystems. The cause of the problem with approach i s of course that it does not ‘kn the triangle figure and the frustr~tednature of its interdctions. Numerically also, the CVM contional entropy i s much more accurat~than the (MF), particularly if ~iigher-c~us~er approximations are used, accuracy being judged by ~ o m p a r i s o ~ with Monte Carlo simulations (Finel, 1999). The last sentence of this parag~aph begs the question: if the Monte Carlo ~ a l c u l a t i o ~are s the standard by which to judge CV results, why bother with clusters at all? Some resear~hers(see for example Wolverton and Zunger, 1995) ha to bother with the CVM at all. reasons not to rely entirely on comput~r~ i ~ u l a t i(1) o~: analytical methods are Ear less c o m ~ u t e r - ~ n t ~ and ~s~ve therefore much faster, (2) simulation methods are accurate but imprecise, analytical m e ~ ~ ~ are ods but inaccurate; that is to say that give results which are in principle correct but will be given to within a certain degree of ~ r e c i s i o whereas ~, the CVM will produce well-de~nedn u ~ b e r s(3) ~ if several competing phases are present, it is not often a simple matter to determine by simulati~nwhat phases are being examined, (4) c o ~ p ~ t a t i oofi i e ~ t r o ~ y hence free energy is not conveniently performed in simulation; simulation does not readily provide traceable free-energy curves, ( 5 ) ~ e t a s t a ~e~uilibria ~e can be calculated by cluster methods, not by the method. A technique which inay advantages of both cluster and suggested by Schlijper and mid (1989) but not used much since because of large storage r ~ ~ ~ r e n ion ei~t computers, a problem which is hardly restrictive ~ o ~ a ~ aThe y s .eth hod in question may be ~ a ~ a l e ~ hybrid one, as it combines aspects of both the C and MC simulation: the analytical free sion (23) is used, as in the classical cluster probabilities cCn, instead of bei minimization o f the free energy functional, are obtained by simply c o u n t i ~tlie ~ frequency of occurrence of the clusters of interest and inserting these values in foriiiulas such as (23). Avoidance o f the mini~izationstep, with its c o ~ v e r ~ e n cproblems, e is a
196
Bonding and Stability
big advantage of the hybrid method but, precisely because of its hybrid character, it is not as internally coiisistent as the ‘pure’ CVM. This undesirable feature was recently illustrated (Tepesch et al., 1998) for the case of ordering on the fcc lattice. In order to obtain correct phase equilibrium results, the authors found needed to be used, in int cluster combination. ver group-theory techiiiyues were wed to handle cluster algebra. The results agreed perfectly with , but entropies and free energies were given explicitly by an analytical expression, which is the great advantage of this combined MC/CVM procedure, in addition to the elimination of the minimization stage of the calculation. There are other benefits for using the cluster approach: the cluster expansion provides a rigorous formulation for the effective cluster interaction (ECI); see equation (20) and extensions thereof for multisite d multicomponent systems. That means for the replacive (or strictly configurational) energy can now be considered not only as fitting par~meters,but as quantities which can be rigorously calculated ab initio. Such atomistic computations are very difficult to perform, as they involve electronic structure calculations~ but much progress has been realized in recent years, as briefly summarized in the next section.
method. The methods also difYer by the approximations used to represent the potential that an electron sees in the crystal or niolecule. Pseudopotential methods define separate ‘potentials’ for s, p, d, . ., valence states, whereas other techniques are ‘allelectron’ methods. Finally let us note that planewave-basis methods (pseu~opotent~als,augment~d plane waves) are the most convenient ones to use whenever forces on atoms need to be calculated. This is an important consideration as correct cohesive energies require that total energies be minimized with respect to lattice ‘constants’ and possible local atomic displacements as well, best handled by Fourier techniques. There is not universal agreement concerning the definition of ‘first-principles’ calculations. To clarify the situation, let us define levels of ab itznitio character:
(a)lst Icw‘I: only the 2 values (atomic number) are needed to perform the electronic structure cnlculation, and electronic self-consistency is performed as part of the calculation via the density functional approach in the local density approximation. This is the most fundamental level. One can also use semi-empirical pseudopotentials the parameters of which may be fitted partially to experimental data. (b)2nd level: electronic structure calculations which do not feature electronic self-consistency. Then the Namiltonian must be expressed in terms of parameters which must be introduced ‘from the outside’. An example is the tight binding method ne of most important breakthroughs in condensed where indeed the Schrodinger equation is solved atter theory occurred in the mid-1960s with the d e v e l o ~ ~ e noft the (exact) density ~ u n c theory ~ ~ ~ ~ a(the ~ Hamiltonian is diagonalized, the eigenvalues suinnled up), and the TB parameters are calcuF) and its local density approximati~n(LDA) which lated separately from a level-one formulation. provided feasible means for performing large-scale (c) 3rd level: empirical potentials are obtained dielectronic structure calculations. It then took almost 20 rectly by fitting properties (cohesive, formation years for practical and reliable computer codes to be -energies, elastic constants, lattice parameters) to developed and implemented on. fast computers. Today experimentally measured ones or to those calcumany fully developed codes are readily accessible and lated by level-one methods. At level 3, the run quite well on work stations at least for moderate‘potentials’ can be derived in principle from size a~plications.The reader may consult Chapter 6 of electronic structure, but are in fact (partially) the first volume of ~ n t @ ~ ~ ? ~ Compounds e t ~ l l i c (Singh, obtained by fitting. No attempt is made to solve 1995) for information on density functional theory and the Schrodinger equation itself, Example of such references therein. approaches are the embedded atom method hat the various methods have in common is (EAM) and the tight binding method in the chrodinger equation (sometimes the second moment approximation. The ern bedded irac equation) in an electronically self-consistent atom technique is a very useful one at this level, manner. The methods differ from one another by the and is described in detail by Daw et al. (1993) and choice of basis functions used in solving the relevant by Voter rn Chapter 4 of Volume 1 of Interlinear differential equation: plane waves, muffin tin metallic Compounds (1995), for example. orbitals, or atoniic orbitals in the tight binding (TB)
(d)$th level: here the interatomic potentials are strictly enipirical and the total energy, for example, is obtained by sui~mingup pair energies. olec~i~ar dynamics ( ~ simu~ations ~ ) operate with 4th or 3rd level approaches, with the exception of the famous Car-Parrinello method forms ab i ~ i t niolecular ~Q dynamics, with level 1 approach. In the phase diagram context, the aim of the electronic structure calcul~tionsis to obtain reliable values for the effective cluster interactions, as exemplified by equation (20) for pairs, which are themselves central to the c a l c ~ l a t i oof~ c ~ i i ~ g u ~ a t i ofree n a energies, ~ as per equation (23), for example. ince the ECls are well to obtain precise defined o p ~ r a t i o n ~it~is~ possible y~ energy values for the V, by some form of ‘structure inversion’ or other method, combined with electronic structure calculations for which standard codes now exist. These various techniques are described in some detail elsewhere (de Fontaine, 1994), where original literature is also cited. Tlie ~ r o ~ for e ~doing ~ ~‘first-priiici~les e tlieriiiodynamics’ may thus be summarized as follows: Start with an CDA electronic structure method to obtain cohesive or formation energies of a certain number of ordered structures on a given lattice, then set up the appropriate system of linear equations and solve it to obtain the relevant EGIs V,. Or perform a DGA ver electronic structure inethod ake sure that convergencc is attained in the ECll co~~Litation§, then insert these interaction parameters in an appropriate CVM free energy, minimize with respect to the correlation variables, and thereby obtain the best estimate of the free energy of the phase in question, ordered or o ngeneral disorde~ed.A more complete ~ e s c r ~ ~oft ~the procedure is given at the end of the next section.
In the previous section the computation of ECls was disc~ssedas if the only contri~Litionto the energy were the strictly repiacive one, as defined in equations (13) and (15), yet it is clear from the latter equation that ECIs could in principle be calculated for the value of the combined function XP itself. But for now, let us still consider the ECfs limited to the replacive (or configurational, or Ising) energy, leaving for the next section a brief description of other contributions to the energy and free energy. 111 the present context, it is possible to break up the energy into different contributio~s.First
of all it is convenient to subtract off the linear term in the total energy, according to the definition of equation (4), applied here to E rather than E;, since we are going to c o ~ ~ i donly e r zero t e ~ ~ e r a t u~r e~ e c t s , i.e. ground state energies. In equation ( classical and M F notation was used, in the ‘cluster’ world of the physics c o i ~ m u n ~ it t yi s customary to use the notation ‘formation energy’ rather than ‘mixing energy’, but the two are exactly the samc, Also the symbol A will be used to emphasize the ‘difference’ nature of the forination energy; it is by d e ~ n i t i othe ~ total (replacive) encrgy minus the concentrationweighted average of the energies of the pure elements. Thus we have, for binary s y s t e ~ s ,
If the ‘linear’ term in equation (24) be taken as tlie reference state, then the f o ~ ~ ~ t ei noenr ~ ycan be plotted as a function of concentration for the hypothetical alloy A- as shown schei~atic~lly in Figure 6 (de Fontaine, 1994). Three (or compounds), of stoichiometry A3 and only those, are assumed to bc temperatures, in addition to the pure elements A and B. The broken line linking those five ~ t ~ “ ~ ~forms ~t~ires the so-called G O I P Z V ~hull X (heavy polygonal h e ) for the system in question. The heavy dashed curve r e ~ r ~ s e n t s schematically the formation energy of the com~letely disordered state, i.e. the mixing energy of the hypothetical random state, the one that would have been obtained by a perfect quench from a very hightemperature state, assuming that no melting has talccn place, The energy distance between the disordered energy of mixing and the convex hula i the ( z e r o - t e ~ p e ~ ~ t uor e~) ~ e r energy. i~g peting structures are identified, say the one whose ~ o ~ a t i oenergy n is indicated by an open square symbol in the figure, the energy di~erencebetween it a i d the one at tlie same stoichiometry (filled square) is by definition the st~ucturfflenergy. of energy d i a ~ r a such ~ s as that found for example in Chapter In~ermetall~c C o m ~ ~ (Perepezko, ~ ~ n ~ s 1995). All the zero-temperature energies illustrated in Figure 6 can in principle be calculated by cluster expaiision (CE) tecliniques, or, four the structures, by first~principles~ e t h o ~ § . energy of the random state can be obtained from equation (21) by replacing the correlation variable for cluster a by ncctimes the point correlatio~t l . Since the
198
~
o and ~tubility ~ ~
Schematic coilvex hull (full line) with equilibrium ground states (filled squares), energy of metastable state (open square), and f ~ ~ ~ a tenergy i o n of random state (dashed line); from cie Foiitaine (1994)
disordered state energy in Figure 6 is a continuous curve, it necessaril~follows that A and €3 must have the same structure and that the compounds shown can be considered as ordered ~Upers~rLictures,or ‘ddiccorations’, of the sanie parent structure, that of the elemental solids. ut which decorations do we choose to construct the convex hull, and can we be sure that a11 the lowesteiier~ystructures have been identified? Such questions are very difficult to answer in all generality; what is required is to ‘solve the ground state problem’. That expression can have various meanings: (a) among given s t r ~ ~ t u r eofs various stoichiometries, find the ones which lie on the true convex hull; (b) given a set of Cls, predict the ordered structures of the parent lattice (or ~tructure,such as hcp) which will lie on the convex hull; (c) given a maximum range of interactions adniissible~find all possible ordered ground states of the parent lattice and their domain of existence in ECI space. Problem (c) is by far the most difficult one. It can be attacked by methods of linear prograixniiiag: the requirement that the cluster concentration lie b e t ~ e e n0 and I provides a set of linear constraints (via the configuration matrix, mentioned earlier) in the correlation variables, which are represented in <-space by sets of hyperplanes. The convex polyhedral region which satisfies these constraints, i.e. the configuration polyhedron, has vertices which, in principle, have the ect 5-coordinates for the ground states sought. method can actually predict totally new ground states, which one might never have guessed at. The problem is that sometimes the vertex determination
i
~
~
produces [-coordinates which do not correspond to any coiistructible structures, i.e. such coordinates are internally inconsistent with the requi~ementthat the prediction will actually correspond to an actual decoration of tlie parcnt lattice. Another liniitatioii of the method is that, in order to produce a non-trivial set of ground states, large clusters must often be used, and then the dimensioiis of the 5-space become too large to handle, even with the help of group theorybased computer codes (Ceder et al., 1994). Readers interested in this problem can do no better than to consult the book by Ducastelle (1991), which also contains much useful information on the CVM, on electronic structure calculation (mostly TB-based), and on the generalized perturbation method for calculating EPIs from electronic structure calculations. Another very readable coverage of ground~stateproblems is that given by Inden and Pitsch (1991)’ who also cover the derivation of m u l t i c o m ~ o ~ eCVM ~ t equations. Accounts of the convex polyhedron method are also given in de Fontaine (1979, 1994). Problem (b) is more tractable since in this case the EGIs are assumed known, thereby determining the ‘objective function’ of the linear programming problem. It is then possible to discover the correspond~ng vertex of the configuration~lpolyhedron by use of the siinplex algorithm. Problem (a) does not necessitate the knowledge or even use of the ECIs: the potentially competitive ordered structures are guessed at (in the words of Zunger (1994), this amounts to ‘rounding up thc usual suspects’), and direct calculations determine which of those structures will lie on the convex hull. In that case, however, the real convex hull structures may be missed altogether. Finally, it is also possible to set up a relatively large supercell and to populate it successively by all possible decorations of A and atoms, then to calculate the energies of resulti structures by cluster expansion (Lu et al., 1991). There too, however, some optimal ordered $tructures may be missed, those whose unit cells will not fit exactly within the designated supercell, Let us close this section by sui~marizingthe general method of calculating a phase diagram by cluster methods. The solution o f the ground-state probleiii has res sum ably given. us the correct lowest e n e ~ ordered ~y structures for a given parent lattice, which in turn determines the sublattices to consider. For each ordered structure, write down the CV minimize it with respect to the config~~ation variables, then plot the resulting free energy curve. Now repeat these operations for another lattice, for which the required ordered states have been deter~ined. Of
course when comparing structures on one parent lattice with another, we must know the structural energies involved, for example the difference between pure A and pure in the fcc and bcc structures. Finally, as in the ‘classical’ case, construct the lowest common tangents to all free energies in presence, with the locus of common tangency points generating the required phase diagram, as illustrated for example in the MF approxii~at~on in Figure 4. The whole procedure is of course much more complicated than the classical or MF one, but consider what has in principle been accomplished: with only the knowledge of the atomic numbers of the constituents, we have calculated the (solid-state portion of the) phase diagram, and also such useful thermodynamic quantities as the long-range, short-range order, equilibrium lattice parameters, formation aiid structural energies, bulk modulus, and metastable phase boundaries as a function of temperature and concentration. Note that calculations at a series of volumes must be performed in order to obtain quantities such as the equilibrium volume and the bulk modulus. A summary of those binary and ternary systems for which the program has been at least partially c o m p l ~ t up e ~ to 1994 was givcn in table form in de Fontaine (1994). Up to now, only the strictly replacive (fsing) energy has been considered in calculating the ECfs. What about the other contributions, displacive and dynamic, which have been mentioned in previous sections? These have so far been left out of the picture for si~plicity,as indeed they often have been in early calculations. It is now becoming increasingly clear tliat those contri butions can be very large and, in case of large atomic misfit, perhaps dominant. The nk initio treatment of displacive interactions is a difficult topic, still being explored, so will be summarized only briefly here (next section).
In a 1994 review article, Ztmger explained his views of displacive effects and stated that the only investigators who had correctly treated the whole phenomena, at least the static ones, were members of his group. At the time, this claim was certainly valid, indicating that far too little attention had been paid to this topic. Today, other groups have followed his lead, but the situation is still far from settled, as displacenients, s dynamic, are diacult to treat. Let us adopt definitions and examine in more detail the term EdtsPl which appears in equations (13) and (15) of Section 2.3, We have, from equations (13), (15) and (24),
where Edlspl of equation (I 3) has been ~ r o k e nup into a volume dcformation (V ) contr~bution,a ‘cell external’ deformatioii (AEdef) and a ”cell Internal’ defor~ation(relax). The various t e r ~ ~ins e ~ ~ ~ t i o (25) are explained schematically on a simple s ~ u a r e lattice in Figure 7 (Morgan et al., 1998 u~~publislied): from pure A aiid pure on the same lattice but at their ~) equilibrium volumes (the refereiice state E I ~ we expand the smaller latticc parameter and contract that of the larger to a common intermediate parameter, assumed to be that of the h o ~ o ~ e ~ e o ~ s mixture, thus yielding the large contributi~nQEy,. Now allow the atoms to rearran~eat constant volume while r e ~ ~ i n i nsituated g on the sites of the averag lattice, thereby recovering e previous section under the ”Ilsingenergy’ was ~ o n t r i ~ ~ t ~ redistribution will lower th states, so that the lattice parameters will extend or contract (such as c/a relaxation), and the cell vary, at constant volunie, a c o ~ t r i b ~ t i odn AEdef in equation (25), but not indicated in Figure Finally, the atoms may be displaced from their positions on the ideal lattice, produc~nga contr~bution denoted as AErelax. Equation (25) also indicates the order in which the calculations should be perforined, given certain reasonable assumptions (Zunger, 1994; 1998 unpublished). Actually, the cycle should bc repeated until no signi~cant atomic d i s p l a ~ e ~ e n t s take place. Such an iterative procedure was carried out in a strictly ak initio electronic structure ~ ~ l c u l ~ ~ i o n and
DO23
in Al3Ti and
atomic displacements wcre small. For more coniplicated unit cells of lower s ~ m ~ e t ~ r yi i~i ~ n i i z ~ n total energy for all possible types of ~ ~ s p l a c e ~ e n t ~ would be a toa time-consuming process, so that actual atomic forces should be computed. For tliat, planewave electranic structure techniques are ~ r e ~ e r a b l e , such as given by F L 01- pseudopotent~al ~ ~ ~codes. Let us mention here the importance of calculating energies (or enthalpies) as a function of c/a, b/a ratios, . . . at constant cell volume. It is p o s s ~ ~ lby e , a so-called ‘Bain transforination’ to go coiititiuously from the fcc to the bcc structur~s,for example.
200
Borzdiag and Stability
re 7 Schematic ~ w o - d i m e i i s i oillustrati ~~~ Morgan et al. (1998, unpublished work at UC Berkeley)
important for calculating structural energies, essential for complete phase diagram calculations, but also for ascertaining ~ h e t ~ the e r higher-energy-structures are metastable with respect to the ground state, or actually unstable. In the latter case, it is not permissible to use e x ~ ~ a ~ o l a t i of o n phase boundaries (through extrapolation of empirical vibrational entropy functions) to obtain energy estimates of certain non-stable structures (Craievich et al., 1997).
utions a ~ ~ ~ a r in i nequation g (25); according to
Other more general ‘cell-external’ deformations have also been treated (Sob et al., 1997). Elastic interactions, resulting from local relaxations, are c~aracte~istically long range. Hence it is generally not feasible to use the type of cluster expansion described in Section 2.3, since very large clusters would be required in the expansion, and the number of correlation variables required would be too large to handle. One successful approach is that of Zunger
Calculation of Phase Diagrams
(Laks, 1992) who developed a k-space SIM with effective pair interactions calculated in Fourier space. A novelty of this approach is that the Fourier transforms feature a smoothing procedure, essential for obtaining reasonably rapid convergence; another is that of subtracting off the troublesome k = 0 term by means of a co~ic~ntration-dependent function. Previous Fourier-space treatments were based on second-order expansions of the relaxation energy in terns of atomic displacements and socalled ~ a ~ z fai rkc m~ (de Fontaine, 1979; Khachaturyan, 1983), wliich forces were calculated by heuristic means, or from the embedded atom method M) (Asta and Foiles, 1996). Another option for taking static dlsplacive effects into account is to use brute-force computer simulation techniques, such as molecular dynamics (MD), or molecular statics. The difficulty here is that firstprinciples electronic structure inet too slow to handle the millions of to reach e~uilibriumover a sufficiently large region of space. Hence, empirical potentials are required, such as those provided by the EAM (Daw et al., 1993). The vast literature that has grown up around these simulation techniques is too extensive to review here; furthermore, these methods are not particularly well suited to phase diagram calculations, mainly because the average frequency for ‘hopping’ (atomic interchange, or ‘replacement’) is very much slower than that for atomic vibrations. Nonetheless, simulation techniques do have an important role to play in dynamical effects, to which we now turn. 2. ~ero-temperature contributions were described in Sections 2.5, 2.6, and 2.7, and Section 2.4 dealt only with configurational free energy. As was explained, the latter can be dealt with by the cluster variation me~hod which basically features a cluster expansion of the configurational entropy, equation (23). But even there, there are difficulties: how does one handle long-range inter~ctions,for example those coming from relaxation interactions? Et has indeed been found that, in principle, all interactions needed in the energy expansion (21), must appear as clusters or subclusters of a ~ acluster ~appearing ~ in the~CVM configurational ~ l entropy. But, as mentioned before, the number of variables associated with large clusters increase exponentially with the number of points in the clusters: one faces a veritable ‘combinatorial explosion’. One approximate^ but none too reliable solution is to
20 1
(supertreat the long-pair interactions by the position, mean-field) model. A more fundamental question concerns the treatment of vibrational entropy, thermal expansion, and other excitations, such as electronic and magnetic ones. First, let us see how the static and dynamic ~ ~ s p ~ ~ c e mcan e n be t s entered in the CTd Several methods have been proposed; i (Kikuchi and Masuda-Jirido, 19971, the CVM free energy variables include continuous displacement$~ and summations are replaced by integrals. For computational reasons, though, the space is then made discrete, leading to separate cluster variables at each point of the fine-mesh lattice thus introduced. Practically, this is equivalent to treating a multiComponent system with as many ‘compo~~iits’ as there are discrete points considered about each atom. Only model systems have been treated thus far, mainly containing one (real) component. In another atomic positions are considered to be distributed about their static equilibrium Such is the ‘Gaussian CVM’ of Fine1 (1994), simpler, but somewhat less general than that of Masuda-Jindo (1997). Again, mostly dynamic displacements have been coiisidered thu perform cluster expansions of the framework (Asta et al., 1993a), in a ~ebye-~runeisen or of the average value of the logarithm of vibrational frequencies (Garbulsk~and Ceder, 19 of selected ordered structures. As for treating vibrational fre configurational one, it was sufficient to include a Delay for each of the phases present (Sanchez et al., 1991). However, Fultz and co-workers at Caltech have recently suggested (Anthony et al., 1993; Fultz et al., 1995; Nagel et al., 1995), based on e ~ p e r i ~ e n t a l evidence, that the vibrational entropy difference between ordered and disor~eredstates could be quite large, in some cases comparable to the configurational one. This was a surprising result, indeed, which perhaps could be explained away by the presence o f n u ~ e r o u ~ extended defects introduced in the method of preparation of the disordered phases, since in defective regions vibrational entropy would tend to be higher. Several quasi-harm~niccalculations were recently u~dertaken to test the experimental values obtained in the NiSAX system; one based on the Finnis-Sinclair (Ackland, 1994), and one based on E (Althoff, 1977, 1978). These studies gave results agreeing quite well with experimental values (Antliony, 1993; Fultz et al., 1995; Nagel cf al., 1995) and with 9
202
Bonding and Stability
handbook (Binary Alloy Phase Diagrams, 1990) is incorrect (a correction appeared in an addendum, reproducing the older version). Tlie CVM-fitted Al-Li phase diagram, including the metastable L12 phase region (dashed lines) and available experimental points, is shown in Figure 8b (Sigli and Sanchez, 1984). The Li-rich side of the %/a’two phase region is shown to lie very close to the A13Li ~toichiomet~y, along, as expected, with the highest disordering point, . A low-temperature nietastablemetastable miscibility gap (because it is a metastable feature of a metastable ~ q L i i l i b r i ~ ~) is~predicted to lie within the a/a’two-phase region. Note once again that the correct ordering behavior of the L12 structure by a first-order transition near xL, = 0.25 could not have hen could m to have perimental points by using the method of concentration waves, which is completeF model? The ans ly e~uivalentto the that these authors ntly left out the temperature portion of their calculated phase boundaries so as not to show the inevitable outcome of their calculation, which would have incorrectly produced a triple second-order transition at concentration 0.5. A first-principles calculation of this system was attempted by Sluiter et al. (1989, 1990), with both fccand bcc-based equilibria present. The tetrahedron approximation of the GVM was used and the EGIs escribed a comparison between ~ r ~ t ~ t ywere ~ eobtained by structure inversion from FLAPW (hill potcntial augmented plane wave) electronic approaches. What will structure calculations. The liquid free energy was be done in this section is to present an application of taken to be that of a sub-regular solution model with rious types of calculations to a real system, the fitted to experimental data. Later, the same -Li binary, in both Jirted and . ~ r s t ~ ~ ~ i n c ~parameters les AI-Li system was revisited by the same first author and s~~iei~es. other co-workers (Sluiter et al., 1996). This time the uniinum~lithiumalloys are strengthened by the LMTO-ASA (linear muffin tin orbital-atomic sphere able coherent A13Li phase of approximation) method was used to calculate the this phase is less stable than structures required for the SIM, and far more the two-phase mixture of an fcc Al-rich solid solution structures were used than previously. The calculated -based ordered structure AlLi (p), it formation energies of structures used in the inversion on the cxperimeiitally determined are indicated in Figure 9a, an application to a real diagram (Binary Alloy Phasc~Diusystem of techniques illustrated schematically in Figure ~~~~~, 1986, 1990). It would therefore be useful to 6. All fcc-based structures are shown as filled squares, late the %,/a’ metastable phase boundaries and bcc-based as diamonds, ‘interloper’ structures (those which are not manifestly derived from fcc or bcc), unrelaxed and relaxed, as triangles pointing down and up, respectively. The solid thin line is the fcc-only, and the dashed thin line the bcc-only convex hull. The heavy broken line represents the overall (fcc, bcc, interloper) convex hull, its vertices being the predicted note that the newer version of the AI-Li ground states for the Al-Li system. It is gratif~ingto rn in the second edition (1990) of the
each other, namely LAS,,~~ equal to about 0 . 2 k ~versus 0.2 to 0.3 k~ reported experimentally. Integrated thern~odynam~c ~uantitiesobtained from ~ A ~ - b a s e d calculations, such as the isobaric specific heat (C,), agree very well with available evidence for the L12 valev et al., 1974). Note that in the roaches, the d~sordered-~tate strucaxed (via molecular statics) and the leinperature-dependence o f the atomic volume (by minimization of the quasi-harmonic. vibrational free disordered state were fully taken into is one to make of this‘? If the empirical s are valid, and so are the experimental ones, that means that vibrational entropy, usually neglected in the past, can play an important role in diagram calculations and should be 1 more work is required to confirm both and theoretical results. Other contributing influences to be considered could e tkose of electronic and magnetic degrees of freef o ~ ~ ethe r , reader should consult the ertoii and Zunger (1995). For the latter, an example o f a ‘fitted’ phase diagram calculation with magnetic interactions was given by Dahmani et al. (I 985).
203
t
AI
ure 8a ALL1 phase diagram from the first edition of the phase diagram handbook (Binary Alloy Phase Diagrams, 1986)
AI-Li SYSTEM 1100
LIQUID
so0 n
5
w n: I3
700
L11: 4:
w a, c W
500
3OO 0.0
0.1
0.2
0.3
0.4
0,s
0.6
0.7
0.8
0.9
1.0
A T O N I C ~ O ~ C E ~ r ~ A rOFI OL Ni
b AI-LI phase diagram calculated from a GVM fit (full lines), L1, rnetastable phase regions (dashed lines) and corresponding e x p e ~ i ~ e n tdata a l (symbols); (Sigli and Sanchez, 1986)
Bonding and ~ ~ a ~ i l i ~ ~
204
-I
AI
Li
Conclentration Li
i~~~~9a Convex hull (heavy black line) and formation energies (symbols) used in first-principles calculation of ALL1 phase diagram; from (Sluiter et nl., 1996)
I
0
1
0.2
0.4
0.6 CLi
0.8
1
Li
igure 9b Solid-state portion of the A1-Li phase diagram (full lines) with metastable Ll, phase boundaries and metastablemetastable miscibility gap (dashed lines, Sluiter et al., 1996)
Calculation qf’ Phase Lliagmms see that the predicted (‘i.ounding up the usual suspects’) ground states are indeed the observed ones. The solid-state portion of the phase diagram, with free energies provided by the tetrahedron-octahedron CVM approximation, is shown in Figure 9b. The etas stable a/&’ (Ll2) phase boundaries, this time calculated completely from first principles, agree surprisingly well with those obtained by a tetrahedron CVM fit, as shown in Figure 8b. Even the metastable MG is reproduced. Setting aside the absence of the liquid phase, the ~rst~principles phase diagram (no empirical parameters or fits employed) differs from the experimentally observed one (Figure 8a) by the fact 32 ordered phase has too narrow a ty. However, the calculations predict such things as the lattice parameters, bulk moduli, degree of long- and short-range order as a function of temperature and concentration, and even the influence of pressure on the phase equilibria (Sluiter et al., 1996). The case study just presented is but one amo~gst many, but should give the reader an idea of some of the difficulties encountered while attempting to gather, represent, fit or calculate phase diagrams. Examples of other systems, up to 1994, are listed for instance in the author’s review article (de Fontaine, 1994, Table V, Section 17). See for example the work on noble metal alloys by Mohri and co-workers (Terakura et al., 1987; Mobri et al., 1988, 1991), that of Astti and co-workers on A1 alloys (Asta et al., 1992, 1993’0, 1995, 1997; Johnson and Asta, 1997) and Cd-Mg (Asta et al., 1993a), that of Sluiter, Turchi and co-workers on AI-Ni (Sluiter et al., 1992; Turchi et al., 1991a) and n (Turchi et al., 1991b), and various ternary systems. Recent entries in that category are works on Heusler-type alloys by McCoimack and others (1 996, 1997). For semiconductor-material systems, see the publications of the Zunger school, and for oxides those of Burton and of Ceder and co-workers, for example urton and Cohen, 1995; Tepesch et al., 1995; Ceder et al., 1995; Tepesch et al., 1996; Kohan and Ceder, 1996), In the oxide category, phase diagram studies of the High-T, superconductor YBCO (de Fontaine, 1987, 1989, 1990; Wille and de Fontai~e,1988; Ceder et al., 1991) are of special interest, as the calculated oxygen-orderi~gphase diagram was derived by the CVM before experimental results were available, and was found to agree reasonably well with available data. Applications of the methods discussed here are covered for the A1-Ni and Ni-Pt systems in the more extended version of the present article (de Fontaine, 1999). Material pertineiit to the topic at hand caii also be found in the proceedings of the joint NSFjCNRS
205
conference on alloy theory held in France in October 1996 (de Fontaine and Dreyssk, 1997).
ssio Knowledge of phase diagram is a ~ s o l u t ~ essential ~y in such fields as alloy development and alloy processing. Yet today, that knowledge is incomplete: the phase diagram compilations and assessments are fragmentary, often unreliable, sometimes incorrect. It is therefore useful to have theoretical models whose roles and degrees of sophistication differ widely depending on the aim of the calculation. The highest aim, of course, is that of predicting phase equilibria from the knowledge of the atomic number of the constituents, a true first-principles calculation. That goal is far from having been attained, but much progress is being realized. The difficulty is that so many different types o f atomic(e1ectronicinteractions need to be taken into account, and the calculati~ns are complicated and time consuming. Thus far, success along those lines belongs priniarily to such quantities as heats of formation; con~gurationaland vibrati~nal entropies are more difficult to compute, and as a result, prediction of transition temperatures are often considerably off. It is interesting to note that theorists look at alloy system free energies from two very different angles: in the felicitous words of Simon Moss (1978), ‘electronic structure specialists devote all their considerable talents to calculating internal energies, but regard configurational entropy as a piece of cake, whereas statistical mechanics types look upon calculations of confi~urat~onal entropy as worthy of a Nobel prize, but disniiss iaternal energy as another piece of cake’. Clearly, both contributions to the free energy, and many others, must be tackled with the same degree of sophistication, so must be investigated by universalists, rather than narrow specialists, hence the small number of practicing alloy theorists today, in a field which has tended to become highly polarized. Despite the often serious discre~ancies between calculated and experimentally determined phase diagrams, the dream of ah initio phase diagram calculations is we11 worth pursuing: how else may one ever attain a true Understanding of the various contributions to alloy free energies, electronic, vibrational, configurational, in a truly Qtomislic approach? The objective of designing directly from
206
Bonding and § t a b i ~ i t ~
the periodic table may not be as far off as currently believed by inany traditionalists. On the other hand, and for the time being, fitting procedures can produce very accurate results; indeed ‘with enough adjustable parameters, one can fit an elephant’. The exercise can be useful as such procedures may actually find inconsistencies in empirically derived phase diagrams, extrapolate into uncharted regions, or enable one to extract therinodynaniic functions, such as free energy, from experiment: calorimetry, X-ray diffraction, electron microscopy or other techiiiqiies, Still, one has to take care that the sta~tingtheoretical models make good physical sense. We have seen from the foregoing that Gorsky-BraggWilliams-type models, which produce acceptable results in the case of phase separation, cannot be used for order/disorder transformations, particularly in ‘frustrated’ cases as encountered in the fcc lattice, without adding some rather unphysical correction ut even in qualitatively favorable cases, the ropy expression of equations (7) and (9), for n regular, sub-regular, CALPWAD, ragg-~illiams, and concentration wave methods tends to overesti~atethe configurational entropy contribution, so that energy parameters backed out by fitting to experimental data will also ov~~estimate enthalpy and excess entropy parameters, in compensation It rnay be argued that ab i ~ i t i uatomistic methods are too complicated to be used with multicompoiient systems, so that only mean-field approaches are feasible. This rnay be true, and Thermocalc software, for example, can do a good job of plotting out sections through fitted inulticoniponent phase diagrams. The results are obviously only as good as the empirical parameters input to the codes, however, and there exists a lot of ~ncertaintywithin the large but still incomplete data bases presently available. Ideally, this drawback does not exist for first-principles calculations: the only inputs are known integers, the atomic numbers of the constituents! In tliis article, both approaches were presented, the empir~~ally driven mean-field, ~ h e n o m e ~ o l o ~ ione cal and the q ~ i a n t ~and m statistical mechanically driven atoniistic one. For now, both approaches must exist side by side, and it will be up to the user to decide which one to use. It is expected that ab irzitio methods will become increasingly useful in the case of intermetallic alloy systems, as such systems often feature e~ and ordered comordering on ~ f r u s ~ r a tlattices pounds which are readily amenable to electroiiic structure calcu~ations.
The author thanks Jeffrey Althoff and Dane Morgan for helpful comments ~once~ni ng an earlier version of the manuscript. This work was supported in part by the US Department of Energy, Ofice of Basic Energy s under contract Sciences, Division of ~ a t e r i a l Sciences NOS.DE-AC04-94AL85000.
eferences Ackland, G . J. (1994). In Alloy Modeli~gand Design (eds 6. Stocks, and P. Turchi). The Minerals, Metals and Materials Society, p. 194. Althoff, J. D., Morgan, D., de Fontaine, D., Asta, M. D., Foiles, S. M., andJohnson, D. D. (1997). Phys. Rev. B, RS70S. Adthoff, J . D., Morgan, D,, de Fontaine, D., Asta, M. D., Foiles, S. M., and Johnson, D. D. (1998). Compt. Mater. Sc., 10, 411. Ainador, C., Hoyt, J. J., Ghakoumakos, B. C,, and de Fontaine, D, (1995). Phys. Rev. Lett., 7 Anthony, L., Okamato, J. K., and Fultz, B. (1993). Phys. Rev. Lett., 70, 1128. Asta, M. D., Wolverton, ntaine, D., and DreyssC, H. (1991). Phys. Rev. B, Asta, NI. D., de Fontaine, D., van Schilfgaarde, M., Sluiteu, M., and Methfessel, M. (1992). Phy,s. Rev. B, 46, 505. d de Fontaine, D. (1993a). and van Schilfgaarde, M. J. M., and Albers, R. C . (1995). Mater. Res. Soc. Symp., 1, 157. Asta, M. D., and Foiles, S. M. (1996). Phys. Rev. B, 53,2389. Asta, M. D., and Johnson, D. D. (1997). Compt. Mater. Sc., 64. er, J. A. (1953). Proc. R, Soc. Lortdon, Ser. A, 216, 45. Binary Alloy Phase D ~ f f ~ r a m (1986). s First Edition. T. 13. Massalski, Editor-in-chief. American Society for Metals. BiEary Alloj? Phase D~agrams(1990). Second Edition. T. B. Massalski, Editor-in-c~~ie~. American Society for Metals. Burton, P. B., and Cohen, R. E. (1995). Phys. Rev. B, C a ~ h Journal a~ (ed. A. T. Dinsdale). ~ e r g a ~ o Elsevier n, Science Ltd., Oxford, UK. Car, E., and Parrinello, M. (1985). Phys. Rev. Letters, 55, 247 1. Carlsson, A. E., and Meschter, P. J. (199s). I n t e r ~ e t a l / i ~ C o ~ p ~ ~ Volume ~ i ~ ~ f.s Princ@les , (eds J. H. Westbrook, and R. L. Fleischer). Ch, 3, pp. 55-76. Cayron, R. (1960). Etude ~ ~ ~ de3 o D~ i a ~ i r ~~ m ~~ e s e d’Equilibre d a m les Systkmes ~ u ~ t e r i ~ uInstifut ~ ~ e s , de ~ ~ t a l / i ~ Louvain. r~ie, Ceder, 6 . (1993). Compt. m ate^, Sc., 1, 144.
Calculation o j Pliase Diugranzs
207
Ceder, G., Asta, M., and de Fontaine, D. (1991). Physzca C, 177, 106. Ceder, G., Garbulsky, 6 . D., A m , D., and Fukuda, K.
Metallur~y(eds C. W. Bale, and G. A. Irons). The Met. Soc. of CIM, Quebec, pp. 57-71. Johnson, D. D., and Asta, M. D. (1997). Cornpt. Mater. Sc.,
, and Tepesch, P. D. (1995). Phys. Rev. B, 51, 11257 Connolly, J. W D., and Williams, A. R. (1983). Phys. Rev. B, 27, 5169. Craievich, P. J., Sanchez, J. M., Watson, R. E., and Weinert, M. (1997). Phys. Rev. B, 55, 787. Dahmani, C. E., Cadeville, M. C., Sanchez, J Moran-Lopez, J. L. (1985). Phys. Rev, Letl., 5 Daw, M. S., Foiles, S. M., and Baskes, M. I. (1993). Maler. Sc. Rep., 9, 251. de Fontaine, D. (1979). Solid State Physics, 34, 73-274. de Fontaine, D. (1994). Sofid State de Fontaine, D. (1996). MRS Bulle de Fontaine, D. (1999). In Method~s E. K a u ~ ~ a nWiley, ). Chapter 2a.2.1-2a.2.23. de Fontamc, D., Wille, L. T., and Moss, S. C. (1987). Phys. Rev. B, 36, 5709. de Fontaine, D., Mann, M. E., and Ceder, G. (1989). Phys. Rev. Lett., 63, 1300. de Fonfaiiie, D., Ceder, G., and Asta, M. (1990). Nature, 343, 544. de Fontaine, D., and Dreysst, H. (1997). Editors, Proceedings of Joint NSFICNRS Workshop on Alloy Theory. Cornpt. Mater. Sc., 8, 1-218. Dreysst, H., Berera, A., Wille, L. T., and de Fontaine, D. (1989). Phys. Rev. 3, 39, 2442. Ducastelle, F. (1991). Order and Phuse Stuhility ~ y 1 Alloys. North-Holland, New York. Ellner, M., and Predel, B. (1995). I n ~ e r r n e ~ a ~~ompounds, ~ic Vulzame 1. Principlrs (eds J. H. Westbrook, and R. L. Fleischer). Wiley, UK, Ch. 5 , pp. 91-125. Finel, A. (1989). Hierarchy of cluster variation methods on 3dimensional lattices and application to the study of FCC phase diagrams. In Proceedings ojthe N A TO-AS1on Alloy Phase Stabi~ity,NATO-ASI Series E, Vol. 163 (eds G. M. Stocks, and A. Gonis). Kluwer Academic Publishers, Dordrecht, The Netherlands, p. 256. Finel, A. (1994). Prog. Theor. Phys. Suppl., 115, 59. Fultz, B., Anthony. L., Nagel, J., Nicklow, R. M., and Spooner, S. (1995). Phys. Rev. B, Garbulsky, G. D., and Ceder, G. (1994). PhJw. Rev. B, 49, 6327. Gibbs, J. W, (1875-1878). On the equ~~ibriumof heterogeneous substances, Trans. Connecticut. A C I I ~ . lIl:lOt? (Oct. 1875-May 1876); ihidem 343 (May 1877July 1878). Hijmans, J., and de Boer, J. (1955). Physics (Utrecht), 21, 471, 485, and 499. Znden, G., and Pitsch, W. (1991). In Phase Trans~orrnal~ons in Materials (ed. P. Haasen). VCH Press, New York, pp, 497-552. Jansson, B., Schalin, M., Selleby, M., and Sundrnan, B. (1993). In C o ~ p u t e rSoftware in C h e ~ ~ c and a l Extractive
ryan, A. G . (1983). Theory of ~ t r ~ ~ c t ~ r ~ ~ T r a n s f o r m ~ t i oin~ Solids. Wiley, New York. Khachaturyan, A. G., Linsey, T. F., and Morris, J. W. Jr. (1988). Met. Trans., 19A, 249. Kikuchi, R. (I 951). Phys. Rev., 81, 988. Kikuchi, R., and Masuda-Jindo, IS. (1997). Gompt. Mater. sc., 8, 1. Kohan, A. F., and Ceder, G. (1996). Phys. Rev. B, 53, 8993. Kovalev, A. I., Bronfin, M. B., Loshchinm, Yu. V., and Vertogradskii, V. A. (1976). High Temp. High Pressures, 581. Laks, D. B., Ferreira, L, G., Froyen, S., and Zunger, A. (1992). Phys. Rev. B, 46, 12587. Lu, 2;. W., Wei, S.-H., Zunger, A., Frota-Pessoa, S., and Ferreira, L. G. (1991). Phys. Rev. B, McCormack, R., and de Fontame, D. (1996). Phys. Rev B, 5 9746. McCormack, R., Asta, M. D., Hoyt, J. J., ~hakoumakos, B. C., Mitsure, S. T., Althoff, J. D., and Johnson, D. D. (1997). Compt. Mater. Sc., 8, 39. Mohri, T., Terakura, IS.,Oguchi, T., and ~ a t a ~ a b K. e, (1988). Acta Metall., 36, 547. Mohri, T., Terakura, K., Takizawa, S., and Sanchez, J. M. (1991). Acta Metall. Mater., 39, 493. Morgan, D., de Fontaine, D., and Asta, M. D. (1998). Unpublislied research a t U. C. Berkeley. Moruzzi, V. L., Janak, J. F., and Schwartz, K. (1988) Phys. Rev. B, 37, 790. Moss, S. C. (1978). Private communication. Nagel, L. J., Anthony, L., and Fultz, B. (1995). Phil. Mag. Lett., 72, 421. Palatnik, L. S., and Landau, A. I. (1964). Phase E~uilibriain M ~ l t i c o ~ i ? o n eSystems. i~t Holt, Rinehart and Winstou, New York. Perepezko, J. H. (1995). I n t e r r ~ e t ~ l lCompoun~s. ic Volume 1. Prmciples (eds J. H. Westbrook, and R. L. Fleischer). Wiley, UK, Cb. 36, pp. 849-870. a. Prince, A. (1966). Alloy Phase ~ 4 ~ i l i ~ r iElsevier, Amsterdam. Sanchez, J. M. (1993). P h p . Rev. B, Sanchez, J. M., Ducastelle, F., and Gratias, D. (1984). Physica, 128A, 334. Sanchez, J. M., Stark, J., and Moruzzi, V L. (1991). Phys. Rev. B, 44, 541 1. Sanchez, J. M., and de Fontame, D. (1978). Plzys. Rev. B, 17, 2926. Schlijper, A. G., and Smt, R. (1989). J. Stat. Phys., 56, 247 Shockley, W. (1938). J . Chern. Phys., 6, 130. Sigli, C., and Sanchez, J. M. (1986). Acta Metall., 3 Singh, D. J. (1995). Interrnetallic Comp(~unds,Volume 1. Principles (eds J. H. Westbrook, and R. L. Fleischer). Wiley, UK, Ch. 6, pp. 127-147.
Sluiter, M. €3. F., de Fontaine, D., Guo, X. Q., Podloucky, R., and Freeman, A. J. (1989). Mat. Res. Soc, Symp. ., de Fontaine, D., Guo, X. Q., Podloucky, R., and Freeman, A. J. (1990). Phys. Rev. B, 42, 10460. Sluiter, M. H. F,, Turchi, P. E. A., Pinski, F. J., and Johnson, D. D. (1992). Mat. Sci. Eng. A, 15 Sluiter, M. H. F., Watanabe, Y., de Fontaine, D., and Kawazoe, Y. (1996). Phys. Rev. B, 53, 6137. Sob, M., Wang, L. G., and Vitek, V. (1997). Compt. Mater. Sc., 8, 100. Tepesch, P. D., Garbulsky, C. D., and Ceder, G. (1995). Tepesch, P D,, Asta, NI. D,, and Ceder, G. (1996). J. Am. Cer. Soc., 79, 2033, Tepesch, P. D., Asta, M. D., and Ceder, G. (1998). Modelling Simul, ~ ~ Sci. tEng.,e 6, 787. ~ Terakura, K., Oguchi, T., Mohrr, T., atid Watanabe, K. (1987). Phys. Rev. B, 35,2169. Turchi, P E. A., Siuiter, M. H. F., Pinski, F. J., and Johnson, D. D. (1991a). Mat. Res, Soc. ~ . ~ m p186, . , 59. Turchi, P. E. A., Sluiter, M. W. F., Pinski, E. J., Johnson, D. D., Nicholson, D. M., Stocks, G. M., and Staunton, J. B. (1991b). Phys. Rev. Lett., 67, 1779.
Turchi, P. E. A. (1995). ~ ~ t e ~ m eC~o am lp lo~u n~~Volume ~ I: Priraciyles (eds J. H. Westbrook, and R. &. Fleischer), Wiley, UK, Ch. 2, pp. 21-54. ~e , Voter, A. F. (1995)' ~ ~ ~ ~ r m ~e o~ um ~~ l o~ Vuco ~~ ~~ m1. Principles (eds J. 1-1. Westbrook, and R. L. Fleischer). Wiley. UK, Ch. 4, pp. 77-90. Vul, D., and de Fontaine, D. (1993). Mat. Res. Soc. Symp. Proc., 291, 401. Westbrook, J. IT., and Fleischer, 3%. L. (1995). Editors of I ~ t e r m e ~ a l lC~ co ~ i ~ o ~ n Volume dsj I , Pri~ciples. Wiley, UR. Wille, L. T., and de Fontaine, D. (1988). P h y , ~Rev. ~ B, 37, 2227. Wolverton, C., Asta, M. D., Dreyss6, H., and de Fontaine, D. (1991a). Phys. Rev. B, Wolverton, C,, Dreyss6, E., and de Fontaine, D. (1991b). Mat. Res. Soc. Symp. Proc., 193, 183. e Fontaine, D., and Dreyss6, W. ~ o l v e ~ t oC., n , Cede (1993). Phys. Rev. 726. Wolverton, C., and r, A . (1994). P h j ~ .Rev. B, 50, 10548. Wolverton, C., and Zunger, A. (1995). Phys, Rev. B, 9 Zunger, A. (1994). In Statics and ~ ~ n ofa Phase ~ ~ T ~ a ~ s ~ o r ~ a (eds tioi~ P ~ E. s A. Turchi and A. Conk). Plenum Press, New York, pp. 361-419.
~
s
t appears that the human mind cannot arrive at siniplicity except by passing through the complex; it is like a mountain more or less elevated, whose heights must be overcome before the plain on the opposite base can be reached; and when rea ed, the level seems to be that of the plain left behind. when a simple solution to a problem is arrived at, and almost self-evident. 6
iscovery consists of seeing what everybody has seen and what nobody has thought.
If you do something once, people will call it an accident. f you do it twice, the third time and you’ve
The atomic-scale structure and properties of the free surfaces of metals and interinetallics have been intensively studied in recent years. The availability of new measurement techniques has improved the knowledge of surface structures substantially. Atomically clean surfaces can be prepared and modified, due to the availability of ultra-high vacuum techniques. Interfaces and free surfkces are of technological importance in the fields of catalysis, corrosion, oxidation and friction and wear. Surface structure and energy are important in fracture processes. Most interinetallics present brittle fracture behavior, wliere knowledge of the energy of the created surfaces IS essential in understanding and improving the fracture properties. The influence of interfaces and free surfaces on the ~ e c h a n i ~ abehavior l of inter~eta~licsis especially important in nanostructured materials. For the purpose of this review we define the surfiace as a region of space limited to a few atomic layers. Different terminations at the surface may affect the adsorption properties of various species. Many adsorbates can be deposited on a metallic surface, which influence the structure and properties of that surface. However, only clean surfaces of intermetallics will be discussed in this work. A general discussion of metal surfaces can be found for example in the book by Howe (199'7) or the review by van Hove (1993). It is well established that ordering in intermetallic bulk crystals has a great influence on the properties. The current interest in intermetallic compounds for high-temperature applications and fknctionai materials has led to a large effort in understanding ordering effects on bulk properties. However, the structure and
properties of surfaces do not necessarily reflect those of the corresponding bulk materials. which means a rearrangement o f atoms at the surface, may occur. Segregation of different atomic species at surfxes i s often related to these recoi~str~~ctions and can change the surface properties s i g n ~ ~ c a n tThe l~. ordering properties of surfaces are often diRcrent from those in the bulk. Also, the concentration of point defects may change due to the presence of free surfaces and interfaces (Aust et al., 1966, 1971). Experimental and theoretical methods of studying surface phenomena have been greatly irnpro recent years. Scanning tunneling microscopy allows direct irnaging of surf'itces with atomic tion, and even chemical contrast between different atom species has been achieved recently on some metallic and intermetallic surfaces. Surface ordering and disordering effects, therefore, can be s t u ~ i e d directly for the first time using ost experiniental characterization techniqucs yield information from the outermost surface area, which makes a careful extrapolation to the bulk ~ropertiesne~essary. The techniques for sui-face investigations will be described in the first part of this chapter. surface structures and their possible recoiistru~tions will be discussed. In section 4 the structures of s simple binary intermetallic compounds with B2(c and Ll~(cP4) structures will be reviewed. The influence of different constituent elements in isostructural compounds on surface structure and relaxation will be compared. Surface phase transitions and orderin~-disor~ering effects at surfaces will be treated in section 5. The surface transitions in a few intermetallic com~ouiids with a solid-state order-disorder tr~~nsition such as
~ ~ ~ ~ rCompounds: ~ e t ~Vol. i ~3, Principles ~ ~ c and Practice. Edited by J. H. Westbrook and R. L. Fleisclier. 0 2 0 0 2 John Wiley & Sons, Ltd.
Properties and Phenomenology
212
CU~AU have been studied thoroughly in the past. ection 6 deals with segregation properties, and in section 7 the properties of surfaces aiid some practical influences are considered.
xpesi erties
n important point for experimental techniques used surface science is achieving an extremely high surface sensitivity, to get information from only a few atomic layers of material. Van Hove (1999) e major techniques for structure e preparation is a critical step for surf'ace studies, since most surfaces are not stable in the clean and cry§tallographically ideal manner necessary for investigations. Most surface investigations, therefore, require preparation and storage of the specimens in ultrahigh vacuum (UHV) chambers at a base pressure close to 10-s pascal. Such environments allow even reactive surfaces to stay clean for hours. ~ r ~ p a r a t i oofn such surfaces is usually performed by a combination of annealing and sputtering with noble gas ions. However, preferential sputtering of one compoiient in a binary system can significantly change the surface coniposition from that of the bulk. For exaniple, A1 is preferentially sputtered in the B2(cP2) compounds FeAl and NiAl. Clean. surfaces of several brittle intermetallics can also be prepared directly by cleavage in vacuum (Heatherly et al., 2997).
iffraction techniques, inaiisly electron and X-ray ffraction, are widely applied to study the structure of bulk materials and surfaces. Low-energy electron diffraction (LEED) is one of the most successful techniques and has been applied in many dies of surfaces, see for exaniple the book by Van ove et al. (1986). Elastically diffracted electrons with energies in the 20-300 eV range are used to probe the surface structure. Since, in general, only elastically scattered electrons with a limited penetration depth iii the range of 1 nm, are recorded, LEED is a very surf~ce-sensitive technique. The diffraction patterns are displayed easily on a fluorescent screen and serve as a convenient monitor of the surface condition. Surfwe preparation procedures are usually controlled with this technique. The long-range-ordering parameter of intermetallic surfaces is determined from the intensity of the diffraction spots. Multiple scattering of electrons
complicates the interpretation of the experimental diffraction data. Therefore, the diffraction patterns of the entire multiple-scattering process are simulated for all plausible surface geometries, from which the surface which gives the best fit to the experimental results is chosen. X-ray diffraction is very useful in studying the structure of bulk materials. However, the long meanfree-paths of X-rays in solids makes them less sensitive for surface studies. Only grazing incidence or emergence techniques (with angles within a fraction of a degree) permit the desired surface sensitivity. These of course are challenging experimental tasks and extremely flat surfaces are necessary (Birgeneau and Horn, 1986), The sufficient photon flux required is often achieved only at synchroton radiation facilities. X-ray diffraction is particularly suitable for investigation of disordering phenomena such as surface roughening and phase transitions. Grazing-incidence X-ray scattering (GIXS) is based elow a critical angle, the X-rays are reflected out of the sample rather than into the sample. By varying the incident angle, it is possible to sample at different depths into the surface. X-ray diffraction at grazing incidence has better sensitivity at short distances parallel to the surface, thus in-plane strain measurements are possible directly and at various depths.
Field ion microscopy (FI ) is a well-established technique to study surfaces with atomic resolution. Nevertheless, the requirement to shape the specimen into a tip with a very small radius restricts its use to a few cases. A new development is the 3D atom probe FIM, "with which full three-dimensional atomic resolution with chemical contrast can be achieved. So, chemical mapping of 3D volumes is possible. A chaiinel plate is used as a detector for the fieldevaporated atoms, to get information as to where these atoms are evaporated from the tip. This, of course, is a very powerful tool to study segregation at interf.a e s or surfaces. Since the invention of the scanning tunneling microscope (STM) by Binnig and Rohrer, 1982, this technique became one of the most powerful and most frequently applied tools for surface scientists. The unique capability of visualizing atoms in direct space on a surface allows new insights into surface properties, which previously were accessible only by
Free S u ~ ~ S~t rcuec t ~ &and ~ e Properties theoretical simulation techniques. Steps, defects and atomic corrugations can now be studied directly. TM readily delivers geometrical data from surfaces, information about the chemical structure was obtained first of all on semiconductors with tunneling spectrsscopy (Tromp, 1989). For example, on a cleaved GaAs crystal either the Ga or As atoms can be made to appear in the STM image, by switching the bias voltage (Cox et al., 1990). On metals it appeared at first impossible to differentiate between different metallic species on surfaces, but recent results demonstrate atomic resolution with
213
chemical contrast on several metallic systems, see for example Schmid et al. (1993); Vargx and Schrnid (1999) and Sprunger et al. (1996). This, of course, offers the possibility of studying ordering and disordering effects on intermetallic surfaces with previously unattainable resolution. applied to an interrnetallic, an example STM image with ch surface is sliovvn in atoms appear at different gray levels and the amount of Pt/Rh concentration at the surface could be determined from the histogram. The difference in. the
Figure 1 STM image of Pt,,Rh,,(lll) with chemical contrast. The surface was annealed at 800°C after Ar sputtering. Thc histogram of the atom heights results m two separated peaks from which the concentrations caii be determined (from Hebenstreit et at., 1999), reproduced with permission
Properties and P ~ ~ e ~ o ~ ~ e ~ o l o g y
14
applied to study surface composition. A very high surface sensitivity is obtained with ion-beam techniques such as secondary ion mass spectrometsy (SIMS) and ion scattering s p ~ ~ t r o s ~ o(ISS). p y Static SIMS, which uses low ion intensitics for no~~estructive surface investigations, coupled with a time of flight (TOE;) mass spectrometer, is probably the most sensitive analytical technique. Ion scattering at low energies below l keV (LEiS) yields information from the top one or two atom layers. ~ ~ a t t e r with i n ~ alkali ions provides particularly good structural and compositional sensitivity, when monitored near the 180' scattering direction. This method is also called ALICISS, al~ali-ion-i~pact collision ion scattering spectrsscopy (Niehus and Comscz, 1984). Another scattering technique, NICISS (impact collision ion s ~ ~ t t e rspectroscopy in~ with neutral particle ~ e t e c t i ~ n ) was demonstrated to be very useful in studying surface structures (Niehus, 1991). These latter methods are parti~ular~yappro~riate in analyzing the surface composition in the very topmost layer and are very useful in studying the different possible surface t ~ ~ i n a t i o nof s inter~etallics. In high-energy ion ) ions can channel between rows scattering ( H ~ I S the
apparent height between h atoms depends on the tip condition. A maximum height difference of 25 pm was r ~ p o r t e ~ which , is much larger than the difference in atomic radii of Pt and Rh (Table 1). kerefore, the contrast cannot be explained purely etrically. The reason for the chemical contrast on surfaces is attributed to an adsorbate on er unstable tip conditions were reported by Schmid et al. (1993) as necessary to get tlie chemical ugh these images were obtained at room STM images can be tare, a ~ o ~ i c a l lresolved y obtained nowadays also at high te~peraturesup to 1200 "C and at low te~peraturesdown to a few elvin. This, of course, is very interesting for the study of phase tE.ansitions at surfaces.
number of techni~uesare capable of reflecting the emical composition at or near the surface. Scattering and electrons are the most Auger electron spectroscopy all, 1983), has been extensively
Periodic table for some metallic elements. The atomic radius and heat of vaporization are given for some metals (all data from Schulze, 1974) 1
0.143 nm 251 kJ/mol .
-
~
2
27
0.146 nm 66 kJ/mol
425 kJ/mol
3
0.125 nm
45
I
46
49
I
0.132 nrn 420 kJ/mol
0.137 nm 556 kJ/mol 77
81
I 0.136 nm 629 kJ/mol
I
Free ~ u ~ f Structure a ~ e and Properties of atoms in the crystal. efects or interfaces in the channeling direction repel the ions, which will be scattered strongly back out of the surface.
The physical properties of surfaces are, of course, essential for applications. Catalytic reactions for exaiiiple take place only at surfaces. The niagiietic properties of materials are largely influenced by surface or thin-film effects, as for example in the giant magnetoresistance (CMR) effect. Mcasurenieiits of the physical properties of surfaces are of great relevance for new developments. STM, as described above, was the first technique in the family of scanning probe microscopies (SPM) (Bonnell, 1993). The other scanning probe technique^ such as atomic force microscopy (AFM) or magnetic force microscopy (MEM) are especially useful for measurements in laboratory air (AFM) or to gain i~formation on material properties (MFM). Scanning probe microscopy is a very useful tool in this context as it allows m e ~ s ~ ~ ~ e mof e nthe t s magnetic and electric forces on a very local scale. A. combination of an AFM with a naiioiiidenter (NI-AFM) allows deterininations of the mechanical properties of very thin surface layers or close to interphases (Goken et al., 1999).
The theoretical re diction of surface structures and energies in atomistic siniulations is based on the iiiteratomic potentials, which in the simplest case is a pure pairwise potential. However, simple pairwise potentials fail to predict the structure of intermetallics accurately. More realistic potentials go beyond the pure pairwise description o f interaction and are called em~eddedatom potentials (Daw and Finnis-Sinclair potentials (Finnis and Sinclair, 1984). They all have the following form in the secondm o ~ e n a~proxii~ation t for the energy of an atom i at zero temperature:
where the first term represents a simple pairwise int~raction,that accounts for the repulsive interaction of the two ion cores. The second term is an attractive inaiiy-body potential that models the effect of electronic cohesion, also called the glue term. The square root
3t
functional form for the cohesive energy comes from the second-moment a p p r o x i m ~ t i oin~the t i g h t - ~ ~ n d ~ ~ ~ description. Potentials of this kind are very successfLi~ for inodelling most fcc metals aiid also many iiitermetallic compounds. Howevcr, they fail to predict tlie negative Cauchy pressure of certain cubic inetals (Pettifor and Aolsi, 1991), since they take no a~count of the angular forces in the total energy. Even more realistic inodels are three- and four-body potentials or tight-bind in^ models beyond the secondmoment approximation. Atomistic s i i ~ ~ ~ l aare ti~~s very successful iii calculating surface energies o f different structural models and iii predicting the relaxations of atoms in the topmost layers by niolecular statics at 0 K temperature. ~ynamicsand Moiite Carlo $imulations can be used to study segregation effects. In principle, the most accurate predictions are expected from ab-initio or ~ r s t - ~ r ~ ~ i cmethods. i~les ~ b - i n i t i o techniques have been used to calculate surface and cleavage energies of ilnaiiy ~nteri~etallic compounds. They are useful to study the surface electronic properties and to u n ~ e r s t a nsurface ~ rippling effects. However, they have not been appli~dto complex structures at interfixes in nietals, due to their high coi~putationalcosts.
A fair number of clean metal surfaces exhibit a bulk termination with only some minor relaxations of the surface atoms. This behavior is most p r o m i ~on~ ~ ~ metal surfaces with low Miller indices, such as fcc and bcc (100) and (1 11) surfwes or basal surfaces in hcp crystals. Clean (130) surfaces of the fcc m e ~ ~i lr si d i u ~ , platinum and gold are known to reconstruct to a (I x 2) surface in order to lower their surfwc energy by reconstruction rather than undergo a simple relaxation process (Niehus, 1984; g et al., 1983). The (1 x 2) reconstruction has a iig-row geometry, where rows of atoms from the ideal (110) surface are removed, thus forming very narrow (1 11) facets on the surf'ace. Surface structure and termination are more ambi uous in the case of lattices with more than one atom per primitive unit cell, such as the intermetallic compounds. Most of the crystallo~raphicorienta~ions imply a choice between two or more non~equivale~t terminations, depending on the exact depth at which the surface is cut. Even in pure metals, as for example on (10 1 surfaces in hcp cryst~ls, two d i ~ e r e n t ~ ~
terminated surfaces are expected from the bulk structure. Since the coinplexity of internietallic structures is very high, only simple binary systems will be discussed in this work. More examples on other systems may be found in the review by Vasiliev (1997). The surface structure of quasicrystals is er et al. (1999) and Shen et al. tructure of an AI5 (cP8) structure inharoy et al. (1986) and liquid metals are discussed by DiMasi et al. (2000). 4.1 Reconstruction and Terminations of
Some clean surfaces take on structures that differ substantially from the ones predicted from the bulk. uch surfaces are said to be reconstructed, as for example the well-known anid complicated (7 x 7) reconstruction of the Si(l11) surface, where a complex rearrangement of atonis in the several topmost layers is involved. In semiconductors the broken bonds at the surface, called dangling bonds, are directed out of the surface. Lowering of the surface energy occurs by re~ybridi~atioii~ that reduces the number of dangling iinpler reconstructions or ideal bulk terniiiias occur on most metals and intermetallics. ertheless, relaxations of atoms near the surface occur frequently, even on ideal, unreconstmctured surfaces. Since the surface atoms are located in an asymmetric environment, they are driven into new locations that correspond to minimum free energy. In the simplest case, only small displacements are needed to achieve the e~uilibr~um structure. However, relaxations have proven to be relatively large in intermetallic compounds.
t
The arr~ngementof atoms at the free surface is, in general, rather complex. Even if the surface is simply as in the bulk, which nieans it is unrecanstructed, the surface unit net is not necessarily a simple projection of the three-dimensional unit cell of the bulk onto the surface. Consider for example the (100) surface of an fcc metal, Figure 2 shows this situati~nwhere bolded filled circles denote atom positions in the top layer and gray circles the atom positions in the next layer. This surface has square symmetry, as expected from the four-fold symmetry. However, the surface is shown by the primitive square unit net s right in Figure 2. This kind of diflerence in description arises, of course, from the use of a noi~~primitive unit cell for the three-dimensional fcc bulk lattice. A convenient notation for surface structures~which is widely used, is that sug ested by Wood (1964). The notatjon defines the ratio of the lengths of the primitive translation vectors of the surface and substrate meshes, together with the rotation angle whicli is necessary to align both meshes. A structure having primitive translation vectors of length a' = p a and b' = gb with a rotation angle 0 is referred to as a (p x q)RB reconstruction. Figure 3 illustrates this notation for the (100) face of L12 ordered intermetallics, where two different surface t ~ ~ i n a t are i ~ ~ s possible. A Cu-Au layer on top of a Cu layer has a (J2 x J2) 45" reconstruction. This reconstruction is typically referred to as a c(2 x 2) reconstruction, where the index c denotes a centered unit mesh, The translational vectors of the centered unjt mesh are two times longer than the primitive translation vectors for the Cu layer. Note also, that already the primitive unit mesh has an atom in its centered position, since the surfxe mesh is built up of two penetrating simple
t
igure 2 (100) surface of a disordered fcc metal. A projection of the 3-D non-primitive unit mesh from the bulk is marked by dashed lines. The (1 x 1) unreconstructed primitive surface unit mesh IS shown on the right
Free ~u~~~~ Structure and Proper ties
2x7
Figure 3 Reconstruction of Llz (100) surface of Cu3Au. The primitjve surface mesh of a Cu layer is marked by continuous lines, the recoristructioa of the topmost AuCu layer dashed (primitive unit niesh) and dashdotted (ten tcred unit mesh)
compositioii Cu3Au, and 110 variations are expected at cubic lattices with different atomic species. Therefore, the surface. only the larger, rotated unit mesh is referred to as a The surface structures of two Llz ordered comcentered mesh. pounds have been examined in detail. The compound Next to structure differences of surfaces from the Wi3Al lias received ~ ~ n s i d ~ r a~b tl e~ e i i t becmsc io~ of binlk, one of the i ~ ~ ~ r features e ~ t ~of ~ordered i ~ its technolo~icali ~ ~ o r t a n case a ca~didatebase for inter~eta~lic surfaces is that they can have a different high-temperat~ire structural materials. The intrinsic termination, For exaniple (100) facets of the B2 brittleness of this compound in its polycrystalline form compound NiAl can have either a Ni- or Al-terminated is the major drawback. Alloying with boron im~roved surface. The number of possible t e ~ i n a t i o n scan be the fracture behavior considerably, due to its segregaobtained from the Miller indices (h, k , I ) in a simple tion to grain boundaries. Cu-Au alloys are a standard manner, For example, surfaces of B2 compounds t e bulk ~ ~and~ ~ whcre h + 12: I is even have one possible t e ~ i ~ ~ ~ a ~example i o n ~ of it binary ~ ~ ~ t i ~t l l~i ~swhose whereas all other surfaces have two possible terminasurface properties have been extensively studied. tions (Farkas and Ran, 1986). Cu-Au alloys have a negative mixing enthalpy and
-+
Figure 4 shows the structure of two i~termetallic compounds based on the fcc ~ t ~ u c ~the ~ r Llz(cP4) e, and the Llo(tP2) structures. The possible terminations o f low-index surfaces can be deduced from the structure. In the case of the ~lz(Cu3Au)structure three different orientations are considered. In the (100) and (1 10) orientations the ideal structure predicts alternating layers of pure Cu and 1:1 mixed Cu-Au layers. The ~oiicentrat~oi~s in the topmost surface layer, therefore are of great interest, In { 111) oriented crystals each plane in the bulk is o f stoichiometric
218
Properties and P ~ i e ~ o ~ ~ i J ~ o l o ~ y
form stable i n t e ~ e t a l l ~phases c at lower temperatures. e their bulk properties and structure have been es~ablislie~, it became a favorite phase for i~ivestigations of surface ordering and disordering 0th intermetallic phases, the tetragonal , phase are CuAu, klo and the cubic C U ~ A ULlz on the fcc struc~ure.The L1z structure in bulk ed by a discoiitinuous first-order transition at a temperature of 663 K from tlie ed fcc structure. The alloy systems Ni-Pt and re other e~ampleswith phase diagrams similar to that of Cu-Au; intermetallic pliases of the same struct~resare f ~ r m e din these systems. A lot of work has been done to clarify the questions of the surface structure and the termination of Cu3Au. Therefore we start with a discussion of it, 4.2.1 Cu3Am SurJiirC~s t is e x ~ e r i ~ e ~ t awell l l y established that the structure u (100) surface corresponds to an ideal bulk termination. LEED data froin Potter and Blakely (1975) showed the c(2 x 2) superstructure expected for the ideal t e ~ i ~ a t i oofnordered bulk (Figure 3). STM ehus and Achete (1993) showed that this surface i s unreactive to oxygen and nitrogen, and it t~ereforeis a model system for kinetic studies of surPxe phase transitions. The crystallographic structure of the oL~ter~iost layer is the same as that in the image of the surface er the surface layer
s found, with pure
detailed study of the te~perature dependence of he intensity of LEED superlattice reflections was evaluated as a function of temperature^ from e order parameter was determined. urface transition will be discussed later. in~estigated the (100) surface of 3Aii with STM and low-energy ion backscatter~~ig at room temperature. This study shows the first direct imaging of individual atoms on this e n t species, Cu and Au, could surface. The ~ i ~ ~ ratomic be ~~iscrimi~ated, although a direct identification in the was not ~ o s s see ~ Figure ~ ~ e 5.~ The NICISS data conceii~~a~ion in the surface layer 0 in tlie second layer. OnIy double were found on the surface with an
STM image of the CU~ALI (100) surface with atomic rcsolutlon. In thc lower right part both atom species are visible in the unit cell at different gray levels (from Niehus and Achete, 1993), reproduced with permiss~Qii
average step height of 0.39 nm, which is only 0.015 nrn larger than tlie bulk lattice parameter, Theoretical calculations by Wallace and Ackland (1 992) confirmed the experimental view and suggested that the stoichio~~trically mixed Cu-Au surface is energetically favored by about 0.3 J/iii2. Therefore, it could be excluded that small Cu-terminated areas exist on the surface. (1 10) surfwes of Cu3Au can exhibit the same possible terininatioiis as the (100) surface, either a pure Cu layer or a 50 at.(%Au,50 at.% Cu layer. As in. the case of the (100) surface, only mixed Au-Cu layers were found as surface terniinations (Morgenstern et al., 1995). LEED showed a investigations ( ~ ~ i ~ a etc al., h ~ 1989) r (2 x 1) diffraction pattern, which corresponds to an ideal bulk termination of the crystal. Further investigations by LEED and STM also revealed a possible (4 x 1) reco~$tructionafter slowly cooling the crystal. The surface composition was found to be again a mixed ALI-CU layer. An interpretation of the second possible rec~ns~ructionwith a detailed model of the surface struct~recan be found in Niehus (1995).
4.2.2 NE'& and
~~~~~
Ll2(cP4) ~ o ~ p o ~ n ~ ~
Next to C U ~ A Uthe , most intensively studied compound is probably Ni3A1. For the (100) surface the
Free Surface Structure and Properties
mixed layer termination is preferred rather than the other possible pure Ni termination (Sondericker et al., 1985). First-principles simulations ( 1986a) confirmed the experimental findings by calculating the cohesive energy, where the mixed layer termination was found to be more stable than the Ni term~nation.A mixed termination was also found for the (110) surface (Sondericker et al., 1986b). As expected from the bulk structure, a surface termination with stoichiometric composition was found on the (111) surface of Ni3A1. A small buckling of the topmost layer occurs 011 all low-index surfaces with the A1 atoms shifted slightly outward (Sondericker et al., 1986~). For the (100) surfaces of the compounds Pt3Ti and Pt3Sn the mixed surface termination was again found to be preferred, although later experiments on the Pt3Ti( 100) surface suggested a Pt termination (Atrei et al., 1992). In Pt3Ti a reconstruction was observed on the (1 11) surface with a pure Pt layer on top (Chen et al., 1993). In this compound the Pt termination may be favored because of the substantially different heats of vaporization of Pt and Ti. Though the top layer consists of pure Pt, buckling was observed in the top layer and stronger so in the second layer, with Ti atoms moved upwards and Pt a t o m moved downwards. The two top layers are contracted relative to the bulk. In Pt3Sn a buckling of Sn atoms out of the surface (Atrei et al., 1992) may be explained by the larger atomic radius of Sn compared to Pt.
The aluminide B2(cP2) compounds (for example: NiAl, FeAl, RuAl and CoA1) gained special attention because of their possible applications as high-temperature materials. Although their crystallographic structure is identical, the bulk and surface properties are quite different. The ordering energy seems to be a very important parameter with reference to their fracture mode and surface structure. NiAl possesses a very high ordering energy, whereas FeAl has a relatively low ordering energy. Compounds crystallizing in the B2 CsCl type ordered structure have bulk (100) planes with an A AB stacking sequence with alternating layers of all A atoms and all E3 atoms. So, for example in NiAl the (100) surface could have either a Ni or A1 termination. Figure 6 shows structural models for the low index surfaces. Results show that a non-bulk mixture or a reconstruction of the (100) surface can be excluded by the observation of (1 x 1)
219
unit cells. AI term~nationsappear to be favored on NiAl(100) surfaces (Davis and Noonan, 19813). The Al termination of the (100) surface was confirmed also by LEIS investigations ( ~ u l l i n s and Overbury, 1988), although a reconstruction of the surface was reported in this study after an annealing procedure between "90 K and 950 IS.The I5g.h ordering energy of NiAl fiwors a structure where each component is surrounded by the other species. A1 has a lower heat of vaporization than Ni, and a larger atomic radius (see Table 1) which may favor segregation of A1 to the surface and cause the AI termination of the NiAl (100) surface. A inore thorough discussion of the NiAl (100) surfzace can be found in the work of (1 10) surfaces in CO have a rectangular unit cell with a mixed stoichiometric composition. No different surface t e ~ i n a t i o n sare expected. The mixed surface termination was confirmed by a quaiititative LEED study (Davis and Noonan, 1987 and 1988). The outennost surface showed a large relaxation or rippled surface structure, which will be discussed in the next section. (1 11) surfaces of I32 compounds again are expected to show two different surface terminations. The relatively open structure of this surface consists of alternating layers of pure A or pu small interlayer spacing of 0.083 NiAI. Therefore, the structure of this surface is inore complex, and different explanations for the surface structure exist. Different possible terminations are expected to have different surface energies, and therefore one should be favored over the others. In contrast, LEED observations on the ( showed both possible terininati diRerent parts of the surface 1988). Noonan and Davis interprete~their investigation based on the assumption that the surface energies are comparable at finite temperatures and that no complete exclusion is to be expected. A model with two different terminations on different parts of the surface implies the existence of domain boundaries with rnonoatomic steps separating the areas with Ni and A1 t e ~ i n a t i o n s .As in the case of the (110) surface large relaxations were reported. STM and LEIS investigations by Niehus et al. (1990) revealed more details. The STM study showed large, flat Niterminated terraces with A1 in the second layer, separated by double atomic steps. No Al-terminated terraces were found on clean surfaces in this study. However, small amounts of oxygen cause the fomation of small triangular A1 domains on top of Niterminated terraces.
220
Properties and ~ ~ e n ~ ~ ~ n ~ l o g y
~ i 6 Struct~ral ~ u models ~ ~for the two topmost layers of low index surfaces of E32 (cP2) compounds. Dark circles are atoms A and light gray circles atoms B. Note the mixed ter~inationon the (110) surface
Relative to the iizv~sti~~tioiis on NiAl surfa'aces, i t is of interest to compare those results with measurements on other I32 compounds based on A1 such as FeAl, CoAl and uch less work has been performed on surface compounds. ~ r a u p n e et r al. (1995) and Hammer et al. (1998) investigated the structure i r ~low-index ~ ~ surfaces of arid s ~ ~ r ~b ~ ~ aa vti oof FeAl by AES and LEE . They found the (100) surface to be Al-terminated and unreconstructed, similar to the case of NiAl. Other surfaces, (1 10), (1 11), (210) and (310), showed no (1 x 1) diffraction pattern, thus indicating the absence of a bulk termination at the surface. ~ n c o ~ ~ e nsurface s ~ r aallays ~ ~ of FeAl2 were found on the (1 10) surface. In CoAl the (1 10) surface shows a bulk-like t e ~ i n a t i o nsimilar ~ to that on NiAl (110) (Hum et al.> 1996b). 1n contrast to nearly bulkterminated surfaces in NiAl and CoA1, FeAl shows significantly different surface structures which are also related to segregation effects and will be discussed in more detail in section 6.1.
Although surface reconstructions have not been observed frequently on i ~ i ~ e r ~ e t a l Isurface i ~ s , relax&tions are a quite common phenomenon. Atoms near the surface are located in very asymmetric environments. Therefore, the bulk positions may not correspond to the lowest energy state, thus driving the surface atoms into new positions of minimum free energy. ~ ~ l a x occur, a ~ ~both ~ ~per~endic~~lar s aizd parallel to the surfiice. Surface relaxations are studied experi~entally with LEED, and a best fit of the
~ i ~ r a ~ pattern t ~ o i iwith an ~ s ~surface ~ ~structure ~ i is~ sought. With such experimental techniques the relaxation of a few topmost layers can be determined. Atomistic simulations also yield detailed knowledge on the locations of atoms in each layer depending on the interatomic potentials used. Farkas (1995) reviewed different i ~ ~ ~ ~ s t i ~of a tthe i o surfatx ~ i s relaxation in the Ni-A1 system. The most striking relaxation process in many int~rmetallicsis a rippling of the surface, On surfaces with a mixed ter~inationthe plane of one atomic species is moved outward, whereas the plane of the other atoms is moved inward, thus producing a buckled surface structure. For example, an outward relaxation of A1 was found on all NiAl and Ni3Al surfaces. In. cases, where the interato~icpotentials can be approximated accurately as pairwise central potentials, the interplanar relaxation is always an expansion. The outermost surface layer is most strongly affected and expansions of several percent are c o ~ ~ m o(Allen n and de Wette, 1969). The magnitude of the expansion decays to zero within a few layers. Contractions, or inward relaxations, of the interplanar spacing also occur, and sometimes an oscillatory dependence of the lattice spacing was found near the surface. Contrdctions at the surface can be explained by a change in the interatomic potential and many open metal surfaces show an inward relaxation (Luth, 1993). A redistribution of electrons can cause such variations near the suiface, see Figure 7 (Finnit; and Heine, 1974). The electrons at the surface tend to smooth the initial boundaries of the Wi~ner-~eitz cell (A
d
22 I
Free SurJlzce Structure and Properties c
c w
together to push the A1 atoms out of the bulk. Surfaces with lower atom densities such as (1 10) surfaces in fccbased structures in general show larger r ~ 1 a x ~ t i ~ ~ than do close-packed (1 11) surfaces or (I 00) surfaces, which often have similar surface energies.
ce
Figure Schematic af electron smootliing 3 s a meclianism for an iiiward surFxe relaxation (from Lath, 1993)
CC indicated in Figure 7. The smoothing out of the charge density gives rise to the f o ~ a t i o nof a surface dipole layer. The positive ions follow this displacement as the result of electrostatic forces. Similar relaxations were found in several other compounds, as for example Cu3Au. Figure 8 shows a cross-sectional view of the relaxations near the (1 10) surface in Cu3Au, with the mixed 50% Cu 50% Au surface t e ~ i n a t i o n Au . atoms are moved outward in the top layer. Wallace and Ackland (1992) compared the relaxations in Ni3A1 and Cu3Au. In both cases the minority element, which has also the larger atomic radius is displaced outward with respect to the smaller atoms. They interpreted this result in terms of the atomic radius and interatomic potential. In Ni3Al Ni t i the o i imany-body term provides a. larger ~ o ~ t ~ i b ~to in the interatomic potentials. Therefore it is expected to reside closer to the bulk, where it might be expected to stabili~ethe surface more effectively. In contrast, Au provides a larger contribution to the interatomic potential in CusAu. In Cu3Au, atomic size and potential oppose one another, whereas in Ni3Al the effects of atomic size and potential work
A F
Layer 2
-
8 ~ r ~ ~view ~of the ~ relaxation ~ ~ of ~atomst in i the topmost layers of the Cu,Au (110) surface. Filled circles represent Au. All relaxations shown are highly exaggerated (from Wallace and Ackland, 19921, reproduced with per~ission ~~~~~~
Many bulk materials show phase transitions between different crystallographic structures. One might expect the surface to show the transition before the bulk material does because of decreased constraint and faster diffusion. However, in pure CO, for example, which undergoes a transition at 723 K from the hightemperature hcp structure to the low-temperature fcc structure, no deviation was found for the surface compared to the bulk. In contrast, the phenomenon of surface premelting has been found on pure metal surfaces, too. At least on some Pb surfaces premelting of a. few outermost surface layers was observed 100 K below the bulk melting temperature. In long-range ordered intermetallics a disordering phenomenon can occur at the surface before the bulk disorders. Phase transitions on alloy surfaces are often related to segregation; for a theoretical discussion based on the lattice gas model see (Teraoka, 1994). (See also chapters by Diniitrov in Vol. 1 on ordering and disordering processes and by Saboungi and Price in this volume on dynamic disorder.) 5.1
~~~~~~~
~ ~ f f s e
As in the case of bulk materials, surface phases are stable over a limited range of thermod~namicvariables such as temperature, coverage (in the case of adsorbates) or gas pressure. The range of tl~erniodynai~ic variables over whch a phase can exist at t h e ~ ~ o dynamic e q u i l i ~ r i ~citn i ~ be described in a phase diagram for the surface. The maxiinurn number of phases p that can coexist is given by Gibb’s phase rule. A useful classifica~ionof phase transit~onsis based on thermodynamic properties. Transitions are called of first order, if any of the first derivatives of Gibb’s free energy G is discontinuous. First-order transitions have discontinuous changes in V , S, chemical potential ,ugof coinponent i, aiid the surface energy ~ Since n the ~ e ~~ t ~ ~ Hl pisyalso ~ i ~ ~ o at~the~ ~ ~ 1). transition, a latent heat of transformation results in a transformation hysteresis. Every region o f the surface i s distinctly in one phase or the other. One example for
222
Properties and ~ ~ e ~ o ~ z e ~ ~ l o ~ y
such a transitio~is the well-st~diedfirst-order (1 x 1> to (7 x 7) transition on the Si( 1 11) surface (Osakabe et al., 1980). ~oexistingregions of (1 x 1) and (7 x 7) reconstructions are found near the transition temperature. In second-order phase transitions the first derivative of G changes continuously, but not the second-order derivatives. Thus, they have no latent heat of transformation and should not show any hysteretic behavior. The way in which ordering changes in the vicinity of a second-order surface transition can be described by a set of numbers called the critical exponents a, p, y, and v. These numbers do not depend on the actual interatomic potentials, rather they are based only on the symmetry properties of the surface. Rules for these critical exponents were developed by Landau and Lifshitz (1980) see also ottman (1981). Due to the variety of symmetries that can occur, there exist several sets of critkal exponents called universal it^ classes (Roelofs, 1996). Every second-order surface transition belongs to one universality class, through which its beliavior near the transition is completely determined. Each critical exponent predicts the variation of specific physical quantities near the critical transition temperature Tc.The te~peraturedependence of the long-range order parameter 11 for example is described by the critical exponent p.
The exponents y and v can be determined from accurate measurements of tlie diffraction-spot profiles as functions of temperature near the transition. Excellent agreement between theory and experiment is found for the surface phase transitions in pure metals (Unertl, 1993; Howe, 1997). For the (2 x 1) --$ ( I x 1> surface phase transition on the (110) surface in pure Au, theory predicts values for the critical exponent p, y, and v of 1/8, 7/4 and 1, respectively. From e~perimentalinvestigat~onsa value for /? of 0.13 wits found (Campuzano et al., 1985). Figure 9 shows the results of an investigation of the surface phase transitio~on the CusAu (100) surface in. comparison with the temperature dependence of the LRO parameter in the bulk. The LWO parameter y of the surface appears to be a continuous function of temp~rature; i.e. a second-order transition occurs instead of the first-order transitioii in the bulk, and the temperature dependence for the phase transition
Figure 9 Temperature dependence of the long-range order parameter for the bulk (triai~gles)(Feder et al., 1958) and surface (circles) (McRae and Malic, 1984) for the Cu3Au(100) surface. The dashed h e i s a fit to the data with ,O =0.23
can be described with equation (2). McRae and Malic (1984) conducted a detailed investigation on the (100) Cu3Au surface with LEED. They considered the exponential temperature dependence of the DebyeWaller factor for the beam intensities and obtained a critical exponent p of 0.23 and a surface phase transition temperature equal within error to the corresponding bulk transition t e ~ p ~ r a t u ifr emultiple scattering is neglected. The order-disorder transition at tlie surface was observed to occur reversibly; in short, none of the hysteresis was observed as would be expected for a discontinuou~transition. everth he less^ they also found weak indications for a partly discontinuous transition. A continuous order-disorder transition at the surface is equivalent to a wettingphase phenomenon at the (100) surface. The surface disorders prior to the bulk. The Cu3Au (100) surface therefore is an example where surface-induced disordering occurs. Though good agreement IS found in the literature on the chemical composition in the surface layers, i.e. 5Q%Au, 50%Cu in the topmost layer and pure Cu in the second layer, some discrepancies exist between different investigations on the nature of the surface t ~ a n s f o r m a t i ~These ~. dis~repa~ciesprobab~y are related to deviations from the stoichiometric composition. Rivers et al. (1995) studied in detail the orderdisorder transition in 3 at. % Au-rich Cu3Au crystals. In contrast to other investigations the order-disorder transformation at the surface occurs at a temperature 20 K above the bulk transition te~perature(644 K) of
Free Surface Structure and Properties this Au-rich composition, No changes in the surface composition were observed with Auger studies near the transition. Since the transition temperature is very close to that of stoichiometric CuliAu, Rivers et al. suggested that a surface layer of stoichiometric C U ~ A U is formed, which led to the increased surface transition temperature. On the (1 10) surface of C U ~ A the U surface transition froni the ordered (2 x 1) reconstructioii to the disordered (1 x 1') reconstruction occured at the same tem~eratureas in the bulk. In contrast to the (100) surface the transition was identified as first-order. McRae et al. (1990) found a significant hysteresis iii the long~range order parameter between cooling and heating cycles.
As well as disordering effects at surfaces of ordered crystals, the inverse: ordering at the free surface of alloys that are essentially disordered in bulk, can also occur. This effect was investigated in Ni-Pt alloys that are disordered in the bulk. Ni-Pt and Co-Pt are systems with phase diagrams similar to that of Au-Cu alloys. Ordered l;lz(cP4), Ni3Pt and CoPt3, and Llo(tP4), NiPt and CoPt, phases appear at lower temperatures. At higher teinperatures Ni and Pt form a coiitinuous series of solid solutions. Ni-Pt alloys gained special interest because of their use as catalytic materials. The surface structure, therefore, is of high interest for these materials. Pt-metal alloys have been extensively studied in the disordered state, see the review by Gauthier (1996). Single crystals of Ni9oPtlo showed an oscillatory Pt enrichment on (100) surfaces (Gauthier et al., 1990). Pt is enriched to a concentration of 24.3% in the topmost layer. The second layer is depleted in Pt to 6.4%. The LEED investigation showed no reconstruction of the surface. A siinilar oscillatory behavior was also found for the interlayer spacings. The topmost layer is expanded from the bulk, the second layer contracted, and the third again expanded. It is interesting to note that this behavior is in contrast to that of pure Ni and most other fcc metals, where a contraction of the topmost layer is found. However, it may be explained by the larger Pt atoms (see Table 1) segregating to the surface. STM investigations on Ni,Ptl-, (100) surfaces confirmed the Pt eiirichrnent of the surface for all investigated compositions (Hebenstreit et al., 1997). The STM study also revealed reconstructions on some surfaces. Strong Pt enrichment was also found on (1 11) surfaces, with a damped oscillatory composition profile. In
223
other words, each of these alloys developed a Pt-Mi sandwich-like structure with Pt on top. But the scgregatioii effect depends on the surface orieiitation and is reversed on (110) surfaces. The first ST investigation of the Ni~sPt25 (111) surface was conducted by Schmid et al. (1993) where the achieved chemical contrast was used to study ordering of Ni and Pt in the surface. Preferential sp~tteringand segregation cause an enrichment of Pt at the surface. The STM images show a short-range ordering eEect on the surface, Chains of up to seven Pt atoms appear in the STM image. Figure 10 shows the positions of atoiiis on the surface calculated with a Monte Carlo method, which results in positions very similar to those in the STM image. This structure is siinilar to an L1o ordered (111) surface, where alternating rows of Pt and Ni atoms are expected. Pt enrichment at the surface causes the formation of this ordering phenomenon at the surfice. The bulk with a 25% Pi composition cannot exhibit this structure. Therefore, a surfaceinduced ordering phenomenon was observed. Surhce-induced short-range ordering was also discovered on (1 11) and (1 10) surfaces of Co75Pt25 alloys, which are expected to behave very similar to Ni-Pt alloys (Gauthier, 1996 and Gautliier et al., 1998). An oscillatory segregation profile is observed, with Pt enrichment in the topmost layer. The surface slab
Figure 10 Atomic arrangement on the (1 1 1) surface of NiPt at 420 K calculated by Monte Carlo siinulations with embedded atom potentials (from Schmid et al., 1993)
stabilizes with a structure and composition quite similar to that of the L12 Co#t phase. Surfaceinduced ordering p~enomenawere also found in the Ag-Cu system, which presents a strong phase separation in the bulk (Meunier et al., 1999).
The composition of a free surface and layers next to the surfiice will generally differ subs~dntiallyfrom the composition in the bulk. Segregation in alloys has been investigated with a large effort. Farkas (1995) reviewed some basic s~gregationmodels applicable to segregation in solid solutions and ordered intermetallics. The ordering properties of intermetallics hinder the diffusion in these alloys. ~ ~ v e r t h e ~segregation ~ss, is also found in ordered compounds, which will be exclusively treated in the following.
Usually, surface segregation is inhibited by strong ordering forces in intermetallic compounds and consequently the surfaces o f these alloys are expected to be bulk terminated. ~~vertheless,in many weakly ordered compounds and especially those where ordering occurs only at lower temperatures, the surface com~ositionis changed largely by segregation. Some
alloys forming solid solutions in the bulls, where ordering effects at the surface are observed, were discussed in section 5.1 and 5.2. Segregation is also found in compounds with an ordered bulk (Heinz and Hammer, 1999). For example, segregation is important in Fe-A1 alloys of various chemical c o ~ ~ o s i t i o n s . Stoichiometric FeAl 2 compounds show strong segregation behavior, very different from that of NiAl, which has a much larger heat of formation. Hammer et aE. (1998) observed segreg~tiono f A1 to all low-index surfaces of FeAl. An in~oi~mensurate surface alloy of FeA12 was found on the (110) surface. Figure 11 shows segregation curves for several surfaces of FeAl. In all cases A1 segregates to the surface. Although the A1 segregation tendency seems to be more pronounced on open surfaces, the segregation sequence (111) > (310) > (100) > (110) can only be understood by considering the complex structure of these surfaces. An e ~ p ~ ~ n a tof i othe n d i ~ ~ r esnet ~ r ~ ~ a t~i o~nh ~ v i o r can be given based on the heat of formation in the broken-bond model (Sachtler, 1984). The broken-bond model predicts segregation of the element that has the lower heat of vaporization, since the heat of vaporization reflects the bond strength. The energy increase of two cleaved surfttces is minimized, if the surface is enriched by the element with the lower bond strength. In most Al-based B2 compounds A1 has the lower heat of vaporization and is expected to be the segregat~n~ element. In contrast the heats of formation of these
igure 11 Segregation curves obtained after quenching from high temperature for low index surfaces in FeA1. The different compositions of freshly sputtered surfaces at low temperatures are caused by the sputtering process. The near-surface concentration corresponding to bulk terminated surfaces is marked by open circles (froni ~ a ~et al., ~ 1998) e r
225
Free Surfizce Struct~reand Properties Surface reconstr~ctionsand term~nationsfor some B2 ordered compound§ Compound NiAl MiAl NiAl FeAl FeAI CoAl FeTi FeTi
Surface
Heat of formation
Reconstruction, termination
References
-58.9 kJ/mol
(I x l), A1 (1 x I), 5O%Ni 50%A1 ( l x l ) , Ni (1x1) A1 (3 x 3) A1 saturated, rich (I x l), 50% CO 50% A1 Fe-rich Ti-rich
Davis and Noonan, 1988 Davis and Noonan, 1987 and 1988 Nooiiaii and Davis, 1988 Graupner et al., 1995 Graupner et al., 1995 Blum et al., 1996'0 Felter et al., 1982 Felter et al., 1982
-25.1 kJ/mol -54.2 kJ/mol
compounds are rather different. For NiAl a heat of formation of 58.9 kJ/mol was reported (Table 2) which is considerably higher than that for FeAl with 25.1 kJ/ mol. Therefore, the driving force to lower the surface energy by segregation i s much stronger in FeAl than in NiA1. CoAl has a heat o f formation comparable to that of NiAl, a fact which is an excellent fit to the observation that similar surface structures are found on CoAl as on NiA1. A1 segregation to the surface was also proven to occur in the compound Fe3AI with the DO3 (cF16) structure (Voges et al., 1992). This occurs well below the bulk order-disorder transition temperature of 800 K. Tn this compound the heat of formation of -13 kJ/mol is even lower than in FeAl, which favors the segregation behavior. SiO
Surface diffusion along craGks or cavities can play the role of a short circuit for bulk diffusion and therefore can influence the high-temperature properties of inte~etallics. A comparison of free-surface and grain-boundary diffusion values gives a better understanding o f the diffusion mechanism. Surface diffusion in general is influenced sensitively by the environment and surface structure. Furthermore, surface diffusion is highly anisotropic. Thus multiple experiments are necessary to gain a thorough understanding of the basic mechaiiisin in intermetallic conipounds. Surface diffusion is studied from a microscopic atomic view with FIM and STN. However, the interaction between the STM tip or the conditions at the sharp tip in the FIM may influence the ~ e a s u r e ~ e n t s~acroscopic . investigations of surface diffusion are performed for example with gra~n-boundarygrooving experinients, which are widely applied to determine grain-bou~da~y energies and surface-diffusion coeEcients. The application of a n atomic force microscope (AFM), instead of convent~onallyused light microscopy allows analysis
of the profiles of rain-boundary grooves or surface undulations with much higher accuracy (Jin et al., 1999; Weber et al., 2000). Recently this technique has been applied to study the grain-bou~darygrooves in NiAl bicrystals (Rabkin et al., ~000). more work lias yet to be performed to study the surface diffusion properties of inte~etallics.
Interfaces and free surfaces have an important influence on the properties of intermetallic cornpounds, as for example m e c ~ a n ~ cproperties. ~l Corrosion and catalytic behavior are obviously dominated by the surface structure and composition of intermetallics. Segregation to free surfaces and interfaces can have a strong influence on the fracture behavior, for example in boron-doped NiJAl, where boron enrichment at the grain boundaries improves the fracture toughness considerably. S SS
The energy of a surface is closely related to the structure of the surf'ace. Surfaces tend to lower their surface energy by reconstructions and relaxations of the surface. Foiles (1987) computed the energies of (1 x 2) reconstructed and relaxed surfwes of many fcc metals using the embedded~atommethod and obtained the largest reduction in surface energy for Au and Pt, which show a (1 x 2) reconstruction. Nevertheless, reliable e x ~ e r i ~ e n t values al for surface energies are available only for a few simple surfaces (Porter and Easterling, 1992; Howe, 1997). Methods for surface-energy measuremen described for example in the book by (1973). Calculations of surface energies are widely made using known interatomic potentials. The surface energies of simple low-index surfaces can also be
calculated with current ab-iEitio calculations. A simple model to describe the surface energy for high-index s~irfacesis to a s s u ~ ea stepped surface ~onsistingof low-index facets. The energy of complex high index surfaces is then calculated as the SUM of all surface facets. Mutasa and Farkas (1998) recently compared simulated surface energies with the results from this simple model and showed that this method gives reasonable results for the compounds NiAl, FeAl and TiAl . For Cu3Au (loo), (1 10) and (1 11) surfxes Wallace and Ackland, 1992, computed surface energies of di~erentlytruncated and relaxed surfaces with FinnisSinclair many-body type potentials and molecular statics techniques. Their results for the surface energies are shown in Table 3. Clearly the mixed-composition Cu-Au surface terminations are energetically favored at zero temper~ture,in both the relaxed and unrelaxed states. These ca~culationscorrespond very well to the ex~erimentalob~ervationof a mixed Cu-Au termination. The close-packed (1 1I) surfaces have nearly the same energy as the open (100) surfaces. In pure fcc metals (111) surfaces would be expected to have substantially lower energies than (100) surfaces. In these calculations a buckled surface structure was also found for CusAu. The Au atoms with larger atomic radii are raised above the copper atoms on each of the three surfaces with the mixed termination. The importance of surface stresses recently became widely appreciated. For metals the sureace stress, which acts within the lateral surface plane, is normally of tensile nature. The surface stress determines the strain ~ ~ ~ e n d ofe nthec surface ~ energy. The influe~ice of surface eiiergy and surface stress is, for example, important in the g r o ~ t h~ e c ha ni s mof thin films, where coalescence of islands occurs (Nix and Clemens, 1999). Many experimental studies have also proven that the lattice constant of small particles is reduced from the bulk lattice constant (Mays et al., 1968; Salomons et d., 1988) due to surface stress, Surface stress can also influence the reconstruction of surfaces, as demonstrate^ by the (23 x J3) reconstruction
observed on Au (111) surfaces (Harten P t al., 1985). Needs et al, (1991) calculated the surface stresses and energies for some fcc metals. The surface stress i s always larger than the surface energy. For example, for A1 a surface energy of 0.06 eV/atoni and a surface stress of 0.078 eV/atom were found. The origin of the surface stress can be described in a simple manner. A t o m at the surface have a lower coordiiiatioii number and tend to iiicrease the charge density around the atoms by an inward relaxation. Since there is an electrostatic repulsion of the atoms, these relaxations are not able to reduce the surface energy to zero. Thus, the surface atoms still sit at a density lower than in the bulk. Therefore, the atoms tend to increase their density to the optimal value through a tension in the surface. Even though few data about the surface stress in intermetallics are available, it is expected that the surface stress makes an important contribution to the properties of fine-scaled materials, where large surface areas are present. 7.2
The details of surface structure, composition and energy are very important for the role of surfaces in the fracture process. ~ e ~ e r m i n a t i oof~ sthe surface energy should lead to a better ~~nderst an~iofngthe fracture process and the prediction of the preferred cleavage planes and fracture mode. Figure 12 for example shows (110) cleavage facets in NiAl imaged with an. AFM. The cleavage fracture in 2 aluminides was reviewed, for example, by Chang et al. (1992). A substantially different fracture behavior was found for NiAl and FeAl. In NiAl (110) cleavage is preferred, but sometimes cleavage on higher-index surfaces has been. reported. In FeAI cleavage occurs on (100) planes. Chang et al. proposed that the difference in the fracture mode can be explained by a broken-bond model. In those B2 aluminides having a low ordering energy such as FeAl, fracture i s suggested to occur on (100) planes. In materials with high ordering energy
Table 3 Relaxed and unrelaxed absolute-zero surface energies of ordered Cu3Au(from Wallace and Ackland (1992) Surface
y, unrelaxed (mJ/m2)
y, relaxed (niJ/in2j
(I00j Cu-Au terminated
896 1192 1051 1240 882
865 1171 1024 1173 863
(100) pure Cu terminated (1 10) Cu-Au terminated (1 10j pure Cu terminated (1 1I j Cu,Au
Free Surface Structure and Properties
227
Figure 12 Fracture surface of a NiAl single crystals. (110) cleavage facets are visible in the AFM scan (from Goken, 1999), reproduced with permission
(for example NiAI) with a relatively large charge transfer from A1 to Ni atoms, the surface charge could be balanced only on (1 10) planes. Ah-initio calculations by Fu and Yoo (1992) shed more light on this point. Their calculations suggested that a strongly directional d-bonding in FeAl is responsible for the high cleavage energy of this compound. The different termination of surfaces discussed so far often has no influence on the ideal cleavage energy determined by simulations. Cleavage in NiAl on (100) planes, which may have either a Ni or A1 termination, would always produce one Ni-terminated and one Al-terminated surface. The cleavage energies must be identical. Calculated cleavage energies for several compounds can be found in Yoo and Fu (1992). They included in their analysis a calculation of the anisotropic fracture toughness and considered also the shear stress on possible slip systems. Thus the preferred (110) cleavage plane in NiAl could be understood. 7.3 Mechanical Properties near Surfaces and Interfaces
With current nanoindentation methods the mechanical properties of very thin surface layers can be investigated. Figure 13 shows an AFM image, where the properties across a grain boundary in a NiAl bicrystal
were determined with extremely small indents. Only the indents performed with a maximum load of 300 pN show a significant increase in hardness at the boundary. Similar measurements at the free surfaces of cracks in NiAl (Kempf et al., 1998) also revealed only very slight deviations of the mechanical properties from the bulk properties at distances above 100 nm from the surface. On the other hand grainboundary hardening effects by solute atoms were observed in several compounds, for example AgMg, NiGa and NiAl (Westbrook and Wood, 1963; Seybolt and Westbrook, 1964). With nanoindentations it is also possible to probe the surface properties in a direction normal to the free surface. Then, even thinner layers near the surface are probed. Several effects contribute to the observed variations of the properties of surfaces in the nanometer range. The well-known indentation-size effect (ISE) increases the observed hardness close to the surfacc by the emission of geometrically necessary dislocations during indenting. Thin films on substrates show a larger hardness or strength than bulk materials (Nix, 1997). Nanoindentations on nearly all metals and intermetallics, as for example Ni3Al TiAl, Ti3Al and MoSSiBz, show a significant onset of yielding (pop-in) in the load-displacement curves (Goken and Kempf, 2001). Dislocation sources are activated first at a stress level close to the
228
Properties and Pdzeizomenology
3 Hardness across a grain boundary iri NiAl probed with nanoin~entation~. The AFM image (above) shows the impressions left from the indentations (from GCSken et al., 1999), reproduced with permission
theoretical shear strength level (Goken et al., 2000). Surface steps and ledges probably are preferred spots for dislocation emission and multiplication (Kiely et al., 1998). These effects are currently under investigation by several groups. The elastic properties near the surf~ceor of thin films seem to match inore or less the properties of the bulk. However, a thorough underof the plastic deformation properties at the surface is still lacking.
n most low-index surfaces of intermetallic alloys a bulk-like termination of the surface is observed.
Reconstruction (rearrangement of atoms at the surface) is not observed an many intermetallic alloys. In contrast, relaxation of atoms in the outermost surface layer and beneath the surface is a rather common phenomenon. Nevertheless, even some pure fcc metals such as Au show a reconstruction of the surface structure, i.e. a non-bulk ~ e r m ~ ~ aexists t ~ o nat the free surface. In intermetal~icsthe ordering energy of the compound may be high, for example in the B2 compounds NiAl or COAL These compo~ndstend to show a bulk-terminated surface since the driving force for a reconstruction is too low. llrr FeAl with a much lower heat of formation, Al. s e ~ r e ~ a t i oton the surface was observed and even reconstructions were reported on sonie surfaces. Ordered compounds can have different possible ~erminations,for example pure Alor pure N~-teri~inat~ons on the NiAl (100) surface. A1 terminations seem to be preferred, which may be understood by the relatively low heat of vaporization for Al. A rippled surface relaxation behavior was found on NiAl and other compounds with the larger A1 atonis moved outward from the bulk. In L12 compounds, for example Ni3Al and CugAu, the mixed surface termination seems to be preferred. A buckling of the ininority element A1 or Au out of the bulk could be predicted from a simple atomic-size argument. Nevertheless, the many-body term in the atomic potentials has a significant effect, which inay either reduce the surface buckling or enhance it. Surface phase transitions were i~vestigatedin detail experimentally and theoretically on. CU~AU, where a discontinuous bulk order-disorder transition occurs. Interestingly, the order-disorder transition in bulk is of first-order whereas the transition at the surface is of second-order. The ordering effects are related to segregation properties (Sanchea and Mora~-Lopez, 1985). Segregation can also influence the surface structure and inay induce a surface reconstruction, as observed in the Fe-A1 system. ~ e g r e g ~ t i oeffects n play a minor role in. conipounds with high ordering energy, for example NiAl. The structure of a free surface determines the properties at the surface. Surface energies were ~alculatedfor several compounds in the relaxed and unrelaxed states. Surface energy and surface stress are important quantities with regard to fracture and ~ ~ c r o s t r u c t u ~The e. anisotropy of surface energy in the fcc-based compound CugAu is significantly different from that in pure fcc metals. Ab-inim calculations of cleavage energies predict the fracture mode successfully. Although simulations and experiments are in relatively good agreement in determining and predicting the surface structure and
Free Surfbnce Structw'e and Pmperties relaxations, inore work has to be done on their influence on surface properties. Few iiivestigations in intermetallics are related to the properties at the surface, for example surface diRusion aiid tlze mechanical properties.
Allen, R. E., and de Wette, F. W. (1969). Phys. Rev., 179, 873. Atrei, A., Bardi, U., Roviadi, G., Torrini, M., and Zanazn, E. (1992). Phys. Rev. B, 46, 1649. Aust, K. T., Peat, A. J., and Westbrook, J. H. (1966). Acta Aust, K. T., and Westbrook, J. W. (1971). Act0 Me~all., 521. Binnig, G., and Rohrer, H. (1982). Helv. Phys. Acta, 55, 726. Binnig, G., Rohrer, H., Gerber, C., and Weibel, E. (1983). Sur$ Sci. Let{., 131, L379. Birgeneau, R. J., and Horn, P. M. (1986). Scrence, Blakely, J. M. (I 973). Introduction to the Properties of Crystal SugfiJces. Pergamon, Oxford. ehreiidt, D., and Niehus, H. (1996a). Surf:
., Castro, G. R., Kottcke, M., Hammer, L., and Helm, K. (X996b). Sio$ Rev. LPtt., 3, 1409. g Microscopy and Bonnell, D. A. (1993). S c a n ~ ~ nTunneiing S'ectroscopy. VCH Publishers, New York. Briggs, D., and Seah, M. P. (1983). Practical Sugjface Analysis by Auger and X-ray Photo-Electron Syectroscopy. Wiley, Chichester, e yIf., , and Marchut L, (1983). Phys. Buck, T. M., ~ ~ e a t ~6. Rev. Lett., 51, 43, Campuzaiic), J. C., Foster, M. S., Jennings, G., Willis, R. F., and Unertl, W. (1985). Plzys. Rev. Lett., 5 anid Lipsitt, H. A. (1992). Act0
Chen, W., Paul, J., Barbieri, A., VanHove, M. A., Camerso S., and Dwyer, D. J. (1993). J Phys.: Condens. Matter., 4585. Cox, G., Szynka, D., Poppe, U., Graf, E(. H., Urban, K., Kiesielowski-Kem~erich, C., Kruger, J., and Alexaiider, H. (1990). Phys. Rev. Left., 62, 2402. Davis, H. L., and Noonan, J. R. (1987). Phys. Rev. Lett., 5 566. Davis, ET. L., and Noonan, J. R. (1988). In The St/~uc~z~re qf Szirfaces II (eds J. F. van der Veen, and M. A. Van Hove). Heidelberg, Springer-Verlag, p. 152. Daw, M. S., and Baskes, M. I. (1984). Phys. Rev. B, D i ~ a s i E., , T ~ s t m a ~ H., n , Shpyrko, 0. G., Deutsch, M., Persbau, P. S., aiid Ocko, R. M. (2000). J. Phys,. Condeeris. ~ a t t e r 12, , ~209. Farkas, D. (1995). In ~ n t e r ~ ~ e t a ~corripomds: lic Vol. I , Przncip~e~~ (eds J. H. Westbrook, and R. L. Fleisclier). Wiley, Chichester, p. 609. Farkas, D., and Ran, A. (1986). Pkiys. Status Solidi, 93, 1.
229
Feder, R., Mooney, M., and Nowick, A. S. (1958). Acra Met., 6, 266. Felter, T., Steward, S., and Uribe, I?. (1982). $U$ Sci., 1 69. Finnis, M. W., and Heine, V. (1974). J. Phys. Fiiiiiis, M. W., and Sinclair, J. E. (1984). Phil. Foiles, S. M. (1987). Surf. Sci. Lett., 191, L779. Fu, C . L., and Yoo, M. H. (1992). Acta Met 703. Gauthier, Y. (1996). Surf. Rev. Lett., Gautliier, V., HoEman, W., and Wutt 233, 239. GautIkr, Y., Dolle, P., ~ ~ ~ U d o i n ~ ~ SR., a vHebenstreit, ~~is, W., Platzgummer, E., Schmid, M., and Varga, P. (1998). Surf: Sci., 396, 137. Goken, M. (1999). Adv. Mater. Process., 1 Goken, M., and Kempf', M. (2001). Z. Me Goken, M., Kempf, M., Bordenet, M., aiid Vehoff, EL (1999). Surf. InmfQce Anal., 27, 302. Golceii, M., Sakid-ia, R., Nix, W. D., aiid Perepezko, 9. 13. (2000). Mat. Sci. & Eng. A , in press.
6.W., and Zehner, D. M. (1998). Surf. SCE,, Hasten, U., Laliee, A. M., Toennies, J. P., an (1985). Phys. Rev. Lett., 54, 2619. Neatherly, Jr, L., George, E. P., Liu, G. T., and Yamaguchi, M. (1997). I n t e r ~ ~ ~ a5,~281. ~~cs, Hebenstreit, E. L. D., Heben ., S c ~ ~ i M., d , and Varga, P. (1999). Surf. S C l . iederniann, A., and Hebenstreit, W., Ritz, G., Sch Varga, P. (1997). Surf. Sci., Ifeinz, K. and Hammer, L. (199 7s.. Condens. Matter, 11, 8377, Mowe, J. M. (1997). 1nterfaces ipt Materials. Wiley, New York. Jin, M., Shimada, E., and Ikuma, U.(1999). J, Mat. Res., 2448. Kempf, M., Goken, M., and Vehoff, N.(1998). Appl. Phvs., ~ 6 6S843. , Kiely, J. D., Hwang, R. Q., and Houston, J. E. (1998). Phys. Rev. Lett., 81, 4424. Krainer, P., Papadopolos, Z., and Teuscher, H. (1999). J Phys: Condens. Matfer., 11, 2729 Krumniacher, S . , Sen, N., Gudat, W., and Ghijsen, J. (1989). 2. Phys. 13, Landau, L. D., and Lifshitz, E. M. (19 Part 1. Pergamon Press, Oxford, p. 459. Liith, W. (1993). Surfaces and Iizterfaccs of Solids. SpringerVerlag, Berlin. Mays, C. W., Vermaak, J. S., and uhl~ann-Wi~sdorf, D. (1968). Su$ Sci., 12, 134. McRae, E. G., and Malic, R. A. (1984). Surf. Sci.,1 McRae, E. G., Buck, T. M., Malic, R. A. Sanchez, J. M. (1990). Surj: Sei. Lc?t Treglia, G., and Legrand,
., Niehus, H., and Comsa, G. (1995). SurJ Mullins, D. R., and Overbury, S. H. (1988). Szryf: Sci., 199. and Farkas, 6). (1998). SurJ Sci ,~ ~ d f r eM. y , J., and M a n s ~ e ~ ~ ,
Niehus, H. (1995). Phys. Stut. Sol. ( k ) , 192, 357.
Slim, Z., Raberg, W., Heinzig, M., Jenks, C . J., Fournee, V., Van Hove, M. A., Lograsso, T. A., Delaney, D., Cai, T., Canfield, P. C., Fishcr, I. R., Colalinann, A. I., Kramer, M, J., and Thiel, P. A. (2000). SwJ Sci,, Sinharoy, S., brag in ski^ A., Talvacchio, J., (1986). Surj: Sci., 16??401. Sondericker, D., Jona, F., (1985). Solid State CO Sondericker, D., Jona, E'. Rev. B, 33, 900. Sondericker, D., Jona, F., and Marcus, P. M . (1986b). Phys. Sondericker, D., Jona, F., and Marcus, P. M. (1986~).Phys. Rev. B, 34, 6770. rd, E., and Besenbacher, F, (1996).
3447. Noonan, 3. K..and Davis, H. L. (1988). J. Vac. Scl. Technol., ., and Honjo, G. (1980). Jpn. J. Appl.
oki, M. (1991). Phil. Tram. Roy. Soc.,
~aster~ing,K, E. (1992). Phase TraFwfori%ati(min Metub and Alloys. Chapinan & Hall, London. Potter, II.C., and Blaltely. J. M. (1975). J. Vac. Sci. Techml., 12, 635. , L., and Semenov, V. (2000). Acta W. N,, Hung, H. H., and Liang, K. S, (1995). Phys. Rm. B, 52, 12601. 6). In ~ ~ a ~ qfd Sucface ~ o o Science, ~ Vol. I , Physicnl Structure (ed. W. N. Unertl). Elsevier, am, p. 713. . (1981). Phys. Rev., B24, 1482.
anchez, J. M., and Morian-Lopez, J. L. (1985). S u r ~Sci., chick, 6 . (19x1). Prog. SurJ Sci., 11, 245. chiid, M., Stactlcr, H., and Varga, P. (1993). Phys. Rev. Lett., 70, 1441. SchuPze, C. E. R. (1974). Metallphysik (2.Aufl.). SpringerVerlag, Wien. Sepbolt, A. II.,and Westbrook, J. H. (1964). Acta ~ e ~ a l 1l . , 449.
Sundaram, V. S., Farrell, B,, Alberi, R. S., and Robertson, W. D. (1973). Plzys. Rev. Le& 31, 1136. Sundaram, V. S., and Robertson, W. D. (1976). SurJ: Scz., 55, 324. Teraoka, Y (1994). Surj. Rev. Lett., 3, 1791, e ~ , Tromp, R. M. (1989). J. Phys. Cond. ~ a ~ 1, ~10211. Uiiertl, W, (1993). In S c ~ n ~n ~u ~~ ~n ~e i ~ ~~o s~ c goand py Spertr(1scop.y (ed. D. A. Bonnell). VCH PLzblisliers, New York. Varga, P., and Schmid, M. (1999). Appl. Surf: Sci., 141, 287. Van Hove, M. A. (1993). In Muterials Sczencc and Technology, Vol. 1, Struct~reof' Solids (ed. V. Gerold). VCH Weinheim. Van Hove, M. A. (1999). Surf erfac face Anal., Van Hove, M. A yWeiberg, vri. H., and Chaa, Low-~nerg3?-ElectroPz-~i~r~~tion. Sprin~er-Verlag,Berlin, Vasiliev, M. A. (1997). J . Phys. D: Appl. Phys., 30, 3037. Voges, D., Taglauer, E., Dosch, H., and Peisl, J. (1992). Swf Sci., 2691270, 1142. and Ackland, 6 . J. (1992). Surf: Sci. Lett.,
x, M., Giiken, M., and Vehoff, H. (2000). Procee~~ngs~ ~ '99, Vol. ~ 4: M ~ O c ~ o s ~ r ~~~ t u r a l A ~nve'stigat~onand ~ n ~ l y s i(eds. s B. JouErey, and J, Svejcar). Wiley-VCH Weinheim, pp. 208-2 13. Westbrook, J. H., and Wood, D. L. (1963). J . Inst. Met., 91, 174. Wood, E. A. (1964). J . AppL Plzys., 3 Yoo, M. H., and hi,C. L. (1992). Mater. Sci. Erg., 470.
Color of a solid material i s usually encountered among dielectrics, semiconductors and insulators but not for ~ e ~with ~ lthes exception of old, silver and copper. But other colored ‘metals, do exist as illustrated by Figure 1 (see color plate section). Current definitions of an intermetallic compound are ~ m b i ~ L i oNesper ~s. (199 1) sug~estsd e ~ n ~ ninterg metallic phases as having two attributes, (a) being compounds of metallic constituents and (b) having metallic conductivity. This definition would exclude for example Go%:!, a metallic ~ ~ a t e r i aofl beautiful dark blue color. It is therefore preferablc to follow the ~ntroductorydefinition of Westbrook and Fleischer (1995) by considering in ad~ition to metal-metal coinpounds also metal-metalloid conipouiids which include silicides, for example. In many cases, these inter~etallic compounds imply a specific stoichiometric ratio between the components and may exhibit long- or short-range order. They might also be binary or i~ulticomponent;the decisive point is the metallic character of their electronic conductivity, which means that valence and conduction bands of the electronic structure overlap, at least in part. Their usually large electrical conductivity originates from rather mobile electronic states. For the nietallurgist, such c h a r ~ ~ ~ e r i s t iare c s typical for single element metals or disordered alloys which melt at a specific teniperature (elemental metals) or over a range of t e ~ ~ e r a t u r and e s compositions (solid solution alloys); however, interi~etal~ic c o ~ p o u n d in. s most cases solidify
congruently at a well-defined teniperature arid composition, and have in additioii a rather na~rowdomain of existence. Early applications of i i i t e ~ e t a ~ l iphases, c such ics mirrors of bronze, coatings for the d e c o r ~ t ~arts, v~ amalgam in dental medicine, and more recent uses as magnetic materials, superconductors, major constituents of superalloys for h i ~ ~ - t e ~ p e r aap~lications, tur~ and so forth, are well known (see e. 1977; SauthoR, 1995). owever, there is another, much less known, but distinctly visible, property of a certain class of ~ntermetallic ~ o ~ ~ o n~a ~ne l~y § , their apparent color. About 100 ~ ~ ~ ~ p oau ~ i~ d s o the 30 000 substances tabulated in Pearson’s ~ u ~ d (Villass and Calve blue NiAl, yellow dark blue CoSi:! Randin, 1994; ~teinemannet aE,, 1997). in ~hapterby Wolff on precious metal ~ntermet~ll~cs this volume.) These materials have metallic luster due to the strong coupling of photons in the range of visible light to c o ~ ~ u c t i ne lge c t r o ~ ~states, c and due to the absence of polarization of the reflected light. appearance and properties of such materials are ~~~c distinctly different from those of d i e l e ~ media. Colored i n t e ~ e t a l ~ ~might c s find a ~ ~ l i c ~ t i for o~is watches and jewelery if they could be properly molded and shaped. For this purpose the i n t e r ~ ~ t a l lniaterial ic must also possess suitable ~ecliaiiicalproperties, in particular they must be sufficiently ductile. For the engineer and jewelery maker it is necessa~yto know related data o f such a material. ~ ~ e r e ~ oatr tlie e , end of
i n ~ ~ r ~C ~ ot a~ l ~~ Vol. ~p ~3, ~ Priumpbs ~ ~ c and ~ Practice. ~ ~ Edited hy J. €3. Westbrook and R. L. Fleischer. @ZOO2 John Wiley & Sons, Ltd.
232
Properties. and P h e ~ o ~ e ~ ~ l ~ g y
our treatise we provide a compact section for applicaecause of the attractive properties of CoSi2 (concerning color as well as the standard mechanical properties), we have chosen this compound as an example, providing partially new calculated data for its electronic structure and optical properties, as well as elastic data and information about cleavage, com~ressionand ductility. Thc present chapter deals with the origin o f color in intermetallics and its relation to the electronic properties, bonding, and the crystal structure of colored intermetallic compounds. For a detailed discussion of many of the physical aspects of color of materials we refer the reader to the most recent book in this field (Tilley, 2000).
lee, 1956, 1921/1922), the great painter, describes color as ‘Gold is a fluctuation between deep te color o f extreme brightness, a er vibrates from dark to very bright and is also defined as a nioving value. Copper is a vibrato from red-orange to over-brightness. These metallic values of color are subtle pictorial agents’. Color, as an impression, is not easily accessible to an objective me~~surement, Most of the attributes used to describe the pressio ion of color such as brilliance, ness, saturation, Luster (see the appended ssary) cannot easily be cast into a definition of objective physical quantities, because they depend strongly on the physiology of the human eye and the stimulation to signal processing by the central nervous system. In fact, the human eye is more sensitive for green colors than for red and blue tints, and the a particular color varies with light wever, it is admitted that the spectral reflection power of a material is the reference and that a preferential reflection is perceived as a distinct color. From a physicist’s point of view, color of intermetallic compounds (which are not transparent to visible light) is connected with the reflectivity of the material. The relation between the perception of color and the p h ~ ~ ~ c a defined lly and measurable optical reflectivity is illustrated by the following examples of colored ‘metals’. For practical purposes, in particular for the jewelery maker, a color scheme was developed by the International Commission on Illumination (CIELAB), namely the so-called L*a*b* scheme, which has the
advantage of describing color in a mathematical way by the three coordinates meiztioned: L* stands far luniinance or brightness, a* and b* for the i~itensities of the color pairs red-green and blue-yellow. All these quantities can be measured by a spectrophotoxneter, which normally has a resol v~n~ power 5 to 10 times greater than the human eye. A more detailed description is found in the excellent article of Cretu and van der Lingen (1999) on colored gold alloys. Figure 2 shows reslnlts of optical reflectivity measurements for gold and the intermetallic conipound PdIn. In the infrared and longer wavelength range of visible light, gold reflects more than 95% of the e l o the ~ wavelength /z = 650 nm or at than 1.9 eV, the reflectivity falls off” rapidly with decreasing wavelength^ The yellow color of gold results from its low r~flectivity,or strong absorption of light for photon energies above 2.4 eV. In the case of Pdln, the absorption edge is shifted to lower energies, and the color of the compound then appears as purplish pink (Jan and Vishnubhatla, 1967; Cho, 1970). The corresponding ClELAB values are L* = 88, a* = + 5 , b* = +35 for Au, and L* =70, a* = 3-12, b* = +6 for PdIn. Other data are shown in Figure 3 representing reflectivities for Ni and the ter~etallicdisilicides NiSi2 and CoSiz; the CIELA values for CoSiz are L* == 66, a* = -2, b* = -7. In the case of Ni, the reflectivity decreases 1110x10tonically with increasing photon energy; thus this metal appears to be of a silvery white color. The curves for the silicides differ markedly because the reflectivity has a steep decrease below the visible range of the spectrum, followed by a peak at higher energies. The compound Nisi2 shows a niaximum at about 3 eV and an average reflectivity over the visible range of about 45% that gives this metal a bluish-gray hue. For CoSi2, tlic peak of the reflectivity is more pronounced and lies at lower energy. Also, the average reflectivity is only 35%’ therefore CoSiz appears to be of a dark blue color. y The optical properties of the H u i ~ e - ~ o t h e r€32 (cP2) aluminides NiAl, COAL and FeAl have been studied intensively during recent decades (Sambongi et al., 1966; Jacobi and Stahl, 1968; Schlemper and Thomas, 1994). The pronounced difference in color between blue NiAl and yellow CoAl is clearly represented by a distinct peak of‘ the reflectivity at 2.6eV for NiAl and for CoAl a shoulder of the reflectivity at lower energies which smoothly drops OR to a ininimuni at 3 eV. Special interest was attracted by the fact that these c o ~ p o u n ~exhibit s dramatic
Color and Q ~ t i c Properties ~l 100 I
100
233 I
1
1
I
I
I
1
I
80
80
s
8 i
5
6
50w U
u
U
w cc
-
20 -
"0
50
Y
. . A -
20
1
2 3 4 PHOTON ENERGY, eV
5
6
Figure 2 Reflectivity of Au and the intermetallic compound PdIn. The visible range and the color scale are Indicated. Two experimental curves are shown for PdIn (reproduced from Steinemaiiii et al., 1997)
changes of color for off-stoichiometry compositions within the existence range. The origin of all these color effects is well understood from an analysis of the electronic structure (Sclilemper and Thomas, 1994). Tlie relation between the electronic structure of intermetallic ~ompoundsand their colored appearance is explored in the next section.
Color is a consequence of the interaction of light with matter in the frt;quency range of visible light (tpproxinlate range of energy (R) 3.5-3.3 c=V or a range of wavelengths (I.) 827-376 nm; conversion factor is 2/01 (eV) = /2/1240nm). The interaction of metallic compounds with light in this frequency range is typically dominated by reflea;tion. Tlie incident photon can be described as a varying electromagnetic field. Entering the metallic material, the photon transfers its energy, inducing oscillations of the materia19selectron system. Due to axwell's laws, these oscillating electrons in turn emit electromagnetic waves, i.e. the reflected photons. In general, the frequency of incident and emitted photons i s the same. And indeed, most metallic compounds exhibit about equal reflectivity over the whole range of visible light and are therefore non-colored. A metallic material will appear to be colored, if over a certain range of frequencies of the visible spectrum light is absorbed rather than reflected. The selective absorption of a photon of a given frequency is caused by electrons in specific electronic states that are able to
0' 0
I
1
2 3 PHOTON ~
~
I
I
4 ~eV R
5
I 6
~ Y ,
Figure 3 Reflectivity of Ni and the two compounds Nisi2 and CoSiz. The visible range and the color scale are Indicated (reproduced from Steinernann et al., 1997)
accept the corresponding energy quanta to junip into an excited quantum state. The absorbed photoiis o f this frequency are then missing in the spectrum of the reflected light; what remains is the so-called complementary color. For example, if a compound has absorbed green light, it appears red. The electronic structure of i n t e r m ~ t ~compounds ll~~ and its interaction with photons can be studied and understood by means of quantum mechanical calculations, coinplemeiiting experimental measurements. Xn terms of quantum mechanics, the above qualitative picture of the interaction process can be put into a more precise formulation which enables calculation of reflectivity and absorption data from first principles, i.e. without introducing empirical parameters. In quantum mechanics, light is an electromagnetic wave, quantized in t e r m of energy packets, hv, with v being the frequency of the photon and h the Planck constant. The photon and its electromagnetic field interact with the electrons of the material. The response of the electronic structure of matter to the oscillating electromagnetic field as imposed by the photons is determined by the tensor of the complex dielectric fknction E ( W ) of the electron system. &(CO)
+
= ~ ( c o ) ic2(o)
(1)
Specifically, the imaginary part of the dielectric function ca (U-)) is directly related to transitions between different quantum states
234
Properties and Phenomenology
Equation (2) explicitly tells us (note the argument of the &function) that a photon intcracts with an electron (of mass 7 8 and charge e ) only if the photon's energy AUJ exactly equals the energy difference necessary to excite the electron from the occupied state with eigeiivalue Et, to, any available unoccupied state with eigenvalue E::
ACO = E: -
= hE
(3)
In the solid state a huge number of electron states, characterized by x-vcctors and band index n as quantum numbers, form quasi-continuous energy ands. The overall dielectric response is therefore given as a sun1 over all possible transitions between occupied Tates n' and unoc ied states y1 integrated over all ?z-vectors in the llouin zone (BZ). In addition, it i s required for an electron-photon interaction that the statc of angular nionientuin is changed by the transition, i.e. that the matrix elements
of the m o i ~ e n t uoperator ~ @ are non-zero, !Q$ denoting the wave f~nctionof the electronic state k , n. En the praxis of quantum nieclianical calculations, equation (2) can be directly evaluated, and the real part of the dielectric constant tensor is obtained from the ronig relations. The optical conductivity tensor 0 ' ' is related to the dielectric function by (5)
From the calculated dielectric function or conductivity tensor, the optical properties can be obtained by a few simple relations. The reflectivity is given as
R(w) =
I &&JE(W)-
1
and the absorption coefficient q is defined as 4 = 2k
cu
-
c
(7)
where k is the imaginary part of the complex refractive index and c is the velocity of light. upported by quantum mechanical modeling, it is possible to identify those features of the electronic structure causing the color of some intermetallic compounds. The required electronic structure is shown schema~ic~lly in Fi In order to achieve selective absorption within a narrow frequency rsmge and good reflectivity elsewhere, distinct and sharp peaks of the density of states
Figure 4 Sketch of electronic energy bands and photon absorption process in an intermetallic cornpound with a pseudogap
are required, separated by a pseudogap. A pseudogap is an energy range with only a few available quantum states, i.e. a valley in the density of states curve. The states below the pseudogap have to be occupied and the states above it should be empty excited states. In other words the Fermi energy separating occupied and unoccupied parts of the electronic structure has to reside inside the pseudogap. Photons with energies equal to, or greater than, the width of the pseudogap can be absorbed by the electron system by exciting electrons from occupied states below the pseudogap into excited states above the pseudogap, provided that the matrix elements for these tramitions are non-zero. These frequencies are then removed from the spectrum of reflected light, causing the intermetallic compound to exhibit the complementary color. As an example, Figure 5 shows the density of states of CoSi2 as calculated by a density-functional-theorybased approach (Hoheiiberg and Kohn, 1964; Kohn and Sham, 1965; Wimrner et al., 1981; Jansen and Freeman, 1984). The blue color of CoSi:! can already be understood by studying the density of states. The pseudogap separates two sharp peaks of the density of states, arising from occupied COd states and empty p states. The separation between these peaks is about 2eV and transitions are allowed, since the angular ~ o i ~ ~ n t changes u r n from I = 2 to l = 1 upon excitation. Therefore, photons with an energy of about 2 eV are absorbed by the electronic structure of CoSi2, removing photons of lower frequencies and thereby the
Color a ~ i dOptical Properties 6,
I
-
DOS of CoSi,
-
_---. CO d
-
I
...., ..., Si s “._I_.
-15
I
I
i
1
Si p
-1 0
-5
0
nergy
5
10
c eV 1
Calculated electroiiic density of states of CoSiz. Optical transitions creating the blue color of CO& are indicated by arrows
red colored light from the reflected spectrum. At higher energies, however, the density of states exhibits a second pseudogap at about 3-3.5 eV. Due to the small number of available excited states in this energy range, photons of higher frequencies are not absorbed and blue colors therefore dominate the spectrum of the reflected light. Some representative optical transitions involved are indicated by arrows in the band structure
15
10
5 n
235
of CoSi2 in. Figure 6. The band structure displays energy eigenvalues of the electron system along certain high symmetry directions in the Brillouin zone. Figure 7 shows the reflectivity curve as ~ ~ l c ~ i l afrom ~ e d the electronic structure. Indeed, the reflectivity shows a maximum at about 3 eV and a miniinurn below 2 eV as expected from the inspection of the density of states and band structure, thus reproducing the measured curve in Figure 3.
The bonding in metal disilicides i s well u n d e r ~ t o o ~ in terms of hybridization of transition metal d-orbitals and Si sp-like states (e.g. Pettifor, 1988). Above we showed that the position and width of a pseudogap are crucial for the optical properties of disilicides. They are also ‘good’ nietals as supported by measureiiients of transport properties (Amiotti et al., 1990; Ditchek, 1984; Hensel et al., 1984; ensel, 1986; Hirano and Kaise, 1990; Tiwari et al., 1993). The electrical conductivity at rooin temperature is about 6 x 104 (Ohm tin)-' for CoSi2, and 3 x 104 (Ohm cm>-’ for NiSiz, values which are comparable to lead. Because of the presence of the gap, holes dominate as charge carriers; in fact CoSi2 has only holes (one per unit cell) and NiSi2 has about 0.2 electrons and 0.3 holes per unit cell. The elastic scatteriiig lengths for these charge carriers are 13 and 6nm, for COS& and Nisi2 respectively, which are of the same order as the penetration depth of light, amounting to 17 nm for Nisi:!, a value also quoted for gold. One may ask, whether it is possible to formulate some principles and requirements for the appearance of colored intermetallic compounds. We believe so. First, the crystal s t ~ u c thas ~ i to ~ ~show h g h s y ~ ~ ~ ~ t r y , thus providing distinct features, i.e. sharp peaks and valleys in the density of states. iii particular, a pseudogap must exist, which enables a distinct onset of absorption bands. Strong covalent liybridi~ation 70
1
I
I
1
I
I
Reflectivity of CoSi, (calculated)
-10
-15 W
L
r
x
W K
Figure 6 Calculated electronic eiiergy bands for CoSiz. Representative optical transitions are indicated by arrows
I 0
I
I
I
I
I
2 3 4 Photon Energy [ eV ]
~ i g u r e7 Calculated reflectivity of CoSiz
I
5
6
236
Properties and ~ ~ e ~ ~ ~ e n ~ l o ~ y
may trigger distinct absorption bands with non-zero matrix elements. Siiice the Fermi level has to reside inside the pseudogap, the occupation pattern of the electronic struct~reis of particular importance. For binary intermetallic compounds this requires the involvement of transi~ion-i~et~l elements at the ends of the periods. In the following section we will try to provide some more specific guidelines on how to find colored ~ n t e r m e compounds. ~~~~~c
Following the discussion of the relationship between color and electronic structure of iiiterrnetallic compounds, we now try to relate color to crystal structure. Are there some rules for the appearance of color? It is an old chemist’s dream that a unique number can possibly be attributed to a specific element which also describes its unique behavior in molecules and compound materials, This idea is to some extent fulfilled by Pettifor’s structure maps (Pettifor and Podloucky, 1985, 1986; Pettifar, 1986, 1995), m a ppi ~ g crystal structures of binary compounds A,B, of any stoichiometry to a two-dimensio~~lmap of some ‘coordinates’ for elements A and B. For a particular stoichiometry A,$,, each point on such a twodirnensional map corresponds to a possible compound. The elements A and B are joined together along the xand y-axes, respectively, their sequence being determined by the ‘coordinates’. Existing compounds are marked on the map by a unique symbol for each crystal structure, For an appropriate choice of the atomic ‘coordinates’, well-ordered domains of crystal structures in the two-d~i~ensionalmap occur. In searching for a valid definition of these ‘coordinates’ it became obvious that the framework of classical parameters such as electronegativi~y,atomic size, and electrons per formula unit is not very suitable to generate the required well-defined mapping (Pettifor, 1986). However, the microscopic point of view in Pettifor’s scheme (Pettifor, 1988) is adequate: the local coordination, as determined by the nature of the quantum mechanical bond determines the ordering of atoms in the solid, and equally well the appearance of color. These ‘coordinates>are phenomenologicalparameters but can be justified from quantum mechanical principles for selected cornpou~ds(Beer and Pettifor, 1984; Pettifor and Podloucky, 1985, 1986). The concept af structure maps may help oiie to guess str:~cturesof yet unknown binary compounds, to find
hints for a structural phase transition or it might even ty of pseudo-binary urthermore, as we used to explore the appearance of colors of binary intermetallic compounds. maps for compounds of Figure 8 shows which could be cand~dates stoichiometry AB a for colored interm Table 1. Those domains where colored ~ompoundsmight occur axe l~igl~liglited, and the boundaries are marked by dashed lines. The cubic structure IS rather: frequent for inte~metallic compounds, aiid it dominates for the colored metals (with some known exceptions: violet Cu2Sb, a tetragonal pseudo-cubic compound, is discussed among other defect structures in Cu-Sb alloys by Bahn and Schubert (1949), or a number of minerals like bluish black, hexagonal pseu~o-cubicweissite Gul.9Te,violet, tetragonal rickardite Cu7Te5, or gray-yellow, orthorhombic pseudo-hexagonal dyscrasite Ag,Sb, a colorecl ~ u m ~ ~ o t hphase). e r y In fact, as outlined in Section 3, colored intermetallic compounds require a crystal structure of highest symmetry, which ensures a sufficiently simple electronic structure for distinct absorption bands of high intensity. A list of binary colored intermetallic compounds is provided in Table 1. Only two diflerent structure types occur: the bcc-based €32 (cP2) type for com~oundsof composition AB (left panel of Figure 8) a (cF12) type for composition A Figure 8). Component A is a late transition metal element of group 8-10 or a noble metal. and B is an element of groups 13 or 14 (with the exception of PdMg). It should be noted, that throughout the chapter the group notation of the periodic table is given in terns of the New IUPAC notation. The structure exhibits high coordination numbers typi of for metals, i.e. each A atom has an envi ~onm e~t eight B nearest neighbors arranged at the corners o f a cube and vice versa. In the Cl st cture each atom of type A is surrounded by eight nearest neighbors; however, each B atom has only four bonds to A atoms within the first coordinatio~shell. The typical configuration for metals maximi%in~the coordination number is thus combined with tetrahedral sp3 bonds typical for covalently bonded semiconductors. It i s therefore not a surprise to obtain a pseudogap in the electronic structure of CoSi2 and related Cl compounds. The high iiumber of valence electroiis of the late transition metal elements determine the appropriate location of the Fermi level inside the pseudogap, which mechanism provides absorption bands responsible for the color of these materials,
237
Color and Optical Properties
MA=
70
60 65 70 Tc Mn Fe Ru OsCo Rh Ir Ni Pt Pd Au Ag Cu M B AS CaF, ,c Sb Fe,C,o Bi
n
MB
B 85 Si Ge
CO%, h
MiAs, h HgNln, t FeSi, c
other
Sn
ao
Pb Ga AI
In Tl Be Zn 75 Cd Hg Mg
0
other
CU
Figure 8 Color maps: regions of high-symmetry structures 111 the AB (left) and AB2 (right) Pettifor structure maps (Pettifor, 1995). Mendeleev numbers (i.e. atomic Pettifor ‘coordinates’) and prototypes (c: cubic, t: tetragonal, 0: orthorhoinbic, fi: hexagonal) are shown. Limits for expected colored intermetallic coinpounds are indicated by dashed lines (reproduc~dfrom Steinemann et al.. 1997)
About one-third of the AB compounds of B2 structure within the highlighted region of Figure 8 are colored. The compounds PtGa and PdGa lying on the boundary crystallize in the B1 structure, which is a high-symmetry fcc-based structure with six nearest neighbors, but these compounds show no particular color because of the missing pseudogap. A further cubic compound on the boundary line, PtMg, is not a candidate for color because its I320 (cP8) structure (prototype FeSi), although cubic, is of lower symmetry. For compounds of stoichiometry AB2 eight out of 11 compounds of C1 structure within the highlighted region appear colorful. Other binary colored intermetallics with structures of lower symmetry are not found. However, the appearance of color for intermetallic compounds is not restricted to the short list of stoichiornetric binary compounds shown in Table 1. Due to the metallic coordination, B2 compounds exhibit a rather extended existence rail component can be substituted by a variety of related elements. As a matter of fact, by such substitutions color can be adapted and varied in a specified way, because the color effects are directly related to the valency of the added element, A well-known example is the spectrum of colors exhibited by NiAl within its
homogeneity region, ranging from yellow for Ni-rich samples (60%) to pink (55%), blue (50%) and gray (45%) with decreasing Ni contents. The appearance of colors in B2 aluminides was investigated, apply in^ the rules for Hume-Rotliery phases, by Jacobi and Stahl (1968) and later on by Schlemper and Thomas (1994). Another prominent example to be discussed is the red compound PdIn. Iridium can be replaced by isoelectronic A1 up to 100%. Even replacing 50% of In by A1 does not alter the color, and only substitution of loo%, resulting in PdAl, shifts the color from red to yellow. A similar change of color can also be achieved by increasing the Pd content of PdIn. Substitution of In by elements of higher valence like Sn and Sb enables a gradual colsr modification from red to yellow to silvery. On the other hand, elements of lower valence like Ag, Gu and Mg can be substitute^ for In up to 10% and the red color is thereby more and more intensified. Some of these compounds derived from Pdln are presented in Figure 1. Their outstanding properties, and the fact that considerable hardening is achievable by macroalloying and microalloying, make PdIn a rather interesting material for precious objects. For compounds of C1 structure the variability of color by substitution is quite limited, because these inpart-covalently-bonded materials have a rather limited
able 1 Colored binary intermeta~lic compounds: their structure types, compositions and colors €32 (cP2)
C1 (cF12)
Compoitnd
Color
Compound
Color
FcAl CoAl COG3 NiAl NiGa PdIn PdMg PdAl
(brown) yellow yellow blue bluish red yellow yellow
Nisi, PtAI, PtGa, Ptln, AuA1, ALlCa, AuIn,
dark blue gray blue yellow yellow yellow violet blue blue
range of existence. For instance, CoSi;! exists only for cobalt contents between 33.8 and 34.4% and ternary n, A1 and Ge are soluble only up to a few percent. This imposes limitations not only for the variability of color but also for ways to achieve ductil~zation. However, a whole class of ternary and quaternary oiiiids, structurally related to the C1 c o ~ p o u n d sexist, ~ ~ i may c h be denoted as Zintl phase derivatives, filled zinc blende structures, or simply AB2-derived compounds; and indeed, a considerable number of them are colored. The cry$tallo~raphicrelationships between the involved structures of cubic symmetry are described in Table 2 and the ~ o n s t ~ c ~ of i o the n ~ ~ r i vat i ves starting from the fcc lattice is illustrated by Figure 9 (in color plate section). The fcc lattice (Wyckoff positions a>provides octahedral sites ( ~ y c k o f fpositions b) and tetrahedral sites ( ~ y ~ positi~ns k o ~ c and d ) which can be occupied by different kinds of atoms. Occupation of all 6 positions or all c positions by atoms different from those 011 a positions yields the rock salt or the
zinc blende structure, respectively. The filled-up zinc blende structures of prototypes AdLiSi, CaF2, F3 are generated by occupation of 6, ns by atoms of different types as described in Table 2. Finally, filling all n and Et positions with atom type A and with atom type B results in the centered cubic, protgtype ' CsG above. A large variety o f elernents are able to build up c o ~ p o u n d sof the filled zinc blende struc~ure.As a typical occupation pattern, octahedra~lycoord~nateda positions are occupied by metal or of valence 3-5 ups 13-15: Sb, AI, Ga, In); tetrahedrally coordinat~d c and d positions are to some extent occupied by late transition metals or noble metals (groups 8-1 1: CO, Ni, Ru, Rh, Pd, Ir, Pt, Cu, Ag, Au), whereas metals of occupy b its well as c raphic datta of these terature (Zintl et al., 1937; Schuster, 1966; Pauly et al., 1968a,b; Schuster et al., 1969; Kistrup and Sch ter, 1972, 1974; ~ e e l ~ ~ i ~ 1978; Eberz et al., 1980; etersenn, 1981; ~ e t e r s e n ~ and Schuster, 1981; Eberz, 1983; ~ ~ h i ~ i d1985; t, Drews et al., 1986; Drews, 1989; Czybulka et al., 1990; Villars and Calvert, 1991). Whether compounds of the filled zinc blende structure are colored or not depends 01%s ~ ~ ~ c h i o ~ ~ t r y and the atom types involved, All known ternary and quaternary c o ~ p o u n dof s this type e ~ h i b i t i ncolor ~ are listed in Tables 3, 4 and 5. The binary boundary cases were listed earlier in Table I , Again, the ~ u e s t i ois~ whether there is a simple criterion which predicts the appearance of color for these compounds. It seems that a simple empirical counting scheme of valence electrons is applica~le,at least with some caution.
~~cht Pearson bcc and fcc structures and crystallogra~hicallyrelated itriictures, identified by S t r u ~ t u r ~ ~designation, symbol, and Wyckoff ~otationfor site occupatio~ Site occupation
St ~ ~ ~ ~ ~ ~ r b ~ r i c h t ~esign~tio~ Prototype
w CsCl NaTl BiF, CuHg,Ti AlLiSi CaF, ZnS N2-161 cu
Pearson symbol CE?
CP2 cF16 cF1 6 CA6 cF12 cF12 CF"3
c n cF4
Substance
P-AgCd PdIll CdLi PdLi,Sn PtLiMgSb PtMgSn CoSi, AuAl BiCe CdLi,
a
b
c
Ag or Cd on any site 41n 4Pd 4Pd 4Cd 4Li 4Li 4Sn 4Li 4Pd and 4Sb 4Mg 4Pt 4Sn 4Mg 4Pt 4Co 4Si 4Al 4.Au 4% 4Ce Cd and 3LI on site a
d
4111 4Cd 4Li 4Li 4Si
239
Color and Optical Properties T a ~ l3 ~ Group-1 1-based ternary colored intermetallic compounds of the Zintl type (derivatives of AB, structure). Table ordered according to noble metal component and increasing total number of valence electrons NVaPThe number of valence electrons is set at 1 for noble metal atoms, at 1 aiid 2 for alkaline and alkaline-earth inetal atoms, respectively, and at 3-5 for metals and metalloids, corresponding to the group number of the periodic table
Table 4 Group 8-, 9- and 10-based ternary colored intermetallic compounds of the Zintl type (derivatives of AB, structure). Table ordered according to transition inetal component and increasing total number of valence electrons NYal.The number of valence electrons per t~ansitionmetal atom i s set to zero (Ekman's rule) Compound
Golor
Compound
Color
CO Li, Ge
yellow-brown
6
Cu, Li Si Cu, Li Ge Cu Li, Ge Cu Li, Sn Cu Mg Sn c u Mg Sb
red-violet red-violet bright red dark violet blue-gray blue
7 7 8
Ni, Li Ge Ni, Li Sn Ni Li2 Si Ni Mg Sb Ni Mg Bi
brass-yellow yellow yellow-brown blue-violet blue-violet
5 5 6 7 7
Ag Li, A1 Ag Li, Ga Ag L1, In Ag Li, Sn Ag Li, Pb Ag, Li Sn
yellow-pink yellow gold-yellow dark red-violet blue-violet light blue
6 6 6 7 7 7
Ru LI Al, Ru Li Ga, Ru Li In,
yellow gray si1very
7 7 7
Au Li, Ga Au Li, In Au Mg Sn Au Li Sb ALILI, Sn Au Li, Pb Au Li,, AI, Au Li,, Ga, Au LI,, In, Au Mg Sb Au Li, Sb
green-yellow green-yellow red-violet red-violet pink violet blue blue gray gray bluish
6 6 7 7 7 7 7.3 7.5 7.6 8 8
RI1 Li, Ga Rh Li, In Rh Li Al, Rh Li Ga, Rh Li In,
light-yellow silvery yellow light blue silvery
5 5 7 7 7
Pd, Li Ge Pd, Li Sn Pd, Li Pb Pd Li, A1 Pd Li, Ga Pd LI, In Pd Li, Ge Pd Li, Sn Pd Li, Pb Pd Mg Sn Pd Li AI, Pd Li Ga, Pd Li In, Pd Li,Sb Pd Mg Sb Pd Mg, Sb Pd Mg, Sb
brown- yellow brown-yellow brown-yellow rose brass-yellow browi1-yellow yellow yellow brown- yellow brown-yellow violet silvery silvery brass-yellow violet light violet blue-gray
5 5 5 5 5 5 6 6 6 6 7 7 7 7
Ir Ir Ir Ir Ir
Li, Ca Li, In Li, Sn Li Al, Li Ga,
silvery silvery silvery red-violet light violet
5 5 6 7 7
Pt Pt Pt Pt Pt Pt Pt Pt Pt Pt
Li, AI Li, Ga Li, In Li, Sn Mg Sn Li AI, Li Ga, Li In, Li, Sb Mg Sb
bright-yellow bright- yellow brass-yellow yellow reddish-brown copper-red brown-pink pink brass-yellow violet
5 5 5 6 6 7 7 7 7
N"d
7 7 7
By such a scheme, 1 and 2 valence electrons are attributed to group 1 and 2 elements, e.g. Li and Mg, u ~ electrons to noble metals and and ( ~ g -~10)~ valence further to elements of groups 11 to 15, wherein ATgroup is the corresponding group number. In particular, for an intermetallic compound containing late transition metal elements such as Fe, CO,Ni, Ru, Rh, Pd, Ir, Pt, the number 0 is attributed according to Elsman's rule. Elcman (193 1) stated this rule as an extension of HumeRothery's electron concentration concept. Schlemper and Thomas (1994) have successfully applied Ekman's rule to determine the shift of optical absor~tionbands in transition metal alurninides. Applying the above counting scheme to all cornpouiids of the filled zinc blende type (the number of valence electrons for the colored compounds are listed in Tables 3-5) reveals the general rule that only those conipounds with valence electron numbers equal to or smaller than rn 7 exhibit color effects, as originally suggested by Eberz et al. (1980). As can be observed in Tables 3-5, this rule applies quite satisfactorily for most of the ternary and
NVd
7 8 9
7
Properties and P h ~ n ~ ~ ~ ~ ~ l ~ ~ y
240
Quaternary colored intermetallic compounds of the Zintl type (derivatives of AB, structure). Table ordered according to noble or transition metal cornponeiit and increasing total number of valence electrons Nynr as in Tables 3 and 4. Formulae that comprise partial occupations of WyckofF positions are included as found by chemical analysis and radiocrystallography Compound
Color
Pd Lt Mg Sn Pd Li Mg Sb
red-violet gray-blue
7 8
Pt LI,, Mg Sn l 3 L105 Mgo 5 Sb Pt Li h4g Sn Pt Li Mgo5Sb Pt Li,, Mg Sb Pt Li Mg Sb
dark red pink copper-red cupper-red dark blue-violet red-violet
6.5 6.5 7 7 7.5 8
Ir Li Mg Sn
gray-blue
7
Au Li Mg Sn
gray
7
NYd
quaternary Zintl-derived phases but also for the binary compounds of Table 1. For Sb~contai~ing phases and disilicides the range should be extended to valence electron numbers of up to 8. On the other hand, color effects may appear within a lower limit of x 5 for the valence electron number. It is further found that all four prototype structures, i.e. NaTl, BiF3, CuHg2Ti, AlLiSi, of the cF12 and cFl6 structure types occur among the ternary and quaternary compounds. Furthermore, there is a color trend from yellow to red, violet, and blue for an increasing number of valence electrons. There is some indication that isoelectronic i~aterialstend to show like colors. For instance, the violet compound AuAL2 is isoelectronic to quite a number of compounds like AgLizSn, AgLizPb, b, AuLizPb, AuLizSn, PdLiAlz, PtMgSb, whose color ranges between pink and violet. It should, however, be emphasized that Ekman’s rule until now lacks a satisfactory explanation from the point of view of electron theory, but is applied with some success on a purely empirical foundation. The color of intermetallic cornpounds does change with temperature. MuMawer (1 952) contributed a comprehensive study of this effect for (Cu, Ag, Au) - (Zn, Cd) /3-brasses. At low temperature (liquid nitrogen), these alloys are yellow, red and pale-gold, but appear pink, purple and violet at room temperature, At a p p r o x i ~ a t e 300 l ~ “C, their color changes to red for the common brass and gray for the other alloys. It is found that disordering is not the primary factor for the color change, but that a phase change,
e.g. the transformation of cubic P-AgCd to hexagonal results in the observed differences in appearance. The former compound has a salmon-pink color, the latter is silver-gray, i.e. i s non-colored. suggests connecting the loss of color with the Debye temperature On. In fact, OD is a measure for the onset of strong electron-lattice interactions. The thermal vibrations of the crystal lattice will change the energy levels of band electrons and ‘blur’ the band structure such as given in Figures 5 and 6 . The result is loss of color. A similar effect of ~attic~”e1ectron coupling might equally produce color changes under static mechanical strains. The variety of colors for Ziiitl phase derivatives might appear attractive for applications. However, there is a considerable obstacle, namely due to the large contents of Li, Mg, Al, the cornpounds are prone to corrosion, and yet another difficulty is brittleness. As a conclusion it appears, that applications of colored intermetallic compounds are limited. The following section will treat some of the questions related to applications.
c,
Knowledge about strength, ductility and toughness behavior is critically important for the engineer when shaping colored metals for precious objects. Several specific criteria may help to obtain information about these user properties. Again, we take as an example the dark-blue CoSi2 compound. Elastic properties are fundamental quantities which are insensitive to the processing and microstructure of a solid. This insensitivity makes the data useful to understand interactions among atoms, and it helps to clarify niicroscopic aspects of cohesion, such as yielding, dislocation interactions, and of fracture. Table 6 reports elastic property data for CoSiz, its ionic prototype (fluorite) Cap?, the metal CO and covalent Si. A first criterion by Pugh (1954) utilizes the relations between elastic and plastic properties of metals. It suggests that the ratio of bulk modulus K to shear modulus G is a measure of brittleness or of malleability. Then, a small value of K signifies easy decohesion and a large value of G characterizes high resistance against (shear) deformation, and vice versa. For cubic and hexagonal metals, the limit between brittleness and ductility lies at X/G x 2. Conteniplating Table 6, it is expected that the intermetallic
24 1
Color and Optical Properties
Table 6 Elastic moduli at 298 E; of CoSi,, CaF, (fluorite, ionic), CO(metal), Si (covalent). Columns ‘Poly’ and Single’ refer to experiments with polycrystals and single crystals; column ‘Calc’ lists ub initio data corresponding to 0 K. Subscnpt N refers Hill’s average. CoSi,
Bulk modulus (fo, GPa Shear modulus (GH),GPa Young’s niodulus (EH), GPa References PugWs ratio (K/G,,) Cauchy pressure (PJ, GPa Rice-Thomson criterion (Gb/yS)
Poly
Single
Calc
CaFz
CO
162 64 170 a,b,c 2.5
167 64 169 a,b,c 2.6 + 61 12
173 63 169 d 2.7
85 43 110 e 2.0 + 11
187 81 213 f 2.3 + 32
Si
98 67 I63 g 1.5 - 16
aGu6nin et al. (1990). bGiauque and Qberli (1992). bStemeinnnn (1994). “Yamaguchi et al. (1993). dStadler et aE. (1996). ‘HCIand Ruoff (1967). %andolt-Bornstein (1984). g ~ c S k i ~and i n Andreatch (1964).
compound Cos& has some inherent ductility, because K/G = 2.5 > 2. A second criterion, based on Cauchy’s pressure PC = c12 - c44, is related to the nature of bonding (Johnson, 1988). For central forces, PC amounts to zero. Electrostatic interactions are of this type, and in fact the ionic compound CaF2 has a small Cauchy pressure. The constant c12 represents a dilatation and c~ refers to pure shear. The negative value of PC in the case of Si is typical for strong directional bonding. For a metal on the other hand, the electron gas will not resist shear but will strongly oppose a volume change, and then PC becomes positive. CoSi2 has the signature of both the metallic and the covalent bonds, the local syminetry being octahedral for CO and tetrahedral for Si. A third criterion by Rice and Thomson (19741, relates the resistance to shear (that is Glbl, in which b is Burger’s vector) with the resistance to cleavage the surface or cleavage energy ys). If Glbj/y, < then a material behaves as ductile. For CoSiz, the [ lOO]l = a = 0.537 nm, equal to the lattice ~arameterof the und. No ~ e a s u r e ddata for ys exist, but a calc for the cleavage plane (110) yields a value of ys = 1.44 J/m2 per surface (Stadler et al., 1998; ~ogtenhuberand Podloucky, 1997). The resulting parameter Gb/y,y= 12 indicates brittleness, which is essentially caused by the large Burger’s vector of the nondisso~iateddislocation. Structure, bond strength, and anisotropy determine the active glide systems in crystals. Five independent systems are needed to accommodate deformation in a
polycrystalline aggregate. Only three glide systems exist in CoSiz (Yamaguchi et al., 1993; Anongba and ~teinemann,1993, 1994a,b, 19951, but this i s suflicient for the defomation of single crystals for which only two glide systems are needed. The compression tests of Figure 10 were done with single crystals in an orientation that can activate glide of the family of (001) planes. The rmation is limited at room temperature but extended above 100 “C. It is noted that (yield) shear stresses can be translated for axial stresses of polycrystalline materials when dividing by the Schrnid factor and multiplying by the Taylor factor of about 1.7. The resistance to compr~ssionof CO quite high, which means that CoSi2 is a hard material.
200c.
0
/4297K
10 SHEAR S T ~ A ~ 74 N,
20
igure 10 Compression tests with single crystals of CO%;!. Crystals are ariented along [ 1131 and the Schrnid factor is 0.27 for glide on basal planes. Deformation speed is 5 to 7 x 10-”s (reproduced from ~ t ~ i n e m a netn al., 1997)
242
Properties and Phenomenology
The fracture toughness of a sintered CoSi2 compound was measured by thee-point bending and by a Charpy test. The toughness values obtained were KIC == 3-4 and K z= ~ 7-8, respectively for the two tests. These are not particularly good values but are much higher than the purely elastic (or Griffithj toughness of KG = 0 . $ ~ M ~ a m 1 /It2 .must be concluded that the brittleness of CoSi2 is caused by the insufficient number of glide systems, but not by a lack of inherent ductility.
established. All the colored compounds crystallize almost exclusively in a high-symmetry cubic structure. Selective modification of colors by s~~bstituting ternary elements into ~ u ~ e - R o t h e rphases y is given as another example for the potential of color engineering. For a convenient overview, tables of compounds with their composition, color and valence electron numbers are provided. From an engineer’s point of view, the attractive colored appearance of metallic compounds must be combined with suitable m e c ~ ~ n i cproperties, al in order to enable processing of these materials in industrial appliczt‘ions.
The chapter focuses on the appearance of color for i n t e ~ i ~ 7 ~ tcompounds aZ~i~ which have distinctly different optical properties in comparison to ‘ s e ~ i c ~ n d ~and c~~r~s dieZ~ctrics.About 100 colored intermetallic compo~~nds exist. Apart from being attractive to look at, such hue a measure of the spectral energy materials might be applied for industrial purposes. The distribution of reflected light i2 is taken as a specific example apparent degree of emission of light throughout this chapter because of its beautiful or per unit area bluish-black appearance and its mechanical properties, brightness, creating possible applications in watchmaking. lightness, The definition of color is not unique. Scientists, luminance, engineers and artists have quite different schemes and or value understanding of color. A scientific definition can be the difference between the color pergiven in terms of reflection and absorption properties ceived and a neutral color perception 01which are based on the electronic structure of the of the same brilliance purity, material, whereas a technical definition for color is intensity, provided by the so-called L*a*b* scheme making use or deepness of luminance and intensities of color pairs. luster as perceived by the eye, the overall The origin of the optical properties of a particular impression of the refraction and com~oundlies in its electronic structure which comreAection of incident light from the prises the bonding characteristics of the solid material. solid surface Reflectivity spectra are therefore discussed in terms of tint a color diluted with white, having energy bands and densitie~of states. Indications and little inten~ity,but high brilliance requirements for the appearance of color o f an a color darkened with black, having intermetallic compound are formulated, namely: (a) little intensity, but low brilliance highly sym~etriccrystal structure and partially covalent bonding, both providing distinct features in the electronic structure such as sharp peaks in the densities of states and pseudo~aps;(b) an admixture of late transition or noble metals which places the Ferini energy suitably close to the pseudogap. We gratefully thank for support by the Austrian A further connection between crystal structure and Ministry of Science (grant nr. 651,45446), and the Center color is worked out. For binary compounds, Pettifor’s for Comput~tion~l Materials Science (CMSj in Vienna. structure maps turn out to be useful for locating possible compositions of colored compounds. For more complex compositions such as ternary or quaternary compoun~s (Zintl phase derivatives), a correlation between colors and the number of valence Amiotti, M., Borghesi, A., Giuzetti, G., and Navtz, F. (1990). Phys. Rev. B,42, 8939. electrons according to the so-called Ekman’s rule is
d Steinemann. S. G. (1993). Phys. Stat. Anongba, P. N.B., and Steinernann, S. G. (1994a). K S M A I ~ ~ " S t r e nof~ tMateriu~s h (eds H. Oikawa e i uL). The Japan Institute of Metals, Tokyo, 125. Anoiigba, P. N. B., and Steineiiiann, S. G. (1994b). ZCSMA i ( ~ - S t r e n ~oft hM a t e ~ i ~(eds l s H . Oikawa et al.). The Japan Institute of Metals, Tokyo, 431. d Steinernanii, S. 6. (1995). Acta Met.
edition of lectures 1921/1922, ~ ~ l i ~ v ~ ~ ~ c 19'79 Kohn, W.. and Sham, L. J. (1965). Ph*ys.Rev.
2161. Muldawer, L. (1962). Plzvs. Rev., 1 Nesper, R. (1 991). Angew. Chemie,
t. K. (1969). S c r ~ ~t e~t ~ i i . , eer, N., and Pettifor, D. G. (1984). Eiectronzc
of Cornplex Syxtenzs (eds P. Phanseau, and W. M. Pauly, H.. Weiss, A., and Witte, H. 6196%). ZeitschriJi T e ~ ~ c r n i a nPlenum, ). New York, 769. M e t u l l k ~ $ n59, ~ ~554. , Cho, S. J. (1970). Phys. Status Solidi, Petersen, A. (1989). Thesis, University of Cretu, G., and van der Liiigen, E. (1999). Gold B i ~ ~ ~3e ~ ~ n Petersenn, . A. v. (1981). Thesis. IJniversity 115. Petersenn, A. v., and Schuster, H.-U. Czybulka, A., Petersen, ~ r ~ ~ s t ~ l l ~ g156, r ~91. ~hie, ~ e s s - C o ~ ~ eo tna l s Pettifor, D. 6 . (1986). New Scienrisd, 29 May, 48. Ditchck, €3, M. (1984). J Pettifor, D. G. (1988). Mater. Sci. Technol., Drews, J. (1986). Thesis, University of Koln, Germany. Pettifor, D. G. (1995). I ~ t e r ~ e ~ ~ ~ l i ~ Drews, J., Eberz, U., and Schuster, H.-U. (1986). J of the VOL. I (eds J. H. Westbrook, and Less-Common M ~ t a l s , Chichester, England, 43 9. Eberz, U. (1983). Thesis, University of KGln, Gerxnany. Pettifor, D. G., and Podloucky, R. (1985). Phys. Rev. Letters, W.* and Schuster. H.-U. (1980). Z. 53, 1080. Peltifor, D. G., aiid Podloucky, R. (1986). J PhWys.C. ~ ~ Stale Phy,F., 19, 31 5. Pugh, S. F. (1954). Philos. Mug., dc Lausanne (in French), Randin, J.-P. (1994). Acfes PC; Gttkiiin, G., Ignat, M., and Thonias, 0.(1990). J Appi. Phys., Chronouniirie, Besaiiqon (France), 13 (in Frcixh). 68, 6515. Rice, J. R., and Thompsoii, R. (I Sarnbongi, T., Magiwara, R., and Yaniadaya, T. (1966). J . Phys. Soc. ~ a p a n21, , 923. Sautboff, G. (1995). Inter~etu~lic.s, VCH Verlag, Weinhet Schleinper, K., and Thomas, L. I(.(1994). Phys. Rev. B, 17082. Hohenberg, P., and Kohn, W (1964). Phys. Rev., 136, B864. Schnxdt, P. C. (1985). Z. Naturfom Jacobi, H., and Stahl, R. (1968). Nuturwis.~~l?.schafien, 55, 272 Schustcr, H. U. (1966). N a t z $ r ~ ~ s ~ ~ . , (in German). Schuster, H. U., Thiedeinann Jan, J.-P., and Vishnubatla, S. S. (1967). Carz. J. Phys., (1969). Zeitschrifi J: Anorg. 2505. Seelentag, W. (1978). Thesis, U Jansen, H. J. F., and Freeman, A. J. (1984). Phys. Rev. B, 38, Stadler, R., Wolf, W., Podloucky, R. 561. Furthmuller, J., and Hafner Johnson, R. A. (1988). Phys. Rev. B, 37, 3924. 1729. Kistrup, C.-J., and Schuster, H.-U. (1 972). Z. Nffturf~r'sc~izing, Stadler, R., Podloucky, R., Kresse, G., and Hafiier, J. (1998). 7b, 324. chuster, H.-U. (1974). Z . Anorg. Allg. Steinetnann. S. (1987). European Patent No. 0284699 (March 3 1). Klce, P (1956, 1921/1922) Bauhaus Lectures, original text Steinemann, S. (1990). United States of ~ n ~ e r i cPatent a ~n German: 'Gold 1st ern Hin-und-Her-Vibrieren von No. 491 1762 (March 27). sattem Gelb nach einem WeiB von fiberstarker Helle. Steincmann, S. (1994). .4ctes Sdme Corigrt;s EuropPen de Ein bewcglicls bestminter Wert. Silber vihricrt voii Chronom~trie,Besanqon (France), 7 (in French). Dunkel nach sehr Hell und ist ebenfalls beweglich St~inemann~ S. G., Anongb bestinimt. Kupfer ist ein Vibrato von Rotorange nach . El., and Podloucky, fiberhell. Die Metallwerte sind aparte bildnerische (1997). .I. Phase Equil., 1 Mittel'. In Paul Klee, Ilus bildnerische Denken, Benno Tilley, R. .I. I3. (2000). Colour ancl Optical Properties of' Schwabe & CO Verlag, ~aseI~Stuttgart 1956; Faksimile M ~ t ezuls, r Wiley Chichester, ~ n ~ ~ a i ~ d . a
)
l
24
~
~
0
~
and ~ ~ ~
., and Narayan, J. (1993). Phiios. Mag. 13,
10805. cstbrook, J. M. (1971). ~ e ~ Trans. a ~ A,~8 .
~
~i
e~ s
~
~
~
e
~
o
~
0
~
Westbrook, 1. H., aiid Flcischer, R. L., eds. (1995). I ~ t e r ~ e ~ u l l i~co ~ p ( ~ u nvol. d s 1. Wiley, Chichester,
Zintl, E., Hardcr, A., and Haucke, W. (1937). Z. Phys. Chern.,
y
deconiposed into a sun1 of partial entropies associated with the different disordering processes: eking of a crystalline solid into a liquid is gener~~ly A S = AS, t- AS?. . . -5 A ought of as a disordering process from a state in long with the atomic structure chan which the atoms execute thermal motions about static place in this way, changes in the transport properties e ~ u i l i b r i usites ~ to one in which they all migrate randomly over the whole space occupied by the ~ a t e r i a ~Ln. a coniplex system, several d i ~ e r ~ i i t components of this di~orderinginay be distinguis~ed: these can include traiislational disorder of one subtransition or gradually over a temperature interval lattice, chemical disorder between two or more sublattices, orientatioi~aldisorder of molecules or between transitions. In this review we cover sonie aspects of ~ y ~ i a i ~ i c ~ l c o ~ ~ l eions, x or lattice melt in^ in one or two directions in the crystal. lien such a system is heated disorder in crystalline interim terms require some elaboration. towards the melting point, different disordering to processes set in until complete long-range disorder is disorder we exclude achieved in tlie liquid state (Ubbelohde, 1978), literature on order-disorder transforni~t~~iis in a ~ ~ althou~hsubstantial short- and i n t e ~ ~ e d i a t e ~ r a n g e imitrov in Vol. I), wliere the main interest lies in the ~tatistica~ and s t r ~ ~ t u raspects al of ersist well into the liquid state - see Singh (1995) in Vol. I of this treatise. Further the static disorder, as well as states of frozen disorder can be made as to whether a particular such as translational and o~ient~tioiial disordering process in the solid state occurs ~radually, crystals). We will also exclude vibrational d y n a ~ i c sas over an e ~ t ~ te~perature n ~ e ~ range, or suddenly, at a in phonons and phasons where the atanis return to phase transition. I f more than one process is associated static equilibrium positions. The only exception is that with a phase transition^ these transitions may occur at we will cover, in the last section, pre~ursorphases of a single temperature - the melting point (Figure la) some rnartensitic transi lions, since these bear some or a sequence of distinct temperatures resemblance to the phenomenon of nielting that proc ~ i l ~ i n a t i ning the melting point ( vides the backdrop to our present discussion. Tlie terinn 1997). In either case the entropy of melting may be ~ ~ excludes~ not only~ Z i ~ systems ~~ i ~ - ~covered Z ~~~~~~~~
~ ~ t e r ~ i o ~ ~t ~~ Vol. i o~ 3, u~ Principles ~~ ~d ~ and ~ Practice. Edited by J. €3.Westbrook and R. L. Fleischer 0 2 0 0 2 John Wiley & Sons, Ltd.
246
Properties and ~ l ~ e n o ~ e n ~ l o ~ y
igure P Schematic illustrations of thc melting process in which ditrt3rent disordering mechanisms set in (a) at the Same t e ~ p e r a t u ror~ (b) at a sequence of distinct temperatures (Reprinted froin Price, D. L. and Saboungi, M.-L., J. Phys.. ~ ~ ~ f f t9, t2707, ~ ~Copyright (1997), with permission from the Institute of Physics, VK)
h (1995) in Vol. 1 of this work s, which now include a wide range of amorphous alloys that can be quenched in bulk elt (covered by Greer in Vol. 1 of this chwarz in this volume), as well as quusici.ystalZiize compounds (covered by Kelton in Vol. 1 of this work). All these classes of material have their own complex and fascinating dynamical behavior that would take us too far from the priniary emphasis of this review. Finally, the term i ~ ~ e ~ ~ e t requires a Z Z i ~ some consideration since, as we shall see, materials formed from perfectly good metals may, in the compound, have properties more typical of semiconducting or even insulating materials. Some of the interesting types of behavior we shall encounter are more familiar in ionic and covalently bonded solids: the classic work on disorder in crystals arsonag age and Staveley, 1978) includes only scdt,s in the chapter on orientational disorder. In fact, some degree of strongly local or dire~tionalbonding appears to be essential for the kinds of disorder we are considering. From a structural point of view, this generally leads to a substantial degree of intermedia~e~~ange order, reflected in Bragg peaks at low scattering vectors in both ordered and disordered crystalline phases that transform into jirst sharp ~ ~~ e a in k ~the ~ ~liquidr (Price ef~ al., 1989). c Thus, we will include compounds not only of conventional nietallic elements with each other but also ones with metallic elements and hydrogen on the one hand
or with metallic elements and a Group IV (especially carbon), V or VI element on the other, which may exhibit a similar kind of behavior. Some of the most striking examples of this behavior are found in Zintl phases, compounds in which electron transfer coinbined with directional bonding of' the negative species leads to chemical behavior in these species characteristic of elements further to the right in the periodic table (van der Lugt, 1996). IFn addition to a fascinating collection of clusters and other local structures described elsewhere in this volume (Sevov, chapter 6), they exhibit dramatic changes in physical properties as the composition is varied over the intermetallic range. Compounds of alkali metals with lead, for example, behave as typical metals near the end members of the system but around the equiatomic compositions show a dramatic decrease in conductivity, with a positive temperature coefficient, typical of semiconductor behavior (Sabozrngi et al., 1990). As discussed later in this review, these phases often exhibit unusual kinds of dynamic disorder, with a subtle interplay between structure and transport as these diiferent kinds of disorder set in. Most of the experimental results presented in this review were obtained in scattering experiments, altliougli NMR has in certain cases provided useful complementary Furthermore the correla~ ~ information, ~ ~ tion between the microscopic information provided by scattering and resonance experiments, on the one hand, and the macroscopic thermodynamic and
~
lTj2iect.Y o j Mobile Species
transport properties on the other, are often illuminating. Since diffraction and inelastic neutron scattering ineasurements have provided much of the information on the structural and d ~ ~ a ~ iaspects c a l of disorder in intermetallic compounds, we will first in Section 2 give a brief review of these techniques. Those interested only in the experimental results and their significance may turn directly to Section 3, where we investigate some examples of the different types of disorder encountered, passing from translational disorder (hydrides, fast-ion phases) to orientational disorder (rotor phases) and thence to paddle-wheel phases and fullerenes, which exhibit both kinds of disorder. Finally ferroelastic phases exhibit a cooperative l i ~ u i ~ - disorder l~e in restricted directions or pkanes of the crystal.
catteriizg from disordered niaterials is invaluable for btaining two types of imcroscopic infor~ation:the a t o ~ ~ z~~t cr ~ c t uo rb ~t a~i ~ efrom ~ d i ~ r a ~ t i omeasuren ments, and the utomic dynamics, obtained from inelastic scattering measu~emeiits.A ~ z ~ ~ a cexperitio~ ment is designed to total s~atteringas a function of scatterin
(1) are the wave vectors of the incident ticks. These may be neutrons, X-rays or electrons - each has particular advantages and disadvantages. In the present work we present the formalism for neutron diffraction, but it can also be used for the other kinds of particles with rather minor modifications. In a diifraction experiment to fix the direction of priate collimators, det only the magnitude of one of the two, generally IQ. is evaluated from equation (1) under the on that the scattering is elastic, i.e. there is no energy exchange between the neutron and the sample, and so the incident and scattered neutrons have the same energy, and thus
(a)
(b)
Schematic illustration of (a) a iicutron diffraction and (b) an inelastic neutron scattering experiment
important, it can usually be taken care of by appropriate inela~ticity(or Tlaczek’) correcti~ns. After taking into account appropriate factors such as beam intensity, number of atoms in the sample, and detector efficiency, and making ~ o r r e c t i o ~for s absorption and multiple scattering iii the sample, the scattering intensity measured in tlie detector i s reduced to a fund~mentalquantity, the ~ z ~ ~ er r o~ ~ ~Y st- si~ ~ ~li o ~ per atom (Price and Skold, 1986).
where yla is the nuniber of atoms of type a in the sample ) is the partial structure factor for the atom pair (a, h). The structure factors are related to the structure in real space through a Fourier transform:
(2)
In practice this not quite correct since some of the scattering is inelastic, i.e. energy exchange does take place (see below). However, the experiments are usually designed to minimize this error and, if it i s
where gab(r) is the partial pair distribution function in real space for the atom pair (a, b) ~ ~ s h c r and ~ft
248
Properties uizd Phenomenology
angreth, 1 9 ~ 7 ) .The latter quantity has tlie direct physical interpretation that
is the average number of A atonis in a volume dr at a distance r from an a atom at the origin. In polycrystalline samples, isotropic structure factors derived from measL~renientson a powder can be defined by averaging all crystal orientatioiis. In a one-component system, the indices a, b disappear and only a single compound factor S ( and a single pair distribution function exist. In a system with more than one component, a full structural analysis requires a number of different measurements with different coefEicients in equation (3); in favorable cases, this may be accomplished with the use of isotope substitution. If only a single measuremen1 is performed, a weighted average of the partial structure factors is all that can be obtained. his can be defined as:
1 must be defined in tude and direction of both the design of the scattering apparatus. The intensity of this scattering process is reduced to a double ~ i ~ e r e ~c rt ~i ,~~ls - ~ e ~ t i o n ~
, E ) are the coherent and ~ ~ c o / ? e rpartial e ~ t scuttering ,functions (sometimes called dynamical structure factors) defined by:
We see here an additional complication arising from the mixture of colzerent scattering, which measures the correlated motions between pairs o f particles:
where
U
along with the corres~ondingneutron-weighted pair distr~butionfunction
In isotropic systems such as liquids and glasses, the structure factor and pair distribution function depend on the scalar variables and r, respectively. In polycrystalline samples, isotropic structure factors and pair distribution functions derived from ineasurements on a powder can be defined by averaging equations (6) and (8) over all crystal orientations.
and imoherent scattering which describes the correlatioiis in tlie positions of single particles at diEerent times:
The relative contributions of coherent and incoherent scattering will depend on the various elements and ost natural elements, as well stly coherent, whereas natural herent and natural lithium and silver, for example, are a mixture of both. These Pacts must be taken into account in the interpretation of the scattering data. Through the energy sum rules CD
The dynamics of' stystem can be measured in an ~ n e l ~ s~t ~ c ~~ ~ ~u texperiment t t e ~ r ~(Figure ~ ~ ~ 2(b)). ~ ~ In this case, the energy transfer given by
-00
(14)
(9) i s measured in addition to the scattering vector defined
in equation (I). To accomplish this, both the magni-
the scattering functions measured in an inelastic scattering experiment can be related to the structure factors measured in a diRraction measurement, which,
as discussed above, is designed to measure the total scattering integrated over all possible energy transfers. In a multicoi~ponentsystem, as in the difTraction case, several measL~re~ents, for example with separated isotopes, are needed to define the full set of partial scattering functions. 111 a single measurement, average scattering functions weighted by the concentration and scattering length of each elernent, analogous to equation (6), are obtained. owever, in many cases of practical interest, this may already contain much useful i n f o ~ a t i o n . In polycrystalline samples, isotropic scattering functions derived from measurements on a powder can again be defined by averaging over all crystal orientations. It is convenient to distinguish between four dynamical regimes of neutron scattering, illustrated schematically in Fi 3 (Price, 1997). Figure 3(a) shows a typical stru we treat Q as a scalar q either an isotropic s average or one with the values along a specific crystalline direction in the case of a crystalline system. S( Q ) shows delta-fu~ction-likeBragg peaks superimposed on a coiitinuous diffuse background: the first derives from the long-range crystallographic order (and would be absent in an isotropic system) while the second results from the various kinds of disorder the sample. We pick out a particular value of , and discuss the time development I(Q1, t ) of the correlations described by S(Q1). Figure 3(b) shows various time regimes that may appear in I(Q1, the corresponding features in S(Ql , urier transfori~ation.
t
3 Schcniatic illustration of dynamical regimes probed by inelastic neutron scattering: (a) stmctnre factor S(Q); (b) internrrediatc-scattering function I( (c) scattering function S(Q& The numbers denote the (1) recoil. (2) one-~honon, ( 3 ) quasielastic and (4) elastic scattering rcgimes. ~ ~ ~ from Pricc. D. L., Current Opimon in Lshiid State a d , 477, Copyright (1997) with permission from Elsevicr Science) ~~~~~~
1. The conceptually simplest scattering event is one that takes place as if the target nucleus i s independe~tof its neighbors. This is in fact what happens at short times, where I ( & , 2") falls off from its value at t = 0, generally with an approximately ~ a ~ s s i behavior. an In the limit of large (2, this recoil scafferingwill be the dominant contribution to S(Q1, E ) , consisting of a peak on the neutron energy-loss side ( E > 0) centered at the recoil energy ER =h2Q:/2M with a shape that reflects the momentum distribution of the system in its ground state. In particular, the variaiice in energy i s related to the mean kinetic energy K: (15) 2 , If there are vibrational processes, say with a frequency cop, I ( Q l , t ) will have an oscillatory part and Scoh(Ql,E ) will have scattering centered at
~
&&cop, generally referred to as ~ ~ ~ - p l z ~ ) scaffering. If the vibrational motion is liarmonic, this will have a delta-function form 5'1(Ql) 6( E Ifr: h q , ) . In crystalline materials the phonons can be labeled by the wave vector index j , so that measurements on single crystals can be made to identify these delta fu hence map out the phonoii dispersion 3. If there are relaxation processes in which the correlations decay at some characteristic rate a, say, S(Q1, E ) will have a b r o a d ~ n ec ~~ m ~ o n e n t
still centered at E = 0, generally called ~ ~ f f . s i e l a s t ~ carbons c (Winter et al., 19981, a class of alkali metalscatteririg EMS): in the case of an carbon compounds that we shall return to in Section is will have a Lorentzian 3.5. The most extensively studied mixed conductors are the silver chalcogeaides AgzS, Ag2Se and Ag2Te. These were first investigated in the nineteenth century by Hittorf?who observed the large jump in coi~ductiv~ . If there are structural correlations which exist ity at the cx --7\ /? phase transition (177 'C), which he interpreted as an increase in ionic conductivity; in facl for inlini tely long times (more precisely, times the effect is principally electronic, as established by t >fi/A.E where A E is the energy resolution of the e, along with their Wagner (1933). Ag2 experii~ent), for exainple in a solid where the copper counterparts, have in Fkct two disordered atoms execute thermal motions about fixed phases, /j and an fcc y phase from which they melt. , t ) will obviously The inter~ediate-tei~perature (8)disordered phase, ependent term. This shared with the non-electronically cotiducting, fast-ion will give a delta-function S,i(Qi)S(E) term in the conductors AgI and Ag3S1, has the bcc crystal scattering function, generally referred to as elastic structure shown in Figure 4. It is seen from the unit sc~t~eri~~. cell that there are 42 possible silver sites to which the In an actual coiidensed system, especially as the four silver ions in the unit cell can migrate. The low conaplexity increases, there will be coupling between 50 ineV per ion, activation energy for migration, the different types of motion and the simple forms means that the conduction process is almost liquid-like given above will be replaced by more complicated above the transition. Considerable activity, measuring evertheless, the distinction diffuse neutron and X-ray scattering in these phases, between the four dynamical regimes will generally be took place in the late l970s, driven in part by the meaningful. technological demand for advanced batteries and energy storage devices. A Fourier map of the silver rzeulrola
-
s of ses
The history of fast-ion conductors dates from Faraday's (1839) report of high ionic conductivity in bF2. Since then, many ionically conducting solids ve been identified, primarily high-temperature phases of ionic c o ~ p o u n d sin which the current is carried by unusually mobile monovalent cations or anions. A review of neutron scattering studies of their dynainical pro~ertieshas been given by Andersen et al. (1987), and a recent overview of fast-ion {or s ~ ~ e r i o ~ i c ~ conductors has been given by Agrawal and Gupta (1 999). A number of fast-ion iaternietallic compounds have also been identified, often called mixed conductors because both electrons and ions have high mobility. These ~ ~ t e r ipresent a ~ s an additional scientific interest erative mutual enhancement of the two components (Raniasesha, 1982), nit cell of the fast-ion (p) p 6b (octahedral) Ag' sites; as well as to their technological potential for use in Ag' sites; 0:24h (trian~ular)Ag+ sites (Reprinted from advanced battery electrodes. ost of these systems Andersen, N. H,, CLausen, K. N. and Kjems, J. K., ~ ~ a+ or Ag+ as the mobile ion, although Scufferthing,Methods of Experimental Physics Vol. 23, Ed. M. some divalent cases will be mentioned below. An Sltold a i d B. L. Prrce (Academic, NY), Part B, p. 187, importaiit class of Li+ conductors are the lithiated Copyright (1987) with perniission from Acadcinic Press)
~
251
Eflecls qf Mobile Sprcics
through. the 6b and 12d siles of the crystal structure shown in Figure 4. A simple model of a relaxation process in a solid is described by uncorrelated jumps o f a mobile atom to pz adjacent sites at distances with a mean residence time z between juiiips at each site (Ghudley and Elliot, 1961). In this case the half-wi~that h~lf-ma~imum of the scattering function in equation (1 6) is given by h I' El - exp(--i r2.z k=l
) =-
(17)
Broadening of this kind was observed in Q ~ e a s u r e ~ e i i ot sf Ag2S (Crier et d., 1984), reflectin a dynamic disorder on the time scale of the neutron experiments, 7-17 psec. Recent work on Ag2Se has been described by kazabi (19x3) and on Ag2Te by Keen an canonical case of an intermetallic conipound exhibiting fast-ion beh~vior.In the 1970s it was a material o f interest for the negative electrode in high-temperature, molten salt bat (Yao et al., 1971). It can be regar c o ~ p o u n d crystalli%ing in the structure, consisting of two inter~enetratingdiarnoiid lattices. Since this is a rather close-packed structure, Li diRusion relies on the relatively high ~oncentra~ion of mobile vacancy defects on the Li sublattice in the Lideficient side of the equiatoniic composition. Since silver and, even more so, both isotopes o f lithium have a substant~aldegree of ~ncoherencein their cross-sections, the distiiiction between coherent and incoherent scattering ment~onedinl must be taken into account. The i ~ c o ~ ~s cr ae~~t etr i n ~ *
Academic Press)
ion distribution in the (100) plane in the p phase of Ag2S at 325 "C, essentiall~~ ~in the ~sense of~ equation (5), is shown in Figure 5 (Cava et al., 1980). The result indicates an extremely delocalized Ag' ion d~strib~ition,with very ~ i g h - d e ~ s i tbands y passing
Np and unoccupied N , sites (Ander§en et al., 1987). If NI, << N,, the coherent scattering has a form similar to the incoherent:
~ ( ~ ) where SP( ) is a structure factor for an indi~idual particle taking the deforination of the s u r r o u n d i ~ ~ medium into account. On the other hand if Np >> N,,
252
Properties arzd P ~ z e ~ o ~ ~ e r z o ~ ~ g y
the coherent scattering 1s dominated by the motions of the vacasicies and we have (20)
conductor around 600 ture with silicon temperature it has a defect Lij I'orrning an fcc sublattice and distributed over tetrahedral and octahedral sites. An a/? i ~ molecular i ~ dynamics ~ ~ computer simulation (Wengert et al., 1996) shows that both cations are mobile, with the lithium having diffusion coe~cients 3-4 times those of the ma~nesium,and the diKusion taking place via vacancy migratioi~through directed jumps. To our knowledge, neutron scattering experiments have not yet been carried out.
by the (Brun et al., 19S1) found that the observed scattering could be described by the incoherent term, essentially equation (18) modified for encounter effects (Wolf, 1977), with the coherent tern broadened so much that it could not be seen above the background. s Ln addi~ionto ma~erialswith mobile Li4, Na4 and Disordered hydrides represent the other importnn t small number of fast-ion conductors class of solids in which diffusion occurs on a time scale have been observed. Mg3Bi2, which e that in liquids and are therefore her ch~lco~enides already ~ ~ ~ t i o n e ~do n ~ p a r a b lwith readily amenable to investigation with QENS. In view some unusual transport properties in the liquid (Xu et of their intrinsic interest for solid-state physics, and al., 1993b; Enderby and Collings, 1970), transforms at also because of their t e ~ h n o l o ~ i ~i ~ a lp o r t a i ~ cin e 703 "C from the l o w - t ~ ~ p e r a t uanti-La203 re structure energy storage, fwion technology and hydrogen to a B phase whose structure is not known but has been embrittlernent, metal fiydrides represent one of the arms et al. (1994) as a fast-ion early successes of neutron scattering ( ~ ~ r i n and ~er cond~ictiiigbcc phase. As in silver iodide and the Richter, 1987). The predo~inantlyincoherent scattersilver chalcogenides, cation motion appears to take ing cross-section of hydrogen was also an important place between the tetrahedral and tnangulrtr sites in factor in i n t e r ~ r e t i nthe ~ results obtained. the bcc lattice. A classic e ~ ~ e r i m e n t result nl is that of vior is found in the ternary Zintl Nelin (1967) on fcc palladium, reproduced in Figure 6. , which appears to be tt hst-ion
Widths of ~uasie~astic scattering in Pd-M (from Skold and Nelin, 1967). The solid lines represent fits of the ChudleyElliot model with the €4 atoms jumping between octahedral sites. The dashed line shows the poorer fit with the NI atoms jumping between tetrahedral sites (Reprinted from Skold, K. and Nelin, G., J . Phys. C h m . SoZ&ds 28, 2369, Copyright (1967) w t h per~issionfrom Elsevier Science)
253 It was found that the results could be nicely ~ ~ p l a ~ n e d ~ e q u a t i o(18-19)) ~~ for ween oc~ahed~al sites, vibration~lai~p~itudes ~ u r t ~ e r ~the o r iii e erstitial site could be obtained o f the ~ e b y ~ ~ ~factors ~ l l e r for bcc metals were more r e ~ n e ~ e nin t s the theory, for e ~ ~ the ~ p‘ ~ ~o ~e j~m lq ’e model in which the hy~rogenalternates between ‘mobile’ and ~immobile’ states ( ~ o t t et~ a/., ~ r ~979).V ~ n a d i uliy~ride ~ is a lzlt four ~ ~ ~o f iieutron i ~ es cs~ t t e r ~ n ~ tion 2.2 have been b r o ~ g h t into lay. For exaniple, a ~ o ~ b i n aof t ~recoil o ~ scatter in^ Lufr;mn et al., 1987) and vjbr~tion~L1 spectroe ~ ~ e l nefi al., ~ ~ 1989) ~ i has been used to dete~1inethe shape of the interstitia~potentia~weil in which the hydroge~atonis sit (Figure 7). For systems with higher hydrogen c o i ~ ~ ~ n t r aitt i ~ ~ n ~ has been possible to use both the incoherent scatter in^ ~ r o s s ~ s e c t i of o ~ hydrogen and the predomi~~ntly of d e ~ ~ ~to eobtain r i ~ ~ coherent scattering cross~sect~oi1 7 Potential af ~ ~ ~ ~ e r s NI t i tatoms ~ a l in the d~~l~rdered a more profound ~ n d ~ r s t a n d of in~ the dynamics. As ~ ~ discussed above, these often give ~ o ~ pkinds ~ ephase~ of V,H ~ deduced ~ ~€ramt inelastic ~ ~~ ~ ~u~ ~ r o~(from ~ e ~ ~ ~et ~al., e1987). l ~ The a solid n ~ curve ~ for tlzc exc~t~~tion o f in~orniation,At srnafl Q3the first can he related to energy (left and top axes) refers to the c direction in the crystal the s ~ ~ ~( ~ t /~ a c~~e i~~-’u~s i~oconstant n ~ ~D and ~ ~ while ~ thel dotted ~ curve refers to the a direction. The potential the second to the c ~ ~ ~ i (‘co~lective’~ cal d~ffusion (right and bottom axes) shown by the solid curve also refers to the c direction in the crystal and corresponds to the sum of constant Dch ( ~ ~ ~ i and ~ g Richter, e r 198’7; Hempelterms ~ ~ a d ~~a t~ i ~ and ~ hqrrartic e d ( ~d ~ ~ s ~in“ ~the ot~ ~~n~~~2 ~ ~Tlie 0 ~two . are co~nectedby the relation d i ~ p l(~ R ecp r~~ ~€ram ~~d ~~ ~ ~ p R.,eRichter, ~ ~D. a ~ aiid Price, D. L., Pkiys. Rev. Left, , 1016, Copyright (1987) with ~eri~~ssion from the A ~ e r j Physical c ~ Society) ~~~~~
n
where c is the liydroge~concentr~tionandf;. is a factor cracks that are larger than the inverse s~atteringvector ~ e s c r i ~ icorrelations n~ between s~~~cessive jumps. QENS ni~asureii~ents on rx-TaI-l were sizawzz to be of the neutron measurement, ther classes of alloy include the (cF24) Eaves phases such as consistent with equation (21) CO bmed with ( ~ studied ~ / ~ G15 ~ ~ determined by the Gorski relaxation method TriVzfl1.1 ( ~ k r i p oet~ al., 1998~and (Skripov et a l , 1999) al., 1984). A ~ o n s i ~ ~ ~ body a b l e of h i ~ h - r e s o l u ~ i Q ~ observed, one correspo ~ e a s u r e i ~ e has n t ~ been carried out on in~er~etallic hexagons formed by tetrahedral inter~tztialsites and alloys, m ~ of ~it driven ~ ~by ithe ~ ~ t e rin~ ~s ty d r o g e ~ ~ the other to jumps bet~veenthem. ~ ~ t r ~a tst e~~ ~~ ~~a ”j c tion ~ ~ a s u r e ~ e on n t sTi-Zr-Ni q~ts~crystals have storage materials such as LaNigzfh (see Ch. 21 of Val, 2 of this work by ~ c l ~ l a p b aet~al.). h Work up to found that hydrogen motions around 220 K are about 1987 was reviewed by ~ p and Richter ~ ~(19 n ~an order ~ of ~ a~g n i t ~ ifaster d e than in the ~ ~ ~ e s p o ~ i ~ p r ~ ~ ~the i ~dli ~e ~ u s ~c o ~n ~ s ~~ ~e n ~t by s u ~ a~p ~ ~~ o x j ~(Foster a n t et al., ~ ~ ~ 0 ) . should be related to the reaction rate co~stantKr for absorptio~or ~ ~ ~ o r pint ai powder o~ of grain size R: ~~~~
n’D Kr = R2
In practice the measured ~ e r i c ~ ~rate o n may be siowed by surfttce effects or speeded up by d i ~ ~ ~ s along ion
~ r y s t a c~os n ~ a i n~i~olecu1es i~~ or complex ions have 8x1 additional c o ~ p ~ e x i ct yo ~ ~ ~with r e dthose d i s c ~ ~ s e d up to now in that they possess o ~ ~ e n t ~ ~ as t i owell ~ aas l tra~islationaldegrees of freedom. They can t ~ e ~ e f Q r e
exhibit a variety o f types of orient~tiona~and translatio~a~ order, A ~ o ~ p r e l i e n s ~recent v e review spectra have the shape o f t h has been given by KI and he^ (1984). fu~cti~ ofnthe i n s t r u i ~ ~In~ the ~ t .a phase they show a They point out tha n-rotation coupling is ~ ~ ~ p ~ r ~o f ou ~ s ~i rt o~a od e~nand ~ db ~ ~ a d e compo~ed resent in all systems of tliis kind but is relatively more nents that can be well represented by a delta ftinction ~ r n p o r t ~ ~in n t ionic c o ~ ~ p o ~than n ~ in s ~ o ~ e c u ~ a and r single ~ o ~ e n t f~ ui ~a ~~ c t ~~ ~ ~s i~, ~ ~both ~ i v e ~ y crys~a~s since the ions are separ~tedby a lattice of broadene~ by the reso~~~tion f ~ n c ~ o The i~. cou~terioiiswhich wea~ensthe direct interactions ~ ~ ~ n c t i no tne n s ~ tpeaks ~ in the region of the ~ e t w e e ~~i e i g ~ i b o orie~i~~tions. ri~~~ We may expect iit, peaks at Q’ f A-I, but has a ~ i ~ n i ~ c a non-zero this to be even truer in those s e ~ i c o ~ i d u ~or t i n ~ value over the rest of the measured inter~etallic ~rystals~which exhibit o r ~ e n t a t ~ o ~ a lLaren t~ian-functionintensr ty has a n ~ ~the mobile electrons will degrees o f f r ~ ~ d owhere to the left of the Bragg peaks (Q screen the direct ~~ite~actions even further. starts to rise to large values at Two interrn~t~~llic compo~ndshave been identified dramatic rise in inten~i~y above 1 -5 in which ~~~~~~~~j~~~~~ as well as ~ ~ ~djsorder~ ~ of that , found ~ in~NaSn.~ The ~ ~ appears ~ ~ a sum of~ the two to N
-
n ~ i ~ b oefr ~ ~ ~ i sproperties ual arisii~g was a d o ~ t e din which ~ s units~ were~assumed b to~ from the fori~ationof tetrahed~alSni- or Phi- poly~ n ~ eindepe~ident, r ~ o ~ a n j ~u ~ between ~o ~~ s ~the four ions’). ~ r y s t ~structures l of these or~entat~ons found in the p crystal e ~ o ~ y a n i o nofs this type arra~~ged cm succession as one moves up the c ax a k c ~lattice, separa~edby a~kali-met~l ions, each of ~ a b o u n ~1991), i , Since these orientati~ns which is shared by two ~ o ~ y ~ n fi o ~ s spaced, two jump times, “ ~ tand zzlwere a s s u ~ e dfor ~ o e n i a ~1953), e ~ , In the ~ ~ ~ ~~ ~ t~ r~ o~ ~~a i~e tt t ie ornearest~ s~ and nr=xt-nel2rest-ne~gfibor jumps, ~ ~ s ~ e c t i v ~ ~ y f $lie l ~ q u si t ~ r o~n ~~first sharp d i ~ r a c t ~ opeaks n The quasie~astics ~ a t t e rspectra ~ ~ ~ c a l ~ ~ ~ fram ate~ are o ~ s e r v ~ ddue , to dense random p a c k ~ nof~ the this model showed both ~ ~ n ~ r and ~ b~r od~ ~e ~ ~e n e d ~ ~ofyanions( ~ e i j eet r ~al., 1989, 1990), The ~ ~ s o r ~ e r icompoiients ~i~ similar to the measured ones. The is a~sociated with dramatic changes in electrical i n t e ~ r a t ~inte~s~ties d of the two c o n ~ ~which ~ ~ ~ ~ t t r ~ ~ n $ ~~ropert~es. ort are inde~endentof the values of 71 and zzl are in The behavior o f CsPb is a prime example of an excellent ~ g r e e ~ ewith n t the measure orientatiotiaffy-disordered phase in an ii~ter~etallic over the diEerent runs ~ ~ i g 8). u ~ e only Phi- ~ o l y a n ~ o nr se o r ~ i~n~d et p ~ ~ i ~ edo ~ t not ly reproduc~ the first peak ~ o r e ~ ~i ~~ ~~ ~i o ~ atn~ e~ ~ i t cations must therefore be p in the dynamic disorder, apparently on a t ~ ~ e ~ s ~ a l e siniifar to that of the Phi- reo one line width i s observed iii the a phase. This behavior contrast as will be ~ ~ i ~ ~ in u s Section s e ~ a ~ ~ i rwhere, e The s t ~ ~ ~ ~ft u~ r~e t of ~ rthe s l o w ~ t e ~ ~ ~ e r @) p o l y a i i i ~ro~ations ~~ and cation n i o t i o ~a~ ~ ~ to~ a r the liquid were similar to those take place oii diEerent time scales. The d,c. electrical cond~~ctivity of CsPb rises s h a at~ the ~ j3-a ~ transition, indicatin~a s i ~ i ~ i ~ cdecrease a ~ i t in the gap in Q 1 A-1 that appeared to be a B(E) (Fortner et al.? 1995). It i n c ~ ~ slow~y ~ ~ e swith agg peaks with a diEuse, nont e ~ p ~ r ~int the ~ r xe phase, and drops ~ ~ ~at~theh t ~ y er c ~ y s t a ~ I pattern ~ n e at l ~ r ~ Q. point. The overt~ll~ isuggested ~ bytthe d ~~ E ~~ ~data, c t~ i ~melting ~~ More recently, cage structLires based a n a crystalliiie phase with e ~ t ~ ~ s idisorder, ve was ~ ~ ~ ~ e ~e l~es~ e un t ssuch ~ as ~ In~eand Tl have ~ been~repor ~ ~ ~ o n ~ r n by i ~ dii~e~astic neutron s ong and Corbett, 1993, 1994, nients carried out on the IN-6 s ~ e ~ t r o m e at ~ e the r 1995; Sevov, tliis volume). These have ~ o I y a ~ i oofn ~
higher wave vector decrease, p r e s u ~ a bon l ~~ccountof i n c r e ~ ~ idny~n a ~ i c ~disorder, l The ~i~~~~ str~ctul-~ shows a ~~ronoLinced 'first sliarp d ~ ~ r a c ~ peak' i # n at about I A-" typical o f liquids with large polyanions, and similar to that observe liquid KTI, CsTl and GsgT17 ed al., 1 ~ 9 ~ ? * The d i ~ ~ a c tresults i o ~ are s ~ i ~ ~ o rby t eqd~ ~ s i ~ l ~ s t i c
CsgT111 appears to be a rotor phase tei~peratures,while anothe~c o ~ p o ~ n ~ ~ shows no d y ~ ~ ~ ratation ~ ~ i c but a ~ seems liigldy disordered structure in the solid phase at all t ~ m p e r a ~ ~ studied. i ~ e s Tlie a ~ s e ~ cofe a change in crystal s t ~ u ~ ~a L ~~ ~~e ~ ~ the ~ aa ~~ ~ ~~ y~ of~i ~n a~ ~ dynai~icaldisorder in CsgTlll is in ~ i a r k ~contrast d to the belinvior observed in the I-JV CsPb discussed above and N 3.4. The r e ~ n ~ n i e noft the Corbett, 199.5) shows the Tl:, polyhe~ra~ ~ ~ iin ~ i t ~ a l ~ e r n a to~ientations ~ parallel to the c ( T 2 - ~ 2axis. ~ DTA. analysis showed an ind~cnt~un of a small thermal anomaly at 381 "C, but none at 294 "C where a transition is indicated in the ~ u b ~ i phase s h ~ d~ ~ a ~ r a ~ ~ ~ r u c t ufiitctors rc for (a> L o r ~ and ~ (bj ~ ~deltaa ~ ( ~ a s ~ ~ ~l ~9s ~ k ~owever, i ~, ~ . if there is a ~ i s o r ~ e ~ i ~ i ~u~i&tion peaks in the d ~ ~ ~ r ~solid e r ephase d of CsPb (880 K). t xn t ~ a n s ~ titi oappears ~ ~ to take place w i t ~ ua~change Circles: data: solid lines: resuits of the model described in the crystal s t r ~ ~ t u r e . text. The t c a ~ c u ~ structure ~ ~ e d factor in (€3) has been ~ r ~ a d e i ? e ~ with the ~ ~ ~ ~ ~ .Quresolutioa i ~ e n f~ ~a ~i ~~ c iR ~ ~~from ~~ r ~ ~ ~ e ~ Price, D. L. and ~ a b ~ uM,-L., ~ i ~ Phys. ~ " R ~ YB. 44, 7289, ~ o p ~ r(1991) i ~ ~wxtk t ~ ~ e r ~ i s sfrom i o n the American Physical ~ o ~ l e ~ ~ ~
e x t r e ~ ecases the traiistation-rotati013 c o u ~ ~ ~ ~ becomes so strong that the two degrees of Geedoin disorder together at the phase t r a n s i ~ ~ The # ~ "~ r i ~ e s the ~ r o u ~metal - ~ ~and I n example of this hehavior is found in ain the smallest p o ~ y a n ~to o ~ s analogous sulfates ~ ~ u n d e 1994 n ~ 8nd as high as '70, 74 and 78 in fullerene-like cages. These therein), where the motioii of the l i t h i u ~ions has phases bear the same relations hi^ to 1-111 diamondbeen given the p ~ ~ t ~ ~ ~ e s ~a Lof~ ~e ~~ ~e ~ ~ l e struct~~re Zintl phases (e.g. LiAl, NaPii, NaTI) that the migration. Recently similar behavior was found in ful~ere~es have to diamond. the i n t e ~ i e t compound ~ ~ l ~ ~ sodiu~-t~~i. The iso ordered solid and liquid phases of CsgTl11, NaSii has a complex melting behaviar, ~ x l ~ i b iin te~ ~ a and ~ ~a ~ K ~have 2 ~ been T ~~ s~udied ~ ~ ~by thel c ~ l ~ ~~ i ~data ~ e~t ~r ~ ~c o et u ial., i g1994) i shown in ~ ~ b et al. ~ ~~~ O ~~ These O ~) , g~ l i ia s ehave ~ ~ e l a t ~ v e ~ yFigure 10 along with that from CsPb. The ex1 spherical ~ o ~ ~ n i of o nthe s form Tliy, Tlii- and a two solid phases was first establi~~ed by ~ o ~ ~ b i ~ ~ofa 77:t i o naiid Ti;-, r e s ~ e c t ~ v eand ~ y ~are Rothery (1928) who did not, however, s Likely c a n ~ i d a t ~for s ~ y n ~ i c a d~sordered ll~ rotor about their nature.. The crystal struct~reat low igure 9 shows the structure factors in the tei~peratu~e ( ~ u l l and e ~ Yolk, r u o ~ - t ~ ~ p e r a t uand r e ~ i g ~ - t ~ n i ~ e rsolid a t ~ rand e in CsPb and the other Zintl corn liquid phases for CsgTlll (x1i.p. = 408 "C). Tlie with the alkali metals (except t e ~ i i ~ e r a att t ~ tern ~ e is in good a g r e e ~ e nwith ~ ray ~ i ~ r a c t results i o ~ (Dong and Col-bett, 199.5). the ~uth~rford-Appleton eating toward the meiting point, the crystal st~ucture Source (EIS) ( ~ a b ~ ~ ~ ~ ~ i does not cl~angebut the intensities of the peaks at 1995) exhibit Char~Gterist~ 111
256
b a
2.0-
Y
n s
1 5-
P e
1.0-
r
d
m Q
1
0 *5-
e C U
6-
1 e
P e
-0
*
s-
-a
*
0-
r E3
r
-a.
5-
0
2
2
3 4 ~~~E~~~~
5
~~A~~~~~
6
7 :
8
8
9
10
f A * " - l )
Structure factors S(Q) as a fu~unctionof wavc- vector Q for Cs,TI,, at room temperature (lower curve), 390 "C (middle curve) and 450 "C (uppcr crrrve). C~irvesare displaced s u c ~ ~ s ~ upward v e ~ y by 1 (from S~~~~~~~~eTt al., ~~~~) solid and liquid pl~ases.~ s p e c ~ ~remarkable, lly in view of tlie signi~cantincoherent scattering cross-section of Na, is the a~senceof o ~ s ~ r ~ ~diffuse a b l e elastic scatter~ngin the CI pliase, which is zero w i t ~ i nexperimenial error when averaged over all aiil;les except those c o r r e s ~ ~ n d to i n ~Q vahes near 1A-1 wliere there is considerable ragg scattering" The absence of s ~ g n ~ f i incoherent ~~nt elastic scattering indicates that tlie d y n a ~ i cdis~rderassociated with the Na+ ions is t r a n s l a t ~ ~ i ~rather ~ l , thaii o r i e n t ~ ~ i o ~in ~ a lnature. , The widths of the single Lorentzian fL~nction~ fitted s a f ~ ~ ~oft Q~ mo n ET chopper spectr whose width and intensity both increase considerably at higher Q. This component is not seen in the IRIS data at higher Q, beca~seit is too broad to show up within the IRIS window (A 0.4 meV). 0 1 2 the other barid the IRIS data show a much arrowe er component whose intensity
11, the ~ e a s widths ~ ~ of r the ~ 1~ IS data are seen to be in reas~~nablea ~ r e e n ~ ewith n ~ the ~ h ~ ~ ~ ~ ~ y - ~ l d ~ s t a ~ c1 eof 3.75 A ~ s i ~ i l ator ~~~~~~c~~ in the 6 crystal) and a ps. It i s therefore reasonable to associate this rrarrow ~ o ~ ~ o with ~ e the ~ i Ma" t ~ i ~ r a t ~and o nthe broad compone~twith the more rapid r e o r i ~ ~ ~ ~ of t ~ othe n s Sn2- pol yanions. The absence of s~gni ~cant elastic diEuse ~ ~ a t t eindicates ~i~g that these reorjent~~io~is are d ~ u s ~ 111 v en~t ure. Taken together, these results show that the a phase of NaSn is ~ y ~ a ~ i ~ o~r d e~r e dThe ~, ~ d y~n ~~~ ~ yi i c disorder consists of rapid r e o r ~ e ~ t a t i oonf ~the polyanions (fast rotor be1iavior) which enhance the slower cation migration (fast-ion ~ ~ ~ ~ d this ~ ~ ~ it~ i~ oa tis ~i o)n : schei~atica~ly illustr~tedin Figure 12. A l ~ h o u ~the h jump frequencies are different, the two processes must be stro13gly coupled since only one phase ~ ~ - a ~i s s ~ t observed prior to melting, ~ ~ 1 e c ~ ~ ~i rcaai ~ i s p ~ rint aSii is c o ~ p ~to e the d y n a atomic ~ ~ ~ d~ ~ s ~ ? r ~~e ~ * .~ ~of the~elec- ~ trical con~uc~ivity ~ ~ ~ ~ei al., r t 1995) n e show ~ a cfrop in two ~ ~ ~ p u agrees ~ e ~with t s the total s ~ ~ ~ t e r~ ~~ n d~ u ~ ~atithe y i t ~ t ~ ~ ~~ nr ~~ ~~ reflect~~ ~m ~a n~ measured in the diffraction detector and hence ing the additional scattering associated with the a c ~ ~ u n for t s the entire diRuse scattering. In Figure dynamic disorder; in contrast, any ionic ~ o m p o n e n ~
li
~
~
,~
F
** d
e ? +!
4
e
1 f
at h ~ ~ of the ~ ntzian peaks for a-NaSn at A HET (40 m e Q the solid j u ~ ~ ~ d i ~niodet. u s i ~The n dotted and dashed lines joining the points m the upper curves arc a guide to the eye ~ ~ e ~from ~ P~ ~ n t e ~ M.-L. and Howells, W. S., Phys. Rev. B f 1995) with ~ ~ r ~ ~ sfrom s i othe n A ~ ~Physrcal ~ i Society c ~ ~
~
the alkali metals in the disordered phases of C24Rb and C24Cs with ~~~~. The i i i ~ ~ r ~ r e t awas t i oc~ ~ ~ ~ l i ~ by tlie fact that carbon is a coherent scatterer. Usin the simple ~ p ~ r o x ~ ~ i(aV~j ino~ny ~ 1958) ~d,
of the me~~sured conductivity must at the tran~ition.‘The large jump within the ct phase reflects a close c o ~ ~ ~between ~ ~ n ga t o ~ ~mi~ration c and electronic mobility, discussed in depth by Fortner et al. (1995). There appears to be little if any change in they e x t r a ~ ~ e ddiffirsion constants of the condu~ti~ity on melting. NaSn appears to translW5 cm2 s-l, much smaller than the vsrlues observed fomi to a crystal~ograph~c phase of lower symmetry, in the correspoiiding bulk liquids at this temperature apparently moiioclinic ( ichardsoii et al., 1996) at the and sL~ggestiveof a s u b s ~ ~ i i ~ i increased a~ly degree of disorde~iIi~ tran~ition~ an unusital example of lattice local correlation. symmetry lowering at such a transition, The most dramatic change in the c a r b o ~ - ~ ~ t a l A related phenomenon is proton diffusion m hydrogensystems came with the discovery o f C60t C?o and their bonded crystals by means of a ~ r o ~ t h ui~ech~nisrn s ~ o m ~ o uand n ~ alloys s in the M e 1980s. The crystaI~iIi~ al ii~volving a l t e r ~ a t i n ~ rotatio~al and t r a ~ i s ~ a t ~ o ~ phase of C60 itself undergoes an orientational dis4uiiips (Lechner CL al., 1993); however, this situation ordering tr~~iisitjon at 250 K, first studied with QE differs from the present case in that each mobile ion by Neumann et al. (1991). The doped alkali metal a t ~a~single center. site can be a ~ s o ~ iwith ~ ~ ~ ~ iqZiC6o., o u xn== dI > 3,~ 4 6, . e ~ h ~ bai range t of fascinating properties, including SuperconductiQity and ~~etal-~nsulator t r ~ n ~ i t i o nat s low ~ e ~ p e r ~ ~ i 3.5 The AI c60 cornp~und$have ~ i ~ h ~ t ephases ~ ~ ~ r a with the rocksalt struct~re,which exhibit s t r ~ c t L ~ ~ ~ ~~itercaiatio~i conipounds of alkali metals in graphite phase transitions oii cooling to orthorhornbic strucr e ~ r e ~ e an ~t w o ~ d i ~ e ~ i s ~ionnt a~~~ ~ e ~ asystem. llic tures with tiiiear p o ~ y ~chains e r of G ~ o These undergo disorder in^ transition^ in which the et al., 1993; Stephens el aE., 1994). layers of metal atoms behave essent~a~~y as 2d y n a ~ o~ rci ~ n t ~ ~ i odisorder, n a ~ ~ppareiitlyd e c ~ ~ ~ l e d Zabel et al. (1983) studied the diffusional motions of from the alkali metal motions (Christdes e~ aE,, 1992). ~~~~~~~~~
.3
32f ~ c o r n e sites ~ ) fully o c c ~ i ~and i e ~24e (hce center)
tiott model ( a $ s ~ ~ i a11 ng scatteriag to be incohere tfi, r e m ~ ~ k a bthe l y ~same jump time of 2.7 f 0 , 5 ps above and below tile is ordering transitio~~ ~ r i s t o ~ edl i al., n ~ 2 ~ ~ The 0 ~ . i~teyisityof the ~ ~ a t t e~ r~i e~ s~ u~~~~ ~ b l ~y ~ to o r t i ~ the iiumber of mobile ions, i n ~ r ~ a s~e s~ ~ b s t a nand ~~all~ seems to be closely c o r ~ e l ~ ~ twith e d the t e ~ r a g o i ~ ~ ~ dist~rtion of the cubic lattice, which must open
In this last section we address a type of dyiiarvric disorder which does not, like tbe o t ~ c ~o i~~ s~ ~ so de~e~ far, precede actual melting, ~ ~ s t it~ occurs a d near a phase trans~tion~ with an inten~itythat ~ecreaseson m o ~ away i ~ from ~ ~ the tra~sitionon both the highand ~ow~temperat~re sides, eve~th~less,it has features in c o ~ n wth ~ ~ the ~ i~~~t~~~p h e n o i ~ e i i ~ n that justify inclusion in the present review. Furthermore, orientational disorder in^ ~ransitio~s are o a ~ ~ o m p a ~ by i e dfer~oe1a~tic i ~ ~ t a ~ i l i t(i e~s y ~ ~ e n ~ and Michel, 1994). ~ e r r o e l ~ s t itra~sitions c by a ~oftening of ~~aasverse acoL~stjcmodes. ever, in contrast to conventioiial melting, longrder i s ~ e ~ e s ~ ~ b l i s h e d on cooling below the trai~~itio~i by a process of contin~~ous crysta~l~~ation ( arris et at,, 1993). They are classified by s y ~ ~ ~ e type t r y m, d ~ ~ e n on d i ~ ~ whether the acoustic i ~ s ~ a ~ i occurs l ~ t y at points ( m = 0), lilies ( m = 1) or planes (nz = 2) of reciprocal space (Cowley, 1976). A classic case is that of the alkali ~ i ~ane ~ cyanides, where the t ~ n s i t i o nis a c c o ~ i ~ p a by alignnient of the CNdipoles but with head-to-tail S c ~ ~ il~~strntion n ~ ~ tof ~~ a~d d I ~ ” ~ ~mrgrntion hee1 111 a-MaSn (from SnboLi~giet al., 1993) ~ R e ~ r i from ~ i t ~ ~disorder fluty, 1981). A typical case in a metallic alloy is that of I ~ J - ~ T(xI = ~ 0.16-0.31) alloys .-L.,Fortner, J., Howclls, W. S. and Price, D. L., 237, Copyright (1993), with permission from (Cunton and Saunders, 1974) where the elastic ~ ~~ ~~ Limited) 1g ~~ ~~ ~~ n1 ~ ~ - C J~ Z~) /tends modulus (C11 ~ to zero as the niar~ensiti~ traesitioii is approached from either the cubic or the tetragonal side. The best-known ferroelastic transitio~sin metallic While x = 4 appears to be the doping limit for the coinpotxnds such as heavier alkalis, values a s high as 12 have been found , 1966) and Nb3Sn with lithium and sodium. KohanoR et at. (1992) proposed on tbe basis of crb initio computer si~ulations dispersion relation A mP( that Li,,CbO should he a stable, symmetric cluster: this the [ l i O ] crystal direction and was subse~uentlysy~it~esized by ~ristofolin~ et ale rnents along [ 11Oj as the tram ( ~ 9 who 9 ~ also ~ n~easuredi t s crystal structure in both cubic to a s ~ ~ g ~dt il sy t ~ ~ ~tetragonal ed structure, is high- (fcc) and tow- ~ t e t ~ a g o i i atemperature ~) phases, approached from above. A somewhat unexpected In the fcc phase the Li4 ions reside in tlie octahedral result is that phonon s o f t e ~ ~ ~occurs n g not just in the voids of the f~~llerene structure, with 4’0 (center) and
Eflkc~so j Mobile Species
259
ure 13 Intensity of the QENS, after Debye-Waller correction, 111 Li,,C,, at 450 and 600 K. Thc continuous lines represent the th.eo~etica1fit to octahedral JUIII~S with. a jump time of 2.7f0.5 ps t~cprintedfrom Cristofolini, L., Facci, P., Fontana, M. P., Cicognani, C . and Dianoux, A. J., Phys. Rev. I3 61, 3404, Copyright (2000) with permission from the American Physical Society)
ion, although this is what drives the transition, but at large values of Q and also for along the [loo] direction. Furthermore the s phonons develop int 'central peak', i.e. scattering intensity centered at = 0, a phenomenon that is a coininoii feature of many ferroelastic and ferroelectric transitions that is not generally understood (Cowley, 1987).
The body of work reviewed, even though a limited selection of the large amount of re1ated studies in the literature, shows that most known types of dynamic disorder and disordering or partial melting transitions can occur in intermetallic systems, although they may be rarer than in ionically or molecularly bonded compounds. In fact, some degree of strongly local or directional bonding appears to be essential for the
a kinds of disorder we have c o ~ s i ~ e r e dFrom . structural point of view, this generally leads to a substantial degree of interniediate-range order, which survives into the disordered phases as well as into the liquid. Except in particular cases where there is a large cliange in the electron structure ([or example, melting of silicon, which takes it from a four-fold coordinated semiconductor to a six-coordinated liquid metal), the short-range structure does not change significantly either. Usually it is the long-ran changes: even though the disordere crystalline symmetry, in general it will be different from the ordered phase. In fact, often only the Brag peaks at low scattering vector, which c h a ~ a c ~ e ~the ize intermediate-range, remained well pronounced. The directional bonding may lead to a band gap at the Fermi energy when the intermeta~liccompound is formed, producing a serniconducting rather than a metallic system. In maiiy cases, however, this gap fills in, at least to some extent, at the disordering tra~sition.
Agr~iwal*R. C., and Gupta, R. K. (1991). J . Muter. Sci,, 3 1131. Andersen, N. H., Clausen, K. N,, and Kjerns, J. K. (1987). ~ ~ S ~ z a t ~~~ ~c t ~ h o~i d ~sof~ ~r , x p e~ r ~ ~~Physics e ~ i ~t ~ ~ ~ Vol. 23 (eds K. Sklild, and D, L. Price). Academic, Ncw York, Part €3, p. 187 A s ~ ~ r o fN. t , W., and Langreth, D. C . (I 967). PAYS. Rev., 1 685, ~ o w e W, ~ ~S, s(1994). ~ J. Phys.. tt, C. S. (1966). Phys. Rev., P 296.
Phonon dispe~~ion relation A cop long the tll0l crystal direction ~~splacementsalong [I 101 as the transition at 45IC is a ~ p r ~ ~ i cfrom h ~ dabove. The unit along the abscissa IS C where = Qv = <2x/a and a i s the lattice constant. Z. B. refers CO the zone boundary which is at = 0.5 (Reprinted from hir ran^, G, and Axe, J. D., Pkys. Rev. Lctk. 2'1, 1803, Copyrig~t(1941) with permission from the American Physlcal SO~~~tY)
Cava, R. J., Reidinger, F., andU'uensc~i,13. J, (1980). .I. Solid State Chem., 31, 49. ~ ~ i ~ i s t ~ C., d e sNeuniann, , D. A., Prassi~es,I<., Copley, J, R. D., Rush, 1. J.? Rossansky, M.J. and Haddon, R, C. (1992). Phys. Rcv., ~ l ~ ~ dC. ~ T., e yand , Etliot, R, J. (1961). Proc. Roy. Soc., 77, 353, Cowley, R, A. (1976). Pkiys, Rev. B, 13, 4877, Cowley, R. A. (1987). ~ e ~ t r oSn~ a t t e r ~ ~~g ,e t h of ~ ~ s (eds K. Skold, and D. L. ~ x ~ e r ~ Physics ~ e ~ Vol. ~ t 23 a ~ Price). A c ~ i ~ e ~New i c , York, Part C, y . 1. ~ristofoli~ii, L., Ricco, M., and De Rcnzi, R. (1999). P h y . Rev. B, 59,8343. Cristofolini, L., Facci, P., Fontana, M. P., Cico~nani~ G., and Dianoux, A. 3. (2000). Phys. Rev. B, 51, 3404. Dong, 2.-C., itlid COrbett, J. D, (L993). J . AWEY. Che~n.SOC., 115, 11299. Dong, Z.-C., and Corbett, J. D. (1994$,.I. Amer. Chem. Soc., 116, 3429. Dong, 2.-C., and Corbett, J. D. (1995). J , Sct., Enderby, J. E., and Collings, E. U'. (1970). J . ~ o ~ z ~ C ~ ~
~~~.~~~~
4, 161. Faraday, M. (1839). Article 1240, Taylor and Francis, L0ndon. .-L., and E n ~ ~J.r E.~ (3995). y ~ Phy.~. ~~~~~~~
This ~ r ~ mobile ~ ~electroils ~ ~or holes e s which, as we have seen, rimy interwt in ii cooperative fashion with the mobile ions, X t is this inter~laybetween iiiobife charge carriers, the n i o t i o ~of ~ ~ ~ iom n and ~ the ~ e t r a n ~ l a~~ n~ ~ c~~t ~ r o~l er ~ ~ t~i~~unt iao~of n scoinptex ictiis, that leads to the ~ ~ s ~ ~ divers~~y ~ a t i we n ~ encounter in these materials.
This work. was supported by the Ofice of ~ c i e ~ cUS e, rtiiient of Energy, under contract W-3 f-109-38 and by the ~ r ~ n CN ch
,I. P., and Dove, M. T. (1993). Pliys. Rev. Left,, ~ ~ ~ ~ eR. ~(2000). ~ ~Q a n n , Solid State ~ ~ ~ Clarendon, ~ ~ Oxford, ~ ~ UK. 1 ~ ) ~ ~ e ~ p ~ ~ R., ~ aRichter, i i n , D., and Pnce, D. L. (1987). P h y ~ . ~ e ~ p e l m a i i R., n , Price, 17). L., Reiter, G., and Richter, 2). ~ ~~ 9 ~~~ i ~. ~ t ~)~~ ~(ecls ¶ ~R.~~S, I oSilver, n s and e P. E. ~ Sokol). Plenum, Mew York, p. 213,
.
Hume-Rotliery, W. (1928). J. Chern. Soc., 131, 947. Johnsoii, G. K., aiid Sabouiigi, M.-L. (1987). J . Clwm Plzys., and Hull, S. (1998). J . Phys.. Condens. Matt., 8217 KohanoC, J., Andrconi, W., and Parrinello, M. (1992). Clzem.
Saboungi, M.-L., Reijers, € T. I. J., Blander, M., and Johnson, G. K. (1988). J. Clienz. Phys.
wells, W. S., and Price, D. L. (1993). Nature, 365, 2.77 Saboungi, M.-L., Johnson, G. K., and Price, D. L. (1994). s Phase ~ r a n s ~ o r r n a t ~(eds un~ Statics and ~ y n a ~ t ocf ~Alloy , R., and Lamprecht, I. P E. A. Turchi, and A. Gonis), Plenum, New York, p. 195. T. (1979). Z. Phys. B, 3 Saboungi, M.-L., Price, D. L., and Corbett. J. D. (2000). 157. Private communication. Lunden, A. (1994). Solid State Iomcs, 6 Sevov, S. C., and Corbett. J. D. (1993). Sczence, Luty, F. (1981). l)~$cts zn I n s z ~ ~ ~ ~Crystals ting (eds V M. Shimojo, F., Okazaki, €3.(1993). J. Phys.. C ~ ~ n ~ ~ e n Turkevich, and K. K. Schvarts). Springer-Verlag, 3405. Heidelberg, p. 69. Shirane, 6 . . and Axe, J. D. (1971). Phy.s. Rev. Lett., 27, 1803. Lyndeii-Bell, K. M., and Michel, I<. HI. (1994). Rev. Mod Singh, R. N., and March, N. H. (1995). In~errnetallic Bhys., 66, 721. Compou~ds:Princ~lesand Practice (eds J. ET. Westbrook, Marsh, R. E., and Shoemaker, D. P. (1953). Acta Cr-vst., 6, and R. L. Fleischer). Volume 1 , Wiley, Chichester, UK I97 Skbld, TC. arid Nelin, G. (1967). J . Pktys. Chem. Solids, 2 Massalski, T. B. (ed.) (1990). Bitzury Alloy Phase Bmagrams, 2369. ASM Intl., vol. 1. Slcripov, A. V., Cook, J. C., Sibrrtsev, D. S,, Kannonik, C., Miiller, W., and Volk, M.(1977). Z. ~ a t ~ r f o r s c h3. , and Hempelinann, R. (1998). .I. Phys.: Condens. Matt., 1 Nesyer, R. (1988). Z . Krrstallogr., 182, 196. 1787. Neumann, D. A., Copley, J. R. D., Cappelletti, R. L., Skrtpov, A. V., Cook, J. C., K Kamitakahara, W. A., Lindstroin, R. M., Creegaii, IS. M., V. N. (1999). Phys. Rev. B, Cox, D. M., Romanow, W. J., Coustel, N., McCaulay, Springer, T., aiid Richter, D. Neutron Scatterrng, J. P., Jr, Maliszewsky, N. C., Fischer, J. E., and Smith, Methods of Experimental Physics Vol. 23 (eds M. Skold, A. B., I11 (1991). Phys. Rev. Lett., 67, 3808. and D. L. Price). Academic, New York, Part B, p. 131. Parsonage, N., aiid Staveley, L. A. I<. (1978). Disorder in Stephens, P. W., Bortel, G., Faigel, G., Tegze, M., Janossy. Cvystals, Clarcndon, Oxford. A., Pekker, S., Oszlanyi, G., and Forro, L. (1994). Nature, Potzel, U., Raab, R., Volkl, J., Wipf, H., Magerl, A., 370, 636. Salomon, D., aiid W o r t ~ a n G. , (1984). J. Less ~ u ~ ~ o n Ubbelohde, A. R. (1978). The ~ o l t e nState of ~ a t ~ e r , M d f i n g and Crystal Structure, Wiley, Chichester, UM. Opiiziori in Solid State and van der Lugt, W. (1996). J. Phys.. Condens. ~ ~ t t8,. 6115, , 8439. Verkerk, P., Xu, R.. Howells, W. S., de Wijs, 6. A., and van Susniaii, S. (1989). .I. Pliys. der Lugt, W. (1994). Price, D. L., Saboungi, M,-L., Vineyard, G. H . (1958). and White, R. (1991). Phys Wagner, K. (1933). Z . P Price, D. L., Sahomgr, M.-L., and Howells, W. S. (1995). Wengert, S., Nesper, R. (1996). Phys. Rev. Lett., 77, 5083. , M.-L. (1991). Phys. Rev. B, Winter, M., Beseiihard, J. O., Spahr, M. E., and Novak, P. 7289. (1998). Adv. Mater., 10, 725. Price, D. L., aiid Saboungi, M.-L. (1997). .I. P1zy.r.: CoIidem. Wolf, D. (1977). Sulid State Cbmrnuvi., Matter, 9, 2707 der Lugt, W. (1993a). J. Price, D. L.. and Skold, K. (1976). Neutrorz Scatterzn~, Methods of Experimental Physics Vol. 23 (eds K. Slcold, ., de Wijs, G. A., van der and D. L. Price). Academic, New York, Horst, F., and van der Lugt, W. (1993b). J Phys.. Ramaseslia, S. (1982). J Solid State Chevn., ond dens. Matt., 5, 9253. Reijers, H. T. J., Saboun~i,M.-L., Price, D. Yao, N. P., Heredy, L. A., and Saunders, R. C. (1971). J . J. W., Volin, K. J., aiid van der Lugt, W (1989). Plzys. ~ ~ e c ~ r ~ Soc.. ~ i z 118, e f ~1039. . Zabel, H., Magerl, A., Dianoux, A. J., and Reijers, H. T. J., Saboungi, M.-L., Price, D. L., and van der Phys. Rev. Lett., 90, 2094. Lugt, W. (1990). Phys. Rev. B, 41, 5661. Zhu, Q., Zhou, O., Fischer, J. E., McGhie, Richardson, J. W., Saboungi, M.-L.. and Price, D. L. (1996). W. .I., Strongin, R. M Ph-vs. Rev. Letters, 76, 1852. (1993). Phys. Rev. B,
This Page Intentionally Left Blank
ac
Fast-moving charged particles are known to ~roduce t r d s of atomic disorder in an i n ~ ~ i ~ e nvariety se of dielectric solids - including crystals, glasses, and high polymers. The literature on such tracks and their uses is vast - estimated to consist of some 8000 journal papers as of 1998 (~leischer,1998). The fact that tracks can be prod~cedin a special variety of solids, as reviewed by Provost et al. (1995) and Barbu et ul. (1995) came as a distinct surprise (Fleischer, 1995) - the reason for which will be explained after a brief overview of the phenomena and nature of conven~ionaltracks, ~~~~~~~~~~
The easiest, most widely used technique for observing al which nuclear tracks in solids i s by c h e ~ ~ cetching, preferen~~~lly attacks the disordered regions and enlarges the ~ e s ~ l t i n~g ~ l-ep sr o ~ u c ~pn~ r~r n a n e ~ t , optieafly visible features in solids (Figure 1 gives exa~ples~ These etched tracks have a r e ~ i a r k a b l ~ diversity of uses - in science to study nuclear reactions, cosmic rays, the ages of minerals, the distributio~of certain elements in materials, and the history of extraterre~~~ial ~ a t e r i a and ~ ~ ;in ~echnologyfor such pur~osesas ~ i ~ ~e x e prl oa ~~a t i odosimetry ~~ of indoor f i ~ dsizes, radon, prod~~ction of filters of w e l ~ ~ s ~ e c ihole and inferring where thermal conditions in the earth were favorable for f o r ~ a t i o n(and later l~cation)of oil and natural gas, ~verviewsof the subject are given in three books (Fleischer et al., 1975; Durrani and Bull, 1987; ~ ~ e i s c h e19981, r ~ and other books describe the d ~ v ~in~ more ~ ~r e ~~ t ryi c ~topics: e ~ Spohr (1990)
rnicrotechnology, Wagner and van den radon and its uses. dating, and flic and ~ u r r a n(1997) i ~ t- king ~ a relar ~ ~ ~ t h ion u ~tracks g ~in ~ tively new o b s e ~ v ~(Barbu t ~ o ~ et al., 199.5) - have not yet been put to any of the uses just noted, the fact that so many ways exist of u t i l i ~ i ~other g tracks implies that there will be new outcomes from irradiated interiaetallics. Most importantly, tracks in c o i i d ~ c t i n ~ solids present a challenge to our under~t~nding o f how tracks are formed.
Now, consider what ~ a r t i c ~form e s tracks in solids and then how we best describe which of the a ~ e ~ ~ ~ a t particles that are available form tracks and which do not for in~ividua~ solids. And for those p r o j e c t ~ that ~~s can form tracks, at what particle speeds do they do so?
The charged particles that can produce tracks are atomic nuclei from which some or a11 of the e~ectrons have been stripped to produce ions. The simplest ions to work with in the laboratory come from radio~ctive l~s ~ ~ ~ r-ceither e s alpha p a r ~ ~ c (4 variety of heavy radioii~icl~des, fix example ~ i r a n i or ~n~ americium, or fission fragments from 2s2Cfor iiiduced fission of 2J5U (which gives a wide dist~ibutionof rather heavy projectiles, e x a ~ ~ p i ebeing s “OBa and 9 y ~ ~ Other ) . ions caxi now be accelera~edat a ~ i u ~ ~ of h ~ a ~ y - i omachines, n giving project~le~ with atomic 1 ~ ~ y d r o g eton 92 ~ ( u r a n i ~ ~ Prior ~ i i ~ .to n u ~ b e r from s
~ ~ ~~~~~~~n~~~~ ~ r Vol.~ 3, PriPaeipLs ~ t and~ Practice. ~ Edited ~ iby J. ~H. Westbrook and R. L. Fleiscber. 0 2 0 0 2 John Wiley & Sons, Ltd.
~
264
lgure 1 Tracks from fission €ragmeats in three classes of cfjelectnc materials: Left, Lexaii ~ o l y ~ ~ ~ a plastic; ~ o ~ceiiter, ~ ~ at e silicate glass; right, a crystalline. mineral - orthoclase, The shapes of the tracks are the results o f preferential etching and, in the case of orthodase, ~ ~ io f chemical ~ ~ attack t r ~ ~ ~
these machines becoming available, exposure to cosmic rays was the ~ e a n of s obtaining fast, very heavy nuclei (Frier et al., 1948; Fleischer ec al., 1967a; Fowler et aL, 1967, ~ ~The tracks ~ 0of interest ~ .here are from ions at ener~iesper unit mass of more than 100 keV. lower pro~ect~lesof such an energy as is used for ion ~ ~ p ~ a ~- perhags t a t ~ o1 ~keV per atomic mass unit - produce some atomic displacenients, but not tracks, and do their dainage by direct knock-ons of atoms, which is not the way tracks from fast ions form.
Figure 2. The ~ ~ ~ ~difference ~ ~ ~between t a tenergy ~ v ~ loss and primary ioni~ation is that the p r i ~ ~ r y ionization is localized at the core of the track - the region of the most i n ~ ~ ~ ~~ e h is respon~ c ~ ~ sible for the track contrast lu the electron ~ ~ i c ~ o s c ~ p e and for the ~ref~rential etching. Tlie energy loss, in contrast, includes energy deposited several nin from the core and thus includes irrelevant i n € o ~ a t i o n . The data in Figure 2 are for a plastic, Lexan polycar~onate. Filled circles denote ~ a r t ~ c ~that es made tracks, open circles those that did not. To satisfy the idea o f a threshold there is a horizontal dashed line, labeled 'Lexan', that cleanly s~pai~ates the two types of results and thus defines a critical ionization rate. 'T"u avoid confusion, data for other materiats are not given, y taking s a ~ ~ of ~ ae single s solid and b o ~ ~ a r d ~ n but g the ~ ~ r e s h o ~(as d sd a s ~ lines) e ~ are d r a ~ nin for separate pieces with different ions of different energies, other plastics and for minerals. Thus the figure shows many tests clarify the cunditions for the presence or also that solid detectors exist that have a variety of absence of tracks, Then the question becomes one o f sensitiv~tm~s, from which one can choose - d e ~ e ~ ~ i m i g finding a critical parameter that defines a t ~ r e s h o ~ d on the use to be served. Next we consider wby priimry above which tracks are formed and below which they ionizat~onis, after fum-ther thought, an approp~ia~e are not. The first thought was that the density o f parameter to control track f~rmation. d ~ ~ o senergy ~te~ (energy per unit path l e ~ g t hm~i g ~ t be a r e a s o ~ ~ bthr~sbold. le This idea failed the test of experi~ent, but a related q~antity- the primary ion~z~tiondensity ( ~ 2 u ~ b eofr electrons displaced Wide experience with nuclear tracks is consistent with from the narrow path of the ion per unit pat11 length:) their being formed by an ionization spike ~ e ~ ~ ~ ~ worked well (Fleischer et a/., 1967b), as displayed in ~~~~~~~~
265
Primary ionization as a ~ e a ~ ~ofithe r ei~itensit~ of track d~magein various n o ~ c o ~ d ~ c tsolids i n g IS p e n as a ~ ~ ~ i c t i o ~ o€ v ~~o ciand t y of energy per mass for various nticler. The damage increases with i ~ i c ~charge ~ a s(or ~ atomic ~ ~ ~u~~~~~ from hydrogen (M)to curiuni (Cin). It also increases as the particle slows down (untii it 1s going so slowly that it becomes less ionized and, and therefore less ionizing). The dashed horizontal lines give the tliresholds for track-recording materials that range front sensitive plastics at the bottom to mica and typical inineral ~ o n ~ t ~ ~ofumeteorites e ~ t s at the top. The ex^^^^^^^^^^ p ~ l €or ~ t ~ n ~ grven t ~ as open circles for zero ~ ~ ~ s t r a and t ~ ofilled i i circles for 100% ~ e ~ i s t r ~ t i o n accelerator tom in Lexan ~ o l y c a r b ~ are
~ ~ l e i ~ ~etlal., i e rlQSS), which is sketched in Figure 3. The idea is that the moving, positively charged particle ~ ~ With the model described in section 2.2 in ~ expels electroiis from atoms alon behind a narrow region packed with positively charged workers in this field were ~ ~ ~and ~evenr ~ ~i a zs e d~ ~ l~ic~ ions. These repel one another with. strong ~ o u l o ~ when ~ i ~the observations of tracks in i n ~ ~ r ~ e t a(and forces that by inutual repulsion eject tnany of the ions in certain metals) were reported ( from their normal sites - creating a high energy region Dunlop et al., 1993). As example of atomic disorder that is later c ~ e ~ i cetched a l ~ ~to show tracks of Pb and U ions in, display the track. This mechanism c m be regarded as a and NiTi. In these cases the tracks are viewed in sittr massive scaling up of the Varley (19S4) mechanism, in using electron inicroscopy, rather than by cheinical which two adjacent positive ions produce a lattice etching. Most o f the data on tracks in internietallics defect. Note, however, thnt if the ~ lec~ r ons return to (and, as we shall see, in metals) comes froni the group n e u t r a ~ ~the ~ eion spike before the ions have time to that includes A. Dunlop and . Lesueur, often with A. move, the electros~~tic repulsions are quashed, and Barbu, W. Dammak, and co-workers. no track f o ~ ~ a t i oisnexpected. Barbu et al. (199s) report that tracks are seen also in Thiis mater~a~s with good electronic c o n d ~ ~ t ~ ~ i t ~but iiot in AuCu3 or AlZr3. Thus, the list of BNi3, were not expected to allow track f ~ r ~ a t by ~ o this n i~tern~eta~lics in which the presence or absence of ~ e c h ~ n ~ This s m . inference is ~onsistentwith the fact t ~ ~ c k - f a~b ~r l~i ~is~y known n ~ Is not extensive. Yet, that metals and good s e ~ ~ ~ ~ ~ o ~ di o~ ~ r ~c fail ta o~rto~s y from these five c ~ i ~ p o u Barbu ~ d s et al. point out one form ion tracks. major ~ i ~ t that i ~ can ~ tbeimade, ~ ~ The two that do
266
w e 3 Track forma~ion by the ion explosion spike ~ e ~ ~ 1 ~ n Electrons i s m : are ejected from atoms by a ~ ~ o v i i i g charged particle. The r e n i a i n ~ positively ~~ charged region (a] IS unstable and the repulsive, e l e c t r o s ~ coulomb t ~ ~ ~ forces lead to atomic disorder (b)
not show tracks have relatively simple, ~ i g h ~ s y ~ i ~ e t r ~ crysta~struc~ures- cubic with only four atoms per unit Bright-field electron micrograph of NiZrz that was cell. an contrast, I?& and NiZr2 are respectively irradiate^ with 0.7 GeV lead ions to a dose of 10" (;;a) o r t h o r ~ o ~with b i ~ f 6 ato~s/celland tetragonal with along the beam ~ i ~ e c t i ofb) n ~view with ~~~~1~ tilted by e d d, i ~ e r ~ ~ i t lview ~ . iTi IS also c o ~ ~ p ~ c a tbut 20". From Barbu et aL(1995) with the p e r ~ i s ~of~ the o~ a s h ~ p e - ~ e alloy ~ ~ o( r~~ a y ~ and ~ a n ~ ~ t ~ rRescar& ~ a l sSociety noue, 1995) in which two phases can co-exist at room t ~ ~ ~ e r a 1-ua,r esimple, cubic form and a more di~lectricsolids, and its vaftte should be ~ o i i s i d e ~ e ~ . con~plicated~~iartensi~ic, monoclinic phase, The idea Using the same units as in Figure 2 for ~ o n ~ ~ ~ t ~ ~ thresholds for NiZr2 and NiTi are around 14.0, as conipared to 70 for the less sens~tiveof tlie niinerais~ Thus from this v ~ e w ~ the o ~t ~ r~et ~~h 3re o ~d~i s~ e ~ e n ~ by only a factor of about two. Two other i n t ~ r ~ e t a l l iNb&f c s ~ and V3Si, ~ h a n ~ e ~ s u ~ ~ r ~ o n d ~ properties ~ G t i n ~ strikingly as a result of internal fission fragment ~rrad~ation ( ~ l t l i o u ~tracks h I-tave not been seen from fission fragments in these materials, the result implied that 1 56 ketr/im niakes continuous damage was close to the level that would be ~ e ~ u i r e d . ones, In NiTi et al. (1998) have shown that the threshol~lies ~ ~ t w e eXe n at 32 keV~nmand Ta at 46 ~ ~ t r --~a nvery n ~similar value to that for NiZrZ3.1 These authors noted that these energy-~oss threshold^ or lie ~o~isider~b1y above values of about 1 k e V ~ nfor~ plastics and 3 to 6 ~ e ~for/ ~ inn e ~r a ~ s . ~ e ~ e t r ~pa~ticles t i n ~ may not register in solids for As ~ o i n t out e ~ earlier9~ ~ o w e primary v e ~ ~ i ~ n i z ~ t i o ~several reasons. The first and most decisive is when no gives a more ~ o ~ s i s t e ncriterion t for thresholds of charged particles can form tracks because of basic ~~~~~~~~
oxides - have shown tracks. And, more convinc~ngly, consider spiiiel ( ~ g A 1 which, ~ ~ ~because ) ~ it is an insulating material, was an unexpected non-recorder of fission tracks, Spine1now (Wiss and - making shown tracks of Bi ions from an ac~e~erator it clear that the issue was obtaining high enough i ~ ~ i % a t ilevels. o n Another spinel, ~ n ~ + a~ ~ ~a ~ O n e 4t Y3FesO12, and the oxide Fe304 magnet~te(Provost et al., 1995) are additional examples of insulators in which u n ~ s u a l ~l?i& y iun~~ation can ~ r o d tracks. ~ ~ c A~ third sort of reason for not observing tracks might be that tracks fade rapidly at the temperature where they are sought, even t h ~ u g hthey formed and then existed too briefly to be observed. The like~ihoodthat such cases exist is deinonstrated by the observation of Barbu et al. (1995) tliat the tracks which they could observe in ~~i~ s h o ~ t after l ~ i~radiation~ dis~~peare~ after a few months.
.Trac Given that one wants to u n ~ e ~ s t a nhow d tracks form in i n t ~ r ~ e t a l lci co ~ ~ ~ uitnwill d sbe~ h e l ~ to f ~know the c ~ r ~ also~ for~track ~ ~ so r~~ ~~in~ other ~~ t ~~ o ~ i conducting solids. As in Figure 4, except for ~ ~ i # n o NiTi ~ l i ~ irradiated at 90 K with 0.76 CeV Limiium to a dose of 5 x ~ m o ~ ~tracks # u sare seen at 300 K. (a) view along the beam direction; (b>mew with saiixpfe tilted by 27” From Barbu et of. (1993) with the permission of the ~ ~ t e r i ~ Research Society
~ ~
Again, as for i ~ t e ~ e t a l l ~most c s , of the work in this has been done by Dunlop, Lesueur, Barbu, l area s Danimak, and co-workers - and this work is summar-
ized by Barbu et al. (1995)- The only metal 111 which tracks have been seen at the time o f ~ r i t i this n ~ chapter physical principles. Xt is presently widely believed that is titaaiuni (see Figure 6), but there i s Information that the simple face~centeredpure metals Fall in tlZis class implies that several other metals, ~ n c l ~ d i nZr, g CO, Fe, (AI, Cu, Ni, , . .). A second is that particles that ionize and Be, are very close to track ~ o r ~when a tthey ~ are ~ ~ inteiisely enough may not yet have been used, i.e. the irradiated with the most heavily ionizing particles tliat threshold is ~ n u s u a l high. l ~ For exam~le,in the first are now available ~ A u d o u a ~etdal., 1990; Dunlop and few decades of part~cletrack etching the question as to Lesueur, 1993a,b; ainniak ef al., 1993, 1995, 1996; whether a material allowed track formation or not was Dunlop et al,, 1994). %xi fact, two of these metals - Zr resolved e~piricallyby b o ~ ~ b a r d iti ~with g the most (Dammak et al., 1995) and Fe ~ ~ a and~ ~ u ~n l o pa , k heavily i o n i ~ ~ nofg easily availab~eparticles - fission 1998) do show ~ r a ~ ~ - l features ike after irr~diation fragments - and then testing by transmission electron with fullererme (CO) molecules of 10 to 30 MeV. As m i c r o s c o ~or ~ etching. More recently at places like they enter solids, the ~ o m p o n e ~atoms t of such armstadt, Caen, and erkeley heavier projectiles projectiles itre rapidly ionized to their e q u i ~ i ~ r i u ~ have become available, up to Pb and U ions - particles charge (as shown by Lassen (1951) and Steiger (1963) that are about twice as massive as fission fragments e~uilib~ium is produced over distances o f ~enetration and more heavily charged. Since these more intensely of 2 to 20 pg/cm2, i.e. 2 to 25 nni in solids), so that they b e r slow down by ionization as though they were 40 ionizing particles have come into use, a n ~ ~ ~ of previously ‘iion-tracl~-registe~ing’, condttcting classes of separate atoins ~ ~ a u d ieti ial., 1994; T o i ~ ~ ~ ceth l ~ o materials - metals, i n t e ~ e ~ a l land i ~ s s~u ~ e ~ c o n d u ~ t i n1995). ~ Other metals, ~ n c l u dAg, i ~ ~Cu, Nb, Ni, Pd,
268
~ r ~images ~ ofht ~ t t~ i ithat ~i u ~was ~ ~irradiated ~ ~ with ~ 2.5 x 10" cnx2 of 2.2 GcV ~ r ~ ~ iions. i u (a) ~ i view ~ along the beam direction; (b) view with sample tilted by 10". From Darnmak et al. (1995) with the ~ e ~ ~ iQFsthe ~ ~i oa ~t ~Research ~ ~ l s Society
, show no hint of track form~tion~ ~ u n l ei o al., p ~ ~ t e r e s t i ~the g l ytrack phase is not the c o ~ m o nliighand, using fission fra~meiits~ AI, Au, and Zn can temperature bm cIZ structure, but the hexagonal form ( ~ a n i ~ ~eta al., k 1993, hP3, CO, hi~h~pressL~re be added to the no-tracks list ~~leisclier et al., 1975). 1996). The exception, is ~iiusualin two way^^ (1) It Using electrica~ resist~v~ty , has a c o i ~ p i i c a t erh~ behedral (hP2) structure, and sliowed that there was a rapid i (2) its density is decreased strikingly by d~sorder~ng~ duction for dE/dx values above about 40 keV/nm for In addition to crystalline metals, there are r, Fe, and most ~ a r k e d l yTi. And Dufour et al. two observations of tracks in amorphous metals (1993) showed a similar effect in Bi. Thus it is not ~ ~ r a u t ~ i i aeti ial., n 1993 and Dunlop et aZ., f997, 1998). ~ ~ l ~ that k e greater ~ y ioiii~~tioii levels from single ions 1x1 the first case ~ e 8the tracks ~ were~ made ~ ~ would ~ ~ @ ~t r u~ c in ke ~some or all of these metals for g . t h r e s h ~ ~ dw~~ which d i ~ p e or ~ ~isolated ~ d ~ l ~damage i ~- such ~ as e ~ visible by c h e ~ i c a e~ t c ~ ~ ~The reported to lie above 24 k e V / i ~and ~ thus is similar is i ~ e a s ~ by ~ erd~ ~ i s t ~-~isi large. ty to those in the other materials disctissed here, In The ~ h ~ e s ~for o l dtrack f o r ~ a t i oin~Ti is not well the second case, surface craters re o ~ s e ~ eon d d e ~ n e dIn terms of dE/dx it lies near 30 k e ~ / n mbut , with &160 irradiated foils of Ni3B after i~radiat~on ~ t ~ o nonly by a tower the primary ~ o ~ i ~ is~ bracketed ~ o l e c ~ lions a r at 30 iVeV and U ions at 800 MeV. The limit of 50 aad an upper value of f 70. In short, it may not be ~ i ~ e from ~ e the ~ tlevel iii the i ~ t e r ~ e t ~ ~ l i cpsr, o d ~ c t i oof~ surface pits and same effect as that observed by ~ tnearly a factor of two. a l t ~ itocould ~ ~be ~d i ~ e r eby where fission f r a ~ ate low ~ ~ angles ~ eject m ~ ~ e ~ a l The c o ~ m o na s ~ e c of t most of the metals in which from the surface and produce bulges from tracks just d a ~ ~ is g elarge (i.e. except Si) i s that they have below the surface. ~ l l o t r o ~forms. ~ c In short, the fact that a second, or third phase is available with little difference in free energy suggests a route to track production. In t i ~ there ~ ~ is de~nite ~ ~ ~vidence* L ~ ~ Tracks ~ have ~ been Aiiother gro-rrp of c ~ n ~ u c t imaterials n~ in wfikh shown to consist of a dif%erentphase from the room charged particle tracks can be p ~ o d ~ c eare d oxides t e m ~ e r ~ ~h ~ e xr ~e ~ o i iW2 ~ 1 alplia structLire, but
263 tlrat in some cases can be supe~conducti~g" Tracks in such materials can pose potent obstacles to motions of magnetic flux lines, and thus they can be useful in e~ihaiicing~ ~ g n e t properties. ic rovost et al. (1995) They show evidence of tracks in
~ o ~ ~ e m et o~ al.d e ions in "-1 23, with electron n ~ i c r o ~ ~ oobservatioiis pe b ~roject~les. ~ontinuuus tracks appear at dE/d.x values around 36 keV/nm and ionization levels at and above 150. In the same article Toulemonde et al. (1 994) describe eflFects of ion irradiations on supercon~uGtin~ properties. As in the case of intermetallics, prior work with fission ~ r ~ g ~(Fleischer ~ n t s et al., 1989) showed effects on supercon~uctivity from fission f r a ~ i ~ e n-t sthe most ionizing of which give a level of about 130, i.e. just below what is required for formation of continuous tracks in U-123. The effects of ~ r r a d i a t i oon ~ supercond~ictiv~ty has been a busy subject, but further d ~ ~ ~ ~liess outside s ~ o ~thei scope o f this work,
The main subject here is intermetallics. characteristics of the other conducting solids that allow track formation provide clues to factors that may enable tracks to form in these materials that have unusual track properties. First, a concise summary o f physical factors common to the condLt~ti~g track formers i s given, follow~dby some possible explanations. The models are s ~ e c u ~ a t at i v ~best, not well~ound~ co d n c ~ ~ ~(except ~ i ~ n perhaps s as viewed by the respective authors o f the p r u ~ o s a ~ sSome ~ . of the models overlap in part with one another.
Qne common factor among the three classes of materials of interest (iiitermetallics, metals, and oxide cand~ctars)is that the threshold in primary ionization (on the vertical scale of Figure 2) lies around 150 about double the value for the previously known least sensitive track detectors. In part this level defines their special character, but in part it was to be expected that if more int~iiselyionizing particles were to be found, then new detectors that have higher thresholds could be i ~ e n t i ~ e d~.e v e r t h ~ l for ~ s sa~ satis~actoryunderstanding o f these new detectors to be achieved, their t h r e s h ~ ~ must d s be e ~ p l a i ~ e d .
A second comi~onalityis in f w t a pair of properties. Both interrnetallics and oxides are ordered compounds - and hence sensitive to atomic interchanges that are of no i ~ ~ p o r ~ a n inc pure e metals. Alternate phases that are not too different in free energy is a second coninion factor, It has already been noted that the metals that either show tracks (Ti, Fe, and are inferred to be close to track ~ ~ ~ m(, a t ~ o ~ have ~ ~ ~ oe qt u r~ loi ~ ~rphases. i ~ ~ ~n A p a r a ~ l especufrt~ Lion about i ~ t e ~ ~ e t a ~isl ithat c s because there are so inany crystal structures in which a ~ a r t i c u f ~GOMr position might exist, the chances of 3 second phase being very close in energy to the e~uilibriumone are good. As a particle moves through, the locally intense gradients of ionization, heating, and pressure, might cause a phase change. It was noted that titanium does show such behavior.
There are two broad possibil~tiesfor track-fo~ation ~ ~ ~ h ~ ~ S 1 ~ S . 5.2.1 ~
~
e~~~k~ ~ ~
a
l
The jostling, ~ r i r n ~ ~ rfi-oin i l y ~0~1011ib ejection of electrons, but also from energy transfer to ions and
occasio~ialatomic interactions, creates mot~onsthat, new lattice positions~ a time that we equated to a a hot a ~ t h o not ~ ~ og r~i ~ ~ nr~ a~ l~y ~ soon ~ o produce ~ , ~ a ~ ~~iec~ea x a t ~time, o n i.e, abotrt s. That n ~ ~ region with a ~ r e c i ~ ~t teom~~se ~ ~ t grad~eiit. ure Can e ~ ~ c t i v eexcluded ly good condL~ctors. track format~onresult from this brief3 but violent A re^^^^^^: ~ ~ ~ v e v ~e ~ ~, m ef ~ al. ~(1986) ~ ~ e h e a t i ~ This ~ ? m e ~ h a n i is s ~a lon~-s~aiid~iig one (Seitz, correctly note that a more i i ~ ~ a n i ~ gtime f L ~for the 1949). Two recent papers have discussed track formaions IS that for them to gain adequate energy from ized track detectors tion in some of the n mutual repulsio~s. They estimated for three by t~ermalspikes, T et al. (1993) direct a time of ions (they were study in^ ~~~~~~~~) work by K, Izui attention to Si and Ge ~un~ublished 0.7 x s is required to gain energy that would o , and various metals and say that and S . ~ u r ~ n 1986) overcome the theoretica~shear s t r e ~ ~of t hthe solid (as realistic ca~culations need to be addressed to the ~ a t t i c e ~ ~ ~ time, l ~ x alW13 t i ~ s). ~ And thermal spike. The Toulenionde group (Wang et al., ions (considering the ~~ternietal~ic 1994) present such ca~cu~ations and (in the absense of NiZrz) 1 ~aIcu~ate about 4 x 10-l5 s for a motion by direct nieasurements on tracks) correlate them with each ion of half an ionic radius, by whnch time the ions the damage rates from irradiation, as inferred from each have an energy gain of 24 eV, which in an increases in electri~a~ resistivity. In addition, Szenes uiid~sturbedlattice is sufficient to part an ion or atom ~1995)considers a set of six oxides; he does c ~ l c u ~ ~into * a new site (see review by Howe, 19%). In an tisns based on the idea that, if the melting teniperature already disturbed lattice, disp1acements sliiould be is exceeded, disorder is quenched in; and finally he easier; and hence the time calculated is only an shows a good c o r ~ e l a t ~between o~ track di~meterand 'tpper limit. Thus a time of about 1Wi5 s should now . This result i s ~romising~ but interi~eta~l~csbe considered in t r a c ~ - ~ o r i ~models. at~o~ have not been considered, And alter~iativemodels The e v ~ l u a t i omade ~ earlier (Fleischer et ul., 1955) ~ itt is~ possible ~ ~ that other i ~ ~ that ~ ~metals e d failed to form t were not e ~ a l so~ that description^ fit these ~ ~ r ~ ~ cdata u l aequally r well. It a t o ~ r~e Ica x ~ t i ~times n were a factor of long. The discussion of times in the previous parsishould also be n u t ~ dthat thermal spikes have not been slxccessfid in making useful predictions for the great graph showed why the gap should be lowered by a factor of about 100, and the higher ioni~ationabove number o f more conventiona~ track detectors, as ~ i ~ ~byu~ ~ s es~ s~ceth~uE, e r (2965, 1975). ~ e ~ ~ e ~ ~ h e l ethat s s , of fission f ~ a g i ~ that e ~ t BOW ~ is ~ v a i l a bin ~~ in ~ ~ a ~ i t a t i vdiRerent ely materials d i ~ rnechane ~ ~ particles ~ ~ from heavy-ion accelerators supp~iesanother f x t o r of four or five - implying that it is borderline as isms may be ~ n v o l v e ~ * to wliether niettls should record particle tracks or not, The most telling evidence implies that the thermal a behavior which the observations in sections 3 and 4.1 spike is not the primary general cause of track fori~ation.The observation i s that primary ioni~ation~ document for materials that have metallic ond duct ion^ forz wakes: Another factor that is related to the rather than total energy loss rate or restricted energy revision required by laumiinzer et al.'s work has to loss rate, correlates with registration thresholds where do with other energ that is imparted ta the newly et al., 1967b; they have been measured (Fl~ische~ path by its created ions along a charged~partic~e~s Fleischer, 1980). direct coulomb interactions with them - a repulsion that gives each ion a radial impulse away from the 5.2.2 Ion path. The p o s ~ i b ~effect e was noted by ~ ~ e et al. ~ ~ c (1965) and i ~ d e ~ e n d e ~by t l yLesueur and Dunlop This mechanism, i l ~ ~ s t r ain t eFigure ~ 3, is also called on occasion a Coulomb spike, ~ o u l o i n b~xplosiQn, (1993). The former paper estimated energy transfers ~ ~ ~~ r~ ga ~y m to e na t Si atom of or ion spike. For good coiid~ictors from a f u ~ ~ ~ fission i o n i ~ ~ t i ospike, n about 1/30 eV - in short, a little above ambient we c ~ n c l ~ ~ dine dour o r ~ ~ i proposal ~ ~ a l (Fleisclcher csf thermal energies. From the i n c r e a ~ e~~ ~ ~ tbat u l s al., 1965) that such ~ ~ a ~ e ~would i a l snot forni tracks. are available from ~cc~lerated heavy tons, a few times ~e were no doubt in~uencedin ottr thinking by the the t l i e ~ ~energy al are available. Thus these inch~ul~~s lack of tracks from ~ssioi~-fragment ~rrad~atjons of a1 noise, and they can prowill not be lost in metals and the most c o ~ ~ u ~ tsemicon~uctors* ive The pagate as a wave. gh and Cilmaa ( 1 ~ and ~ ~ ) rea~oning given was that the re~axat~ontime for unlop ~ ~ 9 9 ~3 ) ~ ~ on ~the ~resenir ~ e e ~ ~ c t-rthe ~ ~time s to n e ~ t the ~ ions ~ ~that i ~had~ blance to a shock wave, One possible iiiipo~tan~e to been produced by the charged particle - needs to be this organized motion of ions is in assisting phase not less than the time for ionized atom to move into +
~~~~~~~~~
~~~~~
Ion Tr.acks changes that might be encouraged by high pressure or shear stresses in a ~ u l t i - a t o ~ iregion l ~ c - NiTi being an example of where shear tr~sforniatioiisoccur easily, and tracks are known to form. And, as noted earlier, where there lurk allotropic phases or hypothetical crystal structures that are just barely energetically unfavorable, these ‘organi~ed’deforniations may bring them into existence. Dunlop et al. (1991) and Lesueur and Dunlop (1993) consider the i~portanceof soft plzonon modes, wbxh they describe as encourag~ngdisplacive transfo~matlons. This behavior appears to be another aspect of the ion-wake ~ehavior. ~ ~ ~ ~ ~ ~~It iwould ~ be~ helpful t c ~ and ~ Z conve~ nient if there were simple e ~ ~ p i r ~~c agl ~of~merit e sthat would predict relative ease of track ~ o ~ n a t in i othe ~ various conducting solids considered here. For example, iii the case of insulating track detectors it was shown (~leischeret al., 1965) that threshol~sfor detection increased with the value of 5Eb4,wliere t is the ~ i e l e ~cu~stant, t r ~ ~ E Young’s modulus, and h the i ~ t e r a t o m idistance, ~ The cor~es~omi~ence of thresholds is not one to one, but it does divide the detectors properly into groups of similar ~ i a t e r i a ~ with s moiiotonically i ~ c r e a s i ntliresliolds ~ for track production. For intermetallics, only the values of lattice spacings are widely available; for cond~ctingsolids, it is not clear what to use for a dielectric constant (perhaps behavior at ultra-hig~i~ ~ e ~ u which e ~ ~ISc not y ~available). Values of Eb4 can be connpnred for nietafs in w h i ~damage l~ is easy (Ti, r, CO,Fe, and Be)+but this parameter does not correlate with damage production. Hence, this direction i s not f ~ ~ t f u l , The only useful correlation is that of Luzzi and Meshii (1987) for describing the ease of converting intermetallics from crystalline to amorphous. In their Table 7 they list two groups o f inter~etallics- those that electroiis can amorpliize and those they cannot. NiTi and NiZrZ can be disordered, and they are track recorders; AuCus and AlZr3, which have not fonned tracks, similarly are not disordered by electroil irr~diation.The statistics are not great, two structures roup, but this direction i s p r o ~ i s ~ ning that Luzzi and MesGi’s list suggests a dozen other ~ n t e ~ e ~ d l lthat i c s are likely track formers, and 15 others that are unlikely cand~dates,Among compounds that ‘were not made amorphous’ are several that (on the basis of crystal complexity) might be expected to allow track f o ~ ~ z a t i othe n ~fact that they did not is merely a State~~ent that in the p a r t i c u ~ ~ r experim~~ reported, ~s they stayed ~r~stalline* In fact, s , Banerjee et n17, one of the r ~ f e r e ~ c eedx ~ e r i ~ e i i tby
27 1
(1984), showed that for irradiation at 200 (tT10, Bla) did in k c t become an~orphoL~s. Of the 12 c o ~ ~ p o u n d that s can be made a ~ o r p ~ o u s by electron irradiation, 11 are of lower thau cubic symmetry, and 10 have more than four atoms per unit cell. The rei~aiiiinginter~etal~ic is Cu structure - but it is notewo~hyin that it exists as an equilibrium phase only between 715 ”@ and 935 “6, hence disorde~ingniay be easy. that C ~ ~ c ~xper~mentally Z ~ ~ ~ it~has~been ~ show~i : charged particles create damage tracks in some ~ n t e ~ ~ ~ ~and ~ ~not ~ l in l i cothers, s in at least three metals and are close to track ~ o r ~ a t i oinnothers, aiid in s u ~ e ~ c ~ n d u c tthat o r s have been studied. ~ the oxide : ~ec~nsideration of the free electron ~ e n s ~ tand y r e ~ a ~ a t i otime n ~ ~ p l that i e ~the usual ion explosion spike model is s ~ i ~ c ~ to e n~t x ~ l a itrack ii forniation from projectile^ that exceed ion f r a ~ ~ e n in ts ~o~iizing power. ~ r y s t a ~that s can be disordered or caused to change phase are ~ a r t i c u l ~ rsusceptible ly to track forniation.
This chapter i s devoted primarily to d o c ~ ~ ~ n eand ~tin~ understanding track f o r ~ a t i o nin i n t e ~ e t a ~and ~i~s allied materials that might aid in. learning about the behavior of solids in general and i n t e r ~ e ~ ~ linl i c ~ particular. ~ o ~ e vsince e r ~particle tracks in solids have many uses iiz science aiid techiio~o~y, it is of interest to ask whether tlzere are pros~ec~ive appl~ca tioas o f tracks in intermetallics. Since neither alpha particles nor fission ~ r a g ~ e n t s make tracks in intermetallic compounds (or the other conductors) uses for radon ineasure~ents~ alpha or fission ~~appimig, fission track dating, and d o s ~ ~ ~ e t r y are not available. The unique feature of i i ~ t e i ~ ~ e t a l l ~ c s is their insensi~ivity,and thus they would be effective in recording a few? very rare events from inteiisely ionizing particles in the midst of a background of less ionizing particles. Hence, searches of tlie cosmic rays for slow, heavy nuclei or h ~ o t ~ i e t i c aother l particles are scientific possibilities. Two routes may be considered, One is to synthesize sheet material for exposure to present-day cosmic rays. A second is to locate and examine one of the good many natrxral inter~etalliccompounds (De~ries,1995) that may have been exposed over geological time - just as nieteoritic minerals have recorded ancient cosmic rays (Fleischer et al., T967a). DeVries notes that iron meteorites contain two ~ ~ ~ i ~n t edK ~a ~ ~~ a l~~ ~
~ a n i ~ a H., k , Dunlop, A., Lesueur, D.*Brunelle, A., Dellai(Llo, tP4) and F i3 (Llz, cP4); these might be Negra, S., aiid Lc Beyec, Y. (1995). ~ ~ Rev. 2 Lett., ~ 7~ ~ ~ useful provided they do record particle tracks. FeNi E 135. ( t e t r ~ t a e ~ because ~ t e ~ ~ it is of lower s y ~ ~ e t risy ,the nd Lesueur, D. (1996) Nuct. more likely ~ ~ i ~ ~The ~ ~ chance a t e of .tracks ~ o r i ~ i ~ ~ in FeNi is enhanced by the fact that FeNi is disordered to a ~~ce-cent ere^ solid solution by lieating above about 300 "C ( ~ l i ~ ~ ~eti al., b e1979). r ~ ~ DeVrres, R.C . (1995). In I n ~ e r ~~ ~ ~o ~ ~~ -l ~~~ri ~i o ~ ~~ ~ l~ ~ A potential ~racticaluse is iii making filters, which arzd Practice (eds J. H. Westbrook, and R.L. Fhschcr). might be doiie by irradiation across thin sheets aiid Wiley, Chichester, vol. 1, leischer et d, 1964). ~ i i t e ~ e t afilters ~ l ~ c ~ u € o C., ~ rALido~a~d, ~ A., F., Dural, J., Girard, ~ ~ ihave ~ special l ~ t ~ r # ~ ~- for r t example ~ ~ s u ~ ~ i s ~ a l ~ yJ. P., Hairre, A., Lev M,, Palmier, E., and sistanee or h j g h - ~ e ~ ~ ~ ~ e r a t ~u ~o~ u ~ ~ RA[.~ (1993 o n ~J. ~Ph-ys. ~ ,~ ~ ~~~t~~~~5, d s t ~ ~ ~AlRu t ~ rise a ~ ~ t ~ r i a l 4573,
otE9 this voltirne). As yet, it is iiot known whether tracks can be made in any 2 structures, however.
Dm1op, A,, Legrand, P., Lesueur, D,, Loreiizelli,N., Morillo, J., Barbu, A,, and BouEarcl, S. (1391). ~ ~ ~ ~ Lc3tx-,, ~ ~ h . ~ ' s . 765. r, D.> and ~ a r ~ A. u ,(1993). f . Nuct.
unlap for lielpfml correspondeiice and her ~ ~ i ~ for n ds e~ ~ ~ lthe y photos i n ~ for this ~~a~ter.
Dunlop, A., Henry. J., and J ~ ~ s k ~ C. ~ ~(1998)~ w ~ ~ z, *c ~ ,
Ces lanzat, E., Boflard, S., Jousset, J, C,, , Dunlop, A., Lesueur, D., Fuclzs, G., er, J., and Thorn&,L. (1990). Phys. Rev. ~
e C. aP., Fieischer, ~ ~ R. L,, S w a r ~ P,~S., ~ and Hart, N,R., Jr. (1965). J. ~~~~~. Phys." 37,2218. ., arid Averback, R. S.
axnd Lesueur, B, (1995). A., Hardoitin Duparc, A., Jaskier G., and Lor~nzelli~ N.(1998). Nucl. ~ n s ~Meth., r ~ ~ . 354. anerjee, S., Urban, K., arid Willtens. M. (1984) Acts Metall.,
Dunlop, A., and Lesueur, D. (199323). Rnd. ~ ~ ~~ ec f ~t c~t , s~ ~ ~126, 123. r ~ ~ ~ ~ , A., and Lesucur, D. (i993b). ~ ~ t e r SCL i a ~~r~~~~ ~ ~ 9, 553. ,S., and Bull, R. K, (i987). Sokid State ~ ~Track ~ ~ Detection, Pergamon, Oxford. Fleischer, R, L. (1980). In Prog-, iiz ~ ~ Sci., ~ ~ ~ ~ ~ Anniversary Volume (eds J. W, Christian, P. Haasen, and T. B. ~ ~ a ~ Pergamon, a ~ s ~ ~ j ~ . Ffeischer, R, L. (19%). M R S &I, Fleischer?R. L. (1998). ~ r to ~ ~ ~ ~ s in. Scwice and ~ e e h S ~p s~~ ~~ ~ ~ o r ~-New V~e York. r~~ ~~ ~ , F ~ e ~ s c ~R. e r ,L., Price, P B., and Symes, E. M. ( ~ ~ 6 4 ) * B., and Walker, R. M. (1965). J. B., Walker, R.M., M~urette,M., and Morgain, 6. (1967a). .I; ~ e o ~ lRm., ~ ~ 72, s . 355. er, R. M., and Hubbard, E. L.
Fleischer, R, L., Price, P. B., and Walker, R. NI. (1975). Baradin, K., ~ ~ ~ n eA., ~ l Ghabot, e, M., Della-Negra, S., m cl^^^ ~ ~ a c IN l ~ Ss o ~ ~Unw. ~ s of ~ ~ a ~ ~ ~ o Press, rnia ~ e ~ a u.Iw .~ ,~ a r a., ~ ~ ~~ k~~ n, ~P., s oLen Beyec, , Y,, Berkeley. 3~11ebaud,A., Fallavier, M., Remilfieux, J., Poizat, Lay, K. W., and ~ L ~ ~ o r s k y ~ and Thomas, 3. P (1994). ~ ~~ ~ ~ ~Me& s ~ ~ r ~ ~ FJcischcr, ~ ~ . R. r L-, Hart, . 13. F; E. (1989). Phys. Re 341. Flciscber, R. L,, Field, , and €31-iant, C. t. ( ~ 9 9 ~ ~ . B ~ ~ l l o R., u ~ and ~ , Cilman, J. J. (1966). J. Appl. Phys., 37, ~ e t a l l ~ Trims., r ~ ~ c ~ ~ 2283. Fowier, P. H., A ~ ~ ~R,i A., ~ sCowen, , V. G., and Kidd, J. M. C ~ a m b c r ~ A., d , Laugier, J., and Penissoxi, J. M. (1979). J. (1967). M ~ ~ ~ ~ e~t ~ % samMater.,~ 10, 139. ~ ~ ~ ~ c Proc. Roy. So Fowler, P. H., Clapham, V M.. Cowcn, V. Dammak, 15.. Barbu, A., Dtrtilop, A,, LC~LICW, D., and and Moses, R. T. (1970). Proc. Roy. Soc., Loreazelli, N. (1993). Phil. Mag. Lett., 67, 253.
Ion Tracks
273
Szenes, G. (1995). P h j ~ . Frier, P S., Lofgren, E. J., Ney, E. P., Oppenheimer, Fa, Bradt, H. L., and Peters, B. (1948). Plzys. Rev., Tomaschko, Ch., Brand Howe, L. M. (1995). In I ~ t e r ~ ~ ~ ~ tCompounds aLlic Volt, H. (1995). NucL and Practice (eds J. H. Westbrook and R. L. Fleischer). To~lemonde,M., Paurni Wiley, Chichester, vol. I, p. 791. Efeects and 0eject.r in Ilk, R., and Durrani, S. (1997). Rado;k,va ~ e a s u r ~ ~ ebyn t . ~ Toulemonde, M., Bouf-Tard, S., and Studer, F. (1994). Nucl. Etched Track Detectors, World Scientific, Singapore. ~ l a ~ ~ u n zSe. , rHou, , M.-I>., and Schumacher, 6 . (1986). Trautmann, C., Andler, S., Bruchle, W., Kiiorr, T. G. (1964). J. App1. phys., 35, 2753. Lasscn, N. 0. (1951). Dan. Mat. Fys. ~ e d d . 26 , (no. 5), 3. op, A. (1993). Rad Efects Defects i, M. (1987). Res ~ e c h a n i c a , Provost, J., Simon, Gh., Hervieu, M., Groult, D., Hardy, V., Studer, F., and Toulenioiide, M. (1995). M R S Bull., (12), 22. 1949). Disc. ~ ~ r Soc., a d ~ ~ ~ ~ (I 990). Track ~ ~ n d a ~ ~ n ~ ~ s c e ~ ~ a n~e ~o ~z ~l is c ~ t iVieweg, a n s . Braunschwei~. Steiger, N. H. (1963). Lawrence Xrxd. Lab. Report UCRL10888, UC-34 Physics, June 26, 26 pages.
lemonde, M. (1993). Rad. ~ ~ e cand t s 207. Varley, J. H. 0. (1954). ~ ~ t u r e , Wagner, G. A., and Van den Haute, P (1992). Fi.~s~on-Tracl~ Dating. Enlre, Stuttgart. Wang, Z. G., Dufour, Ch., Palmier, E., (I 994). J. Phys. Candens~d~ ~ a ~ t ~ r , Wayrnan, C. M., and Inoue, ?A. R. P, (1995). In I n t e r ~ e ~ ~ ~ l i ~ C o ~ p o i -~ Pn r~~ ~ c i ~and l ~ s Practice (eds J. H, Westbrook, and R. L. Fleischer). Wiley, Chichester, vol. 1, p. 827. Wiss, T., aiid Mntzke, 14j. (1999). ~adi~~tio;k,va ~ea~si~rei~e;k,vats, 31, 507
This Page Intentionally Left Blank
structures or the same type of atoms in the e ~ e s~ruc~ure are further apart, which may ~ m ~vacancy ~cieiitificactivities in the field of ordered inte~~~etallicsmi~ration. For defect studies in ordered i i i t e r ~ e t ~ ~ l l such as the Ti-AI, Ni-A1 OF Fe-AI s y s t ~ i ~ arise s from pounds, which date back to the early work of basic research interest^ in these ~ ~ n s ay srt e~~ ~and ~s defects their potentia~for technical applications ~ e ~ t b r o o k and ~ Taylor (1937), a variety of constit~~tional such as the vacancies YEand Yp on fke two s u b l a t ~ ~ ~ ~ 1993; ~ e s t b r o oand ~ Fleischer, 1995; Deevi and cx and fl as we11 as the antisite atoms Ap and B, are to Sikka, 1996; A ~ p ef ~ lal., 2000; ~ a r o l i a2000~ ~ as be considered in inte~etalliccom~ounds.In ~artic~1ar, l ~ g h t ~ w e materials ~~ht with promising high-temperain AI-rich B2-NiA1 high concentrations of vacancies on ture properties (Yamaguchi and Umakoshi, 1990) and the Ni sublattice are found. V a c a ~ ~ i eon s both corrosion resistance. Stimulated by the growing intersublattices are required for ~tabili%ati~n of the la est far teclinical app~ication,research on intermetallic structure in some cases, as, e.g. in the case of Blcornpolxnds also comprises studies on novel alloys as, (Anderson et nl., 1957; Terauchi et al., 1978; Valeeva et e.g. MO-based silicides ( h i et al., 2000) and transition-metal Laves phase compounds (Liu et al., 2000). al., 2001). For the vacancies in i n t e ~ e t a l l ci ~o ~ ~ o ~ two nds In basic research, order-disorder (see e.g. chapter by different types have to be ~ i s t ~ i i g u i s ~Besides e~. Diinitrov in Volume 1 and Calin, 1994) and diffusiori processes (Larikov cliapter in Volume 1 and Mehrer, comcoi~stitutionalv a c a ~ c i ~ ins ordered ~ n t e r ~ e t ~ ~ llic 1996), which are also related to the t e c ~ ~ i ~relevant all~ pounds which are necessary in some cases to creep resi~~dnce p ~ e n o m e n( ~ a u ~ hf o893; ~ , and^ c o ~ ~ p e ~ s afor t e o ~ - ~ t u i ~ ~ i ico o~ i~ ~t r~~~c s oif t ~ o ~ et al,, 1999), are of i ~ p o r ~ i ~The c e .~ n d ~ r s t a ~of d i ~ gthe i n t e ~ ~ e t a lc~oi cm p o ~ nto~ ~ ~ a i the ~ ordered ~ a i ~ these processes requires a profound k n o ~ ~ e dofg ~ structure thermal vacaiici~salso exist, T t h e r ~ a ~ lformed .y defects in ~ n t e i ~ i ~ t a lcompounds, Iic tion of the const~tutionalvacancies can which is the subject of this chapter. the con cent ratio^ of the thermal vacancies as, e.g. in It will be shown below that crystaf structure appears the case of Al-rich NiAl c o ~ p o u n d sfor whrch several. to play a role in thermal vacancy formation. A nuniber atomic percent of constitutional vacancies have been of binary i ~ ~ t e r ~ e t acrystallize ~ l i ~ s in a close-packed reported (Bradley and Taylor, 1937).A clear ~ i s t i n ~ t i o ~ or nearly close-packed structure such as Llz(cP4) has to be made, however, between ~ o ~ ~ t ~ t u t i o n a l (?&AI), D019(kzPS)(Ti3Al) and Llo(tP4) (TiAl) or in vacancies and quenched-in or remnant vacancies, the bcc-type structures which can remain in the lattice of an i ~ t e r ~ e ~ a ~ ~ i (cFl6) (FQAI, Fe$%) structures the compound due to their low mobility and which can also sublattices of the majo yield fairly high concentratio~sthat then have signifiinterco~nected~earest-ne~g~bor sites of the same type cant csrssequences €orthe ~ n e c h a n iproperties ~~1 (Ckmg of atoms, a ~ituationthat fmors vac an^^ jumps on th' et al., 1993; Pike e~ al., 1997). For details on tlie sublattice, whereas the minority atoms in the A3 of r e m ~ a nvt a c a ~ c see ~e~ s e ~ t i 3.2, o ~ In the case o
ion
*
*
type iiite~~metallic coni~oundssuch as NiAl and FeAl can be derived from the d ~ p e n d e n cof~ CV on the the concent~atio~ of thermal vacancies V, on the pressure p tr~nsition- eta^ sublattice has been predicted (Fu et al., 1993; ~ a h et~al.,~ 1999) e to exceed that 01%the A1 (4) sub~~tttice which r e ~ e n t ~has y been ~ e ~ o ~ i s t r a t ~ ~ From the pressure d e p e ~ ~ e n cofe the e q ~ ~ l i b r a t i o ~ rate, the v a c ~ n ~ c ~ g ~volume ~ can t bei deduced ~ ~
local co~~position (see Scliaefer and B a ~ u r a ~ ~ e r g e n ~ 1997). A review of the literature on defects in €32-type which is the volume change during a difibsive jump. iiitermetallic c o ~ ~ o ~ t was i d s given by Cliang and ~ ~ u (1982) ~ ~ withndatan available at the time (three ~i~ ear in a 1998 MRS s y m p o s ~ on es i ~ t e i ~ e ~ d l In l ~ this ~ s . ~eview ) the data on CoGa and ~ i (see~Figure a 4) obtained by measuring techr-riques For s t ~ d y ti h~ ~ ~ eaql ~ ~ l ~ atomic b r ~ udefects ~ in s ~ ~ c i ffor i c v~cancyd ~ t e c t i oshould ~ be particular^ y solids, several r e q u ~ r e ~ ~ ehave i ~ t sto be ~ ~ l l i l by l ~ the d empliasi~ed.Iiz the present overview we will mainly detection t e ~ ~ ~ ~ ~ q u e : deal with vacancies in thermal equilibrium. For this ~ u r ~ the o s~~h ~ r m o d ~ nand a ~ ~iiietics ~ i ~ s of t h e ~ a l (i) high s ~ e c i ~ c ifor t ~ 7a particular type of defect (ii) high sensitivity e ~ ~ i l i ~ r i uvacancies ni are summarized briefly. By (iii) a p ~ l ~ ~ a bati high ~ ~ t temperatures, y and ~ i i ~ ~ ~of~the r variation ~ ~ e nof~the s thermal equirs that i ~ fast ~ e ~ ~ i ~ i b r a T e (iv) short ~ e ~ st i ~ e so I ~ ~ rv~a uc i~~~~ ~~ ~c i~ ~ n tCv~ with ~ t ti eomn~ e r a t ~ processes of the thermal defect c # n c e n t ~ a t i o ~ after rapid t e ~ p e ~ changes ~ ~ ~ ucan ~ be e ~ e a s ~ in order to study the f o r ~ a t i o nand migr~ttionof the same type o f atomic defects S i ~ ~ l t ~ n e ~ ~ s I y the egective vacancy formation entlzalpy HF can be If we restrict ourselves to m e a s ~ ~ r itne~c ~ ~ n i ~ ~ ~ derived, where 3; denotes the efTective vacancyq uleft~ ~ which fullill criterion (i), only a few t ~ c ~ i are ~ ~ r ~ aentropy t i ~ and n k,, o l t ~ ~ n nconstant. ’s of these t ~ c ~ n ~ may q L ~ exhibit ~ s d i ~ ~ ~ l in ties and some From the ~ i ~ e ~ d e ~~ qe ~ ~i ~ i~b nof r at the t i omean ~ f u l ~ i l i nthe ~ residual criteria (ii) to fiv) as discussed in vxancy ~ o n c e n ~ r a G(i) ~ i o ~ after fast temperature the ~olIowing~ c h a n ~ at ~ shigh t e ~ ~ ~ r a t u r ~ ~ is specific to the ositron Ii~etitiz~ spec~ro~copy i ~~n ~ c a n c defects y ~ ~ ~ ~ e detectio~of the ~ ~ r i i i a tof due to the ch~~racteristi~ increase in the lifetime o f a positron when trapped in a vacancy, A with the vacancy diffusivity di~erentia~jon of k-acancies ctn the two sublattices of binary inter~etallicalloys is, however, hardly feasible because only minor di~erencesof their positron lifetimes are ~ r ~ d ~ c ttheoretically ed ( ~ i r ~ s k oetv al., ~ 1995), A specific i d e n t i ~ ~ a t i o ~ the ~ ~ e c t ~v v e~ c m ~ ~~~ ~ r a yte ~ n to~ ~ a ~ p y is of these d i ~ e r e i types i ~ of vacansies is, ~ o ~ e v e ~ , ~ ~ d for ~ the ~ ecase d of a regular distrib~tionof feasible by e ~ i p ~the ~ ~oincident ~ i n ~ i~eas~ir~nzent vacaiicy sources or sinks. Here, C, and Cv denote the of the ~osit~on-electro~i ~ n n i h i l ~ ~ t iphoton on line equilibrium vacancy c o n c ~ n t ~ ~ t i oati i the s initial and (see Section 3.4). final tenzper~tures T, and T f , respectively, d the The positron trapping rate vacancy jump distance, z the coordination number, ebye ~ ~ e ~ u e n2i$fc ythe ~ vacancy niigratioii a quantity c h a r a c t ~ r i ~the ~ nsource/s~nk ~ ~ion ~ ~ n ~ iand t y 1,/ t e the ~ h a r a c t e ~ i ~et~i cu i ~ i ~ r arate, of thermal v~c~tncy c o ~ c e n t r a t CV ~ ~can ~ s be detern a d d i t i ~ i i~~ ~ ~ ~ a c tpar~meter e r ~ ~ t of ~ cthermal mined making use of the time ~ o i i s t a n tTO ~ and ~ o n VF which v a c ~ n c i ei s~the vacancy f o ~ ~ t volume
HY
detectioii limit for vacancy c~ncen~rations but the z1 and the in tens it^ I1 = I - 10 derived from the ~echniquei s not so sensitive to v a c ~ concen~ ~ y positron lifetime spectra. If the specific positron trations below o t i i ~ e - d e ~ e n ~studies e s ~ t of t ~ a ~ p i nrate g cr is considered to be t e ~ ~ e r a t u r e the defect equi~ibr on used for i ~ v e s t i g ~ t i ~ ~ ~~idependent~ then the vaca~icyforii~~tion enth~tlpy defect niig~ationby this t e c h ~ i i ~are ~ ~ltnown e to ffF, c‘clll be d e ~ n e das an effective ~uantityby the the authors. tei~perature va~iation of Cv as ~ ~ e a s ~ rby ed (c) ~ i g h - p ~ e c i s ~ neutron o~i, d i ~ r ~ c t i o tn e c h ~ ~ ~ u ~ positron lifetime spect~oscopyaccording to equa(Kogachi t?f al., 3996) may be useful fur detect in^ tion (6). For a theoretical understanding p ay er vacancies aiid antisite atoms s p e c i ~ c a ~on ~ ythe and F$ilinle, 1997; Fiihnle et al., 1999) of the particular sublattices, They are, however, limited experimental value of the conco~itantformain sensitivity and no ti~e-dependentdefect equition of the entire ensemble of thermal defects has to be taken into ~ccountin order to ~ ~ ~ c uthe l ~ ~ t e libration studies have been reported. tempe~atu~e variation of the vacancy co~centratio~ ($)A t~me-dependentlen th change, A& ( K , Tf), of a s~ecimen ( t i ~ e - ~ i ~ e r e n t i~aill a t o m e t r ~is~ in analogy to the ex~er~mental p~ocedure.In this expected due to the e q ~ i ~ i b r a ~process ion of the regard the recently reported opinion (the chapter by thermal defect con~en~rations after fast temde Novion in ~ o l u 3) ~ that e positron annihilation perature changes between the initial and final expe~iment~ are ina equate to yield the vacancy tempe~at~~res T, and Tf and can yield direct f o r m a t ~ oe ~n ~ h a is ~ i~isleadin~. p~ and on i~iformationon the defect f o r ~ ~ t ~ Fur the direct study of the eRective ~ i i g r a t i o ~ dies by a length pr ~nthalpy of the therm~llyformed vacaiicies, i n v e ~ t i ~ ~ tati ohigh n ~ te~iperaturesin the vicinity of with the speciineizl were thermal e q u ~ ~ i bafter ~ i ufast ~ ~ te~iiperatu~e changes p~rforniedearlier on CoGa (van ~ ~aiid dei ~ are d e s ~ r a in ~ ~order e to avoid ~1iai~ges of the defect ~ i r a ~1981). d ~ ,~ c c o r d to ~ nits~time dependen~e (Figure 1) the ~ e change ~ AE(t, ~ T,,t Tf) ~ due to pattern due to q ~ e n ~ ~ ia~~ gi n.guse of the timedi~erentia~ niea~urin~ tec ique both q ~ ~ ~ n t ~ t ~ e sthe , defect ~quilibrationprocess may be s e p a r ~ t ~ ~ H$ and H v , can be measured ~ n d e p e ~ d e ~ t l y from the instantaneous a ~ i h a r ~ lattice o ~ ~ expanc o ~and s (3) by the teniper~~ according to e ~ ~ a t i (2) sion AI‘, and may be of particular interest for the lure vari~tionsof the ai~p~itude and o f the time investigation of defects if p ~ ~ i t r ostud n depende~ceof the the defect concentrainapplicable, In the s i ~ case ~ ~ of~ ae tion, resp~ctiv~ly. ts of this type were distrib~tionof defect ~ourcesor siiaks the time perfornied by posit atioii techn~queson a ~ependenceof the s p e c i ~ e nlength time scale of some tens of mi~liseco~ds in Au chmid, 1989) and on the scale of ~ ~ ~ set crtl., l ~ i~ 9~ ~ For b~ ) . thermal vacancy studies irn ~ n t e r n i ~ t ac~o~~i c~ ~ o u n d s by means of posit~on”~ifetinie spect~osco~y the with measuring techn~quesand the specinien reparation l* = I*(TJ 3- M’ (7bl ere described earlier ~ ~ c h a eetf ~al.,~ 1990; r o s s ~ a n net al., 1994; ~ ~ r s c eth d., u ~1995b). after a fast teni~eraturechange i s @veil in a ~ ~ l o g ~ ~1thoL~ the ~ h~ositron-lifet~~~e spectroscopy has to equation (2) by an e x p ~ n e n twhere ~ ~ l the time substantial~yc o n t r i b ~ ~to t ethe ~ investigatio~of the constant f~ contains Dv. From its t ~ ~ p e r a t ~ i ~ properties of thermal vacancies in i n t e ~ ~ e t a ~ l i ~dependence the vacaiic~~ i i g r a ~e ~ ot ~ ~ aH l ~v y ~ ~ (see b ~ ~ this t~e c~~ n ~ i q uis~i~~ a p ~~~ i~c ”, can be~ ~ e t e ~ ~d i in n eanalogy d s to equations (2) and able in the presence of high co~centratjonsof (3)- The q u ~ ~ t i tI0yfT,) deiiotes the ~ ~ u i l i b r a t ~ ~ ~ o n s ~ i ~ u tori o~~~~~le ~ c l i e~d - i~n ~ ~( >~10-3), i c i e state ~ at r,, when only saturatiun trapping occurs, as, e.g. in From the temperature v ~ r ~ a t i oofnthe l e ~ ~clian th a d u r a ~ ~ e r g e3997) n ~ or in quasic~ystals( ~ ~ 7 ~ r s cobl nl., ~ ~ ~1904, n i 199%). he di~erentialt h e ~ expansion ~ a ~ ( ~ i ~ ~ i i arid ons aluffi, 1960) technique, which is capable of ~ ~ e a s ~ ~absolute r i n g coii~entratioiisof vacancies, vacant lattice sites, indrzced, e.g. by theriii~tlly-for~~e~ can o v e r ~ o mthis ~ difficulty. There is no upper the effective v~c‘an~y formatj~s~ e n t ~ H~c lcan ~ be ~ ~~
gure IL ~ ~ ~ eof m the ~ t i ~t ~~~ e~ - ds i length ~ e r ~change ~ i ~ iQE~ ~ Sketch of the ex~erimentalset-up for the time(Tl,Tr,t ) due to the equil~b~at~on process of the thermal defect d i ~ e r e ~ ~ t i alength l change ~ e ~ ~ s ~ r e ~after e n t sfast c ~ n e ~ ~ ~ aafter t i o ant e ~ ~ e r a t ~change i r e from T, to Ti. The Two mirror planes are prepared on ure spon~aneouslength change Qk’ due to the a ~ h ~ ~ rlattice ~ o n ~ ct ~ ~ p e ~ a t changes. ~ x ~ a ~ t iss iindicated, o~~ reproduced with p e ~ ~ ~ from s ~ ~ o nthe specimen front platie and the ledge plane in order to measure the length change by the reflexions of the two laser Elsevier Science beams. The s p e c ~ e nand the ~ ~ ~ s ~ a ~ ~ efurnace - ~ e aare ted suspended in an evac~ated c ~ a ~ b ereproduced r~ with permission from Elsevier Science
derived; here BY 1s the change in the atomic volume Y due to the atomic relaxation in the vicinity of the va~~n~y. (Schaefer et al. l999a). ~ p e c i m ~ nwith s co~~ositions The ~ ~ ~ e c t - i n d uti~e~dependeiit ~ed length change Fe55Af45 and EPJi47A153., as an~lyzedby EDX (energy may be s~bs~dntia~ly more complex than is sketched dispersive analysis of X-ray spectr~sco~y), were here, because of more complex source/sisk structures prepared by co-melting the ~gh-purityc o ~ p o n e ~ t s or the s i ~ u l t a ~ ~ formatjo~~ o~s of ~ ~ ~ ttypes i ~ ofl e in a water-cooled Cu crucible and s u b s e q ~ homa~~t defects as, e.g. vacan~iesand antisite atoms on the genization by zone-m~lting* From these rods cy~i~drical s ~ e ~ i ~with e ~ ia s rin~-shape~ ledge (see various s u b ~ a ~ t i of c ~isn t e r ~ ~ e ~compounds ~ l ~ i c which Figure 2) were spark-cut. also may c o ~ t r i ~to~ the t e length change due to their using this high~y s y ~ n i e t r ishape, ~ ~ ~ e ~ s ur t~~ rib~a t i o~~due s to rel~xat~on volLmes (Fiihnle and Meyer, 1998). s ~ e ~ i m ebendi~g n d~iring~emperatur~ changes are It should be poiiited out, that in the case of such a eli~ii~ated. After etching the specimen with a mixture ~ e f e c t - i n ~ ~ dc ~ e d~ a y change ed of a s~ecimenproperty, o f 2 parts of conc~ntratedHN03 aixd 5 parts of 36% two types of e~~erinients inay be pe~furmed. As two parallel p ~ a n ethe ~ ~front plane and the d e s c r i ~ eabove, ~ after a single t e n i ~ e r a ~ uc~~e~ ~ i g eHCI, , ledge plane (see Figure 21, were polished to a high the time ~ e p e n d e of ~ ~the e response may be observed, ~ i t ~of. a t w o - ~ e alaser Byi menm ~ ~ ~ t e r n ~ t ~the v ~same l y , ~ ~ ~ f o ~ m is a tai ov~~i i ~from ~ ~ l e optiezl ~ e ~ e ~ ~ ~e~i inte~fero~eter, the variation of the s p e ~ ~~~easuring e x ~ e r i ~ ~ ewith n t speriodic t e m ~ e ~ ~ tchanges u r e and the length (20 mm) can be tested directly to an ac~LIracyof phase shift observed for the response function, Specific s atomic defect heat m e a ~ u r e ~dewing ~ ~ s~ ~ g h ~ f r e t~e ~~p~e ~ ~nt cL yt r eabout 20 nni which c o ~ e s ~ o ttoi ~an of about 3 x IQv6 in the case of the o s c i ~ l ~ t ~were o ~ s reported r ~ ~ e n t l (y ~ ~ ~ f t m a ~ h econcentration r, g~neratio~i of vacancy type defects. The specimen 19971, teiiiper~turewas ~ o ~ i i t by ~ rtliree e~chro~e~-~~lu~ For thermal defect studies in interi~~tal~jcs by t ~ e r ~ o ~ oy ~i e~ ~I ~eaisn pical ~ t ~ ~ ~ ~ ~r a~ ta ~d ~i ~r en ~ h i ~ ~ - ~ r ~ c i st i o~n~, ~ ~ d ~ ~~ei r~ ~a tnu tmi eatan l~ yincre, of less than 20 K at 1300 over the ~ e a s u rlength i~~ mental ~ ~ c h ~ laser ~ s in~erferomete~ o n was employed. on the specimen and a temperature eqLiilib~~ti~n means of this contact-free technique, making use time of 200 s after a te~peraturechange; this prese~itly a two-beam laser interferometer (see Figure 21, limits the time range of the l e ~ ~ ~ t h ~ cstudies. ~ange ~ ~ a ~ u r e m eup n t to s li~ghtemperat~resare feasible as For studying defects in intermetallic conipounds demonstrat~din the cases of Fe~~Al45 and Ni47A15~ perturbed angular c o r r e l ~ ~ t i(PAC) o~ ~ ~ ~ ~ s u ~ e ~ e chaefer et al., 1999a,b;Calin, 1999) with an accuracy were also used ( ~ o ~et~al.,i ~ 1997). s in ~ ~ o defect ~ i c~ o n c e ~ t ~ a t iofo nabout 3 x
T h e ~ Defects ~ a ~ and ~
rac I
3.2 Thermal vacancy formation in intermetallic compounds was studied recently by positron lifetime spectroscopy (Wurschuin et al., 1995b; Schaefer and Badura-Gergen, 1997; Schaefer et al., 1990; Brossinaiin et al., 1994; ~ u m m e r l eet al., 1995; VVGrschum et al., adura-~ergenand S c ~ ~ e f e1997; r , Schaefer et al., 1999a). From these studies three main features are evident (see Figure 3): -
At ambient temperature no constitutional vacancy concentration higher than 10-6 was found in nominally stoichiol~etric Ni3Al, TiAl, Ti3Al, Fe3Al, and Fe3Si. This i s concluded from the short positron lifetimes in these coinpounds measured at a ~ b i e ~ t - t e m p e r a t u(see ~ e Wiirschurn et al., 1996a; Scliaefer and Badura-Gergen, 1997) which ai-e similar to the free positron lifetimes in pure metals with similar valence electron densities and
-
-
~
~
u
,
~
i
~
~
279
negligible vacancy concentratio~s(see et al., 1996b). This conclusion is ~ d d i t i o ~ a l ~ y corroborated by the appearance of an S-shaped increase of the mean positron lifetime at elevated temperatures, which - as in the case of pure metals - is ascribed to the trapping of positrons at therma~lyformed vacancies. Positron trapping at thermal vacancies begins in the close-packed intermetallics such as Ni3 Ti3Al (see Figure 3) at mtich hi than in the bcc-type compounds FeAl Fe$%. This indicates lower thermal vacancy concentrations and higher effective HF values in the former case. In NiAl (Figure 3) the high positron lil'etinie that is almost independent of te~peratureis attributed to positron saturation t r a ~ p i n g at i ~ i ~ o b i l e vacancies (see below) or coiistitutional vacancies which obscure the te~perature-de~endent thermal vacancy f o r ~ a t i ~ In n . this case, other t e c ~ n i ~ u e s such as tiiii~~differe~tial theriiisrl expansion studies after temperature changes (see below) can supply specific ~nformation.
igure 3 Temperature variation of the mean positron lifetime 5 due to thermal vacancy f o ~ a t i o in n Ni74 1Alzs 9. Ni76 sA1235 (Badura-Gergen and Sckaefer, 1997), Ti48 SA151 5 (Brossmann et nE., 1994), Ti65 6A1344, Ti77 ,A122 9 ( ~ u r s c h et u ~al., 1996a), Fe75Si25, Fe79Si21 (Kiimmerle ef al., 1995), Fe6lA139, Fe63A177 (Wiirschum et al., 199%) and Ni50A150 (Schaefer and BaduraGergcn, 1997). The close-packed structures are denoted by open symbols and the bcc-type structures by filled symbols. Roundedoff co~pos~tions are glveii in the figure, reproduced with permission from Elsevier Science
The te~peraturevariation of the positron tra~ping rate (Figure 4), which can be derived froni the positron lifetii~e spectra according to equation (6), yields high or low values of the effective vacancy forination entlialpies H c for close-packed or more op~n-structL~red bcc-type compounds, respectively (see Table 1). These results can be understood by nearestneighbor (nn) bond models or ~ ~ cal~ulations ~ e r ~ a d ~ i r ~ ~ ~1997). ~ r gIne n ~ ~ 1, see b ~ c~~ a eef and the simplest picture this may be according to nn bond adura and Schaefer, 19931, ascribed to a and a higher fraction of higher c o o r d i n ~ t i o~~1~~~~ ~ tra~sition~metal bonds to be broken when a vacancy on the t r a n s i ~ ~ o n ~ ~sublattice etal is formed in closepacked AJB compounds in contrast to the more o ~up o~~ n~dThis s~. ~picturee, o p e n ~ ~bcc-type ~ r ~B2~ c ~ however, is ~ ~ s u ~for9 c ie.g. e ~~ ~2 - ~ iwhere ~ l , the e x ~ e ~ m e n t ~ ~ ~ l y - d e t e vacancy ~ m i n e dformation enthalpy (1.5 eV) exceeds the low value (0.93 eV) derived from theory for the ~ r a ~ s i t ~ ~sublattice. ~ - ~ n ~1x1t f~ l f ~
c ~ l c L ~ ~ a tof i o ~the s vac~ncyf o ~ ~ i ~ ~~ i iot ihi~ l ~in ies Ni3A1, Fe3A1, FeAl, and NiAl (see Table 1) the H t values for the t r a n ~ i t ~ o n - ~ esub~attices ta~ are, however, o t ~ e ~ w i sin e uod ~ ~ ~ withr the ~ ~ ~ e x p e ~ i m e ~ t adetermined l~~ values, whereas the WF values for the At ~ ~ ~ b ~ a are t t icc ea ~~ ~ ~tol be a i~iuch ~ e ~ higher than on the transition metal s~blattices.This ~ i ~ predicts thermal vacancy f~ ~ r ~~ a~ t ~ oe ~f one the ~ ~ ~ ~ , t r ~ ~ s i t ~s ou ~b -~ ~a (Fu ~ t~et~~af., ~~ 1993) e ~ as demonstrated recently in the case of B2-FeAl by the coin~identlymeasured ~ u p ~ l ebroadening r of the p o s ~ t ~ o ~ - e ~a~ ~c tir ~ ni ~ photon ~ ~ a ~line ~ o~n ~ ~ l ~ e r 2000; ~ ~ l lete al., r 2001b) (see Section 3.4). When we discuss prefere~~tia~ formation of vacancies on one sublattice of an ordered intermetallic cornpo uiid as d e t ~ c ~ se dp e c i ~ ~by ~ ~p loy~ i t r o~ ~~ n ~ h i l a t i o ~ spec~roscopy,we imply that sim~ltaneousl~ antisite atoms are created to maintain stoichiornetric cornposition in analogy to triple defect formation in B2 * ~ ~ ~ ~r u~ ~ ~ ~~ a~s~i~l e~r~ 1968)~v s s~ ~For , c ~ n i p ~ ~ swith on
Arrhenius ~e~resentation of the positron t ~ a p p i nrate ~ CTC,{ l e f t - ~ a no ~ d i ~ ascale) t e of Ti3A1, P-Ti3Al ( ~ ~ ret s c al., 1996~1,TiA1 ( ~ ~ o s s et~ al., a ~19941, n Nip41 (Badura-Cergcn, 19951, Cu3A~1( ~ ~ m and ~ Shim, ~ u 1996) c ~(measured above the disordering t c ~ p ~ r a t ~re), i r e AI, Ni, Fe (Schaefer, 19871, FqAI (Schaefer et al., 19901, Fe3Si (Kummerle et al., 1995), Feh1Af39~Fe63A€;f?~ W ~ r ets uk., ~ 1935b), h ~ ~CuslZn49 ~ (Shim, 1988) and of the thermal vacancy concentration Cv from ~ ~ 19761, a ~ and ~ ~ i ~ r e r ~ nthennal tial expansion studies ( r ~ g ~ t - ~ aordinate nd scale) of Fe31A149 (Kerl et al., 19991, C o ~ (Berner, ~ i s ( ~ a s~ ~ Ia e 2968). ~~s ~~, Note the high vacancy c o n c e n ~ ~ a ~ ifor o n ~Fe,,Al,,, Ni,,Ga,, and Co,,Ga,, d e t e r ~ i by i~~~ d ~ ~ e r~ ~x ~~ at nistudies. ~~ i ~ nThe actual C, values were deduced by the present authors from the data grven 111 the rcferenms cited, For the vacancy comcentratton in Ni47A153 (Schaefer et al., 1999b) the value Cv = 3AIjlo was uscd. The abscissa IS n o r n ~ ~ to ~ ~the ~~ e de ~ t t~e n~ g~ ~ r a tTitI* ~ r For e s further details see {Schae~erand ~ a d ~ r ~ - C ~1997) rgen~
~
28 1
Thermal Defects and Diflusion
Xe P Effective vacancy formation enthalpies HF derived from cCv(T) according to equation (1) with a te~peratureindependent CT value; vacancy equilibrium concentrations CVat Ti,4/2 (half the inelling temperature TM);self-~iff~~sion or foreign diffusion activation enthalpies Qsy ; and eRectivc vacaiicy- ration enthalpies H v and self-diffusion or foreign atoni difksioii activation enthalpics, Qsu. The values for Cv(at TM/Z)and 25'; are taken from differe~itialthermal expans~oiidata on FeAl and Fe3Si (Kerl et al., 1999) or from positron lifetime data for the other compounds a s s u ~ i n gc = 4 x 1014s-l- Thc o-values are derived from a comparison of the positron trapping rates (Fig. 1) and a€diffe~ent~a~ thcrmal expansion data (Kerl e t id., 1999). Values for the vacancy formation volumc, VF, and the vacancy migration volume, V y , are given for Fc,Al,,. For further references for the present data see Schaefer and Badura-Gergen, 1997
H{ (ev) TM
C o ~ ~ o u n d Structure
(IC)
exp.
theor.
CV (at TM/2) 25'; (w7) (k,)
V ~ ~ : 1 . 8") 7 0.014 V~1.2.651' 0.083 0.21 Vpe:1.25 ') 240 100
FetxA139
B2
1660
MiAl
B2
1911
0.98 1.0 a) 1.5 a)
QsD
(ev)
4.9
Ni 3.15
0.7
Ti 3.03 Ti 3.1 Fe 2.44
5 3.7 3.7
Fe 1.65 Ge 3.25 Fe 2.76
Ni 3.01
EI,M(eV exp.
theor,
@
vy
(a) (a)
CT
(10'5 s-I)
0.M
3.6
1.7 2.14 S ) 1.5 a)
1.7 [ ) , j )4.6 'l
1.3
1.8 a) 2.1
")Schaefer et al., 1999b "Fu and Painter, 1997 ')Mayer and Fahnle, 1997
~ F etUal., 1993 ')Wurschum et al., 1995b fkentziiiger and Schober, 2000 69Fahnle et al., 1999 ~ ) ~ i s hand i i Farkas, ~ 1997 '1 Muller et al., 2001b /)Wolff et al., 1997a
experiments the entire defect ensemble has to be taken into account when theoretical studies are performed. When the thermal vacancy concentration 1s high (see Figure 4) and the vacancy binding energy is high (0.38eV predicted for FeAl (Fiihnle et al., 1999) divacancies may contribute to the defect pattern (Wiirschurn et al., 199513). An association of thermal vacancies to form divaccancies is indicated by ix relatively high vacancy formation volume of 1.42 R (s2 is the mean atomic volume) (Wolff et al., 1997a; iiller et al., 2001b) and by a large vacaiicy migration volume of about 4.6 SZ (Muller, 2000~~ ~ l l ete al., r 2001b) derived for €32-FeAI from positron annihjlation under pressure (see below). With the nn bond energies derived from the concentration dependence of thermal vacancy formation, an estimate of the thermal concentration of antisite atoms - iieglecting tlie forniation entropies can be given. This yields values that in Ni&l, TiA1, and FeAl are higher than the thermal vacancy concentrations (Badura and Schaefer, 1993; Schaefer and ~ a d u r a ~ ~ e r1997). ~ e n The ~ specific studies of
thermal vacancies by positron annihilation spectroscopy and tiiiie-differential dilato~etry (see next section) reported here, were p e ~ f o r at ~ tee~m ~ e r a t u r ~ s well below the critical temperature TCfor disordering, so that the data are characteristic for the nearly fully ordered state. The differential thermal expansion studies of Kerl et al. (1999) on FeA1, which are extended to high temperatures, yield indications of temperat~re-dependentchanges of the vacancy formation enthalpy which may be due to disordering or structural transformations. Calculations of the teniperature variation of the vacancy formation enthalpy with the order parameter were performed for y-TiAl (Badura and Schaefer, 1993) as well as for Ni3A1, NiA1, and FeAl (Badura-Gergen, 1995). Thermal vacancy formation 8s a function o f composition has been studied in the cases of Ni3AX (BaduraGergen and Scliaefer, 1997), €32-FeAl ( ~ ~ r s c et h u ~ al. 1995b; Schaefer and Bad~ra-Ger~en, 1997; al., 1999) and Fe$%(Kiimmerle et al., 1995; Kerl et al,, 1999) and can be understood in terms o f nn bond models with bond energies independe~tof compo~i~ion
282
~
~
f
~
~
equation (2)) were derived from the temperature variation of the time constant f~ ~ e t e r ~ by ~ ~i nt t~i ~ g the e~uil~bration process by an exponential (see Figure 6). The low value o f the pre-expo~ential factor compared to the Debye frequency can be attributed to a total number of vacancy jumps of about for the e ~ u i ~ ~ b r a tprocess ion of the va~ancyc o n c e ~ t r ~ t i o ~ which may yield a ~easonabledensity of sottrees or f ~ u n dfor sinks. A high value of H y was ~ddit~onally the nex~-~earest ~ ~ i g l i b o(100) r vacancy jumps in ~ ca~cu~ations (Fiilinle ~ ~et al,, ~1999) ~ FeAl by ~ (see Table I). 353 40 50 0 The vaca~cy~migration process i s closely related to the recent discussion of the p r i ~ ~processes ~ y of Vacancy formahon eiithalpies in F e 4 compounds self-diffusion in intermetallic compounds, Studies of as a function of the composition as measured by positron ~ l Sepiof, ~ ~ ~ lifetime ~ ~ e c t ( ~~~~s~~~~ o ~ c ~et d., ~ ~f995b, EII ~ c ~ a e F e r the Fe jump vector in FeAl ~ V o and ~ a ~ ~ ~ ~studies a t ia v ie ~at ithe~ atomic ~ d i f f ~ ~ i oonn Shirat et al., 1989), or by ~ i ~ e r e ~ thermal tial ~ ~ ~(A Ho ~ and ~ Dodd, s ~1978,o 0 Kerl i ~et at., 1999). The e r al., 1998), and tlie the two s u ~ ~ ~ t t i c~ e~s e h r et full line is ~ ~ ~ l cinuterms l ~ of t ~a nn ~ bond mode1 (Badttrae ~ t r e ~ elow ~ yvacancy co~ice~itratio~ expected on the Gergen, 1995) by II fit to positron lifetime data ~ ~ u r s c et ~un~ AI sublattice from ca~culations(Mayer and ai.,1995b), ~ e p ~ o ~ with u ~ e~ de r m ~ s from ~ ~ o Ekevier n Science FZhnIe, 1997; Fahnfe et at., 1999) favor the picture that the Fe diffusion occurs by simultaneous jumps of two neighboring atoms giving rise to a ~esu~ting jump to a next-nearest neighbor (100) site or a third~~iear~st neighbor (11 I} site without a measurable occupancy density ~easurementsafter ~ u e n c ~ ~ i n gtime of the AI sublattice by v a c ~ n c i ~ ~ . As conclude^ from the ~ o ~ ~ a rofi H s oy for ~ pure bcc iron (< 1.3 eV Schaefer et al., 1977) to that of ~~~~~~~~~
If
and ~~~~~i~~ can be deduced (see Table I). These values are higher than those for pure metals as, e.g. for s-l (Schaefer, 2987)). vac~nciesin Af (a-= 4 x The ~ i ~ r a t oi fothermal ~~ vacancies at high ternperatures was successfuIly investigated recently by of the e q ~ l i b r ~ t i oprocess n positron l i f e ~ i ~studies e of tlieriiial vacancies after temperature changes at high ~ r ~ c het u al., ~ 1995b). By this ~ r o ~ e ~the u r ~e i g r a t i ~ofn thernzal vacancies, which are i ~ e n ~ i by ~ etheir d pasitran l ~ ~ e t ~ can i ~ ebe s ,studied ~~ ~~ o~~f q~ u ~e ~ ~ losses l ~ i ~or ~ ~ ~ ~ a e v c o t eEects clu~~ering in earlier ~ x ~ e r i m e ~n tS~e i d m and ~ ~Baluffi, ~ ~ z ~ z a1972; n ~ ~ h n and o %no, 1978). 1 a high value H v = 1.7 eV (see Table I) and a ~ ~ e - e x ~ ofactor ~ e n1/t~,o ~ i ~=~1.9 x 108 s-' (see
~ 2 - ~ e 6 ~(1.7 A IeV ~ ~~ ~ r s c h u et n i al., 1995b) the vacancy migration enthalpy tends to increase with ~ h values and high decreasiiig Fe content. ~ i high > 1, the removal of thermal vacancies ratios ~~~/~~ beconies difficult due to easy forination of thermal vacancies at relatively tow temperatures where the vaca~cymigration is slow. This appears to be the case in B2-Ni52A148 where positron s a t u ~ ~ t i otnr a ~ p ~ n g at remnant vacancies occurs at ambient temperature y ~ o n ~a n~~ ~e a i~ n(1.5 ~g ex 1 0 ~ s) at peratures (685 K) and positron lifetime n i e a s ~ r e ~ e n(t s~ ~ ~ s c aiid ~ u Schaefer, n i 1997; Zhang et al., 2001) a time constant t~ < 6.2 x 106s for the removal of vacancies could be estimated. Ths yields, together with the pre-exponential factor of h al., u l ~ ~ E , Q 5 x 10-9 s found for FeA1 ( ~ ~ ~ s c et 199Sb), a value H y = 2.1 eV, This is in good agreement with recent molecular dynaiunzcs studies (see Table 1, Mishin and Farkas, 1997) and with further lengthvacancy studies in NiAl by time~di~erential ~hang~ after s fast ~ ~ ~ ~ ~ ~which ~ will a tbe ~ discussed in the next section. A c o i ~ with ~ ~ respect ~ t to the ~ i ~ r a t of io~ i should be made. Here, according to the values given in Table 1, a value Ny =: QSD=0 ~ is anticipated ~ ~ (QSa e is the ~ activation enthalpy for self diffusion). From this an i s o c l ~ r o nannea~ing ~~ of non-eqL~ilibriunivacancies by long-range m i ~ r a t i oi s~ ~expected in a conventional i s o c ~ ~ r o en ~~l p e ~ ~ on e nat I h time scale at about 35OK; this behavior in fact is observed after lowt e i ~ p ~ r a t electron ~re i r r a ~ i a t ~(see o ~ Figure ?), thus co~firmiiigthe low value of El? in Fe3Si. ~~~~~~~
As ~ugge~ted above, length-changen i e a ~ ~ r e ~ eafter nts fast t e ~ ~ e r a t uchanges re should enable specific studies of t h ~ r ~ ~ ~ ~ -f odremf aet c~ ~ o~. The time-depende~tisothermal contraction o f a Fe55Al45 specimen, after cooling from high temperatures to various slightly lower temperatures is shown iiz Figure 8. The accelerat~onof the e ~ ~ l i ~ r a tprocess ion with an increase of the final t e ~ ~ ~ e r(Figure ~ t u r8%) ~ is clearly visible. For a reasoi~ab~e fit of the time d ~ ~ ~ e n d e of n c the e i~othermalcontraction o f Fe55Al45 in Figure 8 two su~erim~osed e x ~ o n e ~ t i a lare s required. From the l temperature variations of the time constants t ~ and t ~ 2 attributed , to the c o n ~ ~ i b ~ ~with i o nthe s higher or
~ i ~ 7u Isochronal-annealing r ~ (fa 30 rnin) dete~~ination of the mean positron lifetime T in Fe;% after irr~diatioi3with r e electrons o f energy F , = 2.5 M ~ Vand dose #e = s 1 0 ~ ~ e-jrn2, reproduced with permission from Elsevier Science I=
the smaller amplitudes, respective~y(see ~ i ~ u9), r ethe a c t ~ ~ ~ tparai~eters ion = (1,s k0.2) e\l and t& = 4 . 105 s-‘; HE”= (0.6 f0.1) eV and tg& = 0.83 ~~7~
(9a) (9b)
are derived (see equation^ (31, (7)), The data of equation (gal are identified with the vacancy migration ~9arameters (see Table 1j which. were s ~ e c i ~ ~ ~ l l y determined from equilibration studies by positron lifetime s ~ e ~ t r o ~ c~o ~~ y~ r s et~ al., h u1995b) i ~ oy1 FeAl. This coincidence in the case of FeAl dernonstrates that thermal vacancies in 2 ~ ~ t e r ~ e t a can llic~ be studied by sensitive lengt~-chaiige~ieasurements. The second process in tlie lengtli-cliange studies on Fe5sAlit5 (Figure 9) is characteri~ed by a weak temperature dependence that requires further discus~ ~Figure ~ s Wa sion. It should be xioted that the ~ s o t h e rin niay need an even more complex description because of the initial ~ n c u ~ a t ~behavior, on w h i ~ hm i ~ i ~~t ~ ~ ~ a t characteristics of a nucleation and growth process (Rieux and Goux, 1969). The te~iperatu~evariation of the ~ ~ (Al,l /lo) and ( ~ l ~ 2of/ the l ~ length ~ change ofFes5A14~ as derived from two-ex~o~ential fits to the data in Figure 8b are presented in Figure 10 together with the coi~ceiitrat~on of v a c a ~~a~~~~ ~ sites as derived from positron-lifetime spectroscopy (Wiirschum et al., 1995bj and the cr-value given in Table 1, The ~ e ~ ~ i
84
P
Time dependence of the i ~ o t h e I ~ acontract~o~ 1 Al of a FessAl45 s p e ~ ~ I i after i e ~ fast cooling from the initial tei~peratureT, to the fiiial te~~perature T f ,a) variation of the ~ q u ~ l ~ ~ r rate a t i owith i ~ Tf and b) variation of the a ~ p l i t u of ~ ethe 1 the process with the l~ngth-c~iange with the initiai temperature 7;where the time scale i s normalized to the relaxation time t ~ of higher amplitzide. The different values of Tfcan be neglected in the evaluation of the amplitudes. The gray lines denote model fits to the e x ~ ~ r j m edata ~ t a (see ~ text), reproduced with ~ e ~ ~from i ~IEIsevier ~ ~ Science o n
of the ~~~~~/~~~ data at high temperatures (Figure 10) can be e x ~ ~ a by ~ ~defect e d losses during cooting, if the vacancy ~igrationen~halpyof equation (9a) is used. From the length-change amplitudes of Fe55A145in ~ ~ the preigure 10, the activation e n t ~ a l p iand ~ ~ ~ o ~~ ~ ~~ i ~t i ~t l ~ ~ s
N~= * (1.0 ] s0.1) eV and 111-A 6 + s".'/kB 4.3;
v
A
HF+* = (0.9 f0.1) eTT and ln-
=t
Pp
4
P I
-5
(104
v,+ SF-'/kB = 2.7 v
1Ob)
s i ~ ~ to~ ithe~ ~ vacancy r f o ~ a t ~ oentha~~ies n determined specifically by posi~ron-lifetiiiie spectroscopy ( ~ u r s c h u met aE., 1995b) are derived (see equation @)>,The results given in e ~ u a t ~ o n(10%) s and (lob) f ~ t h e r i i i ~d ~e e~ o ~ ~ t that i ~ aboth ~ e processes originate from thermal va~aiicyformat~un,where high annih~lati~n for thermal vacancies in Fe3A1 (Franz e1: f ~ r ~ ~ a ~e ni tor on ~ ~ eSF s > 3ka are derived when a/., 1995). A higher value of AF'/Y= 1.42 (WolfT h V / V = 0.53 (see below) is assumed. From a comOZ., 199%) was derived from the h i g h ~ t e ~ p e r a t ~ r ~ parison of the present data with the positron ~ ~ e s s udr e p e ~ d e nof~posi~ron ~ anni~ilationv ~ ~ a ~ ~ ann~hilationdata ~ ~ u r s et~ al., l ~1995b) u ~ on B2studies on B2-FeAt wlzich was a t t r i ~ u tto~ ~the FeA1, we can ~ o ~ c l uthat ~ e the f o r m a ~ i ~and n of some of the vacant lattice sites to ~ i ~ r ~ of ~ i~ u n ~ vacancies ~ can ~ be ~ s ~ ~ c~~ ~ c aaass~ciation l ~ yl ~iv~cancies. detected by t i ~ e - d i ~ e ~ el en nt ~g ~t ~l - c ~ studies, ~n~e Studies of thermal vacancies by ~ i ~ e ~ d ~ ~ e ~ e n From the c ~ ~ p a r i s oofn the total length change ge may be partic~~l~r1y ( ~ Z s * -+ and the c ~ ~ ~ e i C\I ~ ~ofr vacant ~ t i ~ i ~~e n ~ t h - ~ h a ni~~asurenients 2-NiAI, where s p studies ~ are ~ ~ lattice sites in BZ-FeAf (Figure 10) the volume of e c ~ n ~ t i t ~vacan~~~nat the vacant lattice sites can be d ~ t e to ~ be ~ n scarce ~ ~ and where i m ~ o b i ~and cies obscure the detection of thermal vacancies by AV/ V = 0.53. A value o f A Y / Y = 0.7 for this range means of specific pQsitron"a~nihi1atio~ studies due to was derived from the pressu~~~dependent positron ~~~~~~~~
f 0" 0.
0.3
t ature ~ ~ r ~ a t i oofn St ange 11 Isotlierni~~ ~ € ~ e - ~ ~ ~ ~length e r e nchange t ~ a l Al(t> 1 and A l s ~ (after /~~ 1 or in B~-NLQAI~~ after cooling or heating the specimen to Tf, A145 derived from tw 1 fits reproduced with petmission from Elsevier Science to the tmie differential length changes after temperature changes (Figure8b) and of the concentration CV (0) of vacant lattice sites d e t c ~ ~ i ~ €rom i e d the positron t r a p ~ ~ n ~ rates aC'vfll) iit Fe61A13~ ( ~ u r s c et~ ial.,~ ~1995bj ~ (see Figure4) making use of t7 = 1.3 x 1Oiss-' (see Table I). The deviation at 1 g h ternper~~tur~s of the lengtl~-cl~~nge and amplitude^ from an Arrhenius behavior IS due to vacancy AY fosses on cooling (an increase from the present cooling rate of ln - ~~/~~ = 4.8 (f2b) t Kis IS ~ e ~ i r a reproduced b ~ e ~ ~ with peimrssioti from Elsevier Science ~~~~~~
v+
saturation of ~ositront ~ ~ p p i n g~ .e n ~ t h - ~ mea~an~ on ~ ~ - ~ arei shown ~ ~ in. A Figure i ~l l. ~ In s~reineii~s this case eq~ilibrationexperiments upon cooling as well as upon heating (Figure 11) could be performed. Froin these data on NiAl it is evident that, with about time constants as in B2-FeAl the same eqLii1ibratio~~ (Figure 9j, the a ~ p l i t ~ of ~ dthe e length change is much srnaller than in FeA1 which indicates a lower ratio ~~/~~ and a higher value in the case of ~ i 4 7 A The ~ e ~ ~ g ~ h ~ c~l ~ s a~ nt gh ee~ieasured r~s for Ni47A153 ~ ~ c h ~eti al., e f 1999a) ~ ~ are well described by a single exponenti~land from the teiii~eratu~e variation of the rate c ~ n s t l~/ ~t ~(see t Figu~e12) fur the first time a vacancy ~ig r atio nentlialpy with the high value
This NF value for N i 4 ~ A lis~ ~higher than in eFe61AJ39 and c ~ the earlier ~ ~ ~~n - j e c t~deduced ~~re s from the lengtli~c~i~nge data in Fi~LIre11. fn additio~, a ~el at i velhigh ~ v a c a n c y - f o ~entropy ~ ~ ~ ~as~ ~in other inte~netallics(see Table 1) is observed. The
-
~ ~ 'Ia4 ~ ~ .. 'cn
Y
& -r. 2-
l0"
'I
0.8
and with tbe ~ r e ~ e x ~ o nfactor en~~~l ~~~~~~
1.O
I .2
1 / T [ I a3K"I]
1.4
12 ~ e r n ~ e r V ~ ~a u~ r ~~O~f the t ~ ~~ n
t .Ei ~ rate
l / a in B2-Ni~7A153deter~inedfrom ~ ~ p o n ~ ~fits t i to a lthe isotherms, equation (7j, in Figure I I after cooling ( in f32-NiAI could be determined. These values are heating (U) the s p e c ~ ~ eIn n . addition, the reaction rates (0) similar to those in the case of B2-FeA1. for clustering o f remnant vacancies in NiS2Al48 after lowFrom tlie l e ~ ~ t h - ~ h a ndata g e of Ni47Al53 tlie t c ~ ~ e ~ ~long-term t ~ i r e anneal in^ (Zhang et at., 2001) as t e ~ ~ ~ evar~ation ~ ~ t u of~ the e ~ e i ~ ~ h ~-~ ~ i ~ ~p lai t nL ~ ~~ e ~te ~ ~by t ep do s ~ t ~ o ~ ~~s p ~~ c ~ e~(see o sFigure ~~ o ~~14) ~ are~
3
~ (see ~Figure~ 13) was / derived ~ yielding ~ the values
plotted
~
e
286 N A ,i,I Ta=660 K
z -m
7
10"
m
10"
0.8
0.9
1.0
observat~o~~ that the present e ~ p e r i ~ e n tHF a l value for the time ~ e ~ e n d of e ~the c ~ p u s i t r o n ~ t r a ~ rate ~i~i~ ~ i 4 7 is~ higher l ~ ~ than ~redictedtheoreti~~lly (see ETCV after a fast t e ~ ~ e r change ~ ~ ~ when i ~ a e pressureTable 1) may raise the question as to how many nt p o s ~ t ~ o ~ - ~ rrate a p ~ ~ ~is g antisite atoms are fomed or whether a t e ~ ~ e r a ~ u r ie~ d e ~ e n ~ especific assumed (see Figure 16). The curves in Figure lija dependeiice of the c o n s t ~ t ~ t i o vacancy n~l concentraare characterized by a two-stage ~ ~ process. ~ tion inay c ~ n t r i ~ ~ ~ ~ . In the l o n g t e ~ main ~ process it is e v ~ ~ that e ~the t final to the high ratio of ~~/~~ in NiAl discussed r i u ~ co~cent~ation CV is reduced and v a c a ~ c ~cap1 e ~ be removed only by ~ o n ~ - t e r e~ ~ u i ~ ~ bvacan~y l o w ~ t e ~ ~ e iimea r ~ ~ ling, u r ~As shown recently for y p o s i ~ ~ olifetime n s ~ e c ~ ~ ~after s c ~specipy men ~ ~ ~ p a r a ~m i oo n o, v a ~ a ~ i care ~ ~ savailable as conciuded from the positron lifetime z = (179 rrfI: 2) ps (Z1iang et d, 2UOl). They become mobile during longterm a n ~ ~ ~and l i fomi ~ g clusters of more than 20 cooling ~ ~ c as ~e v i~~ e ncc eby ~~ thee ~ ~~ p e a r a nofc a~ tong intet moled cell wall vacuum positron lifetime (see Figure 14).
n
32 ~
In a d ~ ~ ttoi othe~ ~ n ~ hofav a~~ ~~ n € ~ c yo ~~ ns~ a tand io~ ~ g ~ a t derived i o ~ from temper~ture~de~endent studies, specific defect data on the formation volume V;, which has been ~ t ~ d i ebriefly d earlier (Emrick 1980; ickman et al., 1975; Franz et al., 1995; Wolff et ., 1997a), or the migration volume V v (Emrick, 19&l)of vacancies can be deduced from the pressure ra~ure dependence of h i g ~ ~ ~ e i ~ ~ e pos~tron-l~fe~ime ~ e a s ~ e m e nAn t ~appropriate . h i ~ ~ ~ t e m p e ~presatu~e a /J-Ti90Al6V4 alloy and been described recently 001b) (see Figure 15). Schematic drawing of the experimental setup for The eEect of pressure on the e ~ u i ~ i ~ r ~oft ithe o i ~ positron lifetime n ~ ~ a sas ~a function ~ r ~ of~pressure ~ ~ tand temperature (Muller, 2000) thermal defect concentrat~onCV may be reflected in
-0.56
10"
-0. h
-. -0.60 .c-I" T-
e i;z
2 P
-0.62
-0.64
Q
50
100
150
0
0.05
0.10
200
t Eh1
(a)
Figure I6 (a) Isotlicrmal equilibration with tiine ofthe thermal vacancy concentration CVof F~6~A139 at the temperature T3as a function of hydrostatic pressure p after rapid cooling from the initial temperature 770 K. The figure shows the model curves of two s ~ p ~ r i m p o expon~ntials s~d fitted to the experimental data, (b) pressure dependence of the time co~istaiit1~ af the long-term process for deducing the vacancy migration volume @ (Muller, 2000; Miiller et al., 2001b)
that the equil~brationprocess i s sigiii~cantlyslowed down when pressures up to 0.1 1 GPa are applied. From this behavior a rather high vacancy forination volume V g = 1.7 R similar to the volume reported recently (Wolff et al., 1997a) can be deduced and may be assigned to divacaiicy formation. Furthermore, from the pressure dependence of the equilibration time (see Figure 16b) a vacancy migratioii volume of V v = (4.6 k 1.1) f2 is available for the first time. This turns out to be surprisingly high. Much lower values of about 0.15 R were reported from electrical resistivity annealing experimeats after quenching of pure Au (Enirick, 1961). Effective migratioii volumes higher than this and closer to the present value are conjectured from the pressure dependence of Fe self-diKusion (SD) in FeAl (Eggersmann, 1998; Eggersmann and Mehrer, 1999) which report a lower limit value YsD = YF Y y N 1.6R for the activation of self-diffusion as well as from the pressure dependence of Fe foreign diffusion (FD) in pure A1 yielding YFD = 2.9 R (Kummel et al., 1996). A high vacancy-migration volume, i.e. a relatively strong lattice expansion in the saddle-point configuration of the vacancy jump compared to the vacancy on a lattice site, inay originate from a complex jump mechanism in which a number of atoms are involved for locally maintaining the ordered structure. A high activation volume may be supported by the model of a complex diffusion 2-FeAl ~Fahnleet al., 1999); here, in a correlated process, Fe and A1 atoms are suggested to exchange positioiis by siiiiultaneous jumps in the first step of a modified s i x - j u ~ pcycle.
+
3.
aE
The above detailed compilation demonstrates that data on tlie enthalpies and activation volumes for the formation and the migratxon of thermal vacancies are now available for intermetallic compounds and in particular for B2-FeAl. However, the long-standing question is on which sublattice, the transition nietal sublattice or the A1 sublattice, the formation of thermal vacancies occurs. This quest~oncan now be answered by experiment. To do this, we made use of coincident measurements of tlie Doppler broadening of the 511 keV positron-electron-annihilation photon line (coincident Doppler broadening technique) (Miiller, 2000; Muller et al., 2001b). By this technique the electron momenta up to the high values characteristic of core electrons can be detected. This allows for a chemical differentiation of the atonis in the vicinity of the positron annihilation site (Alatalo et al., 1995). In the case of a positron trapped in a vacancy, the atomic environment (and therefore the sublattice on which the vacancy is located) can be specifically identified. In Figure 17 coincident Doppler broaden in^ spectra of FeslA139 normalized to pure defect-free A1 are shown. The momenta of electrons giving rise to Doppler shifts above 514 keV are characteri~ticfor core electrons (Alatalo et al., 1995). The spectrum of FeblA139 measured with the equilibrium vacancy concentration at 610 IS, where due to a low vacancy concentration only partial positron trapping at vacancies occurs (T,,,, = 147 ps), is between tlie spectra of pure Fe and pure Al, because the ~artially
288 ~acanciesin se~iconductorshave been studied recently s p e ~ ~ c ~by~ l the ~ y coincident DOp13lt2r broadening tec~niquein Sic ~ ~ef al.,~20Ola)l ~ and other ~ o n i ~ soe ~ i~~ o~ ndd ~ c(t ~ 9l r~~ teta at. l ~ 1995; Saarinen et al., 1996). In SIC the vacancies on the C or the Si sublattices were directly id~iiti~ed. For this purpose vacancies were selectively generated on the one or the other sublat~iceby low- or l i i ~ ~ - e n e ~ g y electron irradiation. The vacancies on the different sublattices then were s p ~ i ~ c a l lidenti~ed y by coincident Dopgler broa~eningspectroscopy (Miiller et at., 2001a).
3
$ 2
i
51 1
cili 3
555
517
519
E [keV]
ure 27 ~ o i n c i ~ ~ ~nieasurcd tly ~opple~-bro~~~~~d spectra of the p o s i ~ ~ o n - ~ l ~~~nt n~~o~n~ l a t iphoton oii ~ i o ~ i ~tol purc i ~ eA1~(Mkfiiller ef al., 2001b). For details see
compound can drastically change the mechanical pro~erties (see e.g, Fleischer, 1993a,b; Pike et al., 1998). Ternary interi~etalliccompounds may be of delocalized positrons give rise to aniiihilation with particular interest because of high mechanical creep Fe, as well as with Al, core electrons. After a ~ ~ ~ e ~ ~l ei ~ s ~g s at ~ ~e~eva~ed n ~ e ~ e n i p e r ~ t ~as ~ r ed se m o n s t r ~ t e ~ at 770 K a positron lifetime of zv == (185 f 2)ps was for ~ 2 - ~ i ~~ F ~e ~~'1993). ~ ~~ This lt ~strengthenh~ o ~ ~ observed which i s due to positron s a t u ~ ~ t~i o r ~~ p p i n ~ing . may be affected by the ~ r o ~ e ~ tof i e st h e ~ ~ ~ l The ~ o ~ ~ l e ~ ~ ~~ r~ oe ca t d~~eue m ~a seu ~r efor ~ this v ~ ~ a n c i for e s ~which ~ r a c ~ i c ~ l ldata are availabl~. case is s u ~ s t a ~ t i ashifted ~ l y towards A1 (see Figure 17) Recent ~ ~ e a s ~ r e m eshow nts engel and Schaefer, whicli is clear evidence that the thermal vacancies are 2001), that in ~ i 4 alloys ~ ample F ~v a c a~ i ~ ~~~ formed on the Fe sublattice, In this case, positron concentratioi~sare present at anibieiit tempe~dtures annihil~tionwith high-momentu~ electrons occurs so that positron saturation trapping occurs (positron p r e d o ~ i n a n t ~with y core electroils of the A1 atoms that lifetime T 185~s);this ~ e c ~ therefore i ~ ~ ~ is ine neighbor the Fe vacancy in the ordered a~c~ sensitive to the detection of t h e r m a ~ ~ v a cformation st ~ ~ t ~ ~ r e . at high t ~ ~ p e r a t u rA~ high s . concen~ationaf vacanThe obse~va~ion that the e l e c t I . o ~ i ~ ~ o mdistrie~tu~~ bution at 770 K does not coincide with that of pure AI. can be unders~oudby the deviation of the Fe~lAIsg composition from stoic~~iometry' This gives rise to the al., 1999). In this s i t ~ ~ t i othe n merits of the time~ ~ ~ p e ~ i rofa iFe ~ catoms e in the n e a r e s t - n e i ~ ~ shell ~or d ~ f f e r e ~ i d~ i a~~~ ~ t o ~ meect h~n i ~after u ~ f&st t e n i ~ e r ~ ~ of the Fe v ~ c a i i which c ~ ~ may be even more i ~ ~ o r t a n t changes (as d ~ s c r i ~above) ~ d c m be e ~ ~ ~ o y e d ~ when the short-range order around the vacancy i~h~temperature dilaton~etr~ data are shown in deviates from that of the long-range ordered crystal. Figure 18. The fast initid s~~rii~king of the specimen In a d ~ i ~ i othe n , positron wave function may extend to due to the t e ~ ~ e r a t u r e ~ d e p e n ~lattice e n t anharthe n~xt"nearestvacancy neighbors where Fe atoms ~~onicities is clea~ly sep~rated from the del aye^ dominate. isothermal shrinking originating from the ~ ~ u i l i b r ~ t i o n Vacancy f ~ r ~ ~ aon t ~the o nFe sublattice in FeAl has of the thermal vacancy concent~ation(Figure 18a). been ~ u ~ g from e s ~~ ~ ~~ n~ c i~d eo ~~t ~b rl oea dre ~ i n ~ The tempe~~ture variation of the eq u ~ l i ~ r ~ t time ion studies after quenching ~ ~ o m i e setk ial., 1999) but has a l sthe constants derived from fits of e x p o ~ e ~ ~ t ito now been ~ e i ~ i o n s t r ~directly t e d under thermal equilii s o t h e r ~ a~~~ d ~ u r e i ~is e~i i~~ s~ s ine ~~~ g~ ue18b r ed ~~i~~~c o n d i ~ ~ Ito ~~ ~~.r t h e r m has~ rbeen ~ su~~~sted ~ i ~ l dai re~atively n~ high vacancy ration ~ n t ~ ~ a l c ~ ~ c ~ ~ a t(F~linle i o n s et al., 1999). Mote by ~~~~~~i~~~ of H y = (1.40 IO,1>eV (low v a c ~ ~ c y~ o b i l ~ t y ) again that aiitisite defects must be s i m ~ l ~ ~ n e o u s l yt o ~ ~ t h ewith r a ~ r e - ~ x p o ~ ~ e nFactor tial l / i f ~= ,~ created to i ~ a i n t athe i ~ ~overall ~ o ~ p o s i t i o n . 3 x 105 s-I for thermal vacancies. From the relatively text
289
IV
1
-1
(a) Time-depeiident length change A1 of the N i ~ o ~ e , o specimen ~ l ~ o after fast cooliiig from 1023 K to 723 K. The initial shr~nkingof the specimen due to the lattice aii~iarmon~c~ty, concomitant with the temperature change, can clearly be separated from the tim~~~ependeiit shrinkage caused by the equilibration of the thermal vacancy conccntratton. (b) Temperature dependence of the reciprocal time constant t ' ~as a function f tlie reciprocal equilibration temperature Tf (Arrhenius plot). The 1023 K, Cl 973 IS, V 923 K initial te~iper~tures T, were chosen as follows: A I073 K,
low amplitude of the time-dependent length change in ~ i ~ ~ ~ e ~a low * Aconcentration l ~ * , of thermal vacancies can be estimated which, together with the low vacancy mobility, could explain tlie high creep resistance o f ~ i 4 ~ ~ e at l o elevated ~ l ~ ~ temperatures (Sauthoff, 1993).
esses for
all
The self-di~~sivities D in ordered internietallic compounds have been discussed in the framework of various models (see Mehrer, 1996 and Chapters 23 and 32 by deNovion and by Larikov iii Volunie 1). The tem~eraturevariations of the diffusiviti~splotted in Figure 19 on a scale normalized to TM diger by many orders of magnitude for different compounds. The main features may be understood by the characteristics of t ~ e r malvacancies: - In close-packed structures the transition-metal
diffusivities
L)
are as low as in pure metals, or
lower, due to a low thermal-vacancy c o n c e i i t ~ ~ ~ i ~ n te CV (a high H: value) and an i ~ t e r ~ e d i avacancy diffusivity (i nt e~~edi atHe v value). 2 compounds the In the open-structured diffusivities are also low but for different reasons: H; is low and H v is high (see Table l), In the case of inany pure metals the vacancy-mediated tracer difhsivity is governed by the re~ationship
a* ==f.v*Cv
(13)
where v accounts for the vacancy jump frequency and J' is the te~perature-illdependentcor~~l at i on factor. The activation enthalpies are then related by
In the case of B2-FeA1 the experimental values of the activation enthalpies (see Table 1) fit together pe~fectly~ c ~ o r d i ntog equation (14); this ~ehavior also occurs for the most recent data on €32-NiAl by Divinski et al. (2000). Thus in these interinet~llic alloys a simple v a c a n c ~ ~ ~ ~ e d i aprocess ted may govern the transition-metal diffusivity~in fact without significant temperature dependence of the correlation factor. Here it should be mentioned
t
I
1.5
A ~ ~ ~ d e ~~ ua ofs~ the~ s en i ~~ - ~ i ~ u sNi-Ni3Al iv~~~~ (H s ~ ~ c 1971 o c ~A ~ ~ - ~ (Larikov i ~ A ~er al., 19811, Ni-NilGe ~ o ~ et ~al., k1997), a Ge-Ni3Gc onak aka et al., L997), Ti-TiA1 (Mroll et a).,19921, Ti-Ti3AI ( R ~ s i nand ~ Merzig, 1995), ALP CL~~A (disorde~ed) U (Bena et al., 1965), Co-CoCa, Ga-CoGa (Stolwijk et al., 19801, Fe-FeAl ( ~ ~ ~ e r s et ~ aal., n n1997, Eggcrsmann and re^, ZOOO), AI-FeRl (Larikov et al., 1981>,Ni-NiAl (Divinslci et al., 2000), Fe-FeA1 (Tokci et a?., 199Q AlFe3Al (Larikov et al., t981), Fe-Fe&, Ge-@c3Si, ~ e - F e ~ , § (Cudc i ~ y and Mehrer, 1997), and Ce-MoSi, l.,Ge-MoSi, 11 (1.and 11 to the tetragonal axis) (Salainon et al., 2001) as well as the sel~-di~usi~ities AI-AI (Dais et al., 19971, Fe-Fe (Lubbeh~~scii and Mehrer, 1990), Ni-Ni (Maier p t nl., 1976). Ti-aTi (Koppers et al., 19971, and TI-PTi (Cerold and Nerzig, 1997) in the pure metals, The Au ~ ~ ~ in ~Cu3Au s ~(Bencl o et n ul., 1965) IS ~ e ~ § ~ iinr ettic d d i s ~ ~ ~ phase, c ~ c dThe abscissa is ~ o ~ a l i toz the e ~~ e l ~ j n g t ~ ~ i ~ e ~ ~T‘, t u r e s )$
that a c o ~ i ~ r e h e n ~study ~ v e of f-I,M as a functi~nof the ~ o ~ ~ o s i tof i oFeAl n would be of interest, This o nis valid for study could test how far ~ ~ u a t ~(14) various compositions as Qs” increases (Mehrer et nd., 1998) aad H c decreases (see Figure 5) with ~ n ~ r e a s ~AIn gcontent. - The Fe tracer ~ i ~ ~ i ~ observed i v i ~ i in e ~D03-Fe3Si (see
coiit~ntori~inatesfrom a decrease of the t ~ ~ e v~c~ncy ~ ~ c ~ n t r a t Cv i o n~ ~ u ~ ~ et ~ al., i e 1995). r l e The Si or Ge ~ i ~ ~ s i y iin~ Fe$% i e s which are inuch lower than the Fe diffusi~iti~s ( ~ e ~ r e1996) r , (see Figure 19 and Table 11, may arise from a lower thermal v ~ ~ a n ~once~tration cy and a longer vacancy jump distance with a higher vacancy ~ i g ~ a t i o n enthalpy H v on the Si sublattice.
~
ehrer, 1996) are much higher than in B2FeAl (Figure 15), due to both a high concentr~t~on and a high diffusivity of thernial vacancies. Again, the a ~ t i v ~ t i o~~ithaipies n HF and H y (see above) fit QsD (see Table 1) a c c o r ~ i nto~ e q ~ ~ t i o(14), n indicating a conventional vacancy-mediated As demonstrated recently, thermally formed atomic appear to play an i ~ ~ o r t arole i i ~in the kigtn1 the Fe s u ~ ~ a t t~i i~~~u. s i o n ~defects l ~ i ~ u psr ~~c oe s01~~ temperature ~ e c h a n ~p rco~~ ~e ~ t i eofs i n t ~ r ~ e t a ~ l i c jumps of Fe atoms on the Fe sublattices in sto~chiometr~cFe$% were also inferred by alloys, e.g. in the case of the y i e ~ d - ~ t r eanomaly ~s in ~ ~ Acr ~~ o f the ~ Ejssbataer ~ ~ ~ (Sepiol e slid ~ rF o e where ~ ~ xi1 ~ incrmse ~~ yieXd~stress with ~ n c ~ ~ a sti en ~g ~ e r a ~isu robserved e at 800 K ~ ~ e ~ r Vogl, 1993). Tlie decrease of 6)* with increasing Fe
nietry, and using theoretical studies. vacancy formation in intemetallic coinpounds obtained from pos~tron-lifetime ~pectroscopy yield high effective~formationenthalpies HF in ~ l o s e ~ p ~ c k e d structures, and low values in bcc-type ~ t ~ ~ ~ t u r e s , results that can be well understood theoretic~lly.The vac~n~y~migratioi~ e n t h ~ ~ p ~ha at high t e ~ ~ e r a t ~for r e sB the e~uilibrat~on process a As d e ~ o ~ s t r a t ehere d in a with the shear modulus G and characteri~i~ig the FeAl, the thermal f o ~ ~ a t i oand n ~iigrationof defects strength of the hardening. From a fit of equation (1.5) also be sensitively investigated by ~ii~e-~ifferent~al can to the increase of A(r between 700 K aiid 850 K a length-change studies after t e ~ ~ e r a tchanges ~ r e in the vacancy-formation enthalpy of 0.96 eV, very similar to vicinity of the equilibration t~m~eratures.From the value derived by positron lifetime spectroscopy (see pos~troii-iifetj~~ie measurements under pressure after Table 1) was derived, fast temperature changes, the a c t i v ~ t i ovolumes ~ for It furthermore was shown that, after fast heattng the formation (V: = 1.7 SZ) and, for the first time, for from lower temperatu~es,the yield-stress anomaly at the migration. = 4.6 12) of vac~ncieswere derived, 855 I< emerges as a function of time (see Schaefer et nl,, In B2-FeAl, formation of thermal vacancies is experi1997) and, from an exp~~iential fit to these kinetics, a men tally demonstrated time constant t E = 220 s is derived that fits the vacancy co~ncidentmeasuremen m i ~ a data ~ o in~ Figure 9, From the coi~icidenceof ton tine. First the tenipe~ture dependence and tinie-depei~dent the positron-e~ectron-a studies on the ternary c o ~ ~ o ~~ n~d ~ * yield ~ e ~ * kinetics of Acr (see Schaefer et al,, 1997) with the low mob~lities and low ~ ~ n ~ e n t of ~ ~t~ermal ~ ~ o n s ~ ~ c a ~ ~i coy~ a t i oaiid n ~ i ~ r a t ~kinetics on studied vacancies that may give rise to tlie high creep ~ p e c i ~ c aby ~ l ypositrons ~ ~ u r s c l i uet~al., n f995b), we resistance. The present vacancy data can explain the conclude that the high~teiiiperatureme~haiiicalprowide variation of the transition~ni~ta~ self-di~usivities 2-FeA1 are strongly influenced by thermal in intermetallic compounds. For vacancy formation. At present this streiigtlieniiig is temperature mechanical properties are closely linked discussed in terms of an interact~onbetween thermally to the d ~ (111) ~ dis/ 2 ~ formation of thermat defects, as ~ v i ~ e n c eby formed vacancies with pairs of ~ the temperature variation of the yield stress ~ n o ~ a l ~ l o c ~ ~ t i o ~separated is~ by an antiphase boundaiy, and its time d ~ p e ~ d e nafter c e fast heat~Iig. g~idlngon a (1 to) plane. The e~ico~nter of a vacancy with one of the dislocati~nsgives rise to climb, so that the disordered tube generated by the motion of the d~slocation~ cannot be restored by the second dislocation. This tubular disordering exerts a drag on the dislocation, i.e. it gives rise to the increase in yield We thank X. Y. Zhang for his comments aiid stress. ~ ~ o k e f a l v i ~for ~ ~ technical gy help. The ~ n a ~ c ~ ~ support of Deutsche Forschuiigsgemeinschaft (Projects Scha 428/17-1,2,3) is appreciated.
and Baker, 1998), (see also Schaefer et al., 1997). This increase can be specifically ascribed to strengthening via the interaction of thermally formed vacancies with dislocat~onsin analogy to the hardening of after quenching (see Ceorge and Baker, 1998; Schaefer et al., 1997) where the ~ n c r ~ a in s e yield stress Acr IS correlated to the vaca~cyconce~tration&ii by
(P'y
compounds, and, in Atomic defects in inter~~etallic particular thermal vacaiicies and constit~i~iona~ vacancies, play an i ~ p o r t a n trole for the understanding o f diffusion processes, o r ~ e r - ~ i s o ~ dtransitions, er plastic d e ~ ~ i ~ aetc. t ~ ~o un b s ~ ~ progress i i t i ~ ~has been achieved recently in s t ~ d y i n the ~ the~iodynaniicsand kiiietics of t ~ e r m vacancies ~~ by employ~ngtechniques specific to vacancy detection, for example positron~nnihilationspectroscopy and time"differentia1 dilato-
292
Properties and Phenomenology
Appel, F., Brossmann, U., Chnstoph, U., Eggert, S., Janschek, P., Lorenz, U., Miillauer, J., Oehnng, M., and Paul, J. D. H. (2000). Adv. Eng. Mater., 2. 699. Badura-Gergen, K. (1995). Dr. rer. nut. rhesis, Stuttgart University, Germany. Badura, K., and Schaefer,H A . (1993). Z. MetaZlkd., 84,405. Badura-Gergen, K., and Schaefer, H.-E. (1997). Phys. Rev., B56,3032. Benci, S., Gasparring, G., Germanioli, E., and Schianchi, G. (1965). J. Phys. Chem. Sol., 26, 687. Berner, D. (1976). Dr. rer nut. thesis, Stuttgart University, Germany. Bradley, A. J., and Taylor, A. (1937). Proc. Royal Soc., London, A159. 56. Brossmann, U,, Wikschum, R., Badura, K., and Schaefer, H.-E. (1994). Phys. Rev., B49, 6457. Cahn, R. W. (1994). In Physics of New Materials (ed. F. E. Fujita). Springer Senes in Materials Science, Vol. 27, Spnnger, Berlin, p. 179. Cahn, R. W. (1999). Nature, 397, 656. Chang, Y. A., and Neumann J. P. (1982). Progr. Solid State Chem., 14, 221. Chang, Y. A., pike, L. M., Liu, C. T., Bilbrey, A. R., and Stone, D. S. (1993). Inrerme~aZZ~c.~, 1, 107, Collins, G. S., Fan, J. W., and Bai, B. (1997). In Structural ~ ~ ~ e r ~(ed. e rM.a V~. Nathal). ~ ~ c ~TMS, ~ Warrendale, 43. Dais, S., Messer, R., and Seeger, A. (1987). Mater. Scr. Forum, 15-17,419. Darolia, R. (2000). Intermetallics, 8, 1321. Deevi, S. C., and Sikka, V. K. (1996). Intermetallics, 4, 357. Diclcman, J. E., Jeffrey, R. D., and Gustafson, D. R. (1997). Phys. Rev.,Bl6, 3334. Divinsla, S. V., Frank, St., Nerzig, Chr., and Sodervall, U. (2000). Solid State Phenomena, 72, 203. Eggersmann, M. (1998). Dr. rer. naf. thesis, Munster: University, Germany. Eggersmann, M., and Mehrer, H. (1999). Me~aZZo~z~ka i Noveshie Tekhnorogii. 21, 70. Eggersmann. M., and Mehrer, H. (2000). Phil. Mag., AS#, 1219. Eggersmann, M.. Sepiol, B., Vogl, G., and Mehrer, H. (1997). Defect and D i f . F o r m , 143-147, 339. Emrick, R,M. (1961). Phys. Rev., 122, 1720. Emrick, R. M. (1989). Ph-vs. Rev., B22, 3563. Eahnle, M., and Meyer, B. (1998). Scr. Mater., 38, 1131. Fahnle. M., Mayer, J.. and Meyer, B. (1999). Intermetallics, 7, 315.
Fleischer, R. L, (1993a). J. Mater. Res., 8, 49. Fleischer, R. L. (1993b). J. Mater. Res.. 8, 59 Franz, M., Wolff, J.. and Hehenkamp, Th. (1995). Mat. Scr. Forum, 175-178, 355. Fu, C. L., and Patnter, G. S. (1997). Acta Mater., 45, 481. Fu, C. L., Ye, Y. Y.. Yoo, M. H., and Ho, K. M. (1993). Phys. Rev., B48, 6712. George. E. P., and Baker, I. (1998). Phil. Mug., A77 (3), 737. Gerold, U., and Herzig, Chr. (1997). Defect and Diff: Forum, 143-147,437.
Gude, A., and Mehrer, H. (1997). Phit. Mug., A76, 1, Hancock, G. F. (1971). Phys. Stat. 851. ( a ) , 7, 535. Heigl, F., and Sizmann, R. (1972). Crystal Lattice Dejeczs, 3, 13. Ho, K., and Dodd, R. A. (1978). Scr. Metall., 12, 1055. Jnui, H., Ishikawa, K., and Yamaguchi, M. (2000). Intermerallics, 8, 1159. Jirdskovi, Y., Schneeweiss, O., and Sob, M. (1995). Mater. Sci. Forum, 175-178, 399. Kentunger, E., and Schober, N.R. (2000). J. Phys. Condens. Matter, 12, 8145. Kerl, R., Wolff, J., and Hehenkamp, Th. (1999). Intermetallics, 7, 301. Kim, S . M., Takeda, Y.,and Kogachi, M. (1994). Scr. Mater., 34, 1845. Kogachi, M., and Haraguchi, T. (1997). Mater. Sc2. Eng., A230, 124. Koppers, M., Derdau, D., Friesel, M., and Herzig, Chr. (1997). Defect and D @ Forum, 143147, 43. Krafiwakher, Y. (1997). Defect andDzy. Forum, 14%147,37 Kroll, S., Stolwijk, N. A., Rosenkranz, R., Herzig, Chr., and Mehrer, H. (1992). Z . Metalfkd., 591. Kiimmerle, E., Badura, K.. Sepiol, B., Mehrer, H., and Schaefer, H.-E. (1995). Phys. Rev., B52, R6847. Larikov, L. N., Geichenko, V. V., and Falchenko, V. M. (1981). DzjiTUo~Processes in Ordered Alloys. Oxonian, New Delhi. LIU, C. T., Zhu, H., Britdy, M. P., McKamey, C. G., and Pike, L. M. (2000). Intermetaflics, 8, 1119. Lubbehusen, M., and Mehrer, H. (1990). Actn Metall. Mater., 38, 283. Maier, K., Mehrer, H., Lessmann, E., and Schiile, W. (1976). Phys. Stnt. Sol. (b), 78, 689. Mayer, J., and and Flhnle, M. (1997). Defect m d D ~ fForurn, . 143-147. 285. Mehrer, H. (1996). Mater. Trans. JIM, 37, 1259. Mehrer, H., Zumkley, Th., Eggersmann. M., Gallcr, R., and Salamon, M. (1998). In Dzjiission Mechanzsms in Crystalline Materials (eds Y Mishin, G. Vogt, N. Cowern, C. Catlow, and D. Farkas). Materials Research Society, MRS Proceedings, Vol. 527, p. 3. Mishin, Y., and Farkas, D. (1997). Defect and D@. Forum, 143-147, 303. Muller, M. A. (2000). Dr. rer. nnt. #he,szs,Stuttgart University, Germany. Miiller, M. A., Rempel, A. A., Reichle, K., Sprengel, W., Major, J., and Schaefer, H.-E. (2001a). Mater. Sci. Forum 363-36570. Miiller, M. A., Sprengel, W., Major, J., and Schaefer. &E. (2001b). Mater. Sci. Forum 36336.9, 85. Nonaka, K., Arayashikl, T., Nakajima, H., Almazouzi, A., Tanaka, K., Ikeda, T., Nuinakura, W., and Koiwa, M. (1997). Defect and Difl. Forum, 143-147, 269. Ohno, K., and Kino, T. (1978). Phys. Soc. Japan, 4d, 875. Pike, L. M., Chang, Y A., and Liu, C. T. (1997). Acm. Mater., 45, 3709.
). Pike, L. M., Liu, C, T., and Ckang, Y. A (1998). ~ e ~ ~Stotwijk, ~ l .N. A., van Gent, N., and Bakker, H. ( S 9 ~ ~PEI, Mag., A42, 783. Rxeux, J., and Goux, 6.(2969). Mem. Sci, _Rev. Mdt, 66, 869. Tel-auclit, H., Cohen, J. B., and Reed, T. B. (19781, Actrr.
Mehrer, H, (1996). Z. ~ e t a l l k ~ . , ni, J., Gas, P., and Beke, D. L. (€998). Riasing, J., and Herzig, Chr. (1 995). Acru Mater,, 46, 4821, 561. Valeeva, A. A., Rempel, A. A, Miiller, M, A., Reichle, E;. J., Saar~nen,K., Laine, T., Skog, K., Nakinen, J., ~ a u t ~ ~ a l - v i , Tang, G., Spren~el,W. haefer, E-E. (2001). P., ~ a k e n n ~K., ~ s Uusrmaa, , P., Salokatve, A., and Pessa. PIfysica S t a t ~soli^^ ~ (b) M. (1996). Plzys, Rev. Left,, 77,3407 J, (198t). Phil. M&., van Qmnien, A. H., and de Sadananda, K., Feng, G. A@, 387. (1999). ~ a l : ~Sci. p " Eng. Vogl, C., and Sepiol, B, (1 9941, Acta Me~all.~ f ~ t e r , Snlamon, M., Xto, K., wds~lewsk~, R. J,, Butler, s, R., and Ranlon, .T. E. (1968). J. Eversh nd Mehrer, €3. (200l), Defect aizd 1 3 1 8 . Apple Phys., 38, 4234. F ~ p ~ 523.~ ~ e ~ t b r oJ.o €3. ~ , (1993). Xn S ~ r ~ c f ~~p a~l ~ ~ r(eds~ R.e t ~ ~ S a ~ ~ h ~ ~ ,. In ~ ~ ~ aiz2s Ui r a~ e~ r Allays e~ (eds B. . Warrendale, p. I. Frxltz, W. Caha, and D. Gupta). TMS, W ~ ~ e n d a ~ e ~ Darolia et ~ 1 , ) TMS, Westbrook, J. H., and Fleisclier, R. L., (eds) ~ 1 9 9 ~ ~ ~ p. 205. ~ ~ ~ t e ~ ~ Ce ~ ~a l l ~i-c~ ~ r ~ ~ ~m~d ~ ~ ~ ~~ ~~~ ~~l ts~ Sclia~fer,€%.-E.(1987). P ~ ~Stat. ~ s Sat. . (a), 10 Vol. 1 + 2, Wiley, Chichester. Schaefer, H.-E., Wurschun~~ R., gob, M.,Zak, T., Yu, W. Z,, WoIR, J., Broska, A., Franz, M., ICiilzler B., and Hehenkainp, Eckert, W., and Banhart, F. (1990). Phys. Rev., B41, Tb. (1997a). Mater. Sci. ~ Q ~ ~ ~ , 1 1869. WdT, J., Franz, M., Broska, A., Schaefer, H.-E., Maier, K., Weller, M., Herlach, D., Seegel-, A., and Diehl, 1. (1977). Srr. Metall., 81, 803. (1 997b). Defecl: and DiZ"Forum, ~chaefer,&E., Damson, B., WelSer, M., ~ ~ ~R., ~~ r ~c ~ B., sh Urban h ~ o ~~ , Oeorge, E. P. (1997), Ph.y,s. Stat. Sol. ( a ) , Schaefer, K-E., Frenner, M., and ~ u r s ~ R, h ~(1999a~. i ~ ~ u r s c ~ i u nR., i , Trocv. T,, and Grrtshko, B, (1995a). Phys. In tepB~ ~ t ~ l l 7~ ,c2.77. s, Rev., B52, 6411. ~ c h a e ~ eW.-E., r ~ Frenner, K., aiid ~ i i r ~ R, e ~(199%). ~ z W ~ r s c h ~ mR., , Grupp, Ch., and ~ c h K-E. ~ e(1995b), ~ ~ ~ ~ Phys. Rev, Left., 75, 97. d u r ~ - ~ e r ~K. e n (1997). , Detect ar2.d W u r ~ c ~ R., ~ mKiiinrnerk, , E. A., 3 a ~ ~ ~ aK.,- Seeg~r, ~ e ~ ~ ~ ~ A., Herzig, Chr., and Schwfer, €3.-E. (1996a). J. ~ p ~ l . chmid, G. (1989). J. Phys. Cond. ~h.y.s.,80, 724. ~ a ~ f e 1, r ,SA 49, W ~ r ~ c R., h ~ Badura-Gergen, ~ ~ , K., Kiani~e~le. E. A., ~ e ~ d ~ a CO. n nN,, , and ball^. R. W. (1965). Phys. Rev., C., and Schaefer, €%.-E.(399Sb). Phys. Rev., ~ 1 1824. ~ ~ , 9, Seplol, B,, and Vogl, G. (1993). P~JAS. Rev. Letit,, 71, 731. W , R., and Sc~aefer~H,-E. (1997). Matcr. Sci. Simmons, R. O.,and Ballufi, R. W. (1960), Phys. Rev., 117, Forum, 255-257, 81. 1. Y a m a ~ c kM., I ~ and Shirai, U.~ 1 9 9 ~In) .~ h y s ~~c ~e l t u ~ ~ ~ Shirai, Y . (1988). Ball. Jap Inst. Met., ~ Shirk, Y., Seeger, A,, and Schaefer, %-E. ( ~ ~ 8 9In) .~ a s i t ~ a ~ and ~ r ~ ~of ~~ ~s t .e rs ~ ~~~~a ~~~ l i~c(eds~ N. ~S, o S t o ~ oand ~ , V. K. Sikka), Chapman and Half, New York, ~ ~ ~ ~ (eds ~ i L.~ Cj O~o r~~ lf~ ~j nos -~~ et aal.'). n ~ ~World a~~ p. 3. ScientiEc, Singapore, p. 419. ., and Umakoshr, Y, ( 1 9 ~ ~ P) .r g r . ~ a ~ c r ,and Mullett, L. D. (1999). Phil.
H.-E. (2001). Defect
U M Difl ~
Zhang, X. Y., Sprengel, W., and Schaefer, €3.-E.(2001). To be published,
This Page Intentionally Left Blank
~ ~ i s a t i s f yase the ~ past discourse may have made you nk me with the d o ~ t r ~ of ~ ethe s ~ ~ ~~~o~~ ~ the i s ~ ~~~~~~t~and the p ~ ~ n ~ i 1 p can ~ e syet ~ so little d i s ~ ~ ~ e r what to i l . ~ ~ u i in es~ that ~ p e ~ ~ h ~ Ithe Z ce~ ~ ~ ~ io rf i ~ s others have scarce been more U my QWIZ have been to myself,
ox e ~ ~ ~r yo ~ p ~l e ~ r t ere is ~
mple, neat and wrong-
il.
$ o ~l ~ t i othat ~l is
~
s
~
~
While iiiodeling the te~peratureand stress dependence allows one to extrapolate creep data, it obscures the roles of short-range aiid long-range order, the nature Generally inelastic deforniation at constant uniaxial of the crystal lattice, alloying, aiid iiiicrostructure, all stress is perceived as creep, since tlvs is the way a of which contri~uteto i ~ p ~ o v e m e no tf creep resismateria~is characteri~edin the laboratory and the tance. It is hard to ignore the p resulting data used for design purposes. However, in a improving creep resistance with alloy broader sense any time-dependent inelastic deforma~ i c r o s t r ~ c t u r a~l o d ~ ~ c a t of i o the n ~ three p ~ i n c i p ~ l tion, such as stress relaxation at constant strain or a study of strai~i-ratesensitiv~tyare also ~ ~ ~ i i f e s t a t i o n sintermet allic intermetallic-based phase nickel-base superalloys ( of creep. 1998; Cetel and Duhf, 1988), e ~our t current underThis chapter is an ~ ~ s e s s mof s t ~ i i ~ iofn ~creep ~ e h a ~ i oofr in~ern~eta~lics and and Marlsson, 1997; Lupinc et al,, 1997). See also the inter€~etall~c-based ~ L i l t ~ ~ h aalloys. se The principal objective is to coiiipare creep behavior of i n t e ~ e t ~ l l i c s chapters in V o l u ~ 2e by with ordered struct~res, to that of diso~der~d kiu and Pope ~ ~ i ~ ~ ~ j " and Huang and Chesnti ~ ~ a ~ and e rbring ~ ~forth. ~ sany ~ ~ i ~ i q uaspects e that spite of no systematic pattern o f change in stress contribute to the enlianceflient of creep resistance. exponent or activation energy, This question has rece~vedsome atten~~on, and it kas Outside the conventional fo~iison ~ i n i m u mcreep been concluded that the same constitutive equation rate, impressive progress has been Underapplies to internietallics as to conventioiial metallic o Chen er stand~ngthe primary creep behaQi~r systems ( ~ ~ u t 1991, ~ o 1993; ~ , Jung et al., 1987). The al., 1998: Nemker et al., 1997; eddoes et al., 1997; implied n o n ~ ~ n i ~ u e of n e intermetal~ic$ s~ appears to et d,, 1997, 1998; and ~ u d et ~al., n199?), ~ and ~ ~ be a result of our preoccupation with the temperanickel-base superalloys (Caroa et al., 1388; Bollock ture dependence (a~parentactivation energy) and and Argon, 1992). ft is critical to note that even though stress de~endence(stress exponent) of creep rate, high priinary creep has little inipact on creep mechanrather than the ~ r o ~ o r t i o n a ~ iconstant ty (preisms of steady-state creep, as we now ~ ~ ~ d e r s ~ita n d , e ~ ~ o n e n t ifactor), al which connects these paranieters inhibits the e n ~ ~ i e e rai n~~ ~ l ~of ~ ~a ~ . the~~ ~t ti e or iIn (Larikov et al., 1981; Appel and Wa~iier,1998). To terms of absolute a ~ c u ~ u ~ a of t ~ strain o n at a given some degree the role of inter~etall~cs is better ~ e alloying time, the effects of n ~ c r o $ t r u c t ~and discerned using the difhision coeEcient as a more additions on the kinetics and d ~ a n i i of ~ s~ i s ~ ~ c a t ~ o n fundamental paraniete~. motion are critical but d i ~ c L ~tol tasse~s~ ~ a ~ t i t ~ t i v e l This line of inquiry forces one to re~teratethat creep and hence to i ~ e i ~ t ithe f y role of i n t ~ r ~ e t ~P ~e r~~ ~ acpss. is a p ~ e ~ o i i i e ~por inn i a r i ~of ~ en~~neering interest; of it may be more ~eaningfulto assess the use~ulnes~ wherein ~ i n i m i ~ i nthe g creep rate and strain at a given. l i csecond s phases, rather than as parts of time, in an absolute sense, is of paramount i~portance' i ~ t e ~ e t ~ l as eticr
I ~ ~ e ~ ~ i t Cl t ~ ~ l i c~ Vol. 3, Principles ~ and ~ Pracfictl. ~ Edited by ~ J, €3. Westbrook n and ~ R. L. ~ Fbs&er. ~ 02 0 0 2 John Wiley & Sons, Ltd.
298
~ e ~ l ~ ~Properties n i c a ~
a monoliths recogn~~ing the critical role of semicoherent or coherent interfaces as dislocation harriers. To ~nderscorethe role of interme~~l~ics in creeplimited sit ~ ~ a t ~ owe n s ,will conclude the ~nt~odLiction with a rev~ewof the ~ p p l i ~ a t i oofnintermetallic~.Then to set the stage in s u p ~ o rof t the theme outlined in the preceding para~raphs,we consider the measuremelit and analysis of creep behavior in some depth in ~ e c t ~2.o ~This i i s ~ o ~ l by o a~ ~~i s ec ~~s s i oofn factors ~ n ~ u e n c icreep n ~ resistance of inter~etallics and intermeta~lic-basedsystems in Section 3. roughou out tIGs cliapter we have atte~iipted to bring Corth an ~ ~ ~ d e rc lo iy~im~o~~ a l using ~ t y detailed discussion of more mature ~ ~ s t e ~ s .
1.1.1 Creep , ~
re F ~ ~ i l and ~ r e~ ~
~
~
i ~ ~ ~ ~ ~ i o n ~ ~ ~ ~
Although creep rupture is one of the most commoiily recognized failure modes, d~mensionali n s t a b i ~ and i~~ distortion due to creep are e ~ u ~ coininon ~ly issues ~ ~ e ~ compon~nt t i ~ g ~ u r a ~ j ~ iand t y perfo~~ance. While creep rupture failure niay be of primary interest in ~ ~ ~ ~ G a t ii no vi ios l ~ rotating i ~ ~ parts, h e i ~ d i nand ~ d ~ ~ t o r ~ of i o nthin wall sections can become a critical aspect of design for static components subjected to severe temperature gradients,
I . I .Z Role in
~~~~~~~
~~~~~~~~~~a~ F ~
~~~~~
~
~
g
Creep behavior ofcoatin~s,also plays a pivotal role in. ~of coated ~ ~ systems (Duhl, 1989; Boone and Sullivan, 2973). Here the d i ~ e r e n t i in ~ l both tlierinal expansion and creep resistance of the coat~ngand subst~atelead to re~idual stresses upon thermal cycling, and eventually to failure. However, with the emphasis on c o ~ t ~ ~ g oxidation and co~rosionresistance of c o a ~ ~ as ~i~s~ well as coating appl~cat~on technology, marginal attention is paid to coating creep resistance.
lead in^ to t h e ~ 3 ~a e~c h a n i cfailure ~l (
a ~ ~ a ifew n ~~ x c ~ p t ~ ~~~~g o n s and C ~ u ~ ~19981, re, i n ~ e ~ in ~ s creep t behavior of the vast ma~orityof i n t ~ ~ e t a l lisi ~ s driven by aerospace and industrial gasations (Nabarro and de Villiers, 1995, hile ~e~atively high-stress and short~time e s ttong-time creep creep is of ~ ~ ~in ~the~ former, behavior is of interest in the latter case. However, because o f s t I ~ ~ ~ safety e n t and p e r f o r ~ a n cissues ~ a ~ s o ~ i ~ with t e d the a e ~ o s p a cindustry, ~ it leads in 1.13 Stress ~ e l ~ ~ Notch a ~ i ~o e~h ~~ ~ am! i o ~ , terms of advanced material development. But we are Creep/Fatigue ~ n ~ e ~ ~ ~ t i o ~ w ~ t ~ an e interesting ~ ~ ~ n period ~ where some of the ~ 0 t ~ h " r ~ i p tbehavior ure and the role of dwell time on a~vanced,~irectionallysolidified, single-crys~~l nickelare portan ant aspects o f how f ~ t i ~ u e - ~ i ~ i t ~ d base ~ ~ ~ etechnologies r ~ l ~ are o being ~ ~ r a ~ ~ s fto e ~ r efatigue ~ components can fkil in practice. In both cases creep or as turbine. This is likely to bring forth s~ress-relaxationhe~aviorplay a critical role. Gener1on.g~tiinecreep issues in conju~~ction with large ally, study of these aspects is assunied to be in the ~ o ~ p ~ n e iiat n t sfaced hitherto. It is i ~ i ~ ~ o r t aton t domain of meclianics (Tilly, 1972; Gallerneau, 1999), e, not only that the typical superalloy is 60% hut there is clear evidence to suggest that intrinsic sed intermetal~icby volume but that several material behavior plays a very i~~iportant role. It has metallic c o a t ~ ~systems g also derive their excellent been shown that two specii~en~ of the nicl~e~-base from nickel and ~ l a t i n ~alurniii~ ~ ~ i d a t j ore~ist~nce ii superalloy IN 100, with differing heat-treat~~ent condi~ ~ ~i n t ee~ ~ i~e t a l~~Xfi sufficient ~ s~. s progress e ~ is made tions, can have sig~i~cantly different ~ o t c h ~ r u ~ t ~ r e in refractory-metal-based systems, the future presence lives despite yielding identical smooth ~re~p-rupture of i n t e r ~ e t a lsificides l~~ in both the base niateriaf and lives (Law and ~ l a c k b ~ 1980). ~ i ~ , Swh studies of potential c o a t ~ n ~iss inescapable" fii the case of monolit~icinterine~allicsare n a t ~ i r ~limited ~ i y (Lupinc titan~um alloys, we are already on our way to et a/., 1991). c ~ - ~ i A ~ - b a ssystenis. ed (See at.) Even if the creep behav~or of ~ n t e r ~ e t a l l were i ~ s not any better than that of I . P .4 ~ r o c e ~ ~ s i ~ g disordered alloys, the inherently better oxidation ~ u p e ~ p ~ a sf t~i cr ~ i and n g di~usionbon~ingare used ~e~istance of aluminides and silicides forces one to and investigated (Schuh and Dunand, 1998) as an learn to live with them. ~ h a u g hnot always recognized alternate forming processes. Creep plays an ~ ~ ~ o r t a i i t and used directly in tl-ie design process, creep behavxor role in these processes, but in. an opposite sense. of these materials manifests itself in various failure Nevertheless, correlatio~sb e ~ ~ e ~ e ni i ~ r ~ s ~ ~ uand cture ~Odes* *
299
Creep mechanical behavior would provide additional insight. However, this aspect is beyond the scope of this chapter, and will not be discussed further.
alysis Generally creep behavior is assessed in uniaxial tension and a plot of strain vs. time is generated at various te~peraturesand stresses. Often in industria~practice no strain history is recorded, and the material is assessed solely 011 the basis of its rupture life. For brittle intermetallics, in the early stage of development, compressive creep, or bend creep is used to assess the potential of material. By and large, at high teiiiperatures where creep is primarily diffusion controlled, the creep rates in compression and tension do not differ. However, in highly textured, coarse or columnar grain, and single crystal materials, diKerences in creep rate at lower temperatures can be significant in some orientauhl, 1988). Besides constant stress experiments, stress relaxation and constant strain-rate experiments (Dudzinski et al., 1997), can provide invaluable insight into material characteristics, but few such experiments are carried out. In its most general form, tensile creep behavior may be schematically represented by a strain versus time curve, such as the one presented in Figure 1. The early incubation period, where no measurable creep occurs, is followed by a rapid rise in creep strain or the period of primary creep. To better delineate the primary creep region, creep results are often plotted as strain-rate
versus strain. High primary creep ( > 1%) rarely seems to occur in wrought and fine-grained nzaterial. Generally, high primary creep i s associated with a relatively low density of mobile dislocations, and it i s shown to be sensitive to heat treatment in i~~iltiphase alloys. As shall be discussed later, this aspect has not received s u ~ c i e nattention. t Much of the focus in the literature has been on the second stage of creep or the minimum creep rate.
Et has been well accepted that minimum creep rate can be related to stress and temperature by the relation, i = A exp(-Q,/RT)s”
(1)
where, i is the strain rate, Qc is the apparent activation energy for creep; Tis the temperature in K.; R is the gas constant; and A and n are constants. ( the equation to be dimensionally true, s should be replaced by ( a / E )where E is the elastic inod~ilus,or by (a/oo), where 0-0 is a reference state. This omission, however, does not impact the follo~ingdiscussion.) This relationship seeins to be as much valid for monolithic and multiphase intermetallics as it is for pure metals and disordered solid solutions. Thus it would appear that intermetallics play no special role. This may be so because high-temperature materials of interest occupy the power law creep region with reference to Ashby’s ‘deformation mechanism maps’ (Ashby, 1973).In this region, the power law relationship with a stress exponent of 3 may be phenomenologically
In
igure 1 Most general form of h i ~ ~ - t e ~ p e r ~ tensile t u r e , creep behavior
justi~ed~ i t h o u reference t to any specific dislocation ~ e c h ~ as~ follows. n ~ s ~
t "?.' pbv; but since U QC cr; and p o( cr2$ i 01 g3 (2) where p is the dislocation density, k. is the vector, and L, is the dislocation velocity. Zf dis climb were c o ~ ~ s i d ethe r e ~r a t e - ~ i ~ i t ~step, n g a higher ess e x ~ ~ n eof n t3.5 or 4.5 might be ratio~ia~i~ed. See e ~( 1 9r~for ~ t)~urther ~ discussion. ~ ~ ~ e e r ~ ~ and an~ ~lternatively~ the validity of equation (1) may be nothing more than a recoii~rmationof the usefulness orn, or the more popularly used raiiieter P, for represent~ng creep data for all metallic (Evans and ~ i ~ s h i€993), r ~ , and i ~ t e r ~ e t ~ l 1 i c , ( ~ t o ~ 199S), o ~ , y-TiB1 ( ~ u p i n c et al., 1997), and NiAl (Noebe and ~ a l s t o n ,1997) systems. ~ n d e equation e~~ (1) forms the physical basis for these ~ a r a ~ ~ ~ The t e~,arson-~iller ~s. parameter P is expressed as,
stress can then be g e ~ ~ r aast in ~~, consider the scenarios where we want to increase or decrease the creep rate by two orders of ~ ~ ~ ~ i t L ~ d changing either A, uz or (2, one at a t h e . ~ i ~ i l a rin ~y, Figure 2(b) we can plot stress for fixed strain rate or time to 2% creep vs. 1 of temperature, and
pure tiickel, to I ~ to the ~ most~ creep~ r e ~, i ~ t ~ ~ advanced single-c peraf loys. As the plots in s ~ g ~ ethe s t i~~ ~ ~ r o v einm ~ n ~ creep resistance from pure nickel to an advanced sLiperalloycan be achieved,either by c h a n ~ i n ~ the pre~~x~oneiitial factor A by orders of ~ a ~ n i t u d ~ , by c ~ a n ~ ~Q nby g -50 kJ,hnol or by ~ ~ i P$ by ~ ~ ~ 30%. Note that, deterio~~tion or e n h ~ n c e ~ine creep ~~t ~esistance,achieved with variation in any of the parai~eters,yields curves so pardel to each other that any c o ~ b i i ~ a t i of o i ~change in the three parameters can account for the change in creep behavior~ P = ?-(fog t 4-C) validity o f the empirical (3 depending on the stat~st~cal data. Since Q and n can be ind~pendentlyde~erni~ned, where T i s ab~olutetemperature^ t: i s rupture life or these values are r e ~ ~ r but t e the ~ value of A , which time to fixed strain and 1 M 1/i and C i s a constant for , must chaillge by nn order of i ~ a ~ n i t ~isd e~eiierally a wide variety of ~ i ~ t e r i a l s . ignored. This is so even though the ~ a r a ~ Ae ~ e ~ arsoi~-~iller paraiiiet~rin equation entails, the pre-ex~onentialDo ('art of the d i ~ ~ s ~ o (3) is a r e ~ e x ~ r of ~ an s ~~ ~ r r o~ e~n i u s ~ t re~at~onsh~p ype c o e ~ c ~ as e nwell ~ ~as ~islo~atioii mob~lity,and microbetween time and te~peratureas in equation (1). The structural parameters, which greatly infiuence the successful ~ p p ~ ~ ~oft the i o Earsoii-Miller n parameter absolute sate of creep. A s a m ~ l i of n ~these parameters would seem to su~gesteither that equation (1) is very for a wide variety of i n t e r ~ e t a l ~systems ~C studied in robust, of: that the three para~etersA, &, and pz are recent years indi~atesno rational pattern in the value as i i ~ ~ e p e as n ~wee think. ~ ~ ~ ~x~erimentally, while of YE or in the a c t ~ ~ ~ tenergy, i o n as far as the absolute n t ~ y ~ level of creep resi~tanceof alloys is c~nGerne~. nd M, can be d e t e r ~ i n ie n~ ~ e ~ e ~ i ~Ae cannot. eed, row^ and ~ s h (1980) ~ y have shown that A The first question i s this: can an order of ~ a ~ ~ ~ can vary from less than I to more than IQi5. Such R chaiige occur in A with variation in crystal s t r ~ c t ~ r e s ~ variztion has been reported by b ay as hi et al. (2991) order in^* and a l l o ~ i ~ofginte~metal~ics? Based on the for Ni,(Al, X), simply with stoichiometry. In many Russian compilation (Larikov et al., 198I , Chapter 4) ~ ~ t e r ~ aasl ns increases , A increases also. A~cordingto of d ~ ~ u s i oinnordered alloys, the answer seems to be rown and A s h b ~f1980), A and n are a ~ p r o ~ ~ m a t e l ya ~ r ~ a t i v fet. is known, that while the Rctivat~o~ related by o n be a'~~o~imate1y 36T~ e~iergyfor s e l f ~ d i ~ u s ~can c a l / ~ i o l e where - ~ ~ M is for ~ e l ~ jthe ~ gvalue ? of the p r e - e x p ~ n ~ n tfirae~q ~ ~ e nfactor ~ y Do falls in the range of 0.05 to ~,0cm3~sec. For e x a ~ ~ lthe e, where C and IT' are constant^^ dependence of volunie self~di~usion in aTO~ i n ~ e r ~ t the a n dsensitivity o ~ e q ~ ~(1)t to i othese ~ and Nb is described by the followiiig rehtions. this experiment. Xiimgine a basep a ~ ~ e t e rconsider s, a line alloy with 100 hr to 2% creep, or approx~mate~y ~ ~ ~ icreep i ~rateu of~ 7 ix 1OeS sec-] at 1200 K and 200 MPa. This is a close a ~ ~ r o x i ~ a tto i othe n behavior l y nickel~ba~e of ~~1~~ - one of the c o ~ ~ o n used ed superalloys with 55% ~ ~ 3 ~ l - b a sprecipitates* Further zssurne that for such an alloy, M, = 3.5 and Q = 3 ~ ~ J / m Ao ~plot . of n i i n i creep ~ ~ rate vs..
Creep
30 1
I,
Fi~ure2 Sensitivity of equation (1) to changes in pre~exponentialfactor A , stress exponeiit E , and apparent activation energy Q for (a) two orders of magnitude change in strain rate at constant stress, or (b) for a 200K changc in temperature capability at c o ~ s ~ acreep n t strength
302 These values clearly suggest that Do is affected by the creep occurred at intermediate t e ~ ~ e r a t ~(760 r e s"C) d i ~ ~ r ~b~tween n c e the c~ose-p~cked (fcc) vs. open (bcc) only, and the o r i e ~ t ~ i t ~ o~np e n ~ eof~ pi cr ei ~ ~ creep ry str~cture?and that even in pure metals, tlze presuggested that it was associate^ with s i ~ g ~ edefor~s~i~ t exponential factor can vary by -10. We a u g ~ e n this ased on further an~lysis~ Paslay PC al. {~970), a r ~ u ~ ~ ~L~rther e i i t in Section 3, with a d i s ~ ~ ~ s sof i ohow n asserted that the ~ r i ~ creep a ~ yrate for a va~ietyof degree of ordering, s t o ~ c ~ i ~ i i ~ and e t r yalloy , in^ addiorie~ita~ion~ can be best represented by e n ~ p ~ a s ithe ~in~ tions influence Do, and hence D, of single-phase s in preference to (1 1I)( 110) ~ n t ~ r ~ ~ e t aIndeed l l i ~ s . ~ c c o r d i nto ~ ~ ~ u t (199 h oI),~ f971), however, s ~ o w that e~ the diKusion coefficient is the niost i ~ p o r parat ~ ~ ~ ~ occurred in coarse-graine~~ meter; and Ilie creep rate i s best correlated with and directional~y~ ~ l i d i ~c oe ldu ~ n a ~ - g r amaterial, ~n di~~si~ o no e ~ c ~ errsang n t s the orn e~uation,which is and that while ~ r i n ~ a rcreep y strain d e ~ r e ~ s ewith s similar to equation (I), for dislocation creep, where C decreasing stress, the time to complete p r i ~ ~ a rcreep y is the shear ~ o d u ~ u ~ . increases with decreasing stress. Add~~~onally, they also claimed that primary creep strain IS su~pressedby the c.. = ~ ( ~ ~ ~ ~ / ~ (9)~ )in~rod~ictio~i ~ ~ / ~ ) ~ z of a ~ ~ i e - s c as ~u e~ ~ t r ~ ~upon c t ~ shock ~re Tn contrast to diEusion ~ ~ i e c h a i i i s however, ~~s, it is difiicult to isolate the ~iifluence of slip geometry, than a decade later, Caron et ad. (1988) d ~ ~ ~ o c a~ehavior, t j o ~ ~ and ~crostructuralp a r a ~ ~ t e r s that the ~ehaviorwas even more complex than on tlie pre"ex~o1ientia~ factor A, in the framework of previously envisioned. As seproduced in Figure 3, iheir e q ~ ~ t i o (I) n s and (9)* in any reliable manner, Nonestudy of three major or~entatioiisof' a sing~e~~rystal theless, the ~ o n ~ i i i ~rote n t o f these para~etersin alfoy d e ~ o ~ i s t r ~ t botli e d i ~ ~ u b a tperiod ~ o ~ and rmproviag creep resistance in real applications is a function of orientation and y' e ~ p i ~ i c a ~l ~i ~yd e n i a ~Thus, l e . though largely a valid ntly Kakehi ( 2 0 0 ~Iias ~ corroborated these findings and has shown that p r i ~ a r ycreep model, e ~ u a t ~ o n(I) s and (9) are a poor guide for istance of two-phase systems of' is affected by cooling sates, which influence tlie fine fore we return to these aspects in precipitate structure. Primary creep was also observed of ~ ~ creep ~- a ~to be a~ €unction r of major y (Shah and Cetel, 1996) and ~ e ~ t i o3, nhowever, a discussio~~ minor (Shah and Duhl, 1988>, alloying additions to hitherto ignored aspect of creep behavior - is in order. single-crystalsuperalloys.However, based on extensive analysis of the ( 100) o r ~ e n t es ~i ~ g l e ~ ~sL~~era1loy rys~~~ ( ~ 0 1 atid 1 ~ ~~ r ~g o ~~ 9, 9 ~~o)n c f ~ d ethat d No universal, physical model exists to describe primary primary creep following initial incubation WilS a result creep, which is the most strLzcture-sensitive part of creep of ~ ~ j ~t i ha 1 e r misfit ~ ~ ~s t r e s ~ ~ b se t ~ the ~ e7 ~and y' e s~ ~ ~n rg~i ~ a creep ry ~ etiorm, In ~ an ~on ~ ~ ~~ esense, phases, and that it was related to any workbecomes an issue of concern, when 1-2% creep strain hardening or recovery processes, In contrast, Miura is a c ~ ~ ~ u ~ina at efew d hours, in spite of' a sixstained et ak. (1991) argue that in singte phase ~ ~ - ~XIvi ~ ~ s~ea~y-staie creep stage and ~ ~ n d r e dofs hours of the initial s~~nioidal behavior is a reflection of the creeprupture life. Parcametrically, primary creep is yield-point behavior observed under constant tensile d e s c ~ b eas~~ ~ d r a dcreep, e-~ strain-rate testing. Again in co~itrastto Pollock and i e n tKear a i d Argon (1992), an in ~ ~ r e e ~ ~with Piearcey (1967), hu et at. (1998a), studying ~ m ~ ~nn tg~ ~ e t a f l iand c s , inter~eta~lic-based alloys, Ni.,(Al,Ta) in various orientations, concluded octahethe inc~batioi~ period and ~ r i ~ a creep r y phenomena dral slip to be r e s p o ~ s i bfor ~ ~p ~ ~ creep, ~ r while y have been s p o r a d ~ ~ astudied ~ ~ y in Ni,Al, and twoe ~secondary creep. Further c~zbe-crossslip c o n t r i b ~ t to phase cast s L i ~ e ~ d l ~but o y ~recently ~ have received Zhu et al. (1998b) conclude that the blocking OS edge c o ~ ~ s i d a~te~ition e ~ a ~ ~ in ~ y ~ ~ I These ~ 1 . studies are o n ~ their climb ~ i ~ l o c ~ t i obyn sforest d ~ s l ~ c a t ~causes ~ r i e reviewed ~y in the following. and the termination of primary creep. ~~~~~~
~
l and ~~ ~ ~~ ~y ~ ~ ~ ~ t a2.2.2 k ~k ~ c~ ~~ ~ ~
1x1 tlie earliest work on ~ ~ n ~ l e - c r ysuperalloys sta~ (Kear iearcey, l 9 ~ ~it)was , observed that high primary
t a
~~
i
~ ~
~
The same tlieme recurs with binary ~ ~ L i d ~ i nets kat., i 1997) and a~loyedy - T N ~ ~ ete al., o 1997). ~ r i n i a ~
Creep creep in y-TiAl is observed to be a sensitive function of alloy composition, heat treatment, and microstructure. In contrast to superalloys, however, lamellar refinement leads to a reduction in primary creep in y-TiAl; and Rong et al. (1998), showed that pre-straining increased the primary creep strain as additional twinning was introduced. We return to this aspect in some more detail in Section 3. Study of orthorhombic Ti-23Al-27Nb also shows a strong dependence on the primary creep microstructure. In this coinplex system, aging treatments, which induce the precipitation of additional phases, reduce the prima~ycreep strain ( oehler et al., 1997). These observations are opposite to the effect of fine y’ precipivates in superalloys. Nonetheless, following the hypothesis put forth by Pollock and Argon (1992) this commonality points to a potential correlation between misfit strain and primary creep. A common theme between these two classes of materials seeins to be a low density of mobile dislocations initially causing an incubation period, arid eventually bursting into rapid ~ultiplication,leading
303
to high primary creep. restraining or microstrLLctura1 changes which can sniooth out this transition seems to lower the primary creep, but the opposite can occur instead if dislocations are inhibited. It may be rationalized that variously solute atmospheres, misfit strains between the two phases, transition in slip behavior, or dynamic change in microstructure ( aiid Hall, 1991), may all interfere with a smooth transition to steady-state creep. Load-drop experiments, which allow one to measure the extent of reversible strain, provide a good insight into detailed niechanisms.
Generally, creep properties are measured in ambient environments, and for a given ~ a t e r i a are l expected to be reproducible. However, those dealing with a large amount of creep data for well-characterized alloys know that consistent biases in results occur from test location to test location. This aspect has not been systematically addressed so as to isolate environmental
Creep strain (%)
1
(a)
Q
2d0
Figure 3 Effect of y’ size oil creep bebavior at 760°C and 750MPa of CMSX-2 single crystals for the three main orientations: (a) [OOl], (b) [I 111 and (c) [Oll]. Froin Caron et al. (1988)
factors, such as humidity, and and C1 content in air; but the effect of these factors cannot be ignored. In the extreme, it is known that creep properties are severely degraded in low oxygen and sulfur environments. See tidies of superalloys by Seib (2000), and tlow (1985). ~nfortunately,while environmental effects on creep sorely deserve more attention, they are beyond the scope of this review.
3.1.1 Role of Sl@ Gametry
Even though creep resi~tanceis measured by uiiiaxial strain rate at a given stress, usually the deformation takes place on a number of slip systems. In a truly random fine-grained, polycrystalline material, this is a moot point, as it is ave~agedout and may be qu~ntified by the Taylor factor. In single crystals, however, it is fair to consider to what degree each slip system experiences a signific~ntresolved shear stress, and thus contributes shear strain to the total tensile creep strain, is largely a function of geometry. Unfavorably oriented slip systems will be less stressed, and conseq~entlycontribute less strain. In the early stages of creep with a low total strain, the contrib~~tion from each slip system may be assumed to be linearly additive
(Shah and Cetel, 1996). Thus, ignoring intera~tions, steady-state axial strain rate iii may be expressed as,
for all possible families of slip systems, where ap,k are Schniid factors for thejth slip direction on a lcth slip plane, within a family of slip system, ~ i t h i na given family of slip systems, it is reasonable to assume that the proportionality constant Aj/<would be a constant. A good match between predicted and observed creep anisotro~yfor a disordered, solid~so~ution-~~rdened nickel-base alloy (Hastelloy-X) single crystal, vindicates the validity of this relationship. Such a material with fcc structure is known to deform predo~inantly by octahedral (111)(110) slip. Purely based on a consideration of Schmid Pdctors and number of participating octahedral slip systems, a factor of -8Xdecrease in minimum creep rate in the { 111) direction compared to the (100) direction is predicted and observed. See Table 1 for further details. As further d e ~ i ~ n s t r a t eind Table 1, similar arguments can be made to rationalize the 50 times lower creep rate in the ‘hard’ (100) orientation compared to the ‘soft’ { 111) orientation in binary MiAl (~itabjianet al., 1999), where (100) is observed to be the predominant slip direction. The terms ‘hard’ and ‘soft’ are used qualitatively to describe very low and
able 1 Assessment of creep anisotropy based on strain-rate partitioning over operative slip systems
lip systems
Axial Orientatioii (111)
{ 100)
Proportionality factor
Creep ailisotropy
Exaniple
Min, creep rate along (100) Min. creep rate along (1 1I)
No. of Schmid No. of Schmid systems factor systems factor
FCC/L 12 (1 11}(1l0)
8
0.4083
6
0.2722
A cd
-8 (Octahedral slip only, Acd = 0)
Hastelloy-X Shah and Cetel(1996)
CO01} (110)
0
0.0000
3
0.4714
Acd
-2 (Cube slip participation A,d = 2A&)
Typical Superalloy Ni,(AI, Ta) Shah and Cetel(1996)
B2 (001} (001)
0
0.0000
6
0.3333
A,
(1 10}(001)
0
0.0000
3
0,4714
Ad,
1/50 (predominantly (001) dislocations A,, = Adc = 100 Acd = 100Add)
~001}(110) {ll0](110)
0 4
0.0000 0.5000
3 0
0.4714 0.0000
Add
A cd
-
-1 (both (001) and (110) diSloCatiOns A,, = Ad, Acd = Add)
NiAl Kitabjian (1 999)
Ni-47.5A1-2.5Ti Kitabjian (1 999)
Creep high resolved shear stresses on the operative slip systems. In Table 1, equation (11) has been applied to assess the extent of creep anisotropy, which is arbitrarily defined as the ratio of minimum creep rates in (100) to (111) directions. Note from Table 1 that in contrast to simple systems
additional slip systenis intervene. Specifically these are cube slip-(111) (100) in the case of y’ and y/y’ alloys, and (001) (110) and (1 10) (110) slip systems in the case of NiAl. Generally, increased numbers of slip systeins lead to a decrease in creep an~sotropywith a wider dispersion of strain. Of course the extent of decrease in anisotropy is dependent on the relative ease of glide of various families of slip systems, wliich can be defined in terns of a ratio of proportionality constants A j k , as has been done in Table 1. The interaction between the two slip systems is a function of degree of order, alloying additions, and oreover, because of the extended nature of superdislocatian core structure (complex SF and APB), ease of dislocation cross-slip is facilitated or in~ibited,dependin on the orientation and direction of creep stress - that is, tensile or compressive - at intermediate te~peratures (Pope and Ezz, 1984). Recently a dislocation dynamic based model has en developed to capture these interactions by rehrn and Glatzel, 1998, but we are far from predicting the behavior n priori. Not surprisi~glythe model converges to equation (11), when slip interactions are dropped. Though not fully developed, the models allow one to qualitatively rationalize the trend in creep anisotropy with various alloying additions (Shah aiid Cetel, 1996). In some high~molybdenu~ alloys, where MO is known to partition only to the y-matrix, the creep anisotropy factor, defined as a ratio of creep resistance of (1 1 1)/
1400 F 100KS1
305
(100) orientations, can be as high as in Figure 4 (Dalal et nl., qualitatively resembles that of discussed earlier is compositio equivalent to the y-matrix, and shows very high creep anisotropy. Of course, in light of the fact that that the creep anisotropy i s also affected by the ~recipitatesize as shown in Figure 3, the p~enomenaare obviously complex and should not be oversimplified. Nonetheless, if irnproveme~tin creep resistance were the only objective, it is clear that a (111) oriented single crystal of nickel-base y’or y/y‘ would be at least a factor of2X inore creep creep-resistant than (100). analytical interest in rnodeling the nature of creep anisotropy for predicting the multiaxial of notches in single crystal superalloys ( 2000), its value in discerning the basic dis~ocation structure and alloying effect cannot be overlooked. Similar studies of creep as a function of orientation for highly alloyed and two-phase ~iAl-based single crystals are limited (Noebe and Mechanical properties of sucli alloys have been primarily investigated near the (100) orientation (Darolia and Walston, 1997; Garg et al., 1998). Creep anisotropy studies of other intermetallic single crystals are lacking, perhaps due to processin di~culties.Especially the behavior is anticipated t be very interesting in non-cubic intermetallics, such as orthorhombic Ti,AlNb and i,, with their lower symmetry. The lower s y m ~ e t r yshould present even greater opportunities to orient the active slip systems with reference to the principal stress direction and enhance creep resistance in some ‘hard’ orientations. Note that creep anisatropy studies of polysynthetically twinned (PST) crystals of y-TiAl do not fit the framework of the preceding discussion. As discussed later, a highly aligned lamellar rn~~rostructure aiid perhaps other Factors play more dominant roles than
2000 F 2 0 K 5 f
1800°F/36 KSI
0
TIME (HOURS)
TIME (HOURS)
‘0
100
200 TIME
300
400
500
(HOURS)
Orientation dependence of creep bchavior of SC IN-6-7 (Ni- 13.5Mo-6.1MI-6.8Alwt.%) single crystal superalloy. From Dalal et al. (1984)
orientation of slip systenis in ST crystals (Parthsarthy et al., 2000).
3.1.2 Grain Texlure
Ideally, creep anisotropy is expected to manifest itself if there is a strong grain texture. For example in the simplest case, directionally solidified { 100) oriented columnar-grain superalloys such as Mar M200 and MO02, practically behave like ( 100)-oriented e crystals. It is interesting to note that rapidly solidified and recrystallized, high-Mo superalloys were initially claimed to have high modulus and superior creep resistance (Cox and van Reutl], 1980). Subsequently it was t is covered that both characteristics were owing to a strong (111) fiber texture, and similar
when the grain size is much finer than the thickness of the specimen, it is reasonable to assume a random orien~atioii.In such a case, the anisotropic behavior of individual grains inay be ignored, as it is averaged out, and can be accouiited for by the Taylor factor. The real challenge arises when the grain size becomes compamble to the section size of the specimen or the component: There, unless the grain orientation is accounted for, a wide scatter in creep behavior may uch has been the case with transverse -grained materials and creep properties of col bicrystals of superalloys d O’Hara, 1996; Shah and Cetel, 2000). As sh igure 5, scatter in the erties of (100) oriented columiiar M200 may be attributed to a grain orientation ranging from dist~ibutionin at for (100) oriented columnar800 T--775 750
i
CE
-i IQ000
igure 5 Comparison of stress versus rupture life at 1400 “F for Mar M200 1- Hf for single crystal and columnar-grained (CC) material. From Shah and Cetel (2000)
grain material, any direction in the transverse plane can only be those lyin in the (100) plane. Note also that in Figure 5 , scatte in the data extends from being equivalent to the poor creep resistance of (110) orientated single crystals, to the superior creep resistance of (100) oriented single crystals. The dominant effect of grain orientation on creep behavior is well supported by a detailed study of DS IN738LC uested et al., 1988). Using electron back s~attering patterns to assess creep strains, they clearly demonstrate that cracking and/or cavitation in transversely oriented grains occur most clearly when. the adjacent crystals exhibit evidence of different levels of creep strains owing to different orientations. The problem is exacerbated by significant lattice rotation in the large grains with high creep strain. In the preceding discussion the roles of grain boundary sliding, environmental effects, enhanced grain boundary diffusion, and grain boundary segregation effects, were deliberately ignored to e i ~ p h a s i zthe ~ role of plastic anisotropy and resulting strain incompatibility, However, again McLean and Strang (1984) and Quested et al. (1987) have shown that while grain boundary segregation of low melting elements such as Bi, Sb, Se, and Te reduce creep-rupture life and fracture elongation, the failure occurs only along the grain boundaries normal to the stress axis. Moreover study of Bi-doped Mar MO02 shows that degradation is more severe in equiaxed material than in transverse columnar grain inaterial, and it is always owing to adjoining grains with large differences in creep strengths. In much of the p~~blishedliterature on grain-size effects, the preceding viewpoint is ignored because of a lack of data and knowledge of creep anisotropy, and an inevitable but perhaps ‘distracting’ variation in grain boundary microstructure. Obviously, creep behavior of polycrystalliae intermet~llic-based alloys must be controlled by both intergrann~~lar and transgran~~ar bebavior and ~nteractionsthereof, However, with increasing availability of modern tools such as orientation imaging microscopy (OM) to determine grain texture and strain condition, paying due attention to the transgranular behavior is very fruitful, as has been demonstrated by Quested et al. (1987, 1988).
Global spending of upwards of tens of millions of dollars over the last 30 years on improving the creep resistance of nickel-base superalloys is a testimony to the highly empirical nature of our understanding of the effect of alloying. It is not an exag~erationto say that
307
Creep Dii3bsion coefficient, D, and activation energy, (2, for A1 and Ti, and for ~ i ~ ( A 1 . T ~ ~o~pound _
_
_
Values at 1000 “C Q for A1 D for Ti kJ/mol cm2/sec
cm2/sec
Do for Al cm2/sec
1.7 x 10-I’ 8 . 2 10-j2 ~
6.6 x 10-4 6 . 6 10-4 ~
185.0 211.0
__
__
-
D for Al
Do for Ti crn2/sec
Q for Ti kJ/rnol
-
__
~
Ni,AI Ni,(AI,,Tio2) NiWO 6%) 4) Ni,(Alo,Ti,,) Ni3(Alo4Tio6)
3.1 x 1 x 10-’ 2.9 x 10-12 3.1 x
many of these eflorts were focused on creep strength, because the impact of the gain in low-teniperature strength has been modest compared to the influence of enhanced temperature capability as measured by creep strength. While all this is true in a broader perspective, it i s deceptively simple to characterize the alloying trend as merely increasing additions of refractory elements, especially Ti, Ta, MO, W, and Re. See Erickson (1995) for a compilation of recent developmen ts in single-crystal superalloy compositions. The phenomenology of dislocations and the effects of alloying in L1 intermetallics and Ni-based superalloys has been reviewed by Westbrook (1996), who also defines some of tlie unsolved problems that impede continued development of these materials. Exploration of the creep behavior of inte~etallics also shows the refractory elements to be potent creep strengtheners. Comparison of minimum creep rates of binary and ternary alloyed intermetallics based on FeAl, y-TiAl, Ni,Al, NiAl, Cr,Si, and MoSi, in Table 3, clearly shows that the minimum creep rate can be decreased by orders of magnitude by alloying with any of the refractory metals. Interestingly, note the major effect of the addition of Ir in NiAl by Chiba et al. (1998). However, in-depth alloying studies of various intermetallic-based systems such as y-TiAl and NiAl abundantly show that the latitude of alloying is limited. In ordered intermetallic systems, away from terminal solid solutions, the complexity of phase fields and limited off-stoichiometry of the ordered structure constrain the addition of alloying elements. Nevertheless, in appro~matelya decade - a relatively short time - very modest amounts of alloying additions to MiAl have brought the creep resistance of the singlecrystal alloys AFN-12 and AFN-20 to levels comparable to those of superalloys ( arolia and Walston, 1997).
Whether alloying elements improve creep resistance by solid-solution strengthening or order strengthening is a
242.8 228.1
-
-
__
-
2 . 3 ~10-” 2.1 x 10-l2
1 . 9 10-6 ~ 1.1 x 10-‘
145.3 139 4
7.85 x 10
3.3 x 10-6
137.3
l2
matter of site occupancy. In either case a decrease in diffusivity leads to to an increase in creep resistance. In disordered structures, there is little choice but to characterize the effect of alloying additions as solution strengthening, until a tendency for short-range order (SRO) or clustering is discerned by more sophi~ticated techniques. For example, till the advent of transmission electron microscopy, or the application of neutron diffraction, sluggish ordering of Ni2Cr was referred to as the K-effect in Ni-Cr solid solutions. Since then, both theoretical studies (de Fontaine, 1975), and sketchy data on binary phase diagrams do suggest similar clustering tendencies in many NiNi7C06)aiid refractory-metal solid solutions. Interestingly, often when such clustering tendencies lead to a non-cubic structure within a cubic solid solution, e.g. Ni,Cr in Ni-Cr, a domain structure is formed. It may be argued, though, that in inany of the Ni-X binary systems cited above, the order-disorder transformation temperatures are known to be relatively low, and the influence of SRO on the high high-teinperat~recreep behavior may be questionable. However, in practice we are dealing with multicomponent solid solutions, where the theoretical understanding of short-range order is complex, and einpirical data are limited. Indeed, an atom-probe study of coniponent superalloys does show a correlation be improvements in creep resistance and the presence clusters in the y-solid-solutio (Blavette et al., 1988). ~nfortunate~y, 111 a two-phase system it is difficult to completely isolate the solidsolution hardening effects of alloying additions from other aspects of microstructure, sucli as the matrixprecipitate interface and lattice misfit; but given that an improvement in creep resistance is achieved with additions in a wide variety of the second generation of single-crystal superalloys (Broomfield et al., 1998; Cetel and Duhl, 1988), and in the nickel-base eutectic alloys (Lernkey and Machlin, 1985) with different niicrostructure, suggests something more fundamental to the Ni-Re solid solution. Un~uestionably, while our 9
understanding of such clustering beliavior in multicomponent alloys is poor, the authors believe that such tendencies must in~uencethe long-term iiistabilities of nickel-based solid solutions to the formation of undesirable topologically closed-packed pliases. Interestingly, it nized that in two-phase nickel-base superalloys, the high-~emperaturephase-instability of the solid-solution hardened matrix and high creep resistance go hand in hand. This aspect is further discussed in the next section with reference to stacking fa~iltenergy and electron-vacancy number. The ~lusteringtendency is not limited to terminal solid uper-ordering can occur within an ordered well. An example of such a clustering of G-phase (Ni,,Hf6Si7), tendency i s the p~ecipit~ition kave even a trace amount of Si again, the clustering ratures only, and its -temperature creep is considered to be solid~sol~~tion strengt~ening, as the phase dissolves. The critical point is that, without the knowledge about the cl~steringtendency, it may be difficult to understand why some intermediate solid-solution compositions may show optimun creep resistance. In wellc~aracterizedordered intermetallics, a description of site occupancy is useful, but without a quantitative characterization of the degree of ordering, it may be treated as sublattice solution strengt~ening. 11 discuss this with reference to the Ni,(Al, Ti) ,where it has been shown that the occurrence o f slowest diffusion around 12.5 at.% Ti correlates well with the highest degree of orderin
~nfortunate~y, the pre-exponentia~ parameter A' is found to be about 4 x 1Ol2, instead of unity, and so equation (12) must be regarded as an empirical law without an ade~uatetlieoretical foundation (Mabarro and de Villiers, 19951, p.76. h se~i-theoretical,the preceding correlation allows one to u~derstandwhy most unstable nickelbase superalloys with a solid-solution-hardened yinatrix and an electron vacancy number of N v w 2.5, tend to be the most creep resistant. ~ ~ l c u l a t i oofn the electron vacancy number, Nv, or (10 ininus the number of d electrons), for the y-matrix is traditionally used to assess the stability the composition to the formation of TCP phases ( cker, 1970). A ~hysicsbased model suggests that NV can be related to the g fault energy in transition metals as shown in 6 (Papon ei al., 1979). Very simply, Figure 6 is an attempt to predict the stable crystal structure using a single atomic p~rameter.The two curves represent results of different calculatio~s.The point of interest in Figure 6, is the intersection of the two curves around NV 2.5. Clearly the physics suggests that in solidsolution alloys of transition metals, stacking fault energy is lowest around Nv 2.5, and that is the condition at which the FCC structure starts becoming unstable with respect to other close-packed str~ctures. Essentiall~,if creep resistance is enhanced with low stacking fault energy (SFE), phase instability is inevitable when the SFE approaches zero. Similar fundament a1 physics-based rationali~ationshave not been developed for other less mature systems, but exploitation of phase instability in developing creepresistmt pTiAl is ~ r o ~ i s i n g . N
N
3.2. I. 1 Stack~ngFault From the point of view of dislocation inechanics, improve men^ in creep resistance with alloying may be rationalized in terms of lowering the stacking fault energy in solid solutions with closed-packed disordered structure and ~ o ~ p l e x - s t a c ~ i n ~ - fand a u l tanti~hasey in ordered structures. Low fault extended dislocation-core structures, which inhibit cross-slip and dislocation climb, and thereby slow down the recovery. ~ n ~ e ~e od h a m e dand n (1974) have shown that for many pure metals and ~ u - b a s solid e ~ solutions, creep rate can be correlated to the third power of the stacking fault energy y as
i =~
'
(
~
/
~
~
)
3
(
~
~
~ (12) ~ /
3.2.2.1 Degree of Ordering ~ i ~ u s i obehavior, n which strongly controls creep resistance, is further affected by the degree of ordering in ordered structures. For example Ni,Al (with the closed packed L1, structure) has a first order orderdisorder rans sit ion based on the nearest neighbor (Br agg-W illiam) approximation, whereas NiA1 (wi112 a tructure) has a second . If the respective me assumed to be the order-disorder transition temperatures, the variations of degree of o temperature can be compared, as in 1F basis, it may be argued that ~ i f f u s i o nmobility ~l at the same absolute temperature is likely to be higher in ~ NiAl ~ ~ than 4 ) in Ni,Al. Such a kinetic extrapolati~nbased
309
Creep
ure 6 Theoretical plots of stack~iig-faultenergy versus d-band clectron- or electron-vacancy n u ~ b e r( N v ) (Papon et al., 3979). The solid-solution-hardened y-matrix in mast cre~p~resistant nickel-base superalloys has an NV o f approxiniately 2.5
on thermodyna~icparameters, however, is not well accepted, even though it is known that in C u ~ n ( ~ 2 ) aker, 1995), a large increase in creep rate occurs upon disordering, and that in Fe,Ai (StoloE, 199S), a low order-disorder transition temperature limits the high-temperature creep resistance. Though not rigorous, such thinking allows one to rationalize as to why the creep resistance of ~iA1"basedcoatings with open structure is lower than substrate superalloys containing Ni,Al, with their close-packed structure. 3 2 2 . 2 Effect of ~ t o i c h i o m ~ ~ r y
In ordered i n ~ e r ~ e t a ~ ~the i c s degree , of ordering is further influenced by composition relative to stoichiometry. The situation is even more complex in ternary and ~ u l t i c o ~ ~ o n eordered nt i nt ~ r ~ e t a l l i cwhere s ideally more than one ordering parameter i s necessary to describe site occupancy. However, a loss of order does not nec~ssarilylead to an increase in di~usivity and a loss of creep resistance as one deviates from the ~ ~ ~ ~ cc ~~~ ~~ ~ ~ o In ~s i fact ~ t i oethe ~~ . variations r i c in diffus~vityand creep resistance are functions of crystal struct~reand b o n ~ i ncharacte~istics ~ of the elements. We shall discuss these issues with specific reference to N&A1(and superalloys) with Llz (cP4) structure and NiAl and s i ~ i l ian~t ~ ~ e t a l l i with cs
concentration on either side of the stoichio~etric composition. Three representative results o f their study for Ta- V-, and CO- modified Ni,Ai, are presented in Figure 8 (a), (b) and (c), respectively. Behavior of binary and Ti-ii~odifie~ Ni3Al, are similar to that of Ni,(Al, Ta), but occur at progressively higher creep rates. The results show a sharp isc continuity in creep rate at the stoichiomet~iccomp~sition.Also note that, while the rate of variation of creep rate with constant for Ni,(Al, V), it is weak for Ni,(AI, Ta), ~ lthe , ~ i " r i c hside of the very strong for ( ~ i , ~ o ) ,on stoichiometry. Though the reason for this behavior in the latter two cases i s not well un reasonable to assume that both V and 1
a. a.
l?0.5 CD
.E 0.4 e 0.3 0 Q)
0.2
0.1
a Ni3(A1,X) and ~ ~ ~ ~ r a ~ a yexhaustive, s In l an coinpre~$ive-cree~ study of binary Ni,Ai, and ternary nzodifications with 5 ajo addition of Ti, Ta, V, and CO, Wayashi ef al. (1991) have shown that steady-state creep rates in all cases decrease with increasing nickel
Figure 7 Comparison of degree of ordering in NiAZ vs. Ni,AI, assuming the order/disorder transition occurs near the melting point
Mechimical Propertics
310
1x10*
WO-'
f
I
i -
1x10'
a
\r
v)
M,(AI,T$) ,150MPe
1x10-'
0 1123K
I MO-'
78 75 74 M Concentratton / nt%
13
77
500 [
78 75 Td NI Concentration / sf%
1x10-' 77
73
A
1223K
a
1123~
76 75 74 M Concentration / at%
I
I
\
0N b i A i V i 0 (Ni.CoJ,Al 0
%AI
0
3
8 0 0
A
Q
%
.-
NI Concentration /
51%
Ni Concentralion / at%
Figure 8 Effect of stoichiometry on the compressive steady-state creep rate of (a) Ni,(Al,Ta), (b) Ni,(Al,V), and (c) (Ni,Co),Al, from Hayashi et al. (1991), reproduced with permission of the Materiafs Research Society. Analyzed variation with stoichiometry in (d) stress exponent, (e) activation energy, and Q pre-exponential factor, for all binary and ternary Ni,Al material. evaluated by the authors
311
Creep
strong occupants of A1 and Ni sites respectively, as assumed. Analysis of the data show that while the stress exponent (- 3) does not vary with alloying additions or with deviation from stoichionietry (Figure 8(d)), the activation energy iiicreases with increasiiig creep resistance, ranging from 150-500 kJ/mol (Figure 8(e)). ~nfort~inately, as shown in Figure S(f), the latter is also accompanied by orders of magnitude variation - 10-6 to 107- in the pre-exponential factor of the power-law relationship described by equation (l)! Once again, this brings forth the inadequacy of using oiily the stress exponent and activatioii energy while ignoring the pre-e~ponential factor as a guide to developing creep-resistant alloys. At least one fundamental basis for the variation in the pre-exponential factor has been succinctly demonstrated by a diffusion study of Ni,(Al,Ti) by Larikov et al. (1981). In a teriiary intermetallic, both the site occupancy of the third element and the degree of ordering, influence di~usivity.As listed in Table 2, while both activation energy and Do go through maxim^ around ~ i ~ ( A l ~ ~for T D,,, i ~ ~the ) , inverse is true for DT,. As shown in Figure 9, this composition corresponds to maximum ordering as measured by Xray diffraction, and it also coincides with the miniinum in the diffusion mobility of nickel. It is interesting to note that the apparent activation energy of ~ 4 0 kJ/mol 0 from Figure 8(e) for creep of stoichiometric Ni,(Al,,8 T&), is significantly higher than -211 kJ/mol determined by diffusion experiments in Table 2. However, also note that the variation in the pre-exponential factor with stoichiometry, derived from the creep experiments in Figure 8(f), i s also higher than the variation derived from the diffusion experiments for Ti additions. Clearly diffusion can account for some variation in the pre-exponential factor but it is difficult
to isolate the diffusion contribution from other aspects such as anti-phase boundary (APB) energy (Flinn, 1960), unless careful creep and diffusion experinients are done for the same material. In any case, isolating all the contributions is inherently impossible, as the more creep-resistant nickel-rich stoichiometry leads the ~ ~ a t e r i ainto l a two-phase field more akin to nickel-base superalloys. Here the contribution of the two-phase m~crostructure formed by spinoidal type decoi~positionis major. There is little evidence that the creep resistance or the two-phase alloy can be exceeded by its constituent single phase ordered i n t e ~ e t a l l i c(Shah and 1987). NiAl and other B2 ~ ~ t e r ~ e t a For l l i ~intermetallic ~~ compounds with B2 structures, diffusional behavior mirrors the variation af constitutional vacancy concentration with alloy deviation from stoichio~etry. However, the behavior of constitutional vacancy concentration is strongly dependent on the pair of elements forming a specific €32 compound. The important point is that both activation energy and the pre-exponential factor Do can vary with offstoichiometry, as shown in Figure 10 for But in the case of CoGa(B2), both Do and CO and Ga increase slowly with increasing Ga concentration (Baker, 1995). Once again it is clear that a knowledge of crystal structure, activation energy and stress exponent alone cannot capture the whole picture. Differing atomic boiiding between different elements affects the diffusion mechanisms of the two species (correlated or unco~related),and may lead to different levels of creep resistance. In fact, as shown in Figure 11 (Jung et al., 1987) for stoichiometric (Fe,Ni)Al, the max~mum in creep
able 3 ERect of ternary refractory element addition on ~ i n i creep ~ u ratcs ~ of several binary inter~etallics Alloy
("0)
Fe-28 A1 Fe-28 AI- 1.5W Ti-52.3 A1 Ti-40Al-7.5Ta Ni-24.1 A1 Ni-19.1 A1-5Ta Ni-50 A1 ~i-lOlr-5OAl Cr-23Si Cr-20.3Si-34.1Mo MO-66.6Si MO-69Si-12.6Re
Temperature
Stress
873 K
200 MPa
I255 K
100 M P n
1223K
100MPa
1373K
100MPa
1473K
100MPa
1673K
103MPa
Min. creep rate
Source
2.4 x 10-6 s-' Zhang et al. (1 998) 4.4 x 10-"-' 1.0 x 10-6 s-' Valencia et al. (1990) 2.7 x 10-6 s-' 1 . 6 10-6s-' ~ Hayashi et al. (1991) 1.3 1 0 - 7 ~ 1
312
~ ~ ~ c h a ~Properties ical t 70
4
1 60
I50 I 40
130 1
120
1
from Elsevier Science (a) ~ ~ r i a t i oinn intensity ratio of fundamental I, and superlattice I X-ray diffraction lines and (b) s e l f - d i ~ ~ s ~ o n compared to a simple creep test, these are difficult coefficient of Ni, AI, and Ti, with Ti content in Ni,(AI,Ti) parameters to assess experimentally. ~ u r t h e r ~ o r e , (Larikov et al., 1981) these ~ a r a ~ ~ tdo e rnot s capture the role of dislocation-
resistance at 10 at.% Fe coincides with the minimum in interdiffusion coefficient.~nfortLinatelyin this case we have no i n f o r ~ a t i oon~ the degree of ordering, but it is reasonable to assume that it must be a maximum around 10 at.% Fe. According to Sauthoff (1991), it i s not uncommon to observe a higher apparent activation energy for creep than for diffusion, especially at higher teniperatures, and this difference may be attributed to the temperature dependence of the pre-exponential factor. To avoid this discrepancy, the diffusion coefficient is considered the most important parameter; and, as shown in Figure 12, the creep resistance of binary and ternary 2 aluminides can be correlated with difhsion coefficients. The straight line through the data corresponds to the Dorn equation for dis1ocatio~creep, similar to equation (l),
The plot in Figure 12 provides a rational explanation softer FeAl to NiAl improves the creep resistance. It is clear that determination of the ordering parameter or site occupancy and diffusion coefficient can throw more light on the role of ordering. However,
core structure and microstructLire in two-phase intermetallic alloys. Undoubtedly, a lack of cohesive ~nderstanding o f diffusion and dislocation core structure, in terms of atomic bonding in ordered intermetallics, buries the special role of i ~ t e ~ i e t a ~ l ~ c s within the apparently successful applicatio~of the ~arametricequation.
Among the two-phase alloy systems developed with useful high-temperature creep resistance, ~recipitation~ hardened nic~e~-basesuperalloys, precipitationhardened NiAl-based alloys, eutectic NiAl-based alloys, and ~amellary-TiAl are most notab1e.I We can certainly add two-phase Nb- E-Ti-Si-based alloys (Bewlay et al., 1996, and the chapter by and can also add steel to this list, if we consider Fe$ as an inte~etallic. However, we shall limit our discussion to the foriner two as ~recipitatio~-harde~ied system. systems, and y-TiAl as a 1am~llar"stren~thened 'Naka and Khan in their chapter review design of alloys that are similar to pre~ipitation~hardened Ni-based superalloys.
313
Creep 1
.-c I
0
1
i~~~~ IZ (a) Creep r e s i s t ~ n(at~ ~i = I O - ~s-'> and (b) estimated mterdiffusion coefficient L? as iz function of the iron content for stoichiome~ric(Fe, Ni)Al at various temperatures. From Jung et al. (39871, r ~ ~ r o with ~ ~ p~~ rc~ i~s sdi oonf the ~ a t ~ r ~Research als Society
3.3.1 ~~~~~~~~~~~n ~
~
~
~
d
~
~
i
~
~
~
3.3.1. I ~ ~ c ~ ~ l - perall ~ a salloys e d
iz
In comparison to the d e v e l o ~ ~ ~ of e n ts t r e n g t h e ~ i ~ ~c m e c ~ a n i theories s~ for ~ ~ d e r s t a n d~i in~~h - s ~ r a i n ~ r a t e tensile or compressive strengths, rnodeling of creep beliavior of precipitation-hardend systems is sparse aid not well recognized~In any case, in a series of two pnpers, Carry and ~ ~ r (1977, ~ d 1978) e ~ have ~ e s c r ~ ~ e d the essential elements of creep behavior of precipitat ~ o ~ " h a r superall~ys. d e ~ ~ ~ ~ o w ~ vbefore e r ~ we discuss the conceptual model, a ~ e s c r i ~ t o~foancotigle of wellv) accepted empirical correlations, is in order. ClJ L E~pirically,as shown. in Figure 13 irkin in and Kancheev, 19671, it is known that creep resistance is maxiinized when the € ~ r ~ c ~ ~ ~ i t (7'-y) a ~ ~ lattice - ~ a t r ~ x br L misfit is nearly zero. Since then ~ ~ ~ attempts e ~ o U ~ s have been made to model and r n e a ~ ~the r e misfit, but the issue is fraught with ~ ~ ~ bofl ee ~~ p s~ r i ~ e n ~ ~ ~ ~ e t ~ ~ b ~~l iet ay .s u r lattice ~ n ~ misfit at a given temperature of interest for creep ~ ~ € o ~ ~ ainvolves t i o n hight ~ ~ p e r a t u rX-ray e diffraction, which is a non-trivial eradeavur, Nonetheless, it is generally well accepted that ilzost o ~ t i ~ ~ s%~e~d e r ~with l ~ stable o ~ sc ~ b o ~ ~ a ~ p~~~ig~ i tt ~r ~t ec t ~must r e s have close-to-zero misfit. Secondly, as ~ l ~ in ~ Figure w ~ i 14 ~ ~ ~l ~~ ~iti is ~l ?, , also wet1 known that creep r e s ~ $ t a is ~ c~~ i a ~at ~ ~ i ~ e ~
10"7s"'
\
n CO
If3
m
I
some inter~ediateprecipitate size. As-solutioned and rapidly cooled superalloys with very fine y' size are known to display high yield strength at lower te~peratures~ but poor creep resistance at higher t e ~ p e r a t ~ i ~On e s .tke other extreme, overaged alloys with very coarse preci~itates t r ~ c t ~ have r e low yield strength and are also known to be weaker in creep. 0th of*these results, as well as n L ~ ~ e r o transmisus sion e l ~ c t ~ ~o i~~i ~ ~ o sobservations ~ o p i c of dislocation beliavior d ~ ~ r creep, i n ~ are consistent with a simple
model as follows. In general, much of the ~ i ~ l o c a t i o n glide tinder creep cond~tionsis limited to the disordered y - ~ a t r with i ~ ~screw dislocations bowing out between the precipitates, leaving edge dislocations in the y-y' interf'stces. This is well i l ~ u s t ~ ~ byt ~TEM d micrographs presented in Figure 2 5. The ordered y' p~ecipitateis d i ~ c ~tol shear t because of the formation of antiphase boundaries that require superdislocations, However, if the precipitates are small, the dislocation^ can bypass the ~ r e ~ i ~ iby ~~tes climb. Inhibition of dislo~dtionglide i s a h a r d e ~ i ~ ~ process, and dislocat~o~ cliiftb is a r e c o process, ~ ~ ~ ~ and a balance ~ e t ~ ~ the e e two n may result in steadystate creep. In this respect the large p~ecipit~tes can delay recovery (soften~~g), by requiring a tonger time for the dislocation climb, which occurs via the y-y' interface. Since d i s l o ~ a t i o ~ climb is a d i ~ u s i o ~ ~ controlled process, any significant y-y' misfit (positive i s vexpected to speed the recovery and or i i e ~ ~ t ~ e~ enhance creep rate. If the y' ~ r e c i ~ i t a size t e i s very large, the climb process may be delayed, but at a constant volume fraction of tlie ~recipitatesthe matrix c h a n ~ e ~~si d e nr ~ ~ u i r lower i n ~ stress to glide the screw d~sloca~ions, thus e n h a n ~the i ~ strain ~ rate. Indeed, though, as shown by Pearson et at. (19801, creep resistance cafl be ~igni~cantly improved if the y' precipitate is elongated (y' rafting), without much change in tlie interparticle spacing, and thereby s~gni~cantly slowing down the climb process. Such a structure may na~u~ally form early in the creep process because of large y-y' misfit or can be d e ~ i b e r a t e ~ ~ p r e ~ o ~ eunder d stress. Climb of inter~ace edge dislocatio~sis the principal con troll in^ ~ e c h a n i s ~ ~ ,
creep
3 15
< 0.01 I - . QB-~ Z ~ igure IS (a) initial precipitate structure before crccp tcstlng of a single-crystal superalloy N ~ - ~ C ~ - ~ M ~ - " I ' F ~ - S A Q.01C(b) Edge d ~ s l ~ networks ~ a ~ ~1x1othe~y")~' interface after 2.58 hrs and 1.7% strain of a (103) oriented sample, creep tested at 850 ~ ~ MPs./ From ~ Carry ~ andQ Strudel (1978), reproduced with ~ e ~ ~ i shm s i oElsevim ~ ~ Sclence
and slowing this step - either by lower in^ di~L~sion via all~yin~ low , interph~semis-tit, or elongated structure - all tend to improve creep resistance. As s u ~ i ~ a r i 2 ebyd Carry and ~ t r ude(1978), l for the clinxh ~ e c h a n i ~ to m he c ~ ~ n t r o l l the ~ ~ ishear g ~ stress must be below it critical stress, z,, given by the Orowan spacing L in relationsh~pin terns of the interp~~rt~cle this equation,
on those whose sole objective is the d e v ~ ~ o p ~ of ent
creep-~esjstantalloys. One would assume that, since cutting y' is a difficult process, ~ingle-phasey' should be more creep r ~ ~ ~ s t ~ n t . ~ o ~ e ~ine spite r , of a long sustaine~eflbrt in tliis direction, there is no evidence that sucli i s the case. Xi1 fact Shah and Duht. (1987) have shown tbat even with a highly alloyed 1'' c o ~ p o s ~ tei ~~ ~~ i, v a to ~ eoiie ~ tin a coinplex supera2loy, the creep resistance of the superz,= 2 Gb/L (14) alloy cannot be ~ x c e e d e ~ . studyiiig creep behavior o Above zc, strain would proceed by glide of dislocations noted that even a small volume fraction of Ni solidthat bow out and leave Loops around particles, thereby solution phase enhanced creep resistance. As discussed raising the internal stress of' the material. before, all this is ~onsiste~it with the fkct that creep stra~ningcould take place by climb of dislocations resistance improves with increasing Ni concent rat i ~~ around the particles. One of the theoretical difficulties on either side of stoichiometry (see Figure 8). Yet, it is with this model i s that when interface and pipe an enigma, that while y' ~ r e c ~ p i ~can ~ t inhibit ~s d i ~ ~ s i oare n conside~~d to be c o n t r o l ~ i ~the g ~ stress dislocat~oii~ o t i o iii i ~the weaker ~ ~ ~ i ~and~ the ~ t yr ~ x , expoiiei~tis expected to he around 1 based on the welly' interface caii slow down dislocation climb, tlie single accepted Nczbarro model for grain boundary creep, but s ~ n ~ l e ~ ~y'hcannot a s e slow down the creep process any stress exponent is in the range of 3 4 . better. One potential e x ~ l ~ ~ n a tmay ~ o n be that a trudel ~ l ~ and 7 ~ nxmy ) others have dislocation-free ordered phase is produced through a suggested that this must be rationalized by assuming solid-state reaction. It is known t h t the t h e o r ~ ~ i c ~ l an internal stress gi in tw~-pl~ase materials, such that the egective stress 6, is defi~edas strength can be achieve^ in w hi ske~~, o w i n ~to their d~s~ocation-free nature. It is possible that any presence 6, = (Ta - 6i (15) of a disordered y-tnatrrx allows the ordered phase to be th~reby~ ~ ~ c l any u ~plastic i n ~strain The a r g u t ~ eworks ~ ~ i i i a t ~ ~ ~ a t ~ cin a l lbyr i ~ g ~ n g dislo~ati~~i-free, contribution from the large volume fraction of y', down the stress exponent in equation (1)? and is in~ conceptually appealing, but it lacks a wel~~defined except at higlier stress where dislocations o r i g i n a t i ~ the~ y'~ ~h r e ~ ~ ~ ~ ~ physical origin. ~ x ~ e metkods ~ ~ ehaven been ~ ~ the ~ y-matrix can shear t h ~ o ~ Alternat~v~ly~ it may also be a r ~ ~ that e ~ the , most suggested to ~ e t e r ~ i nthe e internal stress, creep-resistant Ni-rich y' with tlie lowest difrustvlty methods fail to provide Q priori guidance. cannot exist as a single phase w i t ~outthe presence of i ~ g basis, such a r g ~ e n tare s lost with no g ~ ~ i dphysical
i-rich ~ i s o ~ d e r ey.d onetheies~the pivotal role of the y-y' interfa in ~ ~ ~ r creep o v resistance i ~ ~
th. As we shall discuss with reference to c o ~ p o s i t e stlie ~ coherent nature of this y-f interface i s ~ ~ ~ the h most ~ ~ ~ p si ~ attribute n ~of the ~ nickel~ ~ base s ~ p e ~ a I ~ o ~ s .
~
t
d that many refractory elements, r, Nb, and Ta, have l~mitedsolid NiAl, and c o ~ ~ e ~are~effective e ~ t l ~ ~ r e ~ i ~ i t a t istren~theners, on as two-phase fields are attained even at low conceiitratios (Noebe and arolia and ~ a l s t o n ,199'7; Liu and e 16 Weak beam image taken using g = 002 r ~ ~ e c ~ i o n of NiAI-2Hf aged at I173 K for 100h and crept to 28% at work on two-phase NiAl1073K under a compressive stress of 35OMP'a show^^^ ements, systenmtic eEorts d e f o r i ~ a t i o n ~ i ~ d ~dislocations, ced Precipitates are rotated have been made to u ~ d e r s t athe ~ ~creep behavior of d ~ creep r ~ ~~ f ~o ~ ~From ~ a tOh-ishi ~ o ~ er', nl. (19971, reproduced with permission from Elsevier Science
aNiAl(C14-Laves phase) The most c r e e ~ - r e s i ~ t ~ ~ ~ i t d AFN-20 also rely on. of l o w - t e m ~ e r ~ t ~ prere
to note that ~ ~ p ~reci~itation ~ ~ s eis an. ~inantic~p~ted ~ o n s ~ ~of~ trace e ~ camotrnts e of Si from the ceramic shell molds used for casting. ~ o n e t ~ e l e sthe s , benefit of p r e c ~ ~ i t ~ t hi oan~ ~ ~ n i n T IS well d e ~ o ~ s ~ r ~ t e d ~ r o c e s s eto ~ be free of rders of m a g n i t ~ ~ e i ~ ~ r o v ein~creep ~ ~ i t wrth fine ~ r e c ~ ~ ~ ~ ~ t ~ o ~ Figure 16, is apparen ~l ethat of the scale of the ~ i i c r o ~ ~is rc o~n ~i ~ta ~r ~rbto s u ~ e ~ ~inl Figure ~ o ~ s15, Indeed, the ~ i c ~ ~ s is t r ~ ~ ~ ~ r e of l o w - v ~ ~ ~ ~ e -sraperfrac~~~~ alloy, ~ b v j the ~ ~analogy ~ ~ ~ y cannot be taken too far. since the matrix in the present case is an ordered ~ n t ~ r ~ eNiA1, ~ ~ l and l i ~ the a ~ ~ ~ r e ~ i p iitself t ~ t is~ a t e r ~ inte~nieta~lic.
7 ~ a r i a tof~ stram ~ ~ i rate with the ~ ~ ~stress ~ ifore d binary ~NiAI and MiAl-2Hf con t a~i i ~ i nfinc ~ ~ ~2 , - ~ i ~ ~A ~ ~ ~ precipitates. The previously reported data o f binary N S f In precipitation- and dispersion-ham-deledtwo-phase are plotted for comparison. From Oh-ishi et ul. (1997), o i ~ Elsevier ~ ~ i e ~ c ~ ~ a t ~ r i athe l s ~ e f o ~phase, ~ ~ i ~g o ~ ~ called o ~ the l y ~eproducedwith ~ e r ~ i s s ifrom
3J.2
~~~@~~~~
~
~
r
~
Creep
3 17
urgers vector ~ r o v i d ~ a sscale factor matrix, is general~ycontinuous^ and the niicrostructure acing. owe^^^^ note also Il can be t ~ ~ ~ast largely e d i s o ~ ~ ow~~i ~c t~hthe e r~ a t e r ~ a ~ be insen~itiveto L for L > l pm. ~ n t ~ nstr s~c is po~ycrysta~l~ne or siiigIe crystal. Thus creep behavior only results when the scale of the ~icrostructureis is well understood in terms of the scale of the sufficiently fine to influence dislocat microstructure and the orientation of the single crystal the same way as interparticle spaci owever, in y-TiAl, the deformiiig y superailoys i s in equation (14). Lac a single grain or a lamella within a this aspect has stimu~atedthe simplistic iiotion of grain flanked by a2 lamella in polycrystai~inematerial, composites to appear very ~ t t r a c ~ i vtame e and a or the entire la~el1ars t r ~ c ~ ucan r e be aligne~tfiroughout the body in a po~y"synt~etica11yt ~ ~ n ( ~~ i ~e T~ )but in vain. Only composite ap~roachesmore akin to dispersion harde~iing have met with some success le crystal. The d e ~ o r ~ a t i oisnno longer contiiiuous (Sadananda et al., 1993). t h r o ~ ~ g l i othe ~ t grain but is continuous within each lamella extending across the entire grain. The dominant role of anisotrop~cor highly aligned lamellar micro3.3.3 Composites structures 1s we21 d e i ~ o n s ~ r ~ ~ int ePST d crystals by ~ a ~ t h s a r a t hety al. (2000~~ Their results clearly show fn the zeal to improve the fracture t o ~ g ~ n eof s sbrittle that ~ a~ r e ~ ~ep s ~ ~ is~i ~~ nc c~ e~e when v e ~the ~ ~ i~t s es ~ m ~~ ~~ ~p man^ ~ e t ~i ~i ~ i n t~e ~ ~ e t a i ~ori c~§ e c a of lamella are oriented parallel to the stress axis. Though different composite approaches have been explored strongly dominated by microstructural alignment for a number of inter~etallicsystems. A variety of compared to precipitation~hardeningsystems, the effect reinforce~entshave been considered, i n c l u d i ~parti~ of a fine-scale ~icrostructure on creep behavior is culates, platelets, fibers, and whiskers. identical. Limiting defoi~ationto the narrow channels Petrovic (1995) for MoSi,, Bowman of the d e f o r m ~ nphase ~ improves creep resistance~This ~ r ~ et nal. (1992) ~ a for ~ Ni,Al, and ous to the i ~ ~ r o v e iin~ creep e ~ i ~ ~ ~ i s t a of n c e (1993) for r ~ f ~ a c t oir ~ y t e r ~ i e t aftl is ~ a~ ~ s~~a rthat ~ nbyt et ~ ~ ~ p e r a ~upon l o y s~ ~ ~ f tof i ny'g precipit~~~es and large no s i g ~ ~ i ~ c iniproveinent airt in creep s t r ~ ~ g t ~ al., 1980). has been achieved with the c ~ m ~ o s iappro~ch. te Since the two-phase y structure can be produced The broader scenario i s best reflected by twin studies through solid-state rea n in polycrystalline y-TiA1, a of y-TiAl by Rosler et al. (1990) and Valencia 01 al. variety of microstructures can be produced through (1990) and by a study of based coinposites by judicious process~nga ~ dheat t r e a t ~ e n ~For s ~ the ~ a d a ~ a et n ~al.a (1993). first case, a Ti,AtC purpose of disc~ssionhere, creep behaviors of duplex platelet dispersed coiap (DP), fully l ~ ~ ~ e ~l c~rao rs t r u c ~ ~ rwith e s narruw observed that while platelets give rise to a ~ r o n o u ~ c ~ d spa~ing(FEn) and those fully lamella^ with wide strengthen~~g effect at room temperature, the same spacing (FLw) are coinpared in Figure 18. The data was not maintained under creep conditions at high are from Beddoes et al. (1997) and Ghen et aE. (1999), temperature. The loss of strengthening was attributed for binary ~-TiAl(Ti-48 %Al). The correspond in^ to diffusional transport along the particle-matr~xintermicrostructLires are presented in Figures 19(a)(b), and face. There is no surprise if this behavior resembles a (c). It is obvious tfrat a fully lamellar structure with fine textbook case of Coble creep in f i ~ e - ~ r a ~nia~eri~f n~d ~ n ~ e ~s ~~ a a~ ~eand~ planar n ~ grain bound~ries at low t e ~ ~ e ~ ~ ~ u r ~ s ~ enhan~escreep resistan~es i ~ n ~ ~ cmore a n t than ~ ~ can Similar loss in creep re~is~ance for ~ i ~ - ~ ~ r ~ i c ~ be achieved with wide lamellae and interlocked grains. reinforced MoSi, was observed by Sadaiianda ei d. It is iiiteresting to note, however, that creep behavior (1993), but was attributed to attendant grain size of material with wide lamellae approaches that of reduction with increasing particle population, inhibitmaterial with a duplex ~ i ~ r o s t r ~ i c t uItr eis. tempting to ing grain growth. As shown in Figure 20, up to 20 vol. % conclude that the l a ~ e l l amo~phology ~ may not be Sic, creep rate is higher compared to ~ ~ r e i n ~ o r ~ r i t 1 ~ and a l ~ perhaps ~ i ~creep l ~resista~ce r could be ~ ~ ~ ~ MoSi2.~ i creep e rates appear to he s i ~ n ~lower a ~ ~ ~ e vif ean d extremely fine, duplex ~ ~ ~ c r ~ s t r ~ c twith u r e 40vol.% Sic, the creep bel~dv~oiof all r e i ~ ~ ~ ~ could be produced. The creep deformation is shown to and unrein€orced i ~ a t e r i a ~as~ p e a rto ~ c~nverge at be controlled by dislocation glide, with finer lamella higher stresses. Sadananda et al. (1993) have proposed requiring a higher Orowan stress rC to bow a modified shear lag theory to account for these dislocations between interfaces with spacing L, as observations, and have shown that increasin~the aspect shown by equation (14). ratio of rei~orcementis beneficial.
earson on
In the second study, of Ta ~ber-reinforcedy-TiAl, as presented in Figure 21, Valencia et a/. (1990) showed that, althou~hTe2 w a s added to ~ r o d ~ ~fibers c e on ~ o k i ~ ~ ~ c athe tion i ~~ ~ ~ ~ o inv creep ~ ~ ~resistance ~ e n t could not be attri~utedto the fibers, but was due to solid-solution alloying with Ta. This study i s reminis14 ?2 cent of the major t ~ r ia~ d ~ ev~ e l ot ~ i ndirection~1ly~ 10 eutectic systems, back s ~ nickel-base ~ ~ ~ i in the ~ ~ 8 1960s ~2nd 1970s ~ ~ ~ 3990). o Besides d tke ~ cost ~ of ~ ~ 6 ~ ~ o d u c ~ ~i ioi c~k,e ~ - ~ aeutectic se alloys could not $ 4 compete with the ~ ~ e~ ~r e ~ v ~ ik of o~~single~~ e ~ t 2 U crystal nickel-based alloys, because there was no clear 0 $00 200 300 400 500 600 700 ~ ~ ~ ~that e nwc ~e ~ l ~ ~h l ii g~ ~h~ -dcarbide ~~ ~ fibers l ~ ~ ~ ~ were ~ b s o l u t en~e~~ s ~ for a ~ iy n i ~ r creep o ~ ~rews~ ~ tance. Creep bchavior of dirpfex (DP), and wide (FLw) Given that we still do not fully understand the impact and narrow (FLn) fully lamellar y-TiAl (Ti-48%A1) at 760 "C/ t ~ t r~~l l o y i ~ ~ ~ 4 0 Data ~ arc ~ from ~ Beddoes . et at. ( 1 9 ~ ~r)e, p r o d ~ i ~ ~of~ subtle changes in ~ i ~ ~ o s t r u cand with ~ e ~ ~from i TMS s ~ C in ~~I ~~u n i c aand t ~ Chea ~ ~ i et al. behavior of two-phase s ~ ~ ~ r a ~l il ~~l ~ bse,2 alloys, sed (I 9991, reproduced with pcrizissiou of Elsevier Science and or of y-TiAI, it is naive to think that ~ r t i ~ c i ~ l l y
~ i ~ r ~ s ~ r u c ~ofL(a) I r duplex cs (DP), (b) wide fully lamellar, aad (c) liarrow fully tamefiat. y-%Al. From Beddoes et al. (19971, reproduced with permission from TMS ~ o ~ ~ n ~ c a tand i o Chen n s ct al. (1999), reproduced with perinrssron of Elsevier Science
Creep
3 L9
igum 20 Creep rate of SiC-partide-reinforced MoSi, matrix cornposrtes with different volume fractions of SiG phase (Sadaiianda et al., 1993)
When the grain size is coarse ( ~ 3 or~ is ~ ~ adding strong articulates or fibers will necessarily lead ~, a ~ i s o t ~ ~ e the e r o s ~ - s e c t j ~creep It eise ~a b u ~ ~ ~ a n t l yc o n ~ ~ a r a b lto to i n i ~ r o v e ~ e iiii i t creep ~ e s ~ s t a ~ appears to dominate the average creep behavior. clear that unless the scale of the microst~~~cture IS fine Also, in this case intergranular failure can occur enough to influence dislocation glide and climb, no because of environ~enta~interactj ons. When the intrinsic improvement in creep resistance occurs. If the grain size is fine, the increased area of grain boundaries scale of the microstructure is too coarse, the creep s c e i~ ~ ~n r ~ aQ . ~ ~ ~ ~ e s ~ s t ~ of n cthe e d e f o ~ i matrix ~ g d ~ ~ ~the~creep ~ ~ t~ e ~s ~the ~ behavior of the system. In such cases, the overall creep resistance is no better than that given by a rule of 3.4.1 &&ct ojGrain Size mixtures. ~nfortunately,even if this were attractive, the inevitable i~troductiono f additional incoherent Studies of a wide variety o f ~ ~ t e r ~ e t a l l iand cs ~ ~ s ~inter~eta~lic-b~sed interfaces would provide rapid diff~sionp a t ~ w and alloys show that c r ~ ~ ~ - r ~ s i s t ~ escalate the creep d e ~ o r m a ~ ~ oThe n . situation IS decreases as grain size decreases. exacerbated when composites are arti~eial~y made, as studies of MoSi:, by §adananda et al. (1993), of opposed to being produced in situ. Furthermore, creep nickel-base superalloys by Bain et al. (1988), and of ~esistanceis unlikely to be helped, if the matrix/ Fe,A1 as cited by Stoloff (1998). accord^^^ to r e ~ n ~ o r e e ~ e~nt~e rmust f ~ cbe~made weak to improve al.dff993), these results may be best ~ a d ~ ~ eta ~ a fracture t o ~ i g ~ ~ e s s . rationalized by using a creep ~ o d e for l d ~ ~ u s ~ o n ~ mechanisms expressed as i = A(l/d,Y e ~ p ( - ~ ~ / ~ ~ ) (36) ~ F ?
~ 2e i ~ t As noted earlier, the effect o f grain ~ o ~ i n d ISa best ~ ~ e ~ where, d is the graiii size, and grain size ~ ~ ~ o pi is for ~ a b ~ r r ~ - ~ e r creep r i n gand 3 for Coble creep. The r ~ t i ~ ~ ~ lini terms z e d o f two factors based on grain size.
320
stress e~ponentE , is 1 for both processes. A simple c~~lculation shows that because of an inverse power law ~ e p ~ ~ i d e0n11cgrain e size, the creep rate is decreased by an order of mapitLKk up to cl grain size of 300 pm, but than that the rate of change te that in e~uation(171, ( l / d ) is not dimensi , unless it is implicitly a s s ~ that ~ e the ~ grain size is n o r ~ a l i ~ etod f pm a ~ ~ r o x i ~~quivalen~ a ~ e ~ y to a typical ~ e a s L ~of r esome t ~ ~ ~ ~ o ~ at ~r a~ ~c~Note te ~t ethat ~ ~a. in~ all the work s ranged from 30 to 4QOprn. s in this range are g~nerallynot ~ p ~ ~ i ~ ~but ~ i oaret i very s a t t ~ ~ ~ cfor t ~ vfat~~Lte-l~ni~ted e applicat~~ns that require a d ~ ~ u a creep te ~ e ~ ~ s such ~ a nas~ t eu ~r ~ i ~ and ~ ~ d shaft a ~ p l i c ~ ~ i o n s . ~ x c e ~ t i o to i i ~grain-size depeidence have been observed t y ~ i c ~ in l y systems with poor creep resis~ s e Fetance, In t ~ ~ - p h AI-rich et al. (1995) have observed no eep strength of ~ e A l ~ - F e ~and A l ~FeA13-Fe,A15. mitarly, T a ~ a h ~ and s h ~~ i ~ (1~99 1) w also a observed 1x0 effect on grain size in Al-rich single-phase y-TiAl. winan (1992) reported no di~erence in creep NiAl. istance o f (1 00) ori$nted or polycrystal~~~e ese ~ ~may be ~~a t ~ o ~~ aoail the i ~ basis e~~ that t in weak ~clteria~s, the ~ i ~ ~ in~ diffusioii e n ~ chare acte~isticsatid hence the creep c o ~ t r ~ b ~from t ~ o nthe
grain ~ o ~ ~ n ~ and a ~ ithe e s Bulk niaterid can insigni~ca~t.
Ix
In material with close-packed st~uctures,d i ~ ~ L ~ ~ ~ o n alang grain b o ~ n d a r ~ ecan s be s i ~ n i ~ c a ~ faster t~y v 19Sl), ~speci~lly for than iii the bulk ~ ~ a r i l ~etoal., f 0" grain b ~ ~withn~ ~ ~s o r ~ e n rt ag~r~~ i oa ~t than ~~~ rs ~ (Turnbull and ~ o ~ 1954). ~ aThis n limit ~ seems to coincide well with the e ~ ~ ~o ~r ~i~ r~v aat i~othat ns creep ~ ~ ~ eand n ~ d utc h~ i l i~t ~r ~ ~ i ~drop ~ in t osingle ~ ~ l y crystal sLipera~~o~s with high-angle-boundary-type ~ s ~ with t ~ i ~ ~ r i e ~>~10" ~tion defects 1996; Shah and Cetel, 2000). Tole grain boundari~scan be ~ ~ n p r o with v e ~the ~ d ~ i toi"i o ~ ~ minor elements such as C , , Zr and Hf, as is typically done in equiaxed or co~umn~r-grain sL~~e~alloys. While identifying the exact role .of these popularly referred to 'grain boundary s t r e n g t ~ ~ n ielements n~' is a complex issue ~ ~ c and~ Strang, e a 1984), ~ it is apparent, based on the above facts, that these elements must mitigate e n y i ~ o ~ i ~ de ~$ i~t ~~ ~l d aoft ~grain o n boundaries. Once $ ~ ebehsd~ e n ~ ~ r o ~ ~d e~gn~t~adla tisi ~~n~ p p r ~ creep vior is ~ d o ~ i no a t ~by d d~ ~ ~ o ~ ~~ ~i an t~i oo ~n p a ~ i ~ i ~ ~ t across grain boundaries. The role of creep a n i s ~ t ~ ~ ~ y was discussed in this context in ~ e c t i ~3.1.2, n There
the nature of ordering not only alters the ~ctivation energy but also has a significant influence on the preexponential factor as well. of the change in a preexponential factor may be tured by a n a l y z ~ nthe ~ and incorpo~atin~ creep data in terms of di~L~sivity model. that as part of the const i t ~t i v~ ~ e a s u r e m ~ofn ~d i ~ ~ s i may ~ i ~ not y be e ~ s y * f ~ n d a ~ e n t ~ lthe l y ,theoretical or experimental Creep behavior of inte etalfics or alloys based on mination of the iiature of ordering in ~ ~ ~ ~ ~ t i c o n i p o n e ~ intermetallics, follows t same constitutive relationalloys xs even more complex. s such, the c ~ e ~ ~ l o ~ ~ ship as other metals and alloys, which 1s g of an optimum alloy c o ~ i ~ o s i t i with o ~ , the lowest expressed as di ~~usi v~t y highest creep resistance, still remains and Xii that largely an e x p e r i ~ ~ n t ea l~~ p ~ ~ exercise. cal context, the general trend for a c ~ e v ~high i i ~ creep This behavior, however, does not imply that the resistance with the addition o f refractory elements such ordered nature of ~ ~ ~ t e r r n e ~plays ~ l ~ ino c s special role as Ta, MO, W, and to interi~etallicsand interp resistance o f the alloys. On the ~erstandabl~? the most useful ~~etallic-based alloys i in-depth literature review, and a approach. Undoubtedly, ftirther r e f i n ~ ~ e niiit this sensitivity analysis of the constitutive relationship process will occur with a d v ~ n c e ~ e ~ini t di~raction s that an iiiter-relationbetween the ~arameter and imag~ngtechni~ues. ship between A , PI, and scures the origin of the ~reep-res~stance en~ia~cement owing to planar core factors which lead to se in creep rate in an lure of dislocations in alloys with low SFE or absolute sense. Essentially, neither the determination energy is well reco nized. Indeed, in the limited of the apparent activation energy, ~ h tthe in at ion of the stress e x ~ o n eP~I ,~provides , suffic~en~ case of superalloys we also have some i ~ i ~ iinto re~atio~iship between $FE, pli~~se-~iistability and elecinsight into the role of tlie factors coiit~-ibzttingto tron vacancy number, but there is no absolute way to i~iiprove~ent in creep r e s i s t ~ n cSuch ~ ~ factors remain iiiclude this in the c~nst i t ut i v~ relatio~~s~p* buried in the p~e~expoiientia~ para~~ieter A which is has been made in under~ neither ~ ~ ~ s twith a n temperature, t nor i n ~ e p e ~ ~of ~ e n t ~ o n s i d e r a b lprogress standing and ~ o ~ of ethe ~a n i is o t~~ ~ pnature ~~ c of the stress exponent, sLi~niary~ the constitutive creep behavior in single ~rystals. In sin~le-phase model equation for cr serves well for interpolating materials as well as two-phase materials with almost and extrapolatiiig creep data, but provides 110 mechanisotropic ~ i c ~ o s t r ~ c t L Icreep r e ~ anisotropy can be istic iiisight into the I‘actors cont ~ ol l i n~ creep rate, and rnodefed using the ~onstitL~ti~e equation hence fails to ~ l ~ ~ c i dthe a t erole o f interi~etallic~~ with the geometry of the slip systems. ~ i i I ~the~ well-accepted e ~onst~t~~ model t ~ v efor clear that orientation of the aligned tw steady-state creep rate, no universal relationship has structure in ‘single ~rystals’can d o ~ i ~ a the t e creep been proposed for nzodeling primary creep, as this anisotropy, as it does in fully lamellar initial, transitory creep behavior is very sensitive to y-TiAl, and superalloys with prerafted y’ precipitates, material processing history and microstrtict~r~. In any e ~ ~ n ~ s o t ~ino p y ~pplicationo f this ~ n o w ~ e dof~creep case, this aspect is receivin~the needed ~ t t e ~ t i oas ii c o ~ i j ~ n with ~ t ~advances o~ in o p t ~ c a~~ ~ a teclini~ i i i ~ high p r i ~ a creep r ~ strain is r e ~ ~ ~ n i 2toe hinder d the ques is likely to improve our ~ n d e r s t a n ~ ~ofn gthe ~ ~ p ~ ~ cofa the t ~ ~o n~ ~ ~~ ~t ~Li A ~ l r- b~alloys. ~ ns e ~d A low creep behavior of polycrystalline mate~ials, density of mobile ~is l~ cat i ons due to a variety of Based on experience with two-p reasons appears to be a c o ~ ~ oorigin n for high sztperalloys, NiAl-based alloys and y primary creep in many materials. abu~idantly clear that intrinsic creep resist~~nceis It may be concluded that the ~iaxii~ization of the o ~ affected only when the scale of micr~structureis Riie degree of order~ng,which leads to n ~ i n i i ~ i z a t iof enough to i n ~ ~ ~ e dn~c sel o c a t ~glide o ~ or cliiab ~ i e c ~ a ~ d i ~ u s i ~ i taty an o p t ~ m u ci ~o ~ p o ~ i t ~ofo nan alloyed isms. Thus, attempts to improve creep resistance with i n t e ~ ~ ~ e t awill ~ l i cminimize , the steady-state creep rate, composite a p p r o ~ ~ h erarely s seem to succeed, espeIn a sense, the effect of deviat~onfrom the stoichiocially at higher stresses where creep ~ ~ f o ~ a t isi o n metric c o ~ p o s i t i o ~ ,which is cry$tal-structLire expected to be doiniiiated by dislocation motion. depe~ide~it, may be considered 111 the same manner. ~omparisons of the precipitation-hardened alloys With reference to a coIistitutive model, it is clear that
appears to be no paral~elto this kind o f behavior in other intermetallic systenis, but that may be more a reflection of tlieir early stage o f developnient.
Ij
with c o ~ p o s ~s y~set e ~ s ,~ ~ d o u ~ t e point d l y to the Poflock, R. D. Kissinger, R. R,~ ~ o ~ mXI.a A. n ,Green, M, McLean, S. Olson, and 5. J. Schirra). TMS, Warrcndale, critical rote of the interface in controlling creep PA., pp. 515-524. t ~ ~ ~ a v~ ~h oi lrethe . coherent or s e ~ i - c o ~ e r e ninterBeddoes, J., Zhao, L., Au, P., D u d ~ ~D.,~ and s ~ ~ face formed by soli~~state reactions generally helps to ~ r i a n t a ~ ~ f.~(1997) o ~ i , S r v ~ ~ r ~ ~ ~ a l 1997 ~ ~ t (eds e ~ ~ ~ t a i n ~ i b recovery ~t and eiihance creep resistance in the M eNstthal, R. Darolia, C. T. Lm, P, N I ~ ~D.t M ~ ~ l~ ~, ~ ~ ~ former; i ~ i v ~ r ~ a~iicoIi~reiit bIy interfaces resulting from R. Wagner, and M. ~ a i ~ ~ ~ The u c hMiner~tls, i ~ . Metals & ~ r t i ~ c icomposite al processes provide rapid diffusion Materials Society, pp. 109- 118. paths and enhance creep rate in the latter. Bewlay, B. P., Jackson, M. R., and Lipsitt, 1% A. (1996). Otrr ~ ~ d ~ r s oft ~the~ role ~ i of n ~s ~ ~ o r t ~ r a i i ~~~ ~ and ~er. ~~ v1 A (Physical ~ ~ ~ . ~ ~e t a . l ~ and u r ~ ~ ~ ~ d ofe i ~ ~ n~ ~t ~ ~in ~i n~ ~~ ~l r l othe i v ~creep ~ s~ ~ A, no, 12, 3 ~ ~ ~ - ~ 8 ~ 8 r ~ s i s t a ~ will c e come from a study of well-~~aracteri~ed Blavette, D., Caron, P., and T. Khan (1988). S i ~ ~ ~IQ&'~ ~ ~ l o (eds D. N. Duhl, G. Maxrer, S. Antolovich, C . L-und, and microstructures resulting from controlled processing S . Reichman). TMS, W~~rrendnle, PA, pp. 305-314. such as d ~ I : e c t ~ ~s noal~~ ~ i ~ cand a t ~controlled o~ coola ~ Miracle, , ing in solid-state ~ ~ o ~ e s~ s r ~ ~ . ~of advances l ~ ~Boehler, ~ C. J., t ~ a~ ~ ~ o~B. dS.,~a Sre ,e t h a r ~ ~V., in ~ ~ ~ r ~ and c t ~~ o na ~ e cg ~ i i~~~ ~ ~~~ ~~ ~the ~e~us ~d j D.~ B.,g and Wheeler, R. (I 997). ~ S ~ r ~ ~r n~ rl ~~ ~~rr. ~a ~i e ~ a ~ ~ 1997 (eds M. Natlral, R. Darolia. C. T. Liu, P. ~ ~ ~ rD.t ~ n , advent of o r i ~ ~i ~~i ~t a ~ ii n~ g~i c~~ ~o s ~ o ~i sy Miracle, R.Wagner, and M. ~ ~ ~ a gThe u Minerafs, ~ h ~ ~ . likely to shed more light on the funda~entalplastic Metals & Materials Society, pp. 795-804, processes and the nature of creep anisotropy in single Boone, D. H., and Sullivan, C. P. (1973). ASTM STP 520, i n ~creep crystals, arid i t s pivotal role in c o ~ t r ~ l l the ~ m e r ~ c aSociety n for Testing and Materials, pp.40 1-4 t 5. ~ e hoI: ~of ~ po ijycrystalline i ~ ~ t e r i a l s ~ o ~ r n ~a. a nR., (1992j. Mat. Ress Soc. symp, Proc., ~~~~~~
a
155, §
We are in~ebtedto Dr. Chi Law, Dr, Davjd ~ ~ o Dr. w ,
B r o o n ~ ~ e R. ~ d ,W., Ford, D. A,, Bhangu, J. K.* Thornas, M. C,, Fraster, D. J., ~ u r k h o l ~ e P. r , S., Hnrris, M., Errckson, C. I_., and Wahl, 3, B. (1998). J. Er2g. jbr Gas Turb. and POW.,220, 595-608. ~ r o w n A. , M., amid Ashby, M -F, (1980~.~ ~ Met.,r 1 ~ p 1297-1 302, Caron, P., Ohtaa,V., Nakagawa, Y. Gh,Khan, 3'. (1988). S ~ p ~ 1988 ~ ~ (eds l ~D. ~ N.} Duhl, ~ ~ 6. ~ ~aL~~er. S. A ~ t o l o ~ i c6. ~ ~Luiid, , and S. Reichman). TMS, Warrendale, PA,, pp. 21 5-224. ~ ~ c h ~ ni i ~oo r ~~ a t~iservices o i~i ~ for their timely help Carry, C., and Strudcl, 3. L, (1977). Acta ~~~~~~.~ in literatur~search. Carry, C., and Strudel, J, L. (1978). Acra M ~ ~ ~ ~ i . ~ Cetei, A., and Duhl, D. N. ~ 1 9 ~~ ~ ~j . ~ e1988 r (eds a D. ~ N. ~ ~ ~ s Duhl, G. Maurer, S. Antolovich, 6 . Lund, and ees S. Reichman). The ~ ~ ~ ~ l l u r Society, g ~ c a lW a r r e ~ i d ~ ~ ~ ~ PA., pp, 235-244, a. Beddoes, , ~ ~~ l ~oJ., ~.and ~ ~,Zhm,~ L. ~ Allen, J. M., and w hit low, C . A, (1985). ,I. ~ ~ ~ ~ Chen, ~ n W. ~R., ~e r ~r ~ ~~ t.~I ( 1999). r ~ ~ ~ ~7, 171-178. r ~ ~ t a ~ ~ ~ ~ ~ , Power., Tram. @'ASME> Chiba, A., Ono, T., Li, X. C., and T n k ~ ~ ~ sS.~ (1998). lr, Appel, F., and Wagner, R. (1998). Mut. Sc. arzd Engv., ~ Y I t e v ~ e ~ a ~6,~ i35-42 c,s, 187-268. Ashby, M. E;, ( I 973), P r ~ ? of ~the ~TTtzird ~ ~~ n t~~ ~n ~ ~ t ~i ~Cottrell, )~ n a i A. €3. (1964). T ~ e o r ~ t ~ Sr u~l r . u r ~~ ~e~ a~ ~a l l ~ ~ Camelot Press, London, pp. 195-198. ~ ~ ~ OYIa the~ S t er . ~ a~~ tof~~i ~~~u ~ ~ e Cox, A. R. and van Reuth, E. C. (1980). ~ ~ ~~~~~~~~~ t a ~ ~ ~ C a ~ b r i d g eThe j , Institute o f Metals, Vol June 1980, 238-243. Baixz, E(. R., Cambone, M. L,, Myzak, J. M Dalal, R. P., Thomas, C. R.. and Dardi, L. E. (I9845. M. C. (1988). s u ~ e r ~ €988 l ~ ~ (eds ~ ~ sD. N. Duhl, Superalloys 1984 (eds M. Gell, C. S. Kortocicb, R. €3. C. Maurer, S. Antolovich, C. Lund, and S. ~ r ~ c ~ xW z ~ B. l ~ ,Kent, and J. F ~ ~ d a v ~ TMS, ch~. ~ e ~ c ~ i 1988, ~ a n~~S~ ~ , W~~rrendale, PA, pp. 13-22 ~ a ~ r e n d a lFA., c . pp. 185-193. Baker, I, (19535). Mat, Sct. Eag. A, 1 Darolia, R., and Walston, W S , (1997). ~ r r ~ ~ r ~ v a Basoalto, EI. C,, Ghosh, R, N., Ardakani, M. G., Shollock, r ~ t e r . ~ ~ ~ e (eds ~ a ~ M. ~ i cN , s ~ t ~R. a ~ Darolia, , 6. T. Liu, P. B. A., and M c L ~ a M. n ~ (2000). S ~ p e ~ ~2000 l ~ (eds o ~ T. ~ NI. ~s
323
Creep
P. Martin, D. Miracle, R. ~ a g ~ eand r , M. ~ ~ ~ a g ~ ~ c arti in, D. Miracle, R. Wagner, and M. Ya~aguchi).The The Minerals, Metals & Materials Society, p p 461-468. Minerals, Metals &; Materials Society, pp. 585-594. Lupmc, V., Onofrio, G,. Bianchessi, A., and Viniercati, 6. de Fontame, D. (1975). Aeta Met., 23, 553. ~ s ~ ~ Deckes, R. F (1 970). S ~ e e l - S t r ~ n~~ e~ ~~ ~e ~~ ~a ~n~ i ~~( 199 ~ 1~). s P, r a c e ~ 3 ~ofi ~ ~ r~ i~t e r ~ ~ ~ S~ ~yi ~~v l p~ l on ~ ~ ~ ~C~~~~~~~~~~ ~ r ~Sendai~ e t Japan ~ ~(cd. l0.~~~~~~. ~ C l ~ ~ ~a oxl y b d e n LCo., i ~ Gree~iwzch~ USA, pp. 147-182~ t ~ pp. 549---553, Duhf, D. N. (1989). Single ~ s y ~superatloys t ~ l in ~ ~ ~ ~ ~ r ~ l l o The y s ,Japan ~ ~ s t ioft Metals, Ltapmc, V., M ~ r ~ h ~M., ~ n~n ~ ,a M.. ~~ i i ~yfG., r ~~o ~ S ~ p ~ ~ e o ~ n pSo~si~p ~~ e~ ~&~e,r(cds a ~ ~J.c M. s Tien, and T. L., Staubfi, M., and Yin, W. (1997). Str ~ ~ u l f e dAcademic j. Press. New York, pp. 149-282. r ~ ~ t e ~ ~I997 n ~ (eds t a M. ~ ~Nathaf, ~ c ~ R. Darolia, 6. T. Liu, DLi~~zinski, D., Beddoes, J., and Zhao, L. (1997). Creep and Frncmure of E v l g ~ ~ ~ e e~r ~~raz ~t e rand ~ ~ S~ ~ l s r ~ e (cds ~ ~ J.~ ~ e s P. Martin, D. Miracle, R. Wagner, m d The Minerals, Metals & Mater~a~s Society, pp. 5 X 5-522. ~ ~ r t h m a and n , 'E"'. ~ ~ h a n i e dTlie ) . Minerals, Metals & McLean, M., and Strang, A. (1984). Met. ~ e c ~ ~ n o11, l . ,454Mater~a~s Society, pp. 289-298. 464, Erickson, Gary L. (1995). JOM, April, 36-39. Mirkm, E. L., and Kancheev, 0.D. (1967). Met, Scr. eat Treat. nos. 1 and 2, 10-13. Miura, S., Hayashi, T., Takekawa, M., Mish~n~a, Y., and Suztrki, T. (1991). Mat~r€al~s ~ e ~ e a r Soczety eh ~ y r ~ ~ p [ ~ s ~ Gallerneau, Fraiick (1999). ONERA, TP no, 1995-95. Vol. 213, Mater~a~s Research Garg, A., Raj, S. V., Noebe, R. D., Nathal, M. V., and Mohamed, F A., and ~ a n g d o T. ~ ~6. , (1974). AeEn ~ e ~ a ~ l . , Darolia, R. (1998). ~ e ~ and a Mater. ~ ~ .Trurzs. A, 179-1 89. 22, 779. Nabarro, F R, N., aiid de V ~ ~ ~I-f.~ L.~ (r~s~ ,9 5The ? . Ph,~sr~~s Hasegawa, S., Wada, S., Takasugi, T,, and fzunni, 0 ,(1991). London. P ~ ~ o c e eof~ r~ n~ ~~ ~~*rs n ~a ~ ~~ o i~ o~ ~ ~~ ~~ ~ ~ ~o ~ ~r ~i qf~ Creep, ~e ~ ~ Tayfor a l ~ &~Fraiicis, & Noebe, R. D., and ~ a ~ s t o nWe . S . (1947). S ~ r ~ ~ & ~ ~ ~ ~ ~~ e ~~ i japan [d a~ ~ (ed. ~ Osaniu ~ ~ Ezunii), d s The , Japan institute of Metals, pp. 427-43 1, r ~ ~ i ~ r ~ ~ f997 ~ ~ ~(eds f f l M, l i &Nathal, s R. Darolia, C. T Liu, Hayashi, Tohru., S ~ ~ n o dT., a , Mis~ir~ia, Y., and Smub, T, P. arti in, D. Miracle, R. ~ a ~ r i and er~ (1991). at. Res. sac. ~~~~. Proc,, The M ~ ~ ~ ~& M ~ ~~t e rS~~a.o i ~c ~ epp. ~ y5,7 ~ - ~ 8 4 * Hemltcr, K., Lu, M., and Zzipan, 'M. (1997). ~ ~ t ~ u c t i Oh-ishi, ~ r ~ ~K., Horita, Z., ar?dNeznoto, M. (1997)' Mar, SCE.md Ir~ter~etall~cs (eds M. Nathal, R. Darolia, C. T. Lxu, P. Martin, D. Miracle, R. Wagner, and M. Uamagucbi). The Papon, A. M., Sinion, J. P., Guyot, P., aiid Des~onqueres, Minerals, Metals Cci Materrals Society, pp. 147-156. M. C. (1979). Phil. Mug. B, 39(4) P a r t ~ a ~ ~ ~T.a A., t h ~S, u b ~ ~ ~ a n ~ a ~ , ratta, M. G., and Diniidulc, D. M. (2000). A 541-551, 0). Tmns. Paslay, P R., Wells, C, H., arid Leverant, 6. ~ ~ s e f fSockt.y r e ~ S y r v l p ~ ~P' sr ~a ~c ~ e ~ ~ ~81, ~ n263-275. ~~s, Kakehi, Koji (2000). Mater. Sci. a d Evlg., A27 764. Kear, B. H., and Piearcey, €3. J. (1967). T r ~ ~a ~ s~.Met, ~ Soc. ~ e Pearson, D. D., L c d e y , F. Ca., and nf'.41A4E*239, 1209-1215. S ~ p e ~9~~ ~ ~(eds~ J. l K.~Tien, ~ ~A. ~ T, L. ~ Wlodek, ~ i ~ a b j i Pa ~H., , Garg, A., Noebe, nd Nix, w. D. Morrow ITf, M. Ge12, mid G. E. Maurer). Anieri (1999). Met. a d ~~~- ~ ~ uA ,n , ~ ~ -600. Society for Metals, Metals Park, OH, Larikav, L. N., Geichenko, V. V., chenko, V. M. Petrovic, J, (1995). Mat. Sei. and E E ~A, . (1 98 1). Dgfusion Procemes in OrcJered Alloys, Amcrind Pollock, T., and Argon, A. (1992). Rcia Publishing, New Delhi, India. NO. 1, 1-30. Law, C., and Blackburn, M. (1980). S~6perullays1980 (eds J. K. Pope. D. P. and Ezz, S. S. (1984). frzt. ~ ~ t ~ l , ~ ~ ~ v ~ e ~ Tien, A. T. L. Wlodek, H. Morrow 111, M. Gell, and 6.E, 167. Maurer). American Society for Metals, Metals Park, OH, Quested, P. N., Hcndersori, P J., and McLcari, M. (1987). pp. 651-661. Proc. qf the Third I ~ ~ e r ~ ~~ o~ i~ ~ f~onenCreep ar ~ ~nnd~ ~ ~ ~ Lemkey, F. D., and Machliii. I. (1985) Eutectic super~~loy ~ r f f ~of t~ un ~~i v~l e e r ~~nc g~ i e r i aaid l s ~ t r u c t ~ ~pp. ~ 279,s, coi~pos~tions and articles, Patent No.: US 4543235 Patent 294. A s s ~ ~ ~ e e Wiiitcd ~ § ) : Techn~log~cs Corp., Patent Date Quested, P. N,, Hetiderson, P J., and McLean, M. (1988). Filed: Filed date 22 Sep 1982, ~ublicatioiiDate: 24 Sep Arm ~ e ~ ~36,~2743-2752. l . , 1985 p v. Rong, T. S., Jorxes, I. P., and S ~ R. E. ~~~ 9 9 8Acta. ) *~ ~ Leverant, G. R., and Duhl, L),N. (1971), Met. Trms., ~ ~4 ~ ~~ 4507-451 1 ? ~~ ~7. r . ~ 908. RGsler, J., Valeiiei~~ J. Is,Lcvi, C. C., Evans, A. ~ 3and .~ Lnu, C. T. and Horton, Jr,, J. A. (1995). Mat. Set. avld Egr., Mehrablali, R. (1990). Mat. Res. Soc. Symp. Proc., A ~ ~ 170-178. 2, 241-248. Lundstroin, D., and Kirlsson, B. (1 997). Structzn.rr1 Ross, E. W., and O'Hara, K. S. (1996). Super € ~ ~ ~ e r ~1997 2 ~(eds ~ aM. I ~Nathal, ~ c ~ R. Darolia, C . T. Liu, R. D. Kissmger, D. J. Deye, D. L. Anton, A ~~~~~
324
Mechanical Properties
Nathal, T. M. Pollock, and D. A. Woodford). TMS, Warrendab, PA.. pp. 19-25. Rowe, R. G., and Hall, E. L. (1991). Mat. Res. Soc. Symp. Proc., 213, pp. 449-454. Sadananda, K., Feng, C. R., Jones, H. W.,and Petrovic, J. J. (1993). Structurnl Intermetallics1993 (eds R. Darolia, J. J. Lewandowski. C. T. LIU,P. L. Martin, D. B. Miracle, and M. V Nathal), TMS, Warrendale, PA., pp. 809-818. Sauthoff, G . (1991). Proceedings qf ~nternationa~ Symposium on h t m e t a l i i c Compoundr, Sendai, Japan (ed. Osamu Izumi). The Japan Institute of Metals, pp.371-378. Sauthoff, G. (1993). In LX@ion zra Ordered A&ys (eds B. Fultz, R. W. Cahn, and D. Gupta). TMS, Warrendale, PA. pp. 205-222. Schuh, C., and Dunand, D. C. (1998). Acta Mater. 46, 56635675. Seib, David C . (2000). S u p e d l ~ j r2000 (eds T. M. Pollock, R. D. Kissinger, R. R. Bowman, K. A. Green. M. McLean, S. Olson, and J. J. Schirrir). TMS, Warrendale, PA., pp. 535-544. Senba, IEiroyukt, and Igarashi, Masaaki (1997). Structural lntermetullics 1997 (eds M . Nathal, R. Darolia, C. T. Liu, P. Martin, D. Miracle, R. Wagner, and M. Yamaguchi). The Minerals, Metals & Matenals Society, pp. 595.604. Seo, D., Bieler, T., and Larsen, D. (1997). Structural htemetallics. 1997 (eds M. Nathal, R. Darolia, C. T. Lm, P. Martin, D. Miracle, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, pp. 137-146. Seo, D. Y, Beiler, T. R.. An, S. U.,and Larsen, D, E. (1998). Metal!. Mater. Tram A, 29A, 89-98. Shah, D. M., and Anton. D. L. (1993). Structural Intermetallics I993 (eds R. Darolia, J. J. Lewandowslu, C. T. Liu, P. L, Martin. D. B. Miracle, and M. V. Natlial). TMS, Warrendale, PA., pp. 755-764. Shah, D, M., and Anton, D. L. (1991). Mut. Res. Soc. Symp. Proc., Materials Research Society, 213, 63-68. Shah, D. M., and Cetel, A. (1996). Superalloys 1996 (eds R. D. Kissinger, D. J. Deye, D. L. Anton, A. D. Cetel, M. V. Nathal, T. M. Poilock, and D. A. Woodford). TMS, Warrendale, PA., pp. 693-702.
Shah, D. M., and Cetel, A. (2000). Supera1loy.s 2000 (eds T. M. Pollock, R. D. Kissinger, R. R. Bowman, K. A. Green, M. McLean, S. Olson, and J. J. Schirra). TMS, Warrendale, PA., pp. 295-304. Shah, D. M., and Duhl, D. N. (1987). MRS Proceedings Vol. 81 (eds N. S. Stoloff, C. C. Koch, C. T. Liu, and 0. Izumi), pp.411-417. Shah, D. M., and Duhl, D. N. (1988). Superalloys 1988 (eds D. N. Duhl, G. Maurer, S. Antolovich, C . Lund, and S. Reichtnan). TMS, Warrendale, PA., pp. 693-702. Stoloff, N. S. (1998). Mut. Sct. and Eng. A, A258, pp. 1-14. Takahashi, T., and Oikawa, H. (1991). Proceedings of International Symposium on intermetallic Compounds, Sendai, Japan (ed. 0. Izumi). The Japan institute of Metals, pp. 513-517. TiHy, G. (1972). J. of Strain Anal., 7, No. 1 , 61-68. Turnbull, D., and Hoffman, R. (1954). Acta Met., 2, 419. Valencia, J. J., Lofvander, J. P.A., Rosler, J., Levi, C. G., and Mehrabian, R. (1990). Mat. Res. Soc. Symp. Proc., 194, 89-96. Weertman, J., and Weertman, J. R. (1983). ‘Mechanical properties, strongly temperature-dependent’ in Physical Metallurgy, 3rd edition (eds R. W. Cahn, and P. Haasen), pp. 1309-1340. Westbrook, J. H. (1996). Superalloys (Ni-base) and dislocations - an introductlon, in Dislocatioras in Solids, Chapter 48 (eds F R. N. Nabarro, and M. S. Duesbery). Elsevier Science. Woodford, D. A. (1990). JOM, 42(11), 50-55. Zang, Z., and Couture, A. (1998). Scrlpta Mater., 39(1), 4553. Zeumer, B. and Sauthoff, G. (1998). Intermetullics, 6, 451460. Zhang, Zhonghua, Sun, Yangshan, and Shen, Gtianglun (1998). Scrlpta Mater.. 38(1), 21-25. Zhu. W., Fort, D., Jones, I. P.. and Smallman, R. E. (1998a). Acta Mater., &(Ill, 3873-3881. Zhu, W., Fort, D., Jones, I. P.,and Smallman, R. E. (1998b). Phil. Mag. Letters, 77(6), 307-313.
oxygen content at high ten~peratures.The s i g n i ~ c ~ ~ n c e of environ~ienta~~y i n d ~ i c ee~i ~ b ~ t ~ l e m ewill i i t be discussed in detail in a later section.
This chapter is concern~d with the respoiise of iiiter~e~dllic coilnpounds to cyclic defori~atio~i. The ~ r o ~ r e s s i oofn fatigue d a ~ ~ during g e the early stages as is the in~uenceof of cyclic loading is descri~ed~ numerous e x p e ~ i ~ ~ and ~ n~t a~ t~~~r variables ~al on fktigue life and crack growth rates. The vast majority The fatigue response of any material i s a fimction of of the fatigue literature describes experiments under both material ~ r o ~ eand r t ~~x ~~ ~~ r ~ conditions. ~ e i i t ~ l stres~-c~iitrolle~ loading, and this i s reflected in the For example, yield strengt~affects ~~i gh-cycl fatigue ~ data reviewed in this chapter. life, while ~ ~ c t i ~ii ~t y~ u e n lco~ws - c ~ cf ~a~t ~ fife. ~ ~ e The earliest studies of fatigue bchavior of inter~ i c r o s t r ~ fkctors c t ~ ~(grain ~ ~ size, ~r~~~~~~of pre~ e t a ~ ~s li i~o w s ~ ~that long-range order can cipitates) tend to agect fatigue response indirectly, s i ~ i ~ i ~ c a naffect t l y ~ ~ g h - fatigue c y ~ ~fives ~ of Fe-Go-V through their egects on ~ t r ~ n g and th ~~ict~~ity. ttner et at’., 1946). Eater, high-cycle fatigue com~osites, ever, in the case of inter~etallic-~atrix e carried out ~ ~ ~ ~onather two i ~systems: y volume fraction of reinforcement, interface strength Ni3A1 (Doherty et at., 3975; Fuchs and Stoloff, 1987) sand o r i e n t ~ ~ ~ ofo$np e c ~ ~ e relative ns to the remforceand Fe3Al (Fuchs and StofoR, 1988). Data for these iiient ~xi s~axesplay s i ~ n i ~ c a nrotes t in ~~t~~~~ s ~ s t ~ ngene i s rat^^ prior to 1987 can be found in the behavior. External variables differ in their impact review of Stoloff et nl. (1987). Since those investigations were reviewed there have been many studies o f d e p ~ ~ dupon i ~ gthe teni~eratureor other c o n ~ i t ~ oof ~is fatigue behavioi. in al~iiiinidesand silicides reported, test, The ratio of ~ i ~ i m ~ to i ~m a x cyclic ~ stress ~ u ~ with most of the data r e v i ~ w ein~ several ~ ~ b l i ~ a t i o n s(R ratio) is ~ a r t i c u l ~i ~m~ ~y o ~ in t ~c n~ ~ ack-~ru~t~ guchi and ~ ~ a ~ o 1990; s h ~ Stolog~ , 1994a,b; expe~iments,while f ~ e ~ ~ e is n cim~ortant y only wlien 1997). In general, the fatigue b e h a ~ i ~ofr ~ ~ v j r ofactors i ~ ~ or creep ~ e ~ r~o ~c ~~c~o n~t re~sb ~to te intern~etallicsis i ~ t e r i ~ ~ d ito a t that e of conv~ntio~ial fatigue damage. At low t e m ~ e ~ d t ~oE creep ~ ecourse, s, structural metals and cerainics. F is less likely to be a factor. The physical size of prere brittle systems (NiAI, TiAl and existing cracks also is im~ortant,~ s ~ ~in ~the~case a ~ l y high c r a c ~ - ~ r o rates ~ t h and big1 of crack-growth studies. ‘Short’ or ‘small’ cracks tend 0nents ~ o i ~ ~ toa more ~ e dductile i ~ ~ e r ~ e tsuch a ~ as l ~ ~ sto grow much more r a ~than i foiig ~ ~cracks, ~ and often 14- B. Perhaps the single factor niost cannot be described by the ~ a r ~ s - ~ r ~(1863) oga~ characteristic of the fatigue behavior of many interrelationship between crac~~growth rate and stress~ ~ ~ is ~ their ~ significant l f i d~ e p~~ n ~ e n upon c~ i n ~ e ~ sfactor i t ~ range. Finally, surface ~ ~ i ~ dmust i ~ i o ~ environ~entalconditions, including especially h m i d be considered, as f’atigue resistance falls wrth incre~s~ng ity level and hydrogen content at low tei~iperatu~es and surface roughness. ~nfort ~nat el y, the latter variable
Mechatdcal Properties
326
seems to have been studied only in TiAl alloys (Jones and Eylon, 1999).
3. Evolution of Fatigue Damage Damage during cyclic loading evolves from a characteristic defect structure accompanying the development of slip bands, to surface roughening and ultimately the development of fatigue cracks. This process has been widely studied in metallic materials, but little information has been reported for intermetallics. Deformation substructures of fatigued Ti3AI consist of walls of tangled edge dislocations that arise from edge dipoles produced by cross-slip of 2/3{ 1120)screw dislocations (Kerans, 1984). Hsiung and Stoloff (1990, 1992, 1994) have reported observlng dislocation dipoles and vacancy agglomerations in single crystals of Ni,AI that precede the development of surface intrusions and extrusions. Point-defect clusters are continuously generated as cyclic plastic strain is applied. Fatigue damage is then initiated by formation of microvoids at persistent slip band (PSI$)/ matrix interfaces, as shown in Figure 1 (Hsiung and tolofK, 1990). It was shown also that the defect structure was affected by the external environment, with damage developing more rapidly in air than in vacuum. Later, Smith et al. (1992) and Kallingal et al. (1995) compared the development of dislocation substructures in Ni,Al polycrystals and NiAl single crystals, showing that extrusions and intrusions also preceded crack development in NiAl, A detailed analysis of dislocation substructures in these intemetallics, produced by cyclic loading, showed that a high density of dislocation loops and dipoles was formed, see Figure 2. Kallingal (1996) also showed that low-energy, misfitrelieving =rays of dislocations formed. It was shown that low-energy cell structures and veining patterns could be simulated by using a method originally developed by Kratochvil and Libovicky (1986). Cyclic hardening in NiAl single crystals at room temperature was shown to result from dislocation interactions, point-defect formation and the growth of dipole arrays (Kallingal et al., 1995). Cyclic hardening and fracture behavior of polysynthetically twinned TiAl crystals (in which all lamellae are oriented alike without any grain boundaries) have been studied by Umakoshi and co-workers (1995, 1996, 1998). Strain-controlled cyclic hardening and fatigue life were both strongiy dependent on temperature and orientation, see Figure 3 (Umakoshi
Figure 1 Scanning electron micrograph showing formation of surface damage at PSB/matnx interfaces m single crystal Ni,Al (S. A.=stress axis) (from Hsiung and Stoloff, 19903, reproduced with permission from Elsevier Science
et al., 1998), but independent of V and Nb alloying additions. The angle cp is between the loading axis and the lamellar boundaries along a (110) zone in the y phase. Study of dislocation structures during cycling revealed numerous deformation twins and a high density of 1/2[110] dislocations within the y phase but not in the a, phase for crystals oriented with v, = 0". In contrast, specimenscycled with v, = 45"revealed slip and twinning along (1 11) planes in the y phase, parallel to the lamellar boundaries. Vanadium additions decreased the stacking fault energy of y, allowing easy formation of twins. The latter produce substantial residual stresses and coarse surface steps which act as nucleation sites for microcracks. In this fashion vanadium reduces fatigue life (Yasuda et al., 1997). This is in contrast to monotonic deformation, where the forniation of twins promotes higher ductility. Studies of fatigue damage mechanisms and their relation to monotonic deformation properties have been reported by Campbell et al. (1999), Worth et al. (1997), and Chan and Shih (1997) for TiAl polycrystals of different microstructures. These studies have largely comprised observations of the development of cracks as related to the underlying microstructure. Dislocations of 1/2(110) type and twins of 1/6(112) type are seen, but no deformation-induced cell structures are observed (Srivatsan et al., 1995). It is generally agreed from these studies that the superior fatigue crackgrowth resistance of lamellar microstructures results from intrinsic toughening inechanisins such as increased toughness under localized yielding. As a
327
2
B
are 3 Relation between stress a ~ ~ l ~and t u number ~ e of cycles to failure iri ~ o l y s y n t ~ ~ t i ctwinned ~ l l y (PST) TiAl si~lgk crystals at various temperatures: a) 0" orientatioit b) 45" orientation (from Uniakoshi et al., 1998), r e p ~ o ~ u c ewith d permission of TMS and ASM € ~ t e r ~ i ~ s.t p. i ois~an ~ ~ .~ o ~ ~ llow o upoint s ~ y
328
Mechmical Properties
result, crack-growth rates for different microstruct~res correlate well with fracture toughness: higher toughness leads to lower crack-growth rates. The evolution of damage under cyclic loading also lnas been studied at room temperature in CoTi single crystals, which display the sB2 structure (Behgozin et al., 1997), see Figure 4. Cyclic hardening is gradual, and reaches saturation after more than 103cycles at a strain range of 0.1%. A large iiuinber of dipoles and small loops were produced by the motion of dislocations containing jogs, as in the case of Ni,Al single crystals (Hsiung and Stoloff, 1992). Activation of secondary dislocations in addition to the motion of primary ( I 10)(loo} dislocations accelerates cyclic hardening and shortens fatigue lives. Dislocation interscctioi~son primary and secondary planes induce the formation of pileups with strongly concentrated residual stress fields, resulting in the initiation and propagation of niicrocracks.
CF) and low-cycle (LCF) fatigue properties of TiAl have been extensively reported. As in the case of fatigue-crack growth (FCG) studies, limited data indicate tliat a lamellar microstructure is more desirable than a duplex structure for improved CF lives, for an alloy tested at 760°C (Huang and Chesnutt, 1995). Another TiAl alloy tested at elevated t e ~ ~ e r ~ t u showed res s o ~ e w h a tcontradictory behavior, in that a lamellar structure provided better
fatigued at various strains, normalized by the maximum stress at the first cycle (from Behgozin et al., 1997), reproduced with p e ~ i s s i oof~The Iron and Steel Institute of Japan
fatigue resistance than a duplex structure at 8OO"C, with the opposite result noted at 600°C Chesnutt, 1995). Additional evidence for t efTects of a lamellar structure has been provided by Trail and Bowen (1995), who have studied the room beliavior of a gamma alloy temperature, R = 0.1 co~taining2 mass% ~ a t i g u elives were affected by microstructure, re stresses and surface roughness. Fdly lamellar structures were superior to partially lamellar structures; and residual tensile stresses, as well as microcracking introduced by electrodischarge niachining, reduced fatigue lives. On balance, taken together with behavior to be described below, the l a ~ e l l astructure ~ is preferred for fatigue resistance. The HCF behavior of nearly stoichioi~etricTiAl has been studied between room temperature and 900°C (Sastry and Lipsitt, 1977a,b). The ratio of fatigue strength at 106 cycles to the ultimate tensile strength was in the range 0.5-4.8, which compares favorably to the ratio for nickel-base superalloys of 0.3-0.4 in the range 700-800 "C (Sastry and Lipsitt, 1977b). Larsen et aE. (1994) and Ku (1994) have reported einforced TiAl (XD) that the HCF properties of T are excellent. Smooth and notched fatigue strengths at room temperature and at 537 "C and 649 "C are a high percentage of their smooth-bar yield strength, and in some cases actually exceed the yield strength, in contrast to nickel-base superalloys and other precipitation-hardened structural alloys, which display much lower ratios of fatigue strength to yield strength (Stolog, 1993). The superior HCF resistance of the XD alloys was attributed by Larsen et aE. (1994) to strain aging caused by interstitial boron, and in the case of notched samples to the superposition of strain aging and notch-strengthening effects.
Limited HCF data at 650°C have been reported for two a2 alloys and a composite reinforced with SCS-6 S i c fibers (Larsen et aE., 1990). When tested in the longitudinal orientation, HCF lives are increased relative to the base alloy, but are inferior to those of Ni-base IN-100; transverse properties of the composite are very poor. These results are fully consistent with those for FCG of the same composites, as will be shown later. Later work by Wang et al. (1998) on a hybrid SCS-SjTi 6-4/Ti 25-10 composite showed that an experimental S-N curve could be simulated by a computer code that also has been applied to crack-
329
Fatigue
The lives of powder-extru 8 . 5YO ~AI- 7.8 Y0Cr-0.8%22-0. extended, relative to tests in (Gordon et aE., 1994). ydrogen also degrades the lowtemperature HCF resistance o f this alloy ( 1992). in spite of tlii similar alloy, IC-221 than the nickel-ba shown in Figure 5 this type of alloy are affected by pores resulting from solidification of castings; higher cooling rates from the inelt promote fewer pores and iiiiproved fatigue resistance. Low-cycle fatigue data other intermetallics are very sparse. and Unni (1991) have reported the dependence of life on strain amplitude for the polycrystalline alloy IC-218. see Figure 6. The transition fatigue life, N,, defined as the LCF life where elastic and plastic strains are equal, is an important factor in gas-turbine disc life. Nt for this alloy is about 3700 cycles, which i s reasonably close to the value for ferrous alloys of the same
growth data. This work also showed that optimum life was achieved with 25 vol.% fibers and that the presence of the ductile Ti 6Al-4V around the fibers accounted for improved properties.
Surprisingly little is known about the HCF resistance of this class of alloys, and most of the data were published prior to 1987 (Doherty et al., 1975; Stoloff et al., 1987). Fatigue life in stress~controlle~ tests of an alloyed Ni,A1 was shown to be independent of temperature below 800 "C, which correlates well with the t ~ m p e r a t ~ r ei n d ~ p e n ~ e n c of e the microyield strength (Doherty et al., 1975). Other important ftlctors governing the HCF resistance of Ni,Al and its alloys are the composition and the test environment. It i s well known that alloys containing 25a% (13.3 mass %) or more of A1 are brittle when tested in air at room temperature. Consequently, fatigue resistance is expected to be poor in this co~positionrange.
Nominal compositions of nickel alurninides Alloy
Element: weight % (atomic %) Ni
IC-50 IC-218 IC-221
Al
Cr
B
0.02 (0.10) 88.1 (77.9) 11.3 (21.7) 82.9 (74.8) 8.5 (16.7) 7.8 (7.9) 0.02 (0.10) 82.0 (74.2) 8.5 (16.7) 7.8 (7.9) 0.02 (0.10)
f
Zr 0.6 (0.3) 0.8 (0.46) 1.7 (0.99)
Figure 5 Comparison of the fatigue life (NE)of IN 71 3C with an Ni,Al alloy, IC221N1, tested at 650°C In air at a maximum stress of 85 ksl and with stress ratio R = 0.05. IC-221M test bars were produced by investment casting at PCC-Oregon and PCC-Airfoil. See Table I for compositions (from Gieseke and Sikka, 1992), reproduced with permission from Elsevier Science
2 18 compared fmorably 1 alloys such as Ti-6A1-4V steel, especially at high strain n and ~ n n i 3991). , This observation is c o n ~ i s t ~ with ~ t the high ducti~ityand good resistance displayed by TC 218. ve ~tudiedthe i n ~ L ~ e of ~ cboron e
~ ~ ~ e a to r e~dci1itate d crack nu~leation,but fracture stage were similar in the of fatigue cracking in has been carried out by hang et al. ( 1 ~ ~Stage 7 ~ .E cracking simultaneously on
I
tram ~
~
two or more (111) planes and c l e ~ v a ~ f rea c t u ~were ~ observed under cyclic tensile loading. An orieiitationdependent threshold for fatigue cracking from a notch root was noted. s
Very little has been reported on stres~~controlled lives of an alloy etermined at three elevated temperatures, see Figure 7 ( 1994). Very high ratios of fatigue limit (at 107cycles) to yield strength were obtained~e ~ ~ e c iina c~ol ~~~ a r i s o n to nickel-base superalloys such as Astroloy ( ~ s o ~ o t o toloff>1990) (also shown in Figure 7) and Udirnet 700. The higher emhmince ratios of may arise from rapid cyclic har~ening,w h i c ~in , turn, delays crack initiation. In the case of NiA1, however, the high notch sen$itivity at low t e i ~ ~ e r ~ t ushou~d res lead to rapid pro~agation to failure. Noebe et al. (1995) have reported strain produced NiAl alloys at 100 Zr a~ditionsprolong life at low strain amplitudes, see Figure 8. On increasing the strain range for the
Io4
~vs. r le ~ ~i r sto ~~ lfailure s ~ for ~ Mi,A1 e alloy IC-218 (from Gordon and Unni, 1991)
o5
33 1
Fatigue
igure 7 High-cycle lives of P/M Astroloy, Ni,Al+ B and NiAl+ 0.28%Fe normali%edfor yield stress (from ~ a t s etual., ~ 1994)
0.
Fatigue life versus total strain range for P/M NiAl alloys at 1OOOK (from Naebe et al., 1995), reproduced with permission of the Materials Research Society
NiAl+Zr alloy from 0.38% to 0.40%, a drastic reduction in life was noted, due to rapid exliaustion of its already limited tensile ductility. For this reason, no data points are shown in Figure 8 for strain ranges above 0.38%. Fatigue lives of Wf-doped, (1 00) oriented, single crystals, tested at 9 8 0 T , are similar to those of Rent: 80, a nickel-base s u ~ ~ r a l l owith y comparable creep resistance, as shown in Figure 9 ( The behuvior of NiAl appeared to be independent of surface co~dition,i ~ d i c aatlack ~ ~ of ~ notch. sens~tivity
This is in contrast to the situation at room temperature, where notch ~ensitivityis hi
S
4.5.1 Fe,A/ AZluys
HCF experiments on Fe& alloys have shown a complex dependence upon temperature and composition (Fuchs tolofT, 1988). Typical data for Fe-28.7 a%Al are shown in Figwe 10. A h . ~ p # s ~ ~ i e ~ 23.7 ~ ~ a%Al etrie
Mechanicac!Properties
332
5
igure 9 The high-cycle fatigue behavior of a Hf-containing NiAl single-crystal alloy compared to that of R e d 80 (from Wright et al., 1992)
alloy is less fatigue resistant at 25 "C than a 28.7 a%Al alloy, but there is a reversal in resistance at 500°C. This may arise from an aging effect, since a two-phase, coherent, alpha + DO, structure develops in the 23.7 a%Al alloy at the higher temperature, while the other alloy remains single phase. At 25 "C, on the other hand, the presence of superlattice dislocations in the higher AI content alloy probably delays crack initiation, thereby prolonging Fatigue life. HCF data for two Fe,Al alloys are compared to data for quenched (disordered) and slow cooled ( 2 ordered) FeCo2%V, see Figure 11 (~oettneret al., 1966, Fuchs and StoloE, 1987; see also Yamagucbi arid Umakoshi, rdering increases fatigue live for the 0, version of Fe-28.7aYoAl provides longer lives than the quenched (partially B2 ordered) condition. These data show clearly the effects of g on fatigue behavior. ntly, Cieseke et al. (1997) have reported a size effect on HCF life for alloy FA-129 (see Table 2 for compositions of Fe-A1 alloys described in this chapter) at 600"C, apparently as a conse~uenceof differing yield stresses. As tliicltness increased from 12.7mm to 16.5mm, yield stress decreased, thereby reducing life. This is one of the few reports in the literature of the effects of processing 011 fatigue behavior.
4.5.2 FeAl Alloys
om~temperaturecrack-growth properties of Fe%AI have been reported by Castagna and Stoloff life data have iiot been reported. owever, the crack-growth data for Fe-35a%Al are
~igure10 ERects of temperature and type of long-range order (€32vs. DO,) on high-cycle ~ ~ t ~ of gu Fe,Al e polycrysta~s (from Fuchs and Stoloff, 1988). reproduced with p e r ~ s s i o n froin Elsevier Science
siinilar to those for an Fe,Al alloy (FA-129) in tlie 332 condition. The e~~vated-temperature Fatigue strengt~of stressrelieved FeAl was found to decrease sharply between 400 and 750"C, as shown in Figure 12 (Jaske et al., 1998). Life was controlled by a co~binationof fatigue cracking and cyclic creep. In another study, Vasuda et al. (1999) reported a peak in cycles to hilure of Fe48 a%Al at about 400 "C when cycled at d o = 150MPa, see Figure 13. There was no direct corres~ondence between the fatigue strength and the temperature dependence of yield strength. ~ n o ~ a l o strengthenus ing at an intermediate temperature does not necessarily result in increased fatigue life. It was concluded that the to-and-fro motion of (100) dislocations during cyclic defor~ationat high temperatLires induces the formation of excess vacancies and/or cavities, resulting in a decrease in fatigue life.
The only stress-life data for niobium aluminides have been provided by Srivatsan et al. (1996) and Soboyejo et al. (1999) for a ~ b - l 2 ~ A 1 - ~ 4 T i ~alloy, l . 5 ~ as 0 shown in Figure 14. The latter is one of a class of ductile, low density I32 alloys developed for possible structural applications. The endurance limit correspoiiding to a cyclic stress range at which the fatigue life is 107 cycles was about half of the ultimate tensile strength, as is the case for many steels. Crack initiation
333
Fatzgue
11 Coinparison of fatigue resistance of Fe-28.7a%A1 with ordered and disordered FeCo-V at room ~~mperature (from. er et al., 1966; Fuchs and Stoloff, 1988), reproduced with pemission of TMS and ASM ~ n t e r ~ ~ t i o i ~ a ~
Co~positionsof Fe-A1 alloys, a%
FAP-Y Fe A1 Cr Zr C MO
Nb Y Grain size (pm)
FA-129 Fe-35%A1 Ternary
1%Zr-C
65.0 35.0
65.95 28.0
77.07 16.12 5.44
66.12 28.08 5.04
0.11
--
0.13 1.07
0.20
-
0.5 1
0.06 42
-
67.0 28.0
O.5%Zr-C 0.5%Zr 66.45 28.0
66.5 28.0
-
__ __ __
-
385
*Partially recrystallized.
was associated with. slip-band initiation and slip-band interactions/intersections.
loading conditions, crack growth. is usually very rapid, with unusually high stress exponents, up to 100 or inore, in contrast to 3-4 for ordinary structural alloys, as noted in the Paris-Erdogan (1963) equation:
Much attention has been directed towards the fracture bekavior of intermetallic compounds under rnonotonic loading. It i s well known that most interinetallics of interest for structural applications are relatively brittle at low and intermediate temperatures, although in many cases brittleness is the result of an environmental effect (Stoloff and Liu, 1994). ~imilarly,under cyclic
where a is the crack length, N is tbe number of cycles, AK is the stress-intensity range during each cycle; C and rn are experimental constants. Again, the test environment can play a major role, especially in the case of alu~inidesand silicides. However, the study of crack-growth resistance under cyclic loading conditions has lagged investigations of monotonic bebavior.
da/dN= ChK"
(1)
5.f .1 TiAl Alkoys
Elevated temperature fatiguc resistance of stressrelieved FeAl (from Jaslce et al., 1998), reproduced with ~ e r ~ i s sfrom i o ~ Elsevier Science
t e ~ ~ p e r ~int ~anr eFe-48aY~Alalloy cycled at AG = 150 MPa (after Uasuda et al., 19991, reproduced with permission from Elsevier Science
nly in the case of a few intermetallic systems has a comprehensive study of factors in~L~encing crack growth been carried out evious reviews of fatigue beliavior in intermetall have demonstrated just the literature (Stoloff, 1996; there is now an appreciable m alui~inides, some Fe,Al se composites of differing morpliologies. ~nfortunately,data for other intermetallic systems remain scattered and there is insufficient uiiderstaiidin~ of cyclic deformation phenomena in most inter~etallics.
In the case of g a ~ TiAl ~ aa b y s , the best combination of mechanical properties i s obtained in the two-phase region centered around 48 a% Al. ~nfortunately, crack-growth expo~entsare extremely high. Therefore, once initiated, cracks grow extremely rapidly. A wide range of crack-growth exponents, from 8.2 to 56.2, is obtained with different test conditions for a Tin-2Nb alloy, but in all cases the exponents are higher r conventional structural materials (James wen, 1992). Under such conditions, flat S-N cur also are to be expected, so that knowledge and control of stresses is extremely important. High values of the exponent are typical of many brittle ceramcs and intermetallics. For these alloys, high growth rates have been attributed to lowenergy fracture niodes such as cleavage of gamma grains at room te~peratureand trans~ranulardecohesion of the gamma and alpha-two phases at 700 or 80Q"G. It was concluded that inicrostructures with a large proportion of gamma grains have much poorer fatigue properties lamellar gamma/alpha-two alloys (Venkateswar et al., 1995~).Some of the difference between and elevate^ teiii~erature FCC results in Ti-48 a%Al have been attributed to the effects of oxide-induced crack closure at the higher temperatures (Soboyejo et al., 1991). Gampbell et al. (1999) show that for the various inicrostructures (single-phase gamma, duplex and lamellar) the crackgrowth resistance and fracture toughness rank in the same order, see Figure 15. The superior FGC resistance of ~amellarniicrostrLictures IS attributed to enhanced crack- tip shielding arising from crack closure and also have shown a bridging. Campbell et al. (199 deleterious correlation between C ~hresholda i d the percent of equiaxed gamma diRerent TiAl microstructures. C report that a ~ne-grainedTiAl alloy showed a higher tensile ductility but lower fracture toughness and crack-growth resistance when compared to a largegrained material. These workers also reported that sniall cracks in fine-grained lamellar material propagated at K (stress intensity factor) levels below the large-crack threshold reported in the literatu~efor a coarser grained alloy of similar, but not identical, Composition, see Figure 16. Additions of chromium or niobium to binary TiA1 refine the grain size but have no effect on crack-growth resistance at room temperature (Gnaiiamoorthy el al., 1996). Tlie anomalous effects of temperature on crack growth in TiAl alloys
7
400
1
1
Effect of maxiinurn axial stress on high-cycle fatigue life of ~b-12.5Al-41Ti-l.5Moalloy (from Srivatsan et al., 1996), reproduced with p e r ~ ~ s s i oof n The Minerals, Metals and Matcrials Society
Figure 15 Fatigue crack-growth rates for thr~~ugh-thickness long cracks and small surface cracks in duplex and fully lamellar microstructu~esof TiA1 (from Camgbell et al., 1999), reproduced with permission of TMS and ASM International
are discussed extensively by ~ c ~ e l v et e yal. (1997), see below. The superior performance o f lamellar microstructures under both cyclic and rnonotonic loading has as shown in Figure 17, For both microstructures, longbeen attributed to extrinsic shielding from crack crack data can be corrected for the influence of deflection, microcrack nucleation alzead of the crack shielding, in which case long- and short-crack data tip and resultant bridging by lamellar colonies in the coincide. Chan and Shih (1998) have reported in depth crack wake, similar to the behavior of aligned on the ineans by which small cracks grow into large composites (Gnanamoorthy et al., 1996). ones iii a TiAl sheet alloy. Small (relative to the relevant microstructural unit 10 size) crack behavior is important to study because o f potential problems in designing structures based upon long-crack data. Long cracks are defined as having plastic zone sizes sufficiently small to meet the criterion for small-scale yielding while being large compared to the basic microstructural unit (grain size or lamella size). Many structural alloys display much more rapid crack growth when small cracks are present compared to long-crack data. In fact, sniall cracks in a fine-grained, fully lamellar TiAl alloy do not exhibit a growth threshold, but they propagate at rates consistent with extrapolation of long-crack data for the same microstructure, as 1 was shown in Figure 16 (Chan and Shih, 1997). Later work showed, however, that sinall surhce cracks in a fine lamellar microstructure grow faster 10'2 1 than long cracks at the same stress-inteiisity levels; this behavior was ascribed by Campbell et al. (1996) to restricted coiitributioiis from crack closure and Figure 16 Summary of fatigue crack-growth data of large and small cracks in fine-grained, fully lamellar Ti-47A1-2Nbbridgin~ d ~ t eto the limited wake associated with 2Cr-0.2B and c~niparisoii with large-crack data for two small cracks. Camp~eliet al. (1999) clearly demoncoarse- grained lamellar alloys, Ti-47Al-2.hNb-2CCr -t V) and strate the much larger differences in sinall and long Ti-48Al-2Nb-2Cr (from Chan and Shih, 1997), reproduced crack behavior in lamellar and duplex m i ~ r o s t r u c t ~ ~ ~ e swith , permission of TMS and ASM I n t e r ~ ~ t i o n a l I
w e 17 Effect of ~icrostructureon fatigue-crack growth resistance of TiAl (from Campbell et al., 1999), reproduced with permission of Thc Minerals, Metals and Materi~ilsSociety
When TiAl is reinforced with ductile particles (20 vol.% Ti-Nb), FCG resistance is soiiietimes reduced, see Figure 18, and in other cases increased, depending upon the orientation of the faces of the pancake-shaped particles relative to the crack plane ao et al., 1994, 1995a). Specifically, fatigue resistance is enhanced only when the particle
faces are oriented normal to the crack plane. Figure 18 shows also that R ratio is a fixtor in that only for R = 0.1 does reinforced material behave similarly to unreinforced TiAl. The degradation of fatigue resistance with ductile particles, seen also with MoSi, (Soboyejo et al., 1993), is attributed to one or more of several factors: the absence of crack bridging, the influence of the ductile reinforcements in decreasing the crack opening displacement (COD) under cyclic loading, and the accumulation and partition in^ of inelastic strains. In replotting the data of Figure 18, Sadananda and Vasudevaii ( I 995) showed that the threshold is a function of Kmaxas well as 21. Only one study of the effect of brittle reinforce~ents on crack growth of TiAl has been reported; XD T i ~ l + 7 v ~ T ihas ~ , a threshold of 7 ( ~ u m a r ,1994), which is slightly higher than for the TiAl alloy shown in Figure 18. McKelvey et al. (1997) have commented on the a n o ~ a l o u st e ~ ~ e r a t udependence re of FCG behavior of TiAl alloys. Near-threshold growth rates are lower at 800°C aiid higher at 600°C conipared to room temperatu~e.Correspondingly¶ thresho~dsare highest at 800 "C and lowest at 600 "C, with 25 "C behavior in bctwecn. Since the effect is lcss striking in vacuum compared to air, environmental e ~ b ~ i t t l e m e may n t be at least partially responsible for the anomalous
0-'
Influence of load ratio, R, on the cyclic fatigue crack-growth behavior in y-TiAl composite reinforced with 20 v% Ti-Nb particles in the edge (C-R) orientation (from Venkateswara Rao et al., 19941, reproduced with periiiissiovp from Elsevier Science
Fatigue
337
temperature behavior. abru et al. (1997) have shown rates for a nearly fully lai~ellarTiirllloy are lower in high vacuum ( < 5 x 10-4 Pa) than in air, see Figure 19. This embrittling effect was attributed to hydrogen released from water vapor in the air, similar to the environmental e~brittlementdisplayed by iron alurninides. 5.1.2 ~ r t h o ~ h To i ~~ ~A il Alloys ~~ b
This class of alloys has higher tensile strength at temperatures to 760°C as well as comparable fracture toughness and ductility relative to Ti,Al alloys. A recent study (Wang and Yang, 1997) of the cyclic deformation of an orthorhombic composite strengthCS-6 SiC fibers reported the development of matrix fatigue cracks from damaged fibers and fiber/ matrix interfkces near free edges. An integraied computer code was developed to simulate the evolution of fatigue damage, the deg~adationof mechan~cal properties and the ~istributionof fatigue lives. The siniulation was based on a modified J-integral approach in which the progressive damage to fibers was incorporated. In general, the predictions agreed well with the expci-imental results. Tliis technique also has been applied to composites based on Ti-6Al-4V and Ti,Al (Wang et al., 1998). 5.1.3 Ti3Al Alloys
As in the case of TiAl alloys, microstructure plays a major role in fatigue crack-growth resistance of Ti,Al alloys (Aswath and Suresh, 1989; Soboyejo, 1992). A coarse ~ i d r n a n s t ~ t tstructure ~n provides better FCC resistance than does an equiaxed or alpha-two structure in a idm mans tat ten matrix. Roorn-ternperature FCG rates in Ti-2~Al-llNbare strongly affected by heat treatment, with the lowest rates at room temperature obtained with an inhomo weave' structure (Aswath and Suresh, 1989). Growth rates increase sharply with test temperature, with rates at 800°C 100-200 times greater than at room t e i ~ p e r a t ~ r This e . increase in growth rate is accompanied by a change in fracture mode from transgranular cleavage to a tra~sgran~ilar ductile mode at 800 "C, but oxidation at the higher temperature may also contribute to the difference. Davidson (1991) has sliown that the iiiitiatioii and coalescence of microcrac~sare important factors in FCC in Ti-24Al-11Nb. These microcracks are assumed to form in brittle zones ahead of the crack tip. Soboyejo et al. (1989) havc shown that the linkage of
Figure 19 Effect of environment on fatigue crack growth of Ti-48AI-2Mn-2Nb (from Mabru ef al., 1999)
these microcracks can occur below uncracked ligaments even when contact does not occur between the crack tips at the surface. There i s a general tendency for fatigue cracks to grow along brittle phases in these alloys (Aswath et al., 1990). This results in a tortuous path for crack growth, no matter what heat t r e a t ~ e n t is used (Aswath and Suresh,l~89~. Finally, it should be noted that the FCG resistance of similar to that of ill-annealed utilized of the commercial titanium alloys, and to Ti48 a%Al, except that the latter has the highest fatigue threshold (Soboyejo et aE., 1993). Several studies of crack growth in Ti,Al alloys have provided evidence for an envir~~nI~ient~1 influence. There is an effect of frequency on growth rates in lab air as well as in 10-5 torr vacuum for a Ti-24Al-l lNb alloy tested at 649 "C (Parida and Nicholas, 1992). Cracks grow much mor in this alloy as well as i al., 1995). Similar obs temperatures in the range between 650 and 800°C (Aswath and Suresh, 1989; enton et al., 1992; Balsone et al., 1990). There is a significant effect of frequency on growth rates in this alloy; lowering of frequency from 100 Hz to 0.01 €32 causes growth rates to increase by a
factor of 10. In general, growth rates in a, alloys increase owever, transversely oriented fibers result in with moist environments, lower test frequencies, tlie highest growth rate of all, increased hold times and increased test temperatures room and elevated temperature essels et al., 1989; Venkaturaman, 1987). Limited work has been done on small crack 5.2 S L alloys (Davidson et al., 1991; d Larsen, 1992; ~avichandran, 5.2.1 Fe3Al Alloys ). Small cracks in super a2 grow at a slightly higher rate than in the nickel-base superalloy Astroloy, The fatigue belzavior of iron aluminides has been 10-100 times slower than in 7075 A1 extensively studied, over a range of al~iminumcontents et al., 1991). Small cracks in Ti-24AIfrom 16-35 ao/o (Castagna, 1994; Castagna and evelop arbitrary shapes that are affected by 1992; Alveii aiid StolofF, 1996). Effects or ternary microstructure and texture ( avichandran, 1995). solutes, microstru~~ure,te~perature and R ratio These shapes, in turn, affect crack-growth rates. With ( ~ i n i m u mto ~ a x i m u mstress intensity) have been a basketweave inicrostructure, as well as with two other studied under both aggressive (i.e. moist) and inert (i.e. microstructures, (a, and a, + fi), small- and long-crack dry) environments. Most of the work has been carried data correlate well when crack closure is accounted for. out on compositions that exhibit Fe,& type order, owever, small cracl~-growthrates were always higher although some work bas been done on the FeA1 alloy than those for long cracks for an aligned colony struccontaining 35aY0Al. We review here the most recent ture; the latter behavior was ~~ttributed to the nonfindings, e ~ ~ h a s i z ithe n ~ effects of Cr and Zr continuum nature of crack growth in that microstrucadditions and environment on crack-growth behavior. ture. Data for crack growth at tlie surface for the four The Fe-AI alloys studied, a11 of wliich were produced struct~i~es are shown in Figure 20. are by the Oak ~ i d g eNational ~ a b o r a t o ~OR^^) y baps the single factor most likely to in~uencethe listed in Table 2. Note that some alloys are partially growth resistance of titanium aluminides is the llized so as to i ~ ~ x i mtensile i ~ e ductility. Allo use of fibrous reinforcements. A continuous SCS-6 S i c (17a%Al) is disordered, Fe-35a%Al is a fibe~-reinforc~dTi-244.1- 11Nb alloy ( l o n ~ i t u d i ~ a ~ i n t e r ~ e t a ~ ~and i c all of the other alloys orientation) is Far more resistant to crack growth ordered Fe,Al type, tested in either the fully ordered than is nickel-base superalloy I -100 (Larsen et al., DO, conditioii or in the partially disordered
igure 20 A comparison of crack-growth rates of Ti-24Ai-llNb. ~ ~ a v i ~ l i1995) a n ~ ~ ~ ~ ,
and large cracks in four microstructures
Fatigue
339
well as the 0.5 a%Zr-C alloy (Alven, 1997). These data condition. All tests were run at a frequency of 20 have been utilized to formulate a model of enviroiiand an R ratio of 0.5; atmospheres varied betw niental enibrittleinenl, in which the r~te"1iiiiiti~g step is laboratory air, oxygen gas and hydrogen gas, as well as transport of hydr~genre~easedon contact mo~eratev a ~ u u It ~ ,has been shown in ~ u ~ e r ~ udislocation s of water vapor with a clean aluminLi~-richsurface investigations (see StoloK and Liu, 1994 for an (Castagna et al., 1993). extensive review) that under monotonic loading iron Fractographic examination showed that dimpled and nickel a~uminidesare embrittled both by atomic rupture is present in all e n v i r o n ~ e ~ tfor s the hydrogen and by the dissociation of water vapor into containing alloys, except hydrogen and oxygen. The latter effect is particularly hydrogen (Alven, 1997). pernicious, since embrittle~entmay occur even under ternary alloy predomin very low partial pressures of moisture. environments. Dimpled rup lure is oiily evident in A summary of FCG data for the Fe,Al alloys listed in Table 2 appears ure 21 (StolofI and Alven, 1996). The highest rates are e~hibitedby tlie disordered FAP-U while the lowest rates are 2 ordered 0.5%Zr-C alloy. Work on ratlares (Tortorelli, film formation at elevate alloy FA-129 has shown that FCG rates are lower in affects trapping of 1997). It is possible, also, the B2 than the DO, condition, and that there i s little difference between rates for fully and partially recrystallized ~icrostructures(Castagna, 1994). The effects of environ~entare decidedly different for ordered conditions, as shown in 5.2.2 FeAl rllloj)s 1994). Note that there is a much gen gas (1 atm.) on s~ow-~oo1ed~The earliest report of fatigue data for Fe DO, material than in the quenched, partially by Alven and StoloE for Fe-35a%Al; data arc ordered condition. Oxygen gas is considered to be as included in Figure 21. T inert as high vacuum for iron aluminides (Stoloff and e~brittlingeffect of 1 atm. Liu, 1994). pared to either oxygen or air environments, see Figure Frequency efTects are significant for Fe,Al alloys 23 (Castagna and Stoloff, 1995). tested in air, as i s typical under conditions of have reported FCG data for a e n v i r o n ~ e ~ embritt~ement ta~ (Matuszyk et aE., 1990). containing 1wt. O h iianometri~-si~e Y2O3particles and Crack-growth rates increase sharply as frequency is prepared with two inicrostructures by either hipping or lowered, for alloy FA-129 (Castagna et aE., 1993) as extrusion. As in the case of Fe,Al alloys reported
3
Fatigue c ~ a c k ~ g r odata w ~ for ~ several Fe-AI alloys. See Table 2 for compositions (from Stoloff and Alven, 1996)
~ ~ c h ~ nProperties i c ~ l
340
Effect of long-range order on fatigue crack-growth of Fe,Al alloy FA-129 at 25 "C (from Castagna, 1994)
above, the crack-growth resistance was higher in high vacuum than in air, The difference in growth rates for the two environments could not be explained by closure ~rguments.Table 3 coinpares these data to na and Stoloff (1995) for Fe,Al alloys. ote that the thresliold values obtained for the HIP
FeAl were much lower than for the more ductile Fe,Al alloys, perhaps due to closure effects in the latter. Further testing by Tonneau et al. (1998) showed that dissociation of water vapor in air, thereby releasing hydrogen, is responsible for higher crack-growth rates in air than in high vacuum or in oxygen. O xyge~seems to
1
E ~ b r i t t ~ i nefFect g of hydrogen on fatigue crack growth of Fe-35a%Al at 25 "C (from Gastagna and Stoloff, 1995)
34 1
Fatigue able 3 Fatigue crack-growth data for iron aluminides (from Castagiia and StoloE, 1995 and TonneaLi et al., 1998)
HIP-FeAl HIP-FeAI HIP-FeA1 HIP-FeAl FA-129 (B2) FA-129 (B2) Fe-35%A1 (B2) Fe-AI-Cr-Zr-C (B2) Fe-AI-Cr-Zr-C (B2)
Air, 25°C Low vacuum, 25°C High vacuum, 25 "C 0,, 25°C Air, 25 "C O,, 25 "C Air, 25 "C Air, 25°C 0,, 25°C
prevent moisture-induced embrittle~entby reducing the number of adsoprtion sites for water vapor at the crack tip. It should be noted, however, that oxygen does little to prevent moisture-induced enhancement of FCG rates in TiAl alloys under similar experimental conditions, suggesting that there is no universal mechanism of embrittiement in a ~ u ~ i n i d of e s di~erenttypes.
5.3.1 Nip41
In spite of the overwhelming interest in the mechanical behavior of nickel aluininides, little has been published in the past five years on their fatigue behavior. However, for the sake of completeness, we include here a brief summary of what is known about the crack-growth behavior of these aluminides. It has been known for some time that crack-growth rates in Ni,Al alloys are lower than for many superalloys, and that rates of growth increase with increasing temperature (StolofT et al., 1987). Detailed studies of an alloy designated 1C-221 (see Table 1 for cornposition) showed that environmental effects are very important, in that oxygen accelerates crackgrowth at elevated temperatures (Matuszyk et al., 1990). Creep processes also contribute to growth rates at temperatures above 700 "C. Crack-growth rates increase with decreasing frequency in both vacuum and air for this alloy at 800 "C,but the rates always are higher in air. Intergranular fracture occurs in air at all test frequencies, while transgranular fracture is noted in vacuuin. Secondary cracking i s enhanced at lower frequencies, s~ggestingthat diffusion of oxygen contributes to their initiation. 5.3.2 NiAl There is surprisingly little information in the literature about the fatigue crack-growth behavior of NiAl
9. I 12.5 13.7 14.3 23.9 (22) 24.1 23.4 42.3
8.0 5.6 7.2 5.2 6.9 3.0 -
4.7 4.8 6.3 5.5 14.5 20.0 16.3 18.1 18.6
alloys. Flores and Dauskardt (1997) have reported a very high Paris exponent, yy2 = 34, for NiA1-0.25a%Fe single crystals, see Figure 24, as would be expected for a brittle material. This observation suggests that fatigue life under stress-controlled conditions will be extremely sensitive to the applied cyclic stress level. Also, the very low thres~oldof about be eliminated entirely by applying p sive overloads. This effect was first shown for aluminurn and steels by Topper and Yu (1985). The fatigue crack-growth behavior of a NiAl-9% aligned eutectic is strongly dependent on Kmaxand less dependent on 6 K (Wong et al., 1998). There appears to be an R-independent fatigue threshold of about 6MParn112 in the range studied: -0.3<X,<0.35. In this system the ductile MO fibers did not provide a ~~dck-bridging enhancement of fatigue resistance, in contrast ta the beneficial effects of the fibers on 104 rrq
a,
U
G
3%
i
W 1
Fatigue crack-growth data for NiAl + Fe single crystals, showing high Paris exponent (from Flores and Dauskardt, 1997), reproduced with peimssion from Elsevier Science
342
~ e c l ~ a n Proper ~ ~ a lties
monotonic toughness. Similar behavior has been noted in ~b-particle~reinforced MoSi,.
The fracture and Fatigue behavxor of Nb,Al and several composite alloys with Nb particles liave been reported by ~ u r u g e s let~ al. (1992, 1993), DiPasquale et al. (1995) and Hanada et al. (1995). These have generally shown the characteristic hi h crack-growth exponents associated with brittle materials. A study of FGG in a ~ b ~ ~ in-sifu ~ , Acoinposite l showed cracli-growth rates that were much lower than for the unreinforced iiitermetallic, although not nearly as low as for pure esh et al., 1993). Crackrowth rates in 18ao/oA1 alloy, also an z n - ~ i t ~ to those of lamellar TiAl 6w%Al) as well as Nb,Al and Nb (from Murugesh et al., alloys (Davidson and Anton, 1993). Aging of a hot1993), reproduced with permission from Elsevier Scieiice
increasing toughness, cornpared to fibers or layered reinforcements. In fact, coarse laminates with Nb produce drainatic improvements in moiiotonic and cyclic crack-growth resistance of Nb,Al (Bloyer et al., 1997). The effect of laminate orientation is more significant for fatigue crack growth than for monotonic toughness. ~ncreasingthe layer thickness improves both fracture toughness and fatigue crack-growth resistance. The toughness and fatigue threshold values for Nb15ao/o~1-10%~T~ and ~b~15ao/~A~-250/oTi alloys in which Nb,Al precipitates of unknown volume fraction are dispersed in a B2 matrix have been reported asquale et al., 1995). The fatigue threshold of the er alloy was only 3 MPam1I2, compared to a fracture toughness of about 20 MPa rn1/2. Data for l-lO%Ta it^ composites show an i~provernentover Nk,Al, but FCG exponents are y high and cracks in Nb still grow more slowly nada et al., 1995). Recent work shows, Figure 26, that ductile, predominantly I32 Nb-Al-Ti alloys containing 40 or 41a%Ti have comparable or better FCG n mill-annealed Ti-6A1-4V and monoasquale et al., 1996; Ye et al., 1999). The improved crack-growth resistance afforded by high levels o f titanium a to be related to improved fracture toughness. ent toughening, slip band formation at crack tips and crack-tip blunting are believed to account for improved toughness (Ye et al., 1999).
5.5.1 ~~~i~
Several investigations of fatigue crack-growth resistance of MoSi, and its composites with Nb or Sic particles have been published. Soboyejo et al. (1993) reported that at room temperature, stable crackgrowth occurs at AK values as low as 1.'7MParn1/2. The crack-growth exponent, m, was 14, consistent with the high values reported for other brittle intermetal~ics. The same study revealed that the fatigue resistance was lowered by the ad~itionof ~ O V O ~ . ~Nb / O particles, even though the rnonotonic fracture resistance was increased slightly by the particles. This work was important in that it de~onstratedthat improvements in fracture toughness by the use of ductile particles do not necessarily lead to improved fatigue crack-growth resistance. Crack bridging IS observed under rnonotonic conditions but not under cyciic loading. It has been shown by the same group that crack-growth rates increase with increasin N b fiber diameter, aiid that growth rates with 75Opin diameter fibers are 100 times greater than for laiiiinated composites reinforced with 200pm thick Nb lamellae (Soboyejo et al., 1996). The varying effects of different microstructures was attributed to differences in reinforceme~tspacing and cracktip blunting by the Nb fibers. The latter results in much lower Paris exponents for the composites, coi~paredto unreinforced MoSi,. Iii the near- threshold regime, the laminates had similar crack-growth resistance to the particulate
Fatigue
343
. .
1
U
1
Fatigue crack-growth data for Nb-Al-Ti alloys coinpared with data for other engineering materials; (a) as-cast 10Ti, 25Ti and 40Ti alloys, (b) cast and heat-treated l0Ti alloy, (c) cast and heat-treated 25Ti alloy, and (d) cast and hcat-treated 4073 alloy (from ~ i ~ a ~ q ueta al., l e 1996 and Ye et al., 1999), reproduced with perinission from Elsevier Science
344
~ e c ~ ~ aProperties ~ i c a ~
reinforced material. Interestingly, the crack-growth resistance of pure Nb, a weak but ductile metal, IS n i ~ c hgreater than that of any of the composites. By way of contrast, composites with fibers and laminates provide higher monotonic toughnesses than for monolithic MoSi,. Another recent finding is that a wire mesh b reinforceiiient provides much lower crack-growth b particles, see Figure 27 (Badrinarayanan ct al., 1996). A similar effect was noted by the same group on monotonic fracture resistance. Increasing load ratios from 0.10 to 0.70 results in much higher growth rates and reduced fatigue tliresliolds; sucli behavior is characteristic of other metals and ceramics, but the data for this system are much more sensitive to Kmaxthan for metals. A modified Paris power law relation which includes the effects of both AK and Kmax has been established: da/dN = C'K;,, Are"
i, with
~her~s
(2)
ere G' is a scaling constant which is independent of ax, K and R, while y1 and p are e x p e r i ~ e n t a ~ constants, determined to be 13.2 and 7.5, respectively, adrinarayanan e f al., 1996). The oved properties of the wire mesh-reinforced i, have been attributed to extensive crack deflecron at the interface between the matrix and the reinforcements, interface debonding, and frictional sliding between mating crack-face asperities. Therefore, increased crack-tip shielding is believed to result from crack deflection and roughness-induced crack closure. Another feature of the fatigue crack-growth resistance of the composites is that stable crack growth readily occurred, while stable crack growth is very difficult to obtain in monolithic MoSi,. Therefore, a mixed result is obtained with these and other ductilecomposites (e.g. T i ~ l - ( T i , ~ b ) ) ao et al., 1994, 1995a), in that tmce is increased under nionotonic loading for all reinforcement morphologies, but is creased under cyclic loading of part~cle-reinforced oSi, except insofar as stable crack-growth is more readily achievable in the composites. Finally, wire mesh is the most effective reinforcement under both monotonic and cyclic loading, while unidirectional wire reinforcements are intermediate in their effectiveness. In s u ~ i ~ a r fatigue y, crack-growth resistance of oSi,-Nb composites is intermediate between that of many cerainics and unreinforced intermetallics and of most ductile metals. everal studies of crack growth at room and elevated temperatures have been shed for MoSi, and its composites with Sic. a from two sources *
10-6
7 Variation in cyclic fatigue-crack growth rates, dal d N , as a function of the nominal (applied) stress-intensity - K,,,), at R = 0.1 for MoSi, reinforced range, AK (= KmaX with 20 vol.% Nb in the form of high aspect-ratio wire mesh (Nb,/MoSi,) and as spherical particles (Nb /MoSi,) (from Badriiiarayanan et d.,1996), reproduced witi pcimissiaii oT TMS and ASM International
(Soboyejo et al., 1993; amamurty et al., 1994), plotted in Figure 28, sho that cracks grow at very low AK values and progress rapidly in ~ o n o l i t h i c specimens at both 25'C and 1200°C. Creep and environmental effects dominate at the high temperature (Sadananda et al., 1999). Much slower crack growth was noted in the composite because of crack deflections at the matrix-reinfo~cement interfaces. Cracks followed an intergranular path at high temperature, and growth was facilitated by oxygen diffusion through the grain boundaries. It has been shown that the fatigue resistance of ( composites at elevated temperat~rescan be improved by addition of 2w%C ( R a m a ~ u r t y et al., 1994). Previous work had shown a similar improvement in fracture toughness with carbon additions (Maloy et al., 1991). Both eEects seem to result from the prevention of formation of silica during processing as well as during exposure to air and the formation of a ternary Nowotny phase that coiitributes to toughening. The presence of carbon increased the threshold for fatigue crackgrowth initiation and contributed to lowering crackgrowth rates. Further, carbon reduced the incidence of cavitation at crack tips in the composite samples.
345
Fatigue
crack with the layer interfaces may promote crack-tip shielding by debonding, crack-tip blunting and crack bridging. Preliminary data indicated that crack-growth rates were higher than those in ~ e t a ~but s , slower than those in other brittle materials.
0
8 u
n O
0
0
CO
D
0
0
5.5.2 Nb5Si3 a
0 Q 28C
Crack-growth data for MoSiz aid MoSi,-Sic composites (from Soboyejo et al., 1993 and R a ~ a i ~ u r et ty al., 1994; see also Sadananda et al., 1999), reproduced wlth perinission from Elsevier Science
Later work by Lu et al. (1996) reported fatigue crack-growth data for MoSi, composites reinforced with Zr0, particles and Sic whiskers or particles. Fatigue crack-growtli parameters, shown in Table 4, revealed very high crack-growth exponents at 1200 "C for monolithic MoSi, and two of the composites; however, composites with Sic particles displayed exponents more typical of those of metals, in the range 3.8-6.0 at 1200 and 1300 "C. Stable crack growth at 1200°C occurs only within a narrow range of AK values, between 0.9 and 1.4MPai~1'1~. The very high growth rates noted in whisker-reinforced material was attribuied to viscous flow of amorphous glassy phases at the grain boundaries superimposed on mechanical fatigue damage. The fatigue and fracture behavior of layered MO%,/ Nb-15A1-40Ti coniposites with different Nb alloy thicknesses has been reported by Ye et al. (1996). Stable crack. growth was achieved at room temperature with 500pm thick layers, unlike monolithic MoSi,. It was suggested that the interactions of the advancing
Only recently have FCP data for reported. (Zinnser and L e w a ~ ~ o w s 1997). In the former work, a Nb prepared as an in-situ composite, c solid solution in equilibriu~wit metallic. Crack-growth experiments carried out at R ratios of 0.1 and 0.4 showed that crack-growth for two composites were higher than for a 1.24a%Si alloy which is equivalent to the matrix composition. Although the fdtigue behavior of singlephase Nb,Si, was not studied, the of 2 MPam1i2reported for this mat 1992) suggests that stable crack gr
appeared that the increase in growth rates for the Composites relative to Nb were proportional to the volume fraction of intermetallic, although differences in micros~ructuremade this comparison d i ~ c u l t As . crack growth proceeded from the ear-thr~shold region to overload, there was an increasing proportion of cleaved primary Nb observed, suggesting an important role for Kmax in the fatigue behavior. Paris-law slopes of 2 to 5 were observed for the composite, in contrast to much higher values, in the range 30 to 100, reported for a variety of i ~ t e ~ e t a l l i c s toughened with Nb or Nb alloys. It should be noted that the size and distribution of Nb particles or fibers has a major influence on fatigue response, with the best results seen when a Nb mesh is utilized as the
Fatigue crack-growth parameters for MoSi, and coniposites, v = 10 Hz, R = 0.2 (from Lu et al., 1996 and Rarnamurty et al., 1993) Materials nionoll~hicMoSi, MoSi, + 20 vol.% Sic, MoSi, + 20 vol.0/0 TZ2Y* MoSl, + 20 vol. 94 Sic, MoS1, -+ 20 vol.% Sic, MoSi, + 20 vol. % Sic,
C {min/cycle)x (MPa4m)-
9.1 x 10-G 2.3 x 1W6 1.8 x 10-l' 3.3 x 10-9 2.3 x 10-7 1.8 x 10-6
*MoSiQ,+ 20 vol.% ZrQ,, stabiliz~dwith 2mole% Y,O,
rn
AK Range (MPaJm)
Temperature "C
32.1 155.6 40.8 3.9 6.0 3.8
0.9-1.4 0.98-1.3 1.Cl.3 4.1-8.0 2.1-4.7 4.67.0
1200 1200 1200 1200 1300 I300
adrinarayanan et al., 1996). nother iiiteresting finding for the composite was a glier threshold than previously noted for silicide matrices, although the thresliold was not as high as for i solid solution which corresponds to the coniposition of the tough phase in the in-xitzk oom t e ~ p e ~ a t u rcrack e growth rates for Nb-Si composites have been compared with those for two E. (1997). The threshold -Si alloys than for other is exponents are lower, apparently as a consequence of the good fatigue characteristics o f the metallic constituents in these systems. Further, toughness is insensitive to temperature up to 500*6, and the thermal expansion ~ i s ~ a t c ~~~et sw e em~tallic n and i n t e ~ e t a l l i cphases isystem are small (Bewlay, 1996).Therefore, mposites appear to be promising for hightemperature structural applications.
The Fatigue results survey d in this review were (with one exception, for NiAl) determined under stress
control. It is well known that under such loading conditions strength is more important than ductility in determiiiing lives or rates of crack growth. Therefore, one of the most direct means of increasing fatigue life or reducing crack-growth rates is to increase the flow stress of the material, provided that no new low-energy fracture modes are introduced. In general, trends seen in HCF tests are expected to ofYer insight into FCG behavior as well. It has been shown, for example, that the test enviroiiment affects both behavior of the nickel and iron alumi manner (Stoloff and Liu, 1994), with hydrogen detrimental and oxygen beneficial. A study of lowcycle fatigue of B2 Fe-42a%A1 also shows that moisture-induced embrittlement can be prevented by deformation in an oxygen environment or by coating with a protective film ~ ~ a n and e s ~ i ~ ~1997). l a , the major fatigue studies of recent years have focused upon FCG beliavior, the remainder of this discussion will be devoted to a summary of FCG observations. A suininary of crack-growth data for most of the alloy systems described in this chapter appears in Table 5 (Stoloff, 1997). While there has been considerable efYort devoted to o ~ t a i fatigue ~ i ~ data ~ for TiAl alloys, Fe,Al alloys and MoSi, composites, there still remains a serious lack of comprehensive data for most
Fatigue crack-growth parameters for structural interrnetallics (Stoloff, 1997)
Alloy
Ti 48AL Ti-47.3A1-2.3Nb- 1..5Cr-0.4V(d) Ti-47.3Al-2.3Nb- I . SCr-0.4V(l) iAl-20 vol.%TiNb Ni,Al + I3 + Hf Fe-28Al-5Cr-0.5 Fe-28Al-5Cr-0.5 Fe-28A1-5Cr 0.5 Fe-28Al-5Cr(€32) Nb,Al Mb, AI/Nb Nb- 18Al-I 0Ta MoSi,-20 vol.%Nb, MoSi2-20v o l . a / u ~ b ~ ~ ~ ~ o , ~ ) S i ~ - 3vol.%SiCp 0 TiAZ TiAl-20 ~ o l . ~ / ~ T i N h Ti-47.3AI-2.3Nb-1.5Cr-0.4V(l) Ti-47.3Al-2.3Nb-If5Cr-0.4V(d) T i - 4 $ A l - 2 ~ n ~ 2(as ~ bcast)
Temp. (“C,
Frequency (Hz)
25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 1200 800 800 800 800 700
-
“Depends on orientation, heat treatment, **Crack growth across eloiigated grains.
R
25 25 20 20 20 20 20 25 25
0. 1 0.1 0.1 0.1 0.1 0.5 0.5 0.5 0.5 0.1 0.1
20 25 25 25 10 25 25 25 25
0.1 0.1 0.1 0.1 0.2 0. 1 0.1 0. I 0.1
25 25 __
A&, (MPa 5 6.5 10 7 5.5 5 20 14.5 13.5 18 16.9 0.6 3 4 4-7-5* 1.7 2 7.2 4.1 5.5 4.9 12 8
(MPa III~’~) m 8 7.5 20 22 8 9.6 60 29.9 19.8 60 26.4 1 4 5 30 3 3 8.5 9 9 8 18 12
50 15”
6.9 11.2
14 14 3.9 6.3 6.9 10
Fatigue
intermetallics. This is particularly striking for NiAl, for which numerous fracture studies have been published, but for which most of the published Eatigue data are under strain-controlled conditiolls (Noebe et al., 1995; Noebe and Lerch, 1992, 1993). What appear to be the only published crack-growth data for NiAl appear in Figure 24. In order for intermetal~ics or their composites to be seriously considered as commercial materials, more effort needs to be placed upon obtaining databases that include both stress- and strain-controlled fatigue data. However, certain guidelines are available to judge the fatigue resistance of these materials. Unless high ductilities and fracture toughnesses are achieved, FCG exponents will be high and thresholds will be low. Environmental effects are a serious problem; hydrogen introduced directly or by breakdown of water vagor causes severe ~ ~ b r i t t l e m e n t of iron aluminides under both cyclic and nionotoriic conditions (Stoloff and Liu, 1994; Takasugi, 1993). Similarly, NiJ.1, TiAl and Ti,A1 alloys are sensitive to these environments.Therefore, hydrogen contact must be avoided to determine intrinsic FCG rates. Oxygen also causes problems with Ti,Al (Wessels et al., 1989) atuszyk et al., 1990) at elevated temperatures. While oxygen prevents hydrogeninduced embrittlemeiit in iron aluminides, there seems to be no similar beneficial effect in TiAl. The effects of hydrogen aiid oxygen on FCG in other alloy systems has not been studied. Nevertheless, the use of ductile Nb r e i n f o r c e ~ e ~ ttos improve toughizes~of silicides can be expected to lead to serious probleins in both monotoiiic and cyclic properties. This is because of the strong afinity of b and other group Va metals of the periodic table for 0th oxygen and nitrogen and the poor oxidation resistance o f Nb, Furthermore, Nb is readily embrittled by hydrogen (Hardie and Mc~ntyre,1973). The role of microstructure in HGF and FCG has been studied extensively only in titanium aluminides. For TiAl alloys lamellar structures clearly are superior in their HCF and FCC resistaizce to all other microstructures. There is contradictory evidence for the role of grain size in crack growth, but in general grain-size effects on fatigue properties are not expected except through resulting changes in flow ever, the distribution of strengthen in^ phases will be a significant factor. This is clearly the case with composites, wliere it has been shown repeatedly that ductile particles niay actually increase FCG rates while fibers and lamellae can be effective in reducing FCG rates. Orientation effects are significant when non-spherical reinforcements are used
347
et al., 1994). Clearly, much more to clarify the differing effects of second phases on monotonic and cyclic properties. This nec~ssitates the study of crack-sh~eldin~ phenomena, as has been extensively done to date only for some TiAl and MoSi2 systems. Small cracks also grow at rates that are de microstruc~ure,as in the case of Ti 1995), adding another variable to the experimental 1% atrix . Apart from microstrLictura1 and environi~ental control, there are additional variables that can influence HCF and FGG resistance. These include frequency, temperature and R rat general, FCG rates increase and re, decreasing frequency and ver, anoma~suste~perature i~creasingR ratio has been seen in TiAl alloys. Frequency effects are most likely to be observed when an aggressive environment is present or when creep can occur during the fatigue cycle. The influence of R ratio is a result of its effect on crack closure; increasiiig X reduces the likelihood of a crack closin while unloading occurs. The result of this is that the ful~ applied stress-intensity range is felt at the crack tip. The significance of Kmaxhas been discussed authors (Sadananda and ~ a s L ~ d e v a1995; n~ nanan et al., 1996) and this Factor also needs to be considered in dealing with intermetallics, as with other brittle materials. Zn short, the status of our knowledge of fatigue processes in intermetallics remains uiisatisfactory if these materials are to be used in structural applications. When comparing the fatigue behavior of interme tallics and conventional alloys intended for structural applications one finds many si~ilarities. Crack initiation and growth i~echanis~ns are basically the same in both classes of alloys, but the presence 01 long-range order in the i n t e r ~ e t a l l i ~suppresses s slip processes at a crack tip. As a conse~uence,in most intermetallics, it is more difficult than in disordered alloys to initiate cracks under cyclic Isa in an increase in high-cycle fatigue life. difficulty 111 inducing dislocation motion at a crack tip results in stresses not readily relaxed by slip and in a consequent increase of crack-growt~ rates. This is especially apparent in the inore brittle intermetalli~s: TiAl, NiAl, MoSi,. The crack-growth response of the brittle intermetallics, therefore, closely resembles that of ceramics and is a significant impediment to commercialization, especially in critical a~plicatioiis such as aerospace.
348
Mechanical Properties
We close the chapter with a brief summary of research that needs to be done to improve our un~ers~anding of the cyclic behavior of intermetallics. ch more attention should be devoted to mechans of fatig~~e-crackinitiation and growth. location interactions leading to persistent slip ands and extr~sion-intrusion formation have only been reported for NiAl and Ni,Al, in marked contrast to the many such studies on pure metals aiid conventional alloys. The relationship between slip processes at various temperatures and fatigue-crack nucleation ancl growth remains unknown in most interi~etallicsystems. The role of environment in ~ a t i ~processes ~ e also needs further study; i ~ - s i t ~ observations of crack-tip phenomena in the scanning electron microscope are desirable. The influence of surface coatings and surface tr~atmentssuch as shotpeening has not been reported, in spite of their potential importance in commercial applications. levated temperature fatigue properties are lacking , with the exception of TiAl and nately, there i s no alternative to further extensive experi~~ental studies to fill in the gaps in our knowledge of cyclic phenomena in inter~etallic ~~mpo~~n~s.
Alven, D. A. (1997). Ph.D. Thesis, Rensselaer Polytechnic Illstltute. Alven, D. A., and Stoloff, N. S. (1996). Scripta. Metall., 3 1937 oloi-2, N. S. (1997). Mater. Sci. Eng., ejo, W. O., and Suresh, S. (1990). In (eds H, Kitagawa, and T. Tanaka). Materials aiid Components Engineering Publishers, Warley, UK, 1941. Aswath, P B., and Suresh, S. (1989). Muter. Scr. EEg., A l l L5. Bad~iiarayana~~, K., McKelvey, A. L., Venkatesw T., and Ritchie, R. 0. (1996). et all. Tram., Balsone, S. J., Maxwell, D. C., Khobaih, M., and Nicholas, T. (1990). In Futigue YO, v. I (eds H. Kitagawa, and T. Tanaka). Materials and CQmponents Engineering P ~ b ~ i s ~ eWarley, rs, UK, 1905. Balsone, S. J., Larsepl, J. M., (1995). Mater. Sci. Eng., Behgozin, A., Nakano, T., and Urnakoshi, Y (1997). ISIJ Internutional. 37, 1218. R., and Llpsitt, H. A. (1996).
Bewlay, B. P., Lewandowski, J. J., and Jackson, M. R. (1997). J . Metals, ~9(8),44. Bloyer, D. R., Venkateswara Rao, K. T., and Ritchie, R. 0. (1997). Muter. Sci. Eng., Boettner, R. C., Stoloff, N. S . , and Davies, R. G. (1966). Jarnes, A. W. (1995). Mater y, A. L., Lillibridge, S., Vcnkateswara Rao, K. T., and Ritchre, R. 0. (1996). In ~ e ~ o r m a tand i ~ Fracture ~~ of U r d e r e ~1~termetallicsIII (eds W. 0. Soboyejo, T. S. Srivatsan, and H. L. Fraser). TMS, Warrendale, PA., 141. Campbell, J. P,, Kruzic, J. J., Lillibridge, S., Venkateswara Rao, K. T., and Rttchie, R. 0. (1997a). S c r ~ Mater., ~a 37, 707. Campbell, J. P., Venkateswara Rao, K. T., and Ritchie, R. 0. (1997b). Muter. Sci. Eng., A 2 3 9 1 ~ 4 722. ~, Campbell, J. P., Venkateswara Rao, K. T., and Ritchie, R. 0. (1999). et all. Mater. Tralzs., 3 0 4 563. Castagna, A. (1994). P1i.P). Thesis, Rensselaer Polytechnic Institute, Troy, New York. Castagna, A., and Stoloff, N. S. (1992). Scripta. m et all., 2 673. Castagna, A., Maziasz, P. J., and Stoloff, N. S. (1993). In High Te~peratureOrdered ~ ~ ~ e r ~ eAlloys t a lV l ~(eds ~ I. Baker, R. Darolia, J. D. Whittenber€er, and M. H. Yoo). MRS Symp. Proc., Pittsburgh, PA., 288, 1043. Castagna, A., and Stoloff, I?. S. (1995). In High ~ e ~ ~ p e r u t u r e lloys VI (eds J. A. Horton et al.). part 2, MRS, Pittsburg~,PA, 1171. . S. (I 997). Metall. Mafer. Truns,
S. (1998). ~ e ~ a l~l .a ~ Tram., e r %9A,73. Davidson, D. L. (1991). In Proc. Morrzs Fine Symp. (eds P. E(. Liaw et al.). TMS, Warrendalc, PA, 355. Davidsori, D. L. (1997). Metall. Mater. Trans., %A, 1297. Davidson, D. L., Cainpbell J. B., and Page, R. A. (1991). Metall. Truns., %%A,377. Davidson, D. L., and Anton, D. (1993). In High Te~perature O r d e r e ~ I n t e r ~ e t aAlloys l l ~ c V (eds I. Baker, R. Darolia, J. D. Whittenberger, and M. H. Yoo). MRS Symp. Proc. v. Pittsburgh, PA (1993), 807. uale, J,, Gahutu, D., Konitzer, D., and Soboyejo, W. 0. (1995). In IIigh Temperalure Ordered h t e (eds J. A. Horton et al.). MRS Symp. Proc. Pittsburgh, PA, 1347. DiPasquale, J., Ye, F., Lernski, J., and Soboyejo, W. 0. (1996). In ~ ~ ~ ~ a rd ~Fracture ~ ~ of i Ordered o n I n i e r ~ e t a l ~ ~I11 c s(eds W. 0. Soboyejo, T. S. Srivatsan, and €3. L. Fraser). TMS, Warrendale, PA, 437. Doherty, J . E., Giamei, A. F., and Kear, B. (1975). Metall Trans., 6A, 2195. Dowling, W. E. Jr., Donlon, W T., and Allison, J. E. (1991). In High Te~perutui~e Ordered Inter~etall~c Allo.ys VI (eds J. A. Horton et al.). MRS Syinp. Proc., 36 757.
Fatigue Flores, K. Me$and Dauskardt, R. €3. (1997). Scripta Mater., 36, 1377. Fuchs, G. E., and Stoloff, N. S. (1987). Scripta M~tall.,211, 863. FLIC~S, G. E., and StoloE, N. S. (1988). Actn Me~all.,36, 1381. Cieseke, I3. et al. (1997). Oak Ridge National Laboratory, unpublished. Gieseke, B., and Sikka, V. K. (1992). Mater. Sci. Eng., A3 520. oh Y., and Mizuhara, Y. (1996).
i, C. K. (1991). J . Mater. Scz., 6183. Gordon, D. E., Uniii, C. I<., and Stoloff, N. S. (1994). Fatigue Fract. Eng. Mater. Struct., Hanada, S., Tatsuo T., aiid Gnamamoorthy, R. (1995). Acla Metall. Sinica (English Letters), 8, 477 Hanes, D. B., and Gibala, R. (1997). In High Ordered IntermetalliC Alloys VII, MRS Symp Pittsburgh, PA., 97. Hong, M.-H., McNaiiey, J. M., and Ritchie, R. 0. (1998).
349
Soboyejo, T. S. Srivatsan, and D. L. Warrendale, PA, 27 LI, G., Guo, J-T., and Watig, Z-G. (1995). In High
~ e f o r ~ ~ i zand t i oFracture ~ qf Ordered Inter~e~ul1ic.s I11 (eds W. 0. Soboyejo, T. S. Srivatsan, and H. L. Fraser). TMS, Warrendale, PA, 507. Mabru, C., HeiiaE, G., and Petit, J. (1997). I n t e r ~ e t a l l ~ ~ s , 355. Maloy, S., Heuer, A., Le w a i i d o ~ s~J,, i , and Petrovic, J. J. (1991). J Amer. Cer. Soc., 74, 2704. Matsugi, K., Smith, T. R., and StolofY, N. S. (1994). Matls. D. J., and Stoloff, N. S.
, K. T,, and Ritchie, R. 0. (1997). Scripta Mater., 11, 1797. Mendiratta, M. G., Lewandowski J. J., and Dimiduk, D. M. (1991). Metall. Trans.,22 Murugesh, E., Venkateswara Rao, K. T., DeJonghe L. C., and Ritchie R. 0. (1992). In Inter~etallic~ o ~ p o ~II~ i ~ e s Hardie, D., and McIntyre, P. (1973). Met (eds D. B. Miracle, D. E. Anton, arid J. A. Graves). MRS Hsiung, L. M., and S toloff, N. S. (1990). Ac Symp. Proc. 273, MRS, Pittsburgh, PA, 433. Hsiung, L. M., and Stoloff,N. S. (1992). Ac Hsiung, L. M., and StolofT, N. S. (1994). A Murugesh, L., Venkateswa (1993). Scripta Metall., Huang, S. C.. and Chesnutt, J. C. (199 ~ ~ o ~ ~ o ~Principles i z d ~ s , and Practice, v. 2 (eds J. H. Nazmy, M., Noseda, C., SauthoE, G., Zeumer, B., and Westbrook, and R. L. Fleischer). Wiley, New York, 73. Anton, D. (1995). In High ~ e ~ p e r a t u r eOrdered Isonioto, T., and Stoloff, N. S. (1990). Mater. Sci. Eng, A1 Intermetallic Alloys VI (eds J. A, Horton et al.). MRS 171. Syinp. Proc. 364, part 2, 1333. James, A. W., and Bowen, P. (1992). M a ~ ~Scl r . Eizg., A1 Noebe, R. D., Lerch, B. A., and Rao, B. S, (1995). In High 486. Te~peratureOrdered In~ern~~tallic Alloys VI (eds J. A. , and Shademan, S. S. (1998). Mater. Horton ef al.). MRS Symp. Proc. 364, part 1 , 291. Noebe, R. D.,andLerch, B. A. (1992). Scrz~t~lMetall., 27,1161. Jones, P. E., and Eylon, D. (1999). Mater. Sci. Eng., ~ 2 6 3 , Noebe, R. D., and Lercb, B. A. (1993). In High ~ e ~ p e r ~ t u r e 296. Ordered Inter~etallicAlloys V (eds I. Baker et al.). MRS d Lewandowski, J. J. (1992). Symp. Proc. 288, Pittsburgh, PA, 743. Parida, B. K., and Nicholas, T. (1992). Mater. SCZ.Eng., A153, 493. Rajan, K. (1995). Muter, Paris, P. C., and Erdogan, F. J. (1963). J. Basic Eng., Tram. Kallingal, C. G. (1996). PhD Thesis, Rensselaer Polytechnic A S M E Ser. D., 85, 528. Inst, Troy, New York. Penton, R. J. T., Cope, M. T., and Bowen, P. (1992). m at er Kerans, R. J. (1984). ~ e ~ a lTrans., l. AI Sci. Eng., A153, 508. Rratoclivil, J., and Libovicky, S. (1986). Scripta Metall., 20, Ramamnurthy, U., Kim, A. S., Suresh, S., and Petrovic, J. J. 1625. (1993). J. Am. Ceram. Soc., 76, 1953. Kumar, K. S. (1994). In C o ~ p o ~ ~Science i t e ~ ~aizd ~ e c h n o ~ o g y , Ra ~ a m u r t h y U., , Kim, A. S., Suresh, S., and Petrovic, J. J. 52, 127. (1994). J. Am. Ceram. Soc., 77, 2681. Kumpfert, T., Kim, Y. W., ancl Dimiduk, D. M. (1995). Mat. Ravichandran, K. S., and Larsen, J. M. (1992). Mater. Sci. Eizg., ,4153, 499. Larsen, J. M., Williams, K. A., Balsone, S. J., and Stucke, M. A. Ravichandran, K. S. (1995). In ~ e f o r ~ a t i oand n Fracture of (1990). In High Temperature Aluniinidm and I ~ i t ~ ~ ~ m e t ~ ~ l l Ordered ~ c , ~ Intermetallic Mutermis 11(eds W. 0. Soboyejo, T. (eds S. H. Whang, C. T. Liu, D. P. Pope, and J. 0. S. Srivatsan, and R. 0.Ritchie). TMS, Warrendale, PA, 39. Stiegler). TMS, Warrendale, PA, 521. Sadananda, IS.,and Vasudevan, A. K. (3995). Mater. SCE. Larsen, S. E., Behreiidt, M., Walker, N., Thakker, A., Eng., Al921393, 490. Davidson, D., aiid Clctnmens, D. (1994). In Fatigue und Sadananda, I<., Feng, C. R., Mitra, R.. and Deevl, S. C. F r a c ~ ~ of r e Ordered I n t ~ r ~ e t a~~ a~ it ce r ~Ia(eds l ~ ~W. 0. (1999). i at er Sci. Eng., A261, 223.
350
~ e ~ ~ i a n iProperties cul
Sastry, S. M. L., and Lipsitt, H. A. (1977a). m et all. Trans., A8, 299. Sastry, S. M. L., and Lipsitt, H. A. (1977b). Acta ~ e t u l l . ,
. (1996). Oak Ridge Natioiial Laboratory, unpublished.
Tonneau, A,, Henaff, G., Gerland, M., and Petit, J. (1998). 1985). Intl. J . Fatigue, 7 , 159. Tortorelli, P. (1997). Oak Ridge Nationai Lab, unpubl Trail, S. J. and Bowen, P. (1995). Mater. Sri. Eng., A19 427. Tsutsurni, M., Takano, S., Kitamura, T., and Ohtani, R.
H. Y., and Nakano, T. (1995). Mater a, H. Y., and Nakano, T. (1996).
~enkateswaraRao, K. T., Odette, G. R., and Ritchie, R. 0. (1994). Acsa Metull., Venkateswara Rao, I<. T., and Ritchie, R. 0, (199%). In Fatigue and Fracture of Ordered I n t e r ~ e t u ~ ~I1i c(eds ~ s T. S. Srivatsan, W. 0. Soboyejo, and R. 0. Ritchte). TMS, ~ ~ ~ r e ~ i dPA, a l 327. e, Soboyejo, W. O., DiPasquale, J., Ye, F., Mercer, C., V~nkatesw~ra Rao, I<. T., Kin, U-W., and Ritchie, R. 0. Srivatsan T. S. and Konitzer, D. G. (1999). Metall. (19958). Scripta ~ e ~ a l l3. , Trans., Venkateswara Rao, K. T., Kini, Y. W., Muhlstein C. G,, and Srivatsan, angwood, M., and Soboyejo, W. 0. Ritchie, R. 0. (1995~). Mater. S n . Eng., A19 (I 995). and Fracture of Ordered I n t e r ~ e ~ a l l i c ~ e n k a t u r a m ~ nS., (1987). A F W A L ~ T ~ - 8 ~ - 4 1 Air 0 3 , Force ~ ~ 1 II0 (eds ~ sW. 0 . Soboyejo, T. S. Srivatsan, and R. 0. Materials Lab, Wright Patterson Air Force Base, , Warrendale, PA., 117. Wang, P. C., and Yang, J.-M. (1997). Mater. Sci. Erg Snvatsan, T. S., DiPasquale, J., and Soboyejo, W. 0. (1996). 101. In D e f o r ~ a ~ i oand n ~ ~ a c t ~ofi rOrdered e I ~ ~ e r ~ e t a i l iIII c,s Wang, P. C., Her, Y. C., and Yang, J.-M. (1998). Mater. Sci. (eds W. 0. Soboyejo, T. S. Srivatsan, and H. L. Fraser). TMS, Warrendale, PA., 483. Wesseis, J . F., Marquardt B. J., and Krueger, D. D. (1989). Stoloff, N. S. (3992). In U r d e r e ~I n t e r ~ e ~ a ~ l-i cPhysical ~s Presented at TMS- AIM^ Svmp.on Creep and Fracture of M~rallurg9and Mec~anical~ e ~ a v i o(eds r C. T. Liu, R. W. ~itaniA u ~l ~ ~ i n ~TMS, ~ e s IndianapoSis, . IN. Cahn, and G. Sauthoff). Kluwer, Dordrecht, Netherlands, Worth. I3. D., Larsen, J. M., Ba 257. (1 997). Metoll. Mciter. Tram Stoloff, N. S. (1993). In Critical Issues i n the D e v e l o ~ i ~ e nof t Wright, P. K., Maurer, E;. High Ternper~ture~ ~ t ~ r i (eds a l s N. S. Stoloff, D. J. Unpubl~s~ed research, General Electric Aircraft Engines, Duquette, and A. F. Giamei), TMS, Warrendale, PA, 367 Ciiicmnati, OH, see also D. 23. Miracle and R. Darolia Stoloit; N. S. (1996a). In P ~ 9 ~ s~i e~tuu ~f l u r and ~ y Processing rme tallic ~ o ~ p o u n d sPr i n c ~ ~ l e and s uf I ~ t e r ~ e t a ~Cl ioc m ~ o ~(eds ~ i ~N.s S. StolofY, and V K. (eds J. H. ~ ~ s t b r o o kand , R. L. Sikka). Chapman and I-Tall, New York, 126. Chichester, England, 53. Stoloff, M.S. (1996b). In Structural Interrnetallics 1997, Proc. Umakoshi, Y, (1990). Prog, Marer. Sci., Second Intl. Sytnp. on S t r u c ~ u Ir ~~ t~e~r ~ e t a l l(eds i ~ s M. V. Nathal et al.) Seven Springs, PA, TMS, ~ a r r e n d a ~PA, e, no, T., Nakazawa, J., and Urnakoshi, Y 33. (1997). J / s f ~r~ternatiunal, 37(12), 1210. Stoloff, N. S. (1997). ISIJ Irzternat~o~al, 37(12), 1197. Yasuda, H. Y., Behgozin, A., and Umakoshi, Y (1999). ., Fuchs, G. E., Kuruvilla, A. K., and Choe, S . J. ( 1987). In High ~ e ~ ~ e r a t Ordered ure I ~ t e r ~ e t a l lAlloys ic Ye, F., Lederich, R. J., and Soboyejo. W. 0. (1996). In N. S. Stoloff, C. T. Lm, and C. C. Koch). MRS ~ e f o r r n ~ i ~ arzd ~ o n Fracture oj Ordered In~errnetal~~c 1, 247. ~ a t e r ~ III a l (eds ~ W. 0. Soboyejo, T. S. Srivat~an,and Sto S., and Alven, D. A, (1996). In Proc. Symp. on H. L. Fraser). TMS, Warrendale PA., 457. COrros ion- Def ~ r ~ioii~ aIntf erac t ions. Mice, France Sept . Ye, F., Farkas, D., and Soboyejo, W. 0. (1999). M~rter.Sci. 24-26, Abstract TV OR 23. Stoloff, N, S., and Liu, C. T. (1994). Inter~etallics, Zhang, G. P.,Wang, Z. G ., and Wu, S . W. (1997). Takdsugi, T. (1993). In Cvifical Issues in the D ~ ~ v e l o ~qf~ e n t ~ e t u lMater. ~. Trans., J. (1998). Metall. Mater. High Te~iperati4r~ ~ a t e r i ~ ~ 1(eds . s N. S. StoloK D. J. Duquette, and A. F Giainei). TMS, ~arrendale,PA, 399
.Int Although priinary mechanisms of solid solution hardening are known in dilute metal alloys, the mechanisms in inter~etallicc o ~ p o u n d sare only beginning to be established. The availability of such knowledge for high-temperature intermetallics is intended to aid the logical development of useful ordered structural alloys. In this chapter some basic results in solid solution hardening of intermetallic compounds are sunimarized. It is concluded that, within our limited current ~no~vledge, hardeniiig is consistent with that caused by standard types of elastic interactions between point defects and moving dislocations. owever, major gaps in our knowledge exist, and they obstruct adequate quantitative tests. Major sotirces of uncertainty that are discussed are (1) identificat~onof the atomic nature of the defects in individtial cases, (2) deciding what temperature for testing is best to maximize and clarify the effects that are sought, and (3) how to deal with superimposed effects of more than a single type of defect. A step toward uiidei-sta~idingsolution hardening in inte~~etallics is to test whether the same elastic interactions between solute atoms and dislocations dominate as have been identified in dilute alloys of pure metals, but with the effective solute concentration
increased in intermetallics by the substitutions or vacancies that are created by deviations from stoichionietry. Such evaluations have now been done in B2(cP2) structures for the alloy and Ti in AlCo (Fleischer, 199 and Ti in AlRu (Fleischer, C1S(cF24) Laves structures for (Fleischer, 1992) - the last using data by Livingston et al. (1989).
re
Note that solutes can affect a variety of mechanical properties by various secondary alterations that they make in the crystal structure. They can precipitate new phases; they affect phase ecluilibria, solubility limits, and phase distrib~tionsin alloys with more than one phase, profoundly altering strength and ductility; they can pair with other solutes, and efyectively remove them from solution; or solute can pair to form dipole defects that have basic interactions with dislocations. Only the latter (basic interactions of solute with dislocations) are the subject of the main text of this chapter. For example Huang and ~ h e s t n u t t(1995) detail the effects of solutes 111 TiAl - fracture toughness, phases present and their distribution, creep, yield stress, and plastic strain to fracture. These are all vital quantities to consider in designing a useful structural material, but the complexity of the structures and the "Derived from the author's article 'Issues in understa~di~g diverse roles of alloying elements and of stoichiometry solid solution hardening of inter~etallics,' in ~ t r u c t u r f f l confuse any attempt at a scientific test of the I ~ t e i edited ~ ~ by ~ R. t ~Darolia, ~ ~ J. ~ J.~ Lewandowski, ~ ~ iiiechanical role of solutes. Similarly the work of C. T. Liu, P. L. Martin, D. B. Miracle, and M . V. Nathal, The Darolia et al. (1992) shows striking effects of Fe, Ga, PA Minerals, Metals & Materials Society, Warreiisb~~rg~ 1993, pp. 691-698, by per~issionof TMS. and MO on the yield strength and ductility of NiAl. ~ n ~ e ~C ~o ~ pt ~a Vol. ul ~~3 ~, ~Principles s ~~ and Practice. Edited by J. €3. Westbrook and R. L. Fleischer. @2002 John Wiley & Sons, Ltd.
352
~ e c h ~ n ~ Properties ca?
However, the non-i~onotonic nature of the effects makes clear that some secondary effect is present - a likely possibility being that the substitutional solutes are acting to remove the (s~aller)amounts of an interstitial solute, or solutes. Thus initial additions reduce strength and increase ductility. Unfortunately little is known in inte~etallicsabout the mechanical effects of interstitials and, again, from these (important) experiments it is diEcult to derive any fundamental information on solute-dislocation interactions. The best hint of a significant role on interstitials comes from the hardening of NiAl by carbon (Cotton et aE., 1993) at a rate (increase in yield strength per unit solute concentration) 3G, where G i s the shear modulus. Such rates signify dipoles (Fleischer, 1962), and were seen also for (Cotton et al., 1993) and for ishirna et aE., 1986c; Cotton et aE., - suggesting that they too produce dipoles. ver, C, Hf and Zr also have major effects on stic moduli (Lin and Zhang, 1993), so that the conclusions just reached must be regarded as tentative. Other types of interactions - again somewhat indiraffect solid solution hardening were iduk (1991) who suggests that solutes in Ni3A1iinportaiitly affect the stacking fault energy of sup~r-disiocations, thereby altering the difficulty of cross-slip, and hence affecting strength, work hardening, and ductility. This variation (for an ordered structure) of the Suzuki effect (Suzuki, 1952) in metal systems IS of a more fundamental nature. However, as noted elsewhere, since in Ni,Al conventional elastic interactions appear able to explain the hardening ( ~ i s h i m aet al., 1986a), it unlikely that the cross-slip effect is of primary importance. To further clarify what is not attempted here, hight ~ ~ p e r ~ effects ~ u r eOS solutes are often intermingled with the ‘imomalous? increase in strength seen in several intermetallics. Until that eEect is adequately unders~ood, it will be difficult to separate superimposed effects. The simplest ideas of solute ~ardeningare most nearly valid at low temperae - as long as brittle failure does not intervene. nce the discussion here will mostly address observations at relatively low temperatures.
ve~~i
S
ardening of ~ r y ~ t aby l ssolute atoms can be caused by a variety of niechanis~s,For randoml~located solute atoms there are two classes of solute, those that produce large hardeiiing per atom (‘rapid hardeners’)
and those that produce lesser (‘gradual’) hardening (Fleischer, 1962). Rapid hardening i s usually caused by atoms - either interstitial or substitutional - that produce large, asymmetrical strains, and therefore have large shear strains and strong interactions with screw as well as edge dislocations. ~ubstitutional solution hardening, such as is of interest here, 1s generally of the gradual hardening type, most often ascribed to one or both of two mechanisms. In the first of these, atoim that alter the lattice parameter b have effectively a different size from the matrix atonis (Mott and Nabarro, 1940) and will produce hardening that increases with the size misfit ~b(=
d(ln b)/dc),
(1)
where c is the concentration of solute. In the second mechaiiis~?atoms that alter the shear modulus G can be regarded as tiny inclusions of different moduli (Ciussard, 1950), which should produce hardening that increases with cG(= d(1n G)/dc).
Hardening in Cu and Ag alloys correlates better with EG than with E,,. and thus was concluded to be the more important factor for those metals (Fleischer, 1961). However, for quan~itativecompleten~ssit is necessary to consider both effects and appropriately sum them to calculate the hardening and to determine the relative effects of size difference and modulus difference (Fleischer, 1963a). In different s y s t e ~ sa variety of summing rules have been used, as is summarized elsewhere (Fleischer, 1987). Ty~icallyin dilute alloys o f metals, ~ardeningis by an amount Gc3’2c14x,where E is the appropriate, weighted sum of cl, and eG and a 700 (Fleischer, 1964).The value of a varies inversely with the volume per unit cell and with the slip vector; it therefore is likely to be smaller for an ordered compound than for the same structure disordered. N
The view followed here was originated by Westbrook (19651, but was not tested quantitatively: ~ingie~phase alloys that are off stoichiometry have inherelit (con~o~~~ defects - ei~hervacancies or A or or A sites - that lead to hardeiiin~on both sides of stoichiometry at low temperature; at high teniperature the same defects increase the diffusion rates and soften the material, so that it then is strongest at the stoichiometric composition (Westbrook, 1965). Direct measureinents of diffusion rates also support this idea
strongly; see the data tabulated by Hagel (1967). Thus one unique aspect of soficl solution hardening of intermetallic compounds is the inherent presence of superimposed effects of ordinary solute additions and those of constitutional solute. In order to determine whether deviations from stoichiometry are ~ c c o m ~ o d a t eby d vacancies or by substitutioiis of one type of atom on the other sublattice, knowledge of the density and lattice parameter as a function of stoichiometry is pertinent, as Figure 1 illustrates (Hagel and Westbrook, 1962). Vacancies will lower the density, while the effect of substitutions is to approximate a rule of mixtures. The data in Figure 1 make it clear that for cases where adequate ofT-stoichioinetry can exist, the alternative defects considered can be readily identified. For example, 2% to 3% on either side of stoichioinetry would have beeii adequate to determine the slopes on either side of stoichiometry from the data in Figure 1.
Although a number of measurements of hardness vs. composition implied that constitutional defects do produce hardening at temperatures that are less than half of the melting temperature, yield-stress results such as those of Wood and Westbrook (1962) are more
definitive and were the best early indicator of effects. Their data, plotted in Figure 2, follow a squai-e-rootof-the-concentration relation, which i s the most frequently observed behavior in solution hardening (Fleischer, 1963b). This behavior i s regarded as strong evidence that the expected constitut~~nal defects do produce conventional solution hardening.
The literature on solution hardening of ixiter~etallicsis limited, so that few studies exist in which data were adequate to test the existing solution hardening models. Prior literature is listed by Fleischer (1993a,c). The striking, definitive exception is a pair of studies on Mi,Al and Ni by Misliima and coworkers (1986a,b). They showed in the Ll,(c structure Ni,Al that hardening followed a function of c,, and cG for the various solutes just a s would be expected if superimposed size and modulus interactions controlled the solution hardening. Even more important is tliat (as shown in Figure 3) solutes in Ni (inany of them the same solutes) caused hardening that followed the same function almost q~antitatively.Not only do these results imply that conventional elastic interactions control the solution hardeniiig in
\ \
Pigure 1 Lattice parameter and density measurements allow constitutlonal defects to be identified. From Hagel and Wcstbrook (1961)
354
ure 2 Flow stress of AgMg vs. square root of the expected concentration of substitL~tiona1defects (Ag on Mg sites or Mg a n Ag sites). Replotted data of Wood aiid Westbrook (1962). The liiicanty IS consistent with the expectation that point-defect hardening i s occurring
but the fact that the results are nearly identical to those i implies that there are no extra major sources of solid solution hardening that are due to the ordered structure of Ni,Al at room temperature. Thus, for the LI2 structure the initial qL~estioii appears to be answered (i.e. conventional hardening occurs). We therefore turn to other structures - B2(cP2) and C 15(cF24).
The work to be summarized addressed three issues Are the primary effects similar in nature and magnitude to those identified previously? What point defects are formed‘? What eflects are seen from cases where there are simultaneously two or more types of defects? uperimposed defects occur in the general case where there are ternary solute atoms and constitutional defects ~~acaiicies or substituted major-constituent atoms on a ‘wrong’ sublattice), Both vacancies and wrong sublattice subs~itutions can act as ordinary solute in altering elastic moduli and lattice parameters and producing elastic iiiteractions with dislocations.
Measureinents of density p and lattice paraineter b determine the unit cell occupancy number N,. In AlCo aiid AlRu, vacancies result from excess A1 in binary alloys - one vacancy per excess A1 atom in AlCo, fewer than one in Al leisclier, 1993~).Using the formula N , = p Nob3/ where c, is the concentration of atoms of er mole M,, and No is Avogadro’s number, results such as those in Figure 4 are derived. The figure shows that for AlCo, excess CO is mostly in A1 sites, but some vacancies are also produced. Substitutions of ternary elements occasionally produce vacancies also, so that in such cases three or four types of defects inay contribute to hardening. The ductilizing element boron (somewhat surprisin i s inferred to reside in substitutional sites in A. (Fleischer, 1993e), where its addition lowers the shear modulus, does not produce precipitates, yet does not alter the lattice parameter. i n the C15 (cF24) structure the lattice parameter data in Figure 5 show that in Cr,Zr each compoiienl can substitute for the other. Hf can replace Zr, and MO replaces Cr (Fleischer, 1992). Here the defect structure is simpler than in AlCo, but relevant mechanical testing is more difficult.
The nature of hardening can be tested ~~alitatively by whether the magnitude is typical of what a given mechanism is known to cause in other systems and quantitatively if the strengths of the interactions with specific atomic defects are known. Interpretation will naturally be simpler in cases where only a single type of defect is present - whether constitutional or thirdelement substitutions. Hardening in binary AlCo alloys is consistent in magnitude and relative effects wlth conventional size and modulus interactions with vacancies and CO substitutions, and the same i s true for Al-rich 8 2 alloys. The effects of ternary solutes are of the expected average magnitude but they do not correlate well with the observed c,,’s and gG5s in AlCo. The presence of multiple types of point defects that need better characterization is a suggested explanation (Fleischer 1993b) . Figure 6 illustrates the situation in binary AlCo single-phase alioys (Fleischer, l993b). At the ‘low’ temperature of 750°C - 0.53 of the melting temperature - (Figure 6, top) the defects that were documented in Figure 4 produce hardening on either side of stoichiometry. ‘Low’ is as close to as possible,
--1
I I
I I
P
I
I
8
F i 3 ~ a r d~e n i n~gper unit ~ solute concentration as a function of combined interactions from size effects "a( = cb) and elastic ~ o d u l u eG. s The data on the left arc for Ni,Al; those on the right are for Ni (Mishima er al., 1986a.b). Solid dots are given for transition-metal elements
0.050
0.025
C (AI-ri~h)
0
0.025
0.050
0.075
0.1 00
0.1 25
C (Go-rich)
Cell occupancy for binary AlCo B2 alloys calculated froin their density and lattice paramcter. Each excess AI sltoin produces a vacant lattice site on the CO sublattice. Excess CO substitutes for A1 but produces increasing but lesser vacancy coiic~ntra~ions (Fleischer, 1993a)
Mec.hclnical Properties
356
In contrast, Cr2Zr (another C15 c o ~ p o u n d )could only be tested in compression to 950 "C (0.63Tm)due to equipment li~itations,so that except for one sainple with Zr in Cr sites, samples were brittle (Fleischer, 1992). Nevertheless, within the scatter, and consistent with microhardness tests at 950"C, there was hardening from solute and constitutional defects that was consistent with the hardening relation noted earlier for metal alloys, but with a somewhat lower value of a which is expected for intermetallics, since unit cells and slip vectors are larger, Although the results described (I) show solution hardening in intermetallics and in the simpler cases (2) fit the models that have been used to describe metal alloys, full understanding of effects awaits solution of several problems to be considered next.
/ 7.23
7.22
7.21
r
7.20
HI for Zr
inary Cr-Rich
719'
''"0
I
I
'
3.0
I
I
I
I
6.0
I
9.0
CO~CE~TRA~IO~ (ATOMI C %)
Lattice parameter data versus composition as a function of off-stoichiometry and solute additions for Cr,Zr alloys (Fleischer, 1992)
but above, the ductile-to~brittletransition (T&),so that plastic flow can occur. In spite of most of the samples b e ~ ~ v i n~lastically, g tlie two that are highest in CO were brittle, giving therefore only lower limits on the flow stress (as indicated by upward arrows). At the higher tei~peratureof 950"C, which is 0.64 of the melting temperature T,, the constitutional defects are presumably mobile and therefore weaken the alloys, as shown by Figure 6 (bottoi~),The ratios of the liar~eningper (unit concentr~tion)1'2are consistent with the measured q,'s and gG's. The Laves phases are difficult to test in a meaningfu~ way, at least when they have high T,, since plasticity mostly occurs at T- 0.8Tm or above. For the relatively 15(cF24) type-compound Cu2Mg ~ivingstonet al. succeeded in showing effects of four solute species by testing at 0.86T~,~(650 "C). The results were shown to be consistent with summed size and m o ~ u l u s effects (Fleischer, 1992), subject to the inference that valence difference is a surrogate for the dulus effect, as it is in dilute copper alloys eischer, I963a).
There are diverse obstacles to being able to test solid solution hardening effects rigorously in intermetallics. They cluster under topics o f deciding on nie~ningful testing temperatures and procedures, identify in^ the number aiid nature of defects, and modeling how to superimpose the effects of more than one type of point defect.
Figure 6 and the discussion of it noted one of the dilemmas of selecting a temperature where interpretable tests can be run. As high temperature is approached, solution hardening ceases at a temperature at which hardening is replaced by softening if defects aid d i ~ u s i o As ~ . temperature is lowered, most intermetallics become brittle, and as a consequence the more variable fracture stresses of brittle materials may mask the effects sought. It would appear that a reasonable protocol would be to lower the temperature to a small increment AT above Tbdin order to measure stresses for plastic flow - which are what most existing theories predict. A further complicatio~that is evident in Figure 6 (top) is that Tbdmay vary with composition. Does one pick a single t ~ m p e r a t ~ ~as r ewas , done in Figure 6, or should each composition be tested as a function of l' and the stress at a fixed Tb,+AT used? Another related questioii i s how one infers what the flow stress would be at absolute zero t e ~ p e r a t u r e ~ where most theories apply most directly. If TbdIS low, a meaiiingful extrapolation may be reasonably straightforward. However, as noted, T,, for some
357
0.
0.1
0
15 1
0.5
0.3
0.4
0.
OLUTE C O ~ ~ E ~ T ~ A T I O ~ ) ” ~
H~ O ~ ~ T (%)R Y
QFF-
2.5
0.2
0.1
0.5 1 1.5
5
9
11
A I-CO 950” c
tU)
CO-W ICH H
0.2
0.1
0, I
0.2
0.3
0.4
0.
ure 6 Flow stress in compression versus ofr-stoichioinetry in AlCo B2 iiitermetallics at 750°C (top) and 950°C (bottom) (Fleischer, 1993b)
358
~ e c ~ ~Properties ~ i c ~ l
inter~eta~lics may be at a high homo~ogoustemperature with a small temperature interval between Tbdand TD,where TDis the temperature above which increased diffusion causes point defects to weaken alloys. A different, but useful workin assumption would be that the stress at Tbd+ AT is nearly the low-temperature value desired, except that the elastic modulus would need to be adjusted to its value at T=0.
efect identification in i~termet~llics is a major problem. It will become clear that the general case where both solutes and con~titutional defects are present is experimentally unresolvable at this time. omewhat less complex are the cases of binary intermetallics when only constitu present. Figure 7 shows some po in a nearly equiatomic intemetallic. Excess atoms of each type ~ ~ g substitute h t in the other sublattice, reside interstitially, or be compensated by vacancies in r sublattice - six distinct possibilities. , Al~-formulaaluminides only two types of constitutional defects are known - X atoms in A1 sites, and X-lattice vacancies for excess AI (Chang and n, 1982). In other B2 alloys such as AgMg, ions occur on both sides of stoichiometry. In
stress dipoles of many varieties, one of which is sketched. Stress dipoles, examples of which appear in Figures 7 and 8, are known to produce major hardening in metals and ceramics (Fleischer, 1962), but cases have not beeii clearly documented in intermetallics. The existence of vacancy pairs has been described in FeAl (Wurschum et al., 19951, and the differences in hardening brought about by thermal treatments (Nsgpal and Baker, 1990) may be from that source. One conclusion is clear. Only for the simplest situations are we likely in the iiear future to be able to infer what defect is present and what its elastic strength is. ~easurementssuch as are given in F i ~ u r 4e will continue to help in this process. In some cases
0
0
0
0 0 0
0
0
'0
0
0
0
0 L ATOM IN S SITE
S ATOM IN L SITE
0
0
0
0 0 0 0 0
0 both substitutions and vacancies, some of which may are retained by rapid Efect was shown by in their Figure 17. wever, Figure 4 is discrepant with their interpretation of compiled results in that the vacancy concentration increases with. deviation from stoichiometry rather than decreases. Ternary substitutions in binary intermetallics have been made, with complicated cell occupancy numbers r e s ~ l ~ i nin g ,both AlCo and AlRu (Fleischer, 1993a,c). In many cases vacancies are produced simultaneously with the ternary substitutions. In such cases, it is not known which sublattice the vacancies occupy or whether they are in both. A few defect configurations that can be imagined appear in Figure 8. The foreign (F) ternary atom may substitute in either sublattice (or both), may lie in one of several interstitial positions, may lead to vacancies osite or the same sublattice - at least nine s. Finally, defects may interact to form
0 0 0 L ATOM I~TERSTITIAL
0
S ATOM INl-E~STITl AL
0
0
0
0
0
0
0
0
0 0 0
0
0
0 0
0 0
0
0
0
0
0 0
EXCESS OF s GIVFS I- VACANCY
0
0
EXCESS OF L GIVFS S VACANCY
F i ~ 7~ Possible r ~ defects in a binary AB intermetallic. S and L are the smaller and larger atoms
Solution and Defect Hardenirtg
0 0
0 0 0
0
0
0
0
0
0 0 @
0
0
c s o o
L = LARGE ATOMS S=S M A ~ ~ A T O ~ ~
F = FOREIGN ATOMS
0
0
0 0
0 0 o@o
359
0
F ATOM
F ATOM
REQUIRES VACANCY IN SAME SUB LATTICE
REQUIRES VACANCY IN OTHER SUB LATTICE
0
Q
0
U
0 0 0 F ATOM SUBSTITUTES FOR LATOM
FATOM IS INTERSTITIAL
0
0
SOFT F A T O ~ ~ ATTRACT
Some possible defects lrom ternary solute In an AB lntermetallic
defects can be further characterized by electron channeling measurements in electron microscopy ( A ~ C ~ Eliterally ~ I , Atom Location by CHanneling Enhanced Mfcroanalyses), but this inethod i s insensitive to vacancies.
rich compositions of AlCo iiiay be one example of where two types of point defects are present; dipole formation - if incomplete - is another; solute that substitutes on inore than one sublattice is a third.
It is evidently likely that in any real case both constitutional defects aiid ternary (or higher-order) solute are present - perhaps niultiple types of defects. How do their effects combine'? Gypen and Deruyttere (1977) and Pike et d.(1997) have made a beginning of a theory, but no organized, definitive experimental test has been reported. The early predecessor theory is given by Koppenaal and Kuhlman-Wilsdorf (1964). We don't need greater complexity, but we must note a further complication: In measuring effects on lattice parameter and elastic modulus, it is difficult to infer values for individual defects if more than oiie species is present. Where the individual defects can be separately produced in different samples, their effects on b and on 6; can be measured; but where they are structurally associated, the separation is not accessible. The Co-
Existing experiments in relatively simple cases of ORstoichiometry aiid of teriiary substitutions in binary alloys are consistent with the primary strengthening coming from the familiar elastic forces between point defects and moving dislocations due to size and modulus interactions. These statements apply to data on Ni,Al (LlJ, Cu,Mg (C15), Cr,Zr (&115),AlCo ( and AlRu (B2). The conclusion should not be taken as applyiiig necessarily to properties and testing ranges that have not been examined. Other properties of order or of dislocation dissociation into partials may play a role, but they have not yet been demonstrated. In many cases multiple types of point defects are present, and adequate definition of their separate abundances and effects is lacking. complications aiid uncertainties that need to be resolved have been identified.
Mechanical Proper ties
360
The author i s pleased to acknowledge the General esearch and Development Center for its hospitality during the original writing of this inanuscript.
Gypen, L. A., and Deruyttere, A. (1977). .J. Marer. Sci., 1 1028-1 033. Hagel, W. C. (1967). ~ n ~ e r ~ e t Ca ~o ~~i c~ o ~Chapter n ~ ~ s20, , 377-404 (ed. J. H. Westbrook). Wiley, New York. Westbrook, J. H . (1961). Trans. AIM& Chesnutt, J. C. (1995). Inter~etallic C a ~ p # -~Principles ~ ~ s and Practice (eds J. H. Westbrook, and R. L. FIeischer). Wiley, Chichester, UK,
efereiices -WilsdorC D. (1964). Appl. Chang, Y. A., and Neumann, J. P (1982). Pro$. Solid St,
. D.%and Kaufman, M. J. (1993). In Struct~ral ~ f a t e r ~ (eds ~ ~ l ~R.s Darolia, et al.). TMS, Warrenclale, PA, p. 513. Darolia, R., Lahrman, D. F., and Field, R. D. (1992). Scripfa Diniiduk, D. (1991). ,J. dc. Physique, 3 (no.l), 1025. Crussard, C. (1950). M2taux et Corrosion, 25, 205-226. Fleischer, R. L. (1961). idcta Met., 9, 996-1000. Fleischer, R. L. (1962). Acta Met., 10, 835-842. Flcischcr, R. L. (1963a). Acts Met., 11, 203-209. Fleischer, R. L., and Hibbard, W. R. (1963b). NPL Con$ On ~ e l a t i ~Bemeen n Structure and Properties of Metals, 0, London, p. 261-297. Fleischer, R, L. (1964). Chapter TV in The ~ t r ~ ~ gof ~ ~ e ~ f f D. ~ sPeckner, , Ed., R e ~ n h o l93-140. ~, Fleischer, R. L. (1987). Scripfa Met., 21, 1083-1085. Fleischer, R. L (1992). ScriJltu Metall. Mater., 29, 799-804. Fleischer, R. L (1993a). J. Matericds Res., 8, 49-58. Fleischer, R. L (1993b). 3. ~ ~ t e r zRes., a ~ .8,~ 59-67. Fleischer, R. L Fleischer, R. L Fleischer, R. L
In S t r ~ c ~~a ~t ~r r~i (eds a~ ~ s R. Darolia, et d.).TMS, Warrendale, PA, p.445. Livingstoii, J. D., Hall. E. L., and Koch, E. F. (1989). Mater. Res, Soc. Syrnp. Proc., 133, 243. Mishima, Y., Ochiai, S . , Hamed, N., Yodogawa, M., and Suzuki, T. (1986a). T ~ a n sJapan . Ins Mishima, Y., Ochiai, S., Hamed, N., Suzuki, T. (1986b). Trans. ,Japan Ins Mishima, Y . , Ochiai, S., Yodogawa, (1986~).Traits. Japan Inst. ~ @ ~ a27, I s 41. , Mott, N.F., and Nabarro, F. R. N.(1940). Proc. Pkys. Soc., 2, 86-89. Nagpal, P., and Baker, I. (1990). Metall. Trans., A., and LIU, 6. T. (1997). ~ e ~ 2 j ~ ~ orts Res. Insd. ~ o h o Ubziv., ~ ~ ~ i 455. Westbrook, J. EI. (1965). High Strength Materids, 29, 724760. Wood, D. L., and Westbrook, J. €3. (1962). Trans. A M E , 224, 1024-1037. p, C., and Schaefer, H.-E. (1995). Phys.
The theoretical and experimental basis for the phenomenon in metals and alloys of increasing flow stress with plastic straining~ termed strain hardening, has had a glorious history (Basinski and 1974; Hirsch, 1964; Hon966, 1975, 1982; Kocks et lsdorf, 1962, 1966, 1985; Mecking, 1977, 1981; Mecking and Estrin, 1987; Mecking and Kocks, 1981; Mughrabi, 1975; Nabarro, et al., 1964; ~ o ~ l and e t ~Kocks, 1994; Seeger et ctl., 1957; Sevillano, 1993; Taylor, 1938). A great deal of experimental and theoretical insight into the nature of strain l i a ~ d e n i nof ~ crystal~ine materials has been gained over the past 50 years. Unforttznately, a concise, quantitative theory capable of describing all the aspects of work hardening in single and polycrystall~ne materials, whether disordered or intermetallics, for all crystal symmetries remains elusive. This review is aimed at the phenomenon of strain hardening in ordered intermetallic c o ~ p o u n d s .It is not intended to be an overview of work hardening in single- and polycrystalline materials in general, therefore the reader is encouraged to refer to the articles and reviews cited above. The strain-hardening behavisr of i n t e ~ e t a l l i cis~ quite complex, given the broad array of plastic
deformation processes possible over the range o f crystal structures encountered in these systems. details on dislocations in specific intermetallic syst the reader inay wish to refer to a number of recent reviews (Appel and Wa 1993; Banerjee, 1995; Miracle, 1993; Nabarro al., 1993; Pope and Ezz, 1984; Sauthoff, 1989; Sikka et al., 1993; Suzuki et al., 1989; ~ e y s s i ~ and re 1995; Yamaguchi and Umakoshi, 1990; Yoo et al., 1988, 1990). As will be discussed in more detail in following sections, the und~r§~anding of h~rdening processes in inte~ietallicsis rapidly evolving due to numerous recent deformation, electron niicroscopy and dislocation-m~delingstudi~s.~ o n t i n u ad~ances e~ in the u n ~ e r s t a n d ~ of i ~ gdisloc~tionsin the hardening processes is very important, since intermetallics tend to fail in a brittle manner in tension only after a limited amount of plastic deformation accu~ulates.Also for this reason, strain liardening in ordered intermetall~cs must generally be studied under compressive §training conditions. To assess the unique characteristics of interme tallics, it is useful to contrast their strain-hardening behavior with that of disordered inetals and alloys. Accordin a brief summary of strain hardening in sing1e-crystal and golycrystalline disordered materials is first presented. Thereafter the unique features of the strainhardening behavior of ordered intermetal~ics are
I ~ ~ ~ ~ r n eCompoundc t u ~ l ~ c Vol. 3 , Principles arzd Practice. Edited by J. €3. Westbrook and R. L. Fleischer. @ZOO2 John Wiley & Sons, Ltd.
reviewed from the point of view of macroscopic phenomenology and microscopic dislocation processes.
atevials
hear-stress/sl~ear-straincurves (the shear stress and strain resolved on the most active slip system) for tive disordered metal single crystals, includCu, and Fe, are shown in Figure 1. All curves entations such that the resolved shear stress will be thc highest 011 a single slip systeni for the duration of most of the deformation imposed. Given this type of loading, the majority of slip in fcc and hcp metals is known to occur on the primary slip system. his occurs although the applied stress, due to strain s u ~ c i e n that t the critical resolved is exceeded on a secoiidary, less ip system. This process, known as latent hardening, raises the resolved flow stress for shear on all slip systems of a given type when the flow any system is increased by strain hardtaneous slip on several systems will occur if the crystal is oriented such that the resolved shear stresses 011 two or more systems are equal. Wnlike fcc and hcp-type metals, bcc metals typically exhibit slip on two or more planes which share the direction of niaximum stress rather than on a single plane from the start of deformation (called 'pencil glide'). The shear-stress/sliear~stsaiii response of singlecrystal hcp Mg, Figure I, displays a constant low hardening rate that is sustained up to fracture. Note that the hardeniiig rate, 8, is defined as the slope of the resolved shear s t ~ e s s - s t r acurve, ~~ 8 = dz/dy. The single-crystal Cu, which is fcc, exhibits three stages of ilsdorf, 1966; Rollett and age I , termed "easy glide', is characterized by a low linrdening rate, of similar magnitude to that of the Mg. Stage 2, which exhibits a high and a ~ p ~ o ~ ~ i i iconstant ~ ~ t e l yhardening rate, is termed the 'linear hardening' stage. Finally, Stage 3 is the region of the s~ngle-crystalresponse in which the hardening decreases, due to stress-assisted dynamic recovery. itional stages of hardening may also be present at and Haasen, 1993; llett and Kocks, 1994); these are not discussed here, as they are not as relevant to mtermetallic compounds. Finally, in contrast to fcc and hcp materials, the bcc Fe crystal in Figure l exhibits a strain-hardening rate which continuously decreases as deformation proceeds, as is typical for many bcc materials.
The three stages of hardening in fcc crystals, shown in Figure 1 for CL^, also correspond to stages of dislocation substructure developineiit (Nabarro et al., 1964; Rollett and Kocks, 1994). These are worth briefly reviewing, as substructure development in intermetallics is typically very different. During Stage 1 the dislocation density increases, but with ~ i n i ~ a l dislocation entanglement. The strain-hardening rate is low and has been related to the elastic interactions of edge dislocation dipoles and multipoles. Accordingly, the rate of hardening in Stage 1 is insensitive to temperature due to few dislocation intersections and minimal need for thermally assisted processes for continued glide. During Stage 2 deformation there is continued dislocation n~Liltiplicationand storage and activation of secondary slip systems. H~rdeningarises from longrange elastic interactions as well as from ShQrter-ran~e entanglements that begin to develop due to dislocation intersections. Mobile dislocations that glide into sessile entanglements cluster in tlieir vicinity. The degree of clustering increases with increasing dislocation density (and therefore with increasing stress). The tangled clusters eventually evolve to more closed cell structures. This stage of deforniation is characterized by a high, nearly constant strain~harden~ngrate which correlates with increasing hardness. As the cells become fully developed, with high densities of dislocations within their walls and much lower densities in their interiors, Stage 3 cominetices.
n
-0
10
50 in, 5%
Figure 1 Shear-slresslshear-strain curves of disordered metallic single crystals of Fe, Cu, and Mg
Hardening continues during Stage 3, but with a hardening rate, due to the onset of dynamic rocesses. There are several possible dynamic recovery ~ e c h a n ~ s m (Nabarro, s 19891, including cross-slip, cancellation of dislocations in cell walls via glide encounters with dislocations of opposite sign, climb by core diffusion and climb induced by vacancies created during plastic deformation. If these local processes result iii the removal of sessile obstacles, there may also be a subsequent glide collapse of dislocations that previously accumulated due to the presence o f the sessile barrier (Prinz and Argon, 1980). Although less often addressed, hardening in bcc metals and alloys is different compared to fcc-type materials. The entire shear-stress/shear-strain curve o f single-crystal Fe resembles the decreasing strain-hardening rate representative o f Stage 3 deformation in fcc crystals. In spite of the non-planar screw dislocation core structures (Christian, 1983; Mitchell et al., 1963), cross-slip is observed from the initial stages of deformation in bcc single crystals, unlike fcc crystals where widely dissociated partial dislocations limit -slip and dislocatio accumulation processes neville et al., 1988; u h l i ~ a n n ~ ~ i ~ s d1966; orf, tt and ICoclss, 1994). To generalize the flow behavior of rnetallics as function of strain rate and te~perature,Figure shows a scheniatic of the characteristic strain-har ening behavior of polycrystalline fcc, bcc, and hcp metals and their alloys. igh-purity fcc metals in an annealed condition, such as Cu, Ni, AI, and Ag, exhibit nearly strain-rate independent yielding behavior, while their post-yield strain hardeniiig is strongly rate dependent, as shown in Figure 2a. This strong rate dependency of strain hardening in fcc metals i s due to the sLippression of dynamic recovery processes with
increasing strain rate and/or decreasing the temperature. Alloying fcc metals or testing bcc or hcp metals above their intrinsic lattice ~ e s i s t a n cor~ ' dominated regime, as schei~aticallysho 26, i s known to produce an increased dependency of the yield strength on rate and temperature while retaining a divergent strain harden in^ ~esponse. strain-hardening response of bcc metals (Fe, Tn, MO, Nb) (Clien and hcp metals and alloys Kocks, 1975) i s represen streiigtlis of bcc and licp metals and alloys display a strong temperature and strain-rat their high lattice resistance (Chen In this class of metals, strain-hard yielding are nearly invariant as a function of strain rate; i.e. the stress-strain curves are all nearly pa~dllel in slope although offset 111 their initial yields. The variatioii in the dependency of the yield and strainhardening responses of metals, whether disordered or in the current case of intermetallics, forms the basis for development of advanced constitutive models. The bulls of the infori~ationon strain hardeniiig in internietallics is available for polycry~tallineforms of material in Compression. U tiliztrltion o f compression testing, in contrast to tensioii, allows in-depth study of the Work-hardeninE response of the various alu~inides without the preinature termination of the stress-strain response due to intersectio with the failure envelope (George and Raker, 1998; irnura et al., 1994; al., 1993; Yainaguchi and Inui, 1993). As will he shown in the next section, the Strain~hnrde~iinE response of polycrystalline intermetallics differs substantially from the general behavior of disordered metals and alloys shown in Figure 2 in two distinct ways. The first is tlie influence of crystallogra~l~ic order on the
in igure 2 Schematic representations of the influence of temperature and strain rate on the yield and strain-hardenmg response oT: pure fcc inetiils such as Ni, Al, Cu, b) fcc alloys such 8 s 304 stainless steel and/or bcc or hcp metals at higher t e ~ p ~ r 8 t u r eand s , c) bcc aiid hcp metals at lower temperature and/or high strain rate
a> anncalcd
364
Mectzanicd Properties
strain-hardening behavior for a fixed composition, which leads to high inherent hardening. The second is the propensity for high strain-hardening rates to be sustained up to large Bow-stress values. While knowledge of the mechanisms is still incomplete, the role of the ordered crystal structures and the dislocation processes associated with these qualitative differences in intermetallic strain-hardening behavior will be reviewed.
3. Strain Hardening in I A number of cli~racteristi~s of ordered compounds, conipared to their disordered counterparts, have the potential to substantially influence flow and hardening behavior, including: urgers vectors of dislocations in interinetallic coinpounds are typically greater in magnitude compared to their disordered counterparts. This may result in a high intrinsic lattice resistance to dislocation glide and/or increase the probability of dislocation dissociation or decomposition, A high degree o f elastic anisotropy may influence the decomposition andlor glide characteristics of these dislocations, The energy of faults such as antiphase boundaries and complex stacking fiiults may vary widely and have a high degree of variation on different slip planes,
e
Multiple types of slip systems may operate during the course of deformation at a fixed temperature, Slip inode transitions rnay occur as a function of temperature, Ordering usually decreases the diffusivity of the individual species within the compound; this in turn rnay influence static and dynamic recovery processes, Variations i n the degree of ordering may exist.
These properties ot intermetallic compounds and their influence on hardening will be briefly highlighted to the degree that they have been considered in investigations of the flow behavior of various ordered materials. Exaiglples of systems where these features strongly influence the hardening behavior will be given in the following sections.
As first pointed out by Cottrell(1953), the existence o f order should influence the deformation, since dislocation glide in the superlattice is likely to occur via pairs of dislocations coupled by an antiphase boundary. While increasing the degree of order in an intermetallic a s initial ~ flow stress, in may either increase or ~ e c ~ e the most systems an ordered alloy will exhibit a iiiuch higher rate of strain hardening, compared to its disordered counterpart (Flinn, 1960; Stoloff and Davies, 1966; Vidoz et al., 1963). An illustration of the effect of order on strain hardening is shown in Figure 3 for the case of CuJAusingle crystals tested in
igure 3 Strcss-strain rcsponse of ordered aiid disordered Cu,A-u single crystals quasi-statically deformed at room temperature and 77K (Jin and Gray 111, 1995)
Strain ~ a ~ d e ~ i n g
both an ordered and disordered state. The strainhardening rate for the disordered Cu,Au single crystal is signi~cantly lower than that displayed for the ordered Cu3Au at both temperatLires during quasistatic deforniation: 50 aluiiit strain for the disordered versus 565 MPa Cu3Au. Ordering t r a n s f o ~ a t materials also result in such enhancements in work hardening (Hosoda et al., 2000). Ordering also substantially reduces the temp~raturedependence of the stress-strain response of Cu,Au. Decreasing the temperature from ambient to 77 K for quasi-static loading IS seen to increase the flow strength of disordered C~i3Auby -40% while the ordered material increases its flow stress by -20% for the same decrease in temperature. ~icrostrLictura1examination of deformed Cu,Au samples in both states of order has shown fine slip and uniforinly distributed dislocations in fully ordered Cu3Auand coarse, planar slip accompanied by inhomogeneously distributed dislocations in the disordered state (Jin and Gray 111, 1995). It has been suggested that these differences in slip distribution and accompanying differences in hard~ningrate may arise due to retention of local order on the slip plane for fully ordered alloys and d ~ s r ~ ~ tof i oany n local order present on the slip plane in the disordered variant of the material. This progressive destruction o f the local order would promote further flow localization and concentration of sLibsequent glide on the same slip plane, resulting in coarse slip (Flinn, 1940), for a disordered alloy. Clearly a coinplete understand~iigof the influence of order on st~ain-hardeningbehavior requires information not only on the macroscopic aspects of flow, but also on the character of dislocations gliding on a given slip system and their ability to cross-slip, climb and interact with one another and other structural features. As with disordered materials, it IS most straightforward to first consider the slip systems and hardening behavior of i n t e ~ e t a l l single ~ c crystals.
-
3. The shear-stress/shear-straincurves for single-crystal ordered intermetallics are difficult to widely assess due to the lack of con~~rehensive studies on large-strain single-crystal deforniation behavior. Nevertheless, the limited sin~le-crystal studies on inter~etalliccompounds of several different crystal structures, including Ni,A1 (L12), poly thetically twinned (PST) TiAl (L1,) and NiAI ) reveal significantly different
365
hardening behaviors, compared to disordered materials, Figure 4. An obvious difference is the absence of Stage 1 hardening in any of the compounds shown in Figure 4, even for orientations where only a single-slip system IS operating (§hi et al., 1999a), Figure 4(b), This absence of Stage 1 hardening is also apparent for FePt 2) single c r y s ~ ~(l s~ ~ ~ ~ets al., u g i 1998). A coi~parisonof the shearstress/sliear-straiii curves in Figure 4, relative to Figure 1, suggests that the response of intei~etallicsingle crystals is similar to that of a bcc metal. While the stress necessary to iniliate dislocation glide is often similarly high in the bcc and intermet~lliccases, due to a relatively high Peierls stress (or perhaps interstitial impurities), the strain-hardening responses thereafter usually bear little in common. Followin in the intermetallic crystals the strain hardening is characterized by nearly linear hardening that can continue until strain localization, cracking and fracture intervene. This essentially linear train-hardening behavior is observed under both quasi-static and high strain-rate loading cond as shown in Figure 4c for (100) and (1 10) NiAl, aloy and Gray-111, 1995; Maloy et al., 1995) d d under i~u~ti-slip conditions. These differences in the early stages of hardening in single crystals of ordered compounds suggest that the intrinsic inobility of dislocations is limited compared to that observed in fcc and hcp c Figure 5 shows the substru~ture stages of deformation on a single ( system at room temperature. The substructure contains a fairly high density of p r i s ~ a t i cloops, with longer segments of dislocations that contain superjogs created as a result of frequent, double cross-slip of screw segments, wbich have a much higher mobility than the edge segments in this system (Caillard et al., 1999; Field et al., 1991; ~essersc1itmidtet al., 1997; Sfii et al., 1999). Interestingly, cross-slip occurs easily, in spite of the elastic-anisotropy-driven stability of the screw orientation under no applied stress (Loretto and Wasilewski, 1971). The prismatic loop debris is a result of the drag of inter~ediate~~ieight jogs on the screw segments. In spite of the fact that these dislocations with conipact core structLires (Ternes et al., 1995) initially glide with ease, their mobility is quickly reduced by their tendency for cross-slip, The general tendency for the mobility of dislocations to be limited by cross-slip processes appears to be a of deformation in many intermetallics. ple of this is shown for TiAl in Figure 6, where a/2[110] screw dislocations are repeatedly p i ~ n e dalong their
00 500
100
length (Ott, 1998). The progressive ii~mobili~ation of screw dislocations by such pinning processes would serve to suppress cross-slip induced dynamic recovery processes. In fact, the high liardening rates present in many internietallic systems have been attributed to a general lack of dynamic recovery processes (Gray 111 and ~ m b u r y , 1993). Another important point for the specific example of NiAl is the greater degree of spreading o f the core structure o f the (100) dislocations with deviations from esserschmidt et a/., 1997). This that changes in dislocatioii mohility and the resultant hardening could be expected from relatively minor shifts away from the stoichio-
n
formation of cube-oriented crystals o f NiAl, an orientation which activates non-( 100) type slip systerns, also provides an interesting perspective on the unique characteristics of hardening in iiiteri~etallics (Kim and Gibala, 1991; ~ r i ~ i v a s aetn al., 1997, 1998). In this orientation, a(111) dislocation motion is associated with flow stresses that are approximately three times higher than in non-cube-oriented crystals. However, as temperatures are increased, there is a stoichiometry-dependent decrease of flow stress that is associated with a transition in the dominant operative slip system. Associated with this transition i s the decomposition of the a ( 111) type dislocations into a( l10>+ a(001) type dislocations. Although the
106
0
0.
0
(a) Shear-stresslshear-stra~ncurves of [001] single crystals of Ni,A1-0.223 at 298 and 8C)OK after Heredia and Pope and binary PST-TiAl (orielitation B) tested in vacuum after Y a i ~ a g u ~ (b) h i ~[223] NiAl oriented for single slip (Shi et al., 1999a); aiid (c) multi-slip (100) aiid (1 10) NiAl (Takasugi et al., 1990; Whang et d,, 1998) stress-strain data for quasistatic and high-strain rate loading
of Weak bean? ~ ~ c r o g ~ aofp hthe su~s~rL~cture NiAl following room-temperature deformation of a 12231 oriented single crystal oriented for single slip. Alf dislocations are 1100) tvne. Microeranh courtesv of X. Shi \
I
.‘A
U
I
Strain ~ u r ~ e n i ~ g
i~~~~ 6 Ordinary a/2[110] dislocatioiis in Ti-4~Al-2Cr-2~b after 0.5% plastic strain at 760°C at a strain rate of 2 x lW4/s
inotion of the a(110) type dislocations is responsible for subsequent straining at higher temperatures, highresolution traiisiiission electron nlicroscopy studies suggest that they are constrained by local climb caused by decomposition of their edge components (Srinivasaii et al., 1998). Because motion of the edge components would require the operation of local di~usionalprocesses, deviations from stoichiornetry that influence the vacancy content of the material would be expected to influence the hardening and recovery characteristics of the MiAl crystal in this orientation. The dorniiiant presence of edge components following deforination is also observed 111 other intermetallic single crystals, such as CoSi, (Ito et al.,
1992), suggesting that harden in^ associated with a high instrinsic resistance to the motion of edge dislocatioiis is not uiicoiixnon in intermetallic systems. Other substructural features unique to ordered systems that may influence strain hardening are anti-phase boundary (APB) tubes (Sun, 1994; Vidoz and which have been considered in detail for L1:, type systems (Hazzeldiiie aiid Hirsch, 1987; Schoeck, 1969; Shi et al., 1996) and faulted dipoles that contain superlattice intrinsic or extrinsic stacklng Faults (Viguier and Heinker, 1996). Compreheiisive studies documenting substructure evolution along with quantitative nieasuremeiits of changes in dislocation density (or twin density) with
365
~ ~ c h a ~ iProperties cal
stress, are essent~alinput to work-hardening models, but are presently lacking. Nevertheless, studies conducted on iiitermetallic single crystals to date clearly suggest different directions for the development of hardening theories, coinpared to disordered nnaterials. Given that intrinsic limitations on the motioii of either components correlate with high rates of is clear that mechanisms of immobilization and the associated storage rate of dislocations must be accounted for in a comprehensive theory. A e lies in developing better capabilities for predicting dislocation core structures and mobilities and integration of these key aspects of deformation into multi-scale models that quantitatively predict the onset of yielding and the rate of hardening for any given system in its single-crystal form. Finally, there are a number of additional considerations for hardening in polycrystalline intermetallics; these are addressed in the remaining sect~ons.
ver the past 10 years a greater amount of inforniatioii on the hardening response of polycrystalline interme tallics has accumulated, particularly for aluminidetype intermetallics, due to their potential for structural app~ications~ Y a ~ a g u and c ~ iUmakoshi, 1990). Variations in te~nperatLir~~ composition, microstructure, environmental effects. stress-state effects, and strain rate have been broadly utilized to probe the deformation behavior, including the rate dependency of yield and work hardening. Qf particular interest has been the contribution of work-hardening phenomena to the aiiornalous increase in the flow stress with temperature, present in a wide variety of intermetallic compounds. A survey o f the literature (Gray 111, 1996) on tem~erature, strain rate and work-harden~~g effects on the flow of Fe, Ni- and Ti-alurninides yields a significant number of experimental studies: (Albert and
uk, 1993; Kitano and , 1994; Lee et al., 1995; Gray III, 1995, 1996; Maloy et uE., 1995; Mendiratta et al., 1993; et d., 1992; Palm et al., 1997; Shih et al., 1993; and Gray 111, 1992; Sizek and Gray-
3611, 1993; Viguier et al., 1995a; Yoshimi and Hanada, 1993; Yoshiini et ul., 1995). Models for the deformation response of some of these aluiainides have also been developed, with particular emphasis on the anomalous rise of flow stress with temperature (Busso and M c ~ l i i i t o c ~1996; , Ezz and Hirsch, 1995; George and Baker, 1998; and Sun, 1993; Viguier et A comparison of the work-hardening response of Ti-, Ni-, and Fe-alu~iinidesusing com~ressiondata for five intermetallic aluminides (Gray 111, 19961, is shown in Figures 4 and 7. In-depth details of the composition, processing, heat-treatments, and final microstructures have been presented previously for Gray 111, 1995; Maloy et al., 1995), Ni,A1 (Sizek and Gray 111, 1992, 1993), Ti,AI (Albert and Gray 1997, Gray I11 et al., 1993, 1996), TjAl (Gray TIE, 1991; Gray I11 and Enibury, 1993; Jin et al., Maloy and Gray 111, 1995, 19961, Fe,Al ( 19931, and FeAl (Gray 111, 1995). The coinpressive true-stress true-strain beliavior of NI&, NiAl, Fe&, Fe-40,414). 1€3, Ti-24A1-1 ZNb, aad Ti-48A1-2Cr-2Nb all depend on both the applied strain rate, which ranged from 0.001 to 7500 s-', and the test temperature, which was varied between 77 and 1273I< (Gray H I 9 1996). Examination of the individual stressstrain responses of the aluminides as a function of strain rate and teinperature reveals some striking similarities and some distinct di~erences.These can be seen in Figure 7a and 7b for quasi-static and dynamic loading, respectively. The quasi-sta tic yield stresses of ail the polycrystalline aluminides at both strain rates span a relatively narrow range from 250 to a. The work-hardening behaviors o f the Ni-, Fe-, and Ti-aluminides all exhibit: 1) considerably higher hardening rates than that typical for their disordered polycrystalline base metals, and 2) remarkably sustained lineal- hardening behavior, in particttlar for the Fe-4OAl-0.1B, NiAl, Ni,Al, and Ti-45A1-2Gr2Nb compounds, over a wide range of temperatures. The average work-hardening rates for the aluminides are summarized in Table 1. The remarkable linear tage 2 stress-strain behavior , NiAl and Ti-48A1-2Gr"2~bpersists to true strains in excess of 0.20. In some instances, s ~ ~ h as for Ti-48A1-2Cr-2Nb, sequential reloading with relubrication can be used to strain aluminides in compression to flow stresses >2GPa prior to failure by non-crystallographic shear (Gray 111 and Embury, 1993). These stress levels are equivalent to -E/lO (where E is Young's modulus), which is an inipressive flow stress for any material to stably support.
369
2
ure 7 Stress-strain response for polycrystalliiie aluminides at 29s K, a) quasi-statically, and b) d y n a ~ i c ~ ~ l y
Overall, the stable high rate of work h a r d e ~ i ~ g exhibited by inany polycrystalline intermetallics, with hardening rates often >>p/lOO (where p = shear modulus) and absolute flow stresses > 1500MPa, is consistent with the suppression of dynamic recovery processes (Gray IIX aiid Embury, 1993). Althougli there has been very limited study of the sub~tructureof
inter~e~allics following higher strain deformat~oii,the available evidence suggests a lower tendency for the formation of the cell structures that are characteristic of Stage 3 d e f o r ~ a ~ in i o~~i s o r d e r ~crystals, d where dynamic recovery processes become operative. An. example of this is shown for Ti-48A1-2Cr-2Nb in Figure 8, where a fairly u ~ i f o r i~istribu~ioii ~ of
370
MeclznnicmI Properties
Figure 9 Polycrystalline RuAl after 7%) cornpresswe strain at room temperature. Su~struc~ure consists o f a mixture of { 110) and (I 00) dislocations
Bright field TEM inicrograph of (101} supe~dis~ocationswith g = (002) in Ti-48-2-2 after deformation at 900°C and strain rate of 2000/s (after Maloy and Gray, 1996)
superdislocations is present after a compressive strain of 16%. Other examples of materials that show a limited tendency for cell fori~ationinclude orthoranerjee, 1997), L1, nl., 1995) and C15 Cr,Nb (Takasugi et
al., 1995). One interesting exception is exhibits a marked tendency for dislo at compressive strains as Low as 7%, Figure 9. In this (110) high melting point C O ~ ~ Q L (100) E K ~ and , dislocations on (1 XO] planes are observed within the tangled substructures (Lu and Pollock, 1999); if. is worth noting tliat slip on the { I l O ] plane with this urgers vectors provides five independent slip systems. As ~ i g ~ l i g h t e by d the original analyses of Taylor (Taylor, 193&), five independent slip systems are needed for uiiiform deformation of ~olycrystals.~nfortunately,many intermetalli~slack a sufficient number of slip systems at low temperatures.
~ u m ~ a of r yrate sensitivity and w o r ~ - h a r d e ~ ~response ng of intermetallics studied at 298 K
Material
~ o r l ~ - h a r d e nrate i n ~ t) at ~ i t 0.001 s-l [ ~ ~ a / ustrain]
Ti
Fc Ni Superalloy NiAl [100] NiAl [310] Cu,Au
Ni& Fe-40Al-0.1B Fe,Al RuAl Ti-48A~-2Cr-2~b Ti-24Al- 11 Nb
I000 I000 3000 2400 4000 2900 565 5 500 6600 1000 6000 ~ 5 0 ~ 0 0 0 1800
Quasi-static" Stage-2 ratio of p at 25C
Work-hardening rate at 2000 s-' [ ~ ~ a / u nstrain] it 2400
1600 5000
4000 3 560 530 8000 6500 6500 4500 3600
"Nori~aliziiigthe quasi-static work-~~rdening rate t3 (column 1) with the Taylor factor, M , equals 8/M2
Strain-rate sensitivity, m (1/1n> 67 (0.015) 20 (0.05) 100 (0.01) low 41 (0.024) (0.08 5)
low low 21 (0.047)
217 (0.0046) 35 (0.029)
~ o d i f i e dTaylor-type analyses have been performed via finite-element-based crystal plasticity models for deformation in such intermetallics (for example, for NiAl, which has only three independent systems (Ahzi, 1999)). It is apparent that stress levels within the polycrystalli~ieaggre~atewill be elevated due to this lack of sufficient iiidepeiident slip systems (Parks and choenfeld et al., 1995). However, a quantitative assessment of tlie coiitributioii of this constraint to overall wor~-har~ening rates has not yet been consi~eredin detail for any system. During compression testing of niany of the aluminides shown in Figures 4 and 7, microcrac~ingat the scale of the grain size often occurred at strains far below those which Liltini~~tely result in catastrophic failure. This suggests that the characteristically hi hardening rates in inte~ietallics(and the corresponding high stresses) contribute to the early onset of fracture in tensioii. Apparently damage evolution in compression within most aluminides does not significantly alter the stable, high rate of work hardening typical of these materials. The rate sensitivities, ‘WE’(= d l i i ~ / d l n ~for ) , these aluminides were calculated using the flow stress values at a plastic strain of -2% (t strain-rate change data with bar data, without the comp dure of directly measuring rate sensitivity). Tlie rate sensitivities and work-hardening rates available are 1976). Overall, the Fe-, Ni-, d exhibit a wider range of strain-rate sensitivities than their high-purity disordered base metals. The relatively hi sensitivities suggest low activation volumes for deformation (short-range barriers to dislocation glide) in some, but not all, inte~metallicsat low temperat~res. This wide range of behavior is perhaps not surprising given the variety of deforination modes arid the potential array of dislocation core dissociation and decomposition phe~omena in these intei-rnetallics (Veyssiere and Douin, 1994). One notable exception with regard to rate sensitivities is Ni3Al, which displays an extremely low dependence of its yield strength on strain rate at 298IL consistent with a low Peierls barrier for dislocation motion ( izek and Gray 111, 1993). This very low rate-dependence at 298 I( (Sizek and Gray TIT, 1993), along with the anomalous dependence of flow stress on temperature, has been the focus of a great number of experiments and widely contested inechaiiistic models for the rate-limiting flow plienomena in Ni,AI. These are reviewed in d e t d elsewhere
emlcer et al., 1992; Ezz, 1984; Veyssikre strain-rate effects on have played a pivotal role as delimiters for ~ i ~ e r e n t i a t ing new defect model des interesting exception is sensitivity characteristic compound also has an apparently high tougbness in its polycrystalline form and contains s tions of both (100) and (110) type substructure along with a tendency of cell-type structures after large-strain compressive Fleischer and Zabala, 1990; Lu an Because of the wide variation in sli occur within a single class of c cannot easily be predicted.
the hardening behavior, detailed s y s t e ~ ~ tstudies ic are lacking. Comparison of the w o r ~ - h a r ~ e n i n various aluminides as a function of temperature at high rate reveals that the rate of strain hardening is predo~inantlyi~variaiit~ again exception of Ni3AI and to some extent Ti-4 2Nb. In the majority of the aluniinide data p suminarized (Gray 111, 1996), increasing strain rate and/or temperature at high rate was seen only to shift the hardeniiig curves up or down; the nearly invariant hardening rates result in a family of parallel curves. The observation of a pronounced increase in yield and flow stress with decreasing teiiiperatLire and/or increasing strain rate is typical for bcc, Figure 2, and lower symmetry crystal structures and has often been explained by a high Peierls stress 1975), or in some cases due to interstitial impurities. For the case been shown that the interstitial impur material has a significant influence on the flow
rate also prevalent. iiot only influenced also by deformatio in this volume fo increasing strain rate and/or decreasing temperature, the propensity for twin formation in TiAl i s seen increase. Increasin~a ~ o u n t of $ ~ e f o ~ ~ a ttiwoi n~ n i
372
~ e & h u ~ ~Properties cul
are seen to correlate with an increased rate of work hardening (Gray 111 1994; Maloy and Gray HI, 1996). In addition to twinning, long planar arrays of straight < 101 > superdislocat~ons are observed following high-strain-rate t restraining at elevated temperatures as seen in Figure 8. th of these modes of deformation result in a h rate of defect storage and inhibit dynamic recovery, thus producing a high work-hardening rate (Maloy aiid Gray-111, 1996). See the chapter on twinn~ngby Yoo. s mentione~previously, one interesting feature present iii many ordered intermetallics (but absent in their disordered cou~terparts) is the anomalous i n c r ~ ~ in s e the flow stress as a function of increasing temperature. This was first observed in Ni,AI by Flinii (Flinii, 1960)- and was foreshadowed by the hothardness experi~ents of ~ e s t b r o o k (1957). The presence of this behavior for the Ni, Fe, and Tialuminides is shown in Figure 10. The low-strain flow stresses of Ni,Al, Ti-4$Al~2Cr-Z~b, and Fe-4OA1-0.1B all exhibit an anomalous increase in stress level with increasing temperature at high strain rate. The anomalous increase in. flow stress in Fe-40A1-0.1B with increasi~igte~~perature above 600K is preceded by a proiiounced decrease in flow strength upon increasiiig temperature from low temperature. This U-shaped flow stress behavior with temperature is consistent
with the constitutive response of Fe-40Al-0.1 48A1-2Cr-2Nb being dominated by the Peierl low temperature aiid then involving one or more additional hardening mechanisms upon increasing temperature. One of the earliest models of the anomalous tem~erature dependence of the flow stress in systems was proposed by Kear and Wilsdorf ( and Wilsdorf, 1962).They suggested that the increase in flow stress with temperature occurred due to t h ~ r i ~ a l l y activated cross-slip of APB-coupled a/2(110) dislocations from the (1 11) plane, where they are mobile, to the (100) plane where they are immobilized and contribute to hardening (and thus the anomalous increase in the flow stress). Subsequent discussion of this mechanism, more recent experimeiital observations and other cross-slip inspired models are reviewed by Nabarro and de Villiers (1995). Whether a KearWilsdorf mechanism, another cross-slip pinning ~ ~ c ~ (Baker ~ n and i sNagpal, ~ 1993; Hirsch and Sun, 1993; Yoshimi and Hanada, 1993) or vacancyinduced hardening (George and Baker, 1998) is responsible for the anomalous temperatur~d e p ~ ~ d e n c e remains a subject of current debate. Not all iiitermetallics display the anomalous hardening and increase in flow stress at intermediate temperatures. For example, MiA1, Ti-24Al-1I Nb, and
Plot of flow stresses at fixed plastic strains versus temperature for various intermetallics at high strain rate
Struin ~ ~ r d e n ~ n ~
display a decrease in their high strain rate Bow s with increasing temperature, Figure 10. This behavior is also more typical of disordered metals and alloys where fall-off in the yield strength of a material is commensurate with the decrease in shear modulus with increasing temperature. The one unique feature in the current observations of the anomalous yield increase with temperature is the lack of a downturn in yield at very high temperatures. At conventional nturn is well documented (Anton, 1994; e, 1994; Yarnaguchi and Inui, 1994). Unlike the usual fall-off in yield with temperature for samples tested at quasi-static rates at high temperatures, due to the activation of additional slip systems, such as cube slip in Ni,Al, or climb-related processes in FeAl (Baker, 1995), high strain rate loading is observed to prevent this high-temperature deformation mode change in the aluminides exhibiting anomalous yielding. Up to the temperatures attained to date in split-Hopkinso~-bar testing, the high strain-rate yield strengths of Ni,A1 (Sizek and Gray 111, 1993), Fe40Al-0.1B (Gray 111, 1995), and Ti-48A1-2Cr-2Nb ~ ~ a l and o y Gray TIT, 1996) have not been observed to decrease at high ternperatures, However, it is expected that a downturn would be encountered as the melting point is approached. Of the aluminides in Figure 10 that display the anoinalous increase in yield strength with temperature at high strain rate, the overall magnitude of the temperature effect on yielding and post-yield flow stresses is most evident in Fe-40Al-0.lEQ. Previous studies have documented this phenomenon in FeA1, in articular in the work of Yoshirni and Hanada (1993) r et al. (1995a,b), ker (1995), Klein and 94), aiid Yoshiini a Hanada (1993). Fe-39 rigle crystals have been observed to display a 200 MPa from nearly constant yield stress of ambient temperature to 600 I<, followed by a positive ure dependence with a peak at a temperature when tested at a strain rate of 1.7 x 10-4 sit (Yosliirni and I-lanada, 1993). The anomalous slip transition in FeAl in that study was related to the decompositio~of (111) ~islocat~ons, while a change in slip vector from the (111) to (001) type occurs d the peak temperature 1993). It has recently been alous yielding phenomenon in Fe-40Al i s perhaps linked to vacancy hardening To summarize, inte~metallicsdisplay an amazingly diverse range of work-hardening beliaviors and characteristics as a ftnnction of loading conditions, whether
373
they are in single- or polycrystalline form. This depth of response makes generalizations about the physical mechanisms and processes controlling w o r k - ~ ~ r ~ e n i n behavior in intermetallics, and the odel ling of these processes, problematic. While sustained ‘Stage 2’ hardening is often observed for a number of intermetallics to relatively large strains, substructural observations show that the processes controlling this hardening vary between intermetallics with difTererrt crystal structures. Continued work i s required to provide the basis of more generalized models describing the micromechanisms controlling plasticity in intermetallics, particularly for polycrystalline intermetallics. Such models will necessarily be multi-scale in nature to provide an accurate description of intermetallic hardening, spanning from modeliiig the dislocation core up to c a l c ~ l a t i nthe ~ contributions of long-range barriers such as grain boundaries. Future utilization of structural iiiterinetallics in engineered applications will require a ~ r ~ d i c tunderiv~ standing of the yielding, strain hardeiiing, and damage evolution behavior.
Based on this brief review of the strain-hardeniii behavior of single aiid polycrystalline intermetallics, a number of general observations can be made: e Interinetallics, when tested in compression so that
fracture processes are suppressed, typically sustain very high strain-hardening rates to high flow stresses. e A comparison of the work-hardening rates of the various a l ~ r ~ i ~ i as d e as f~nctionof strain rate or temperature often reveals an invariant rate of strain hardening, similar to the response of pure bcc metals. The exceptions to this behavior are low Peierls-stress intermetallics, such as Cu,Au or and systems that deform in part by twiiining, such as TiAl. The high rates of strain harden in^ often observed for intermetallics are due to a ?qgh rate of storage of dislocations coupled with a general lack of dynamic recovery processes. High rates of dislocation storage may arise due to: a) a low tendency for cross-slip that results in a generally low rate of dynamic recovery or b) high rates of storage of screw dislocations that readily cross-slip, but are subject to cross-slip pinning processes that also create debris, or c) accurnLilation of a high content of
344
~ e c h ~Properties n i ~ ~ ~
dislocation^ that are relatively immobile due to a high Pejerls barrier, point-defect pinning andlor a complex core structure. Studies of the deformation and strain-hardening response of intermetallics can provide insight into the details of defect generation and storage processes in ordered i ~ t e ~ e t ~ lconipounds. lic In addition, they are relevant to many operational and nianufkturing questions concerning engineering utili~ationof intermetallics.
The authors are grateful to J. Schneibel of ORNL, R. rolia and C. Austin of General Electric and ischer for materials used in the studies presente
ert Carpenter 11 f ~ assistance r with niechanical testing. The work of 6.71. Gray 111 was performed he auspices of the US Department of Energy. ollock would like to acknowledge the experiE.A. Ott and D.C. Lu BES grant DE-FG020 0 E 1 ~ ~ ~ 8For 2 0the preparation of this manuscript
Ahzi, S. (1999). ~ ~ ~Sirnzd. ~ Mater. l ~ SCL n Eng., g 7, 841-850. Albert, D. E., and Gray, G. T. (1997). Acta Mater., 356. Anton, D. L. (1994). In r ~ ~ t e r ~ e ~ aCompounds, llic Vol. 2: Practice (eds J. H. Westbrook, and R. L, Flcischer) Wiley, p" 3. Appcl, F., Sparka. U., and Warner, R. (1995). In HighTenywrature Ordered I i ~ t ~ ~ r ~ ~ eAlloys t a l l ~VI c (eds J . A. Horton, I. Baker, S. Hanada, R. D. Noebe, and D. S. Schwartz) Materials Research Society, Pittsburgh, PA, pp. 623-628. Appcl, F., Sparka U., and Wagner, R. (1999). Inferrnetallics, 7, 325-344. Appel, F., and Wagner, R. (1998). Mat. Sci. Eng., Argon, A. S . , and EIaasen, P. (1993). Act0 Metall. Mader., 41, 3289. Baker, 1, (1994). In Processing, Properties, and Apj~licationsof Iron Alunziiirclcs (eds J. €4. Schneibel, and M. A. Crimp) TMS, pp. 101-1 15. aker, I. (1995). Mat. Sci. Eng., A19
Baker, I., and Nagpal, P. (1993). In Structural i ~ t ~ ~ r ~ ~ e t a l l i c s (eds R. Darolia, J. J, Lewandowski, C. T. Liu, P. L. Martin, D. B. Miracle, and M. V. Nathal), TMS, Seven Springs, PA, pp. 463473. Baker, I., Xiao, W.,Klein, O., Nelson, C., and Whittenberger, J, D. (1995). Act0 Metall. Mater., Banerjee, D. (1995). In Inter~etallicC o n z ~ o u n ~ Vol. s , 2: Practice (eds J. H. Westbrook, and R. L. Flewher), Wiley, p91. ., and Basinski, 2;. (1979). sn Dhslocarions in Solids, Basinski, Z. (1974). Scripia Met., Bonneville, 3., Escaig, B., and Martin, J. (1988). Acra Metall., 36, 1989. Busso, E. P., and ~ c C ~ i n t oF. c ~A., (1996). Irat. J. Plas., 1 1-28. Caillard, D., Vailhe, C., aiid Farkas, D. (1999). Plzilos. Mug., 7 ~ A 723. , Chen, S. R., and Gray Ill, G. T. (199~).Metal/. Mater. Trans. A, 27, 2994-3006. Christian, J. W. (1983). ~ e ~ a ~ ~ ~Trunsuctions r ~ i c u l A, 1237-1256. Cottrell, A. H. (1953). ASM Smtinar on the Relation of Properties to ~ i c r o ~ ~ t r u c t u rCleveland, e. ASM, Metals Park, OH. Dong, Z., Liming, W., Dunxu, Z., Mei, Y., and Qigong, C. (1992). J . Matls. Scr., 11, 1992. Douin, J., Kumar, K. S., aiid Veyssikre, P. (1995). Mater. Sci. and Eng., ~ ~ ~ 292-96. 1 ~ 9 ~ , , D., and Pollock, T. M. (1998). Scrrpta Mater.,
Ezz, S . S., Sun, Y. Q., and Hirsch, P. B. (1995). In High~ e n ~ p ~ Ordered ~ a ~ ~I n~~re re~ c i ~ a lAlloys lic VT (cds J. A. Norton, I. Baker, S. Hanada, R. D. Noebe, and D. S . Schwartz). Materials Research Society, Warrendale, PA, pp. 695-700. Ezz, S, S. and Hirsch, P. B. (1995). Philos. Mag. A , 72, 383 402. Field, R. D., Lahrman, D. F., and Darolia, R. (1991). In Mat. Res. Soc. Syinp. Proc. (eds L.A. Johnson, J.O. Stiegler, and D.P. Pope), MRS, Warrendale, PA p. 255. Fleischer, R. L., Field, R. D., and Briant. C. L. (1991). Metall, Trans., 22A, 403-41 4. Fleischer, R. L., and Zabala, R. J. (1990). Metall. Trans., 2709. Flinn, P. A. (1960). Trans. A I ~ E , George, E. P., and Baker, I. (1998). P~zilo~~. Mag., 77,737-750. Gray 111, G. T. (1991). In ~icrcistructurejprop~rty Relation~~~ips in ~ ~ t a n i uAl~nzinides m a d Alloys (eds Y,W. Kim, and R. R. Boyer), TMS, pp. 263-274. Gray TII, G. T. (1994). In Twiizizing in Advanced Materials (eds M. H. Yoo, and M. Wuttig), TMS, Warrendale, PA, pp. 337-349. Gray 111, G. T. (1995). In Plasticity '95-Dvnarnic Plasticity and Structural Behaviors (eds S. Taniinura, and A. S. Khan), Cordon and Breach, Osaka, Japan, pp. 527-530.
Strain Hardening Gray 111, G. ‘r. (1996). In Dejkrrnation and Fracture of Ordered InterF~etallicMaterials III (eds W. 0. Soboyejo, T. S. Srivatsan, and H. L. Fraser), TMS, Warrendale, PA, pp. 57-’73. Gray 111, G. T. (1 99 1). In Microstructure/Property Relaticiiiships i ~ i Titanium Aluminides and AlloJis (eds Y. W. Icim, and R. R. Boyer), The Minerals, Metals & Materials Society, Warrendale, PA, pp. 263-274. Gray III, G. T., and Embury, J. D. (1993). In HighTer~perarurc Ordered Intermetallic Alloys V (eds I. Baker, R. Darolia, J. D. Whittenbergcr, and M. H. Yoo), Materials Research Society, Pittsburgh, PA, pp. 585-590. Gray 111, 6.T., Hong, S. I., and Marquardt, B. J. (1993). In Titanium ‘92 - Science and Technology (eds F. H. Froes, and I. L, Caplan), TMS, Warrendale, PA, pp. 1163-1 170. Gray 111, G. T., Hong, S. I., and Marquardt, B. J. (1996).
H. (1995). Metall. and Matls. Trans. A. 26, 113 131. Hazeldine, P. M., and I-lirsch, P B. (198’7). In Mater Xes. Soc. Symp. Proc. (edsN. S. StoloE, C. C. Koch, C. T. Liu, and 0. Izumi). MRS, p. 75. Eleinker, K. J., Mills, M. J., and Nix, W. D. (1992). J . Matls. Res., ‘7, 2059-2069. Hirscli, P. (1964). Discussions of tkw Fara. Soc., Hirsch, P. B., and Sun, Y. Q. (1993). Mutls. Sci. 395-400. Honeycornbe, R. W K. (1984). The Plastic Defor~ationof Metals. Victoria, Australia, Edward Arnold. Hosoda, H.. Miyazaki, S., Inoue, I<., Fukui, T., Mizuuchi, uki, T., (2000). J . Alloys and
, and Yamaguclii, M. (1992). Mater. Sci. and Eng., A152, 153 - 159. Jin, Z., and Gray 111, G. T. (1995). In Metallur~icaland Materials Applications of Shock- IVuw and High-StrumRate Phcrromena (eds L. E. Murr, M. P. Staudhammer, and M. A. Meyers), Elsevier, A m s t e r d a ~pp. ~ 901--908. Jin, Z., Gray 111, G. T., Kim, Y.-W., and Yamaguchl, M. (1997), Strzictural Infermetallics 1997 (eds M. V. Nathal, R. Darolia, C. T. LIU,P. L. Martin, D. B. Miracle, R. Wagner, and M. Y amaguchi), TMS, Warrendale, PA, pp. 225-234. Jin, Z., Gray TIT, G. T., and Yamaguelii, M. (1998). Philos. MU<^. A , 78, 239-253. Juimnji, K., Uela, S . , Kato, M., and Sato, A. (1994). Philos. Mag. A , 69, 1111-1 124. rf, H. 6. F. (1962). Trans. Adetall. Kim, Y.-W., and Dimiduk, D. M. (1993). High-Temperature Ordered Inrcrrnetallic Alloys ’G/(eJsI. Baker, R. Darolia, J. D. Whittenberger, and M. H. Uoo) Materials Rcsearch Society, Pittsburgh, PA, pp. 671-677. Kim, J. T., and Gibala, R . (1991). In Mat. Res. Soc. Synz. Proc. (cds L. A. Johnson, D. P. Pope, and J. 0. Stiegler), MRS, Warrendale, PA, p. 255.
375
Kiniura, A., Koya, A., Morimura, T., and Misawa, T. (1994). Matls. Sct. Eng. A , 1196, 425-430. Kitano, I<., and Pollock, T. M. (1993). In S t r ~ c t z ~ r a l Iriterme~ullics(eds R. Darolia, J. J. Lewandowski, G. T. Liu, P L. Martin, D. B. Miracle, and M. V. Nathal), TMS, Seven Springs, PA, pp. 591-599. Kitano, K., Pollock, T.M., and Noebe, R. D. (1994), Scrzpta Mater., 31, 397-402. Klein, O., and Baker, I. (1994). Scrrptu Meiall. Mater., 863-868. Kocks, U. F. (1966). Phil. Mag., 1 1, 541. Kocks, U. F. (1975). In ~ o n s t ~Equations t u ~ ~ in ~ Plasticity ~ (ed. A. S . Argon), MIT Prcss, Cambridge, MA, pp. 81-1 16. Kocks. U. F (1976) Journal of Engineering Mafcrials and Technology, 98, 76-85. Kocks, U. F. (1 982). In Mechanical Testing for Defiwniation Model Development (eds R. W. Rohde, and J. C. Swearengen), American Society for Testing and Materials, pp. 121-138. Koclts, U. F.. Argon, A. S., and Asliby, M. F. (1975). Prog. Matls. Sci., 19, I . Kocks. U. F., Stout, NI. G. and Rollett, A. D. (1988). In 81h Int. Conj: on Tpxturcs ~ ? ~ M u t e(eds r ~ ~J.lS, .~Kallend, and 6 . Gottstein), The Metallurgical Society, Tampere, Finland, pp. 31-36. Kuhlmann-Wilsdorl, D. (1962). Tram. Met. Soc. AIME, 1047. f, D. (1966). Work ~ a r d e n z nA~ ~I ~ E ~ f, D. (1985). Met. Tram., 66A, 2091. Kuniar, K. S. (1995). In Interrnrtallic Compounds, Vol. 2: Practice (eds J. H. Westbrook, and Wiley, 13.211. Lee, T.-S., Hon, W.-P., Yang, S.-C., Hsu, S.-E., Chin S., and Anton, D. L. (1995). In High-Temperature Ordered Intermetallic Alloys V I (eds J. A. Horton, 1. Baker, S. Hanada, R. D. Noebe, and D. S. Schwartz), Materials Research Society, Pittsburgh, PA, pp. 537-542. Levit, V. I., and Kaufman, M. J. (1997), Str~cturc~l Internzerullics 1997 (eds M. V. Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. Yarnaguchi), TMS, Warrendale, PA, pp. 683-692. Lewandowski, J. 5. and Lowhapha~du,P. (1998). h t . Mat. Rev.. 43, 145-18’7. Lipsill, H. A. (1993). In High Temperuture Ordered Intermeiallics V (eds I. Baker, R. Darolia, J. D. Whlttenberger, and M. H. Yoo), Mat. Res. Soc. Symp., p. 119. Llu, C. T., and Pope, D. P. (‘1994). In Inte~~ietallic Compounds, Vol. 2: Practice (eds J. H. Westbrook, and R . L. Fleischcr), Wiley, p. 17. Loretto, M. H., and Wasilewski, R. J. (1971). Philos. Mag., 23, 1311. Lu, D., and Pollock, T. M. (1999). Acta Mater., Maloy, S. A., Gray 111, G. T., and Dai-oh, R. (1995). Mat. Sci. and Eizg., A1921193, 249-254.
376
Meclzanicul Properties
M a ~ o ~S., A., and Gray 1x1, G. T. (199%). In Gamma T ~ t a ~ A~ l iuu~ ~ n~ i d(cds e ~ Y.-W. Kim, and R. Wagner), TMS, Las Vegas, NV, pp. 307-314. Maloy, S. A., and Gray 111, 6. T. (1995b). In Nigh~emperatureOrdered Inter~etallicAlloys VI (eds J. A. Horton, 1. Baker, S. Haiiada, R. D. Nocbe, and D. S. Schwartz), Matcrials Research Society, Pittsburgh, PA, pp. 549-554, Maloy, S. A., and Gray 111, G. T. (1996). Acta Mater., 44, 1741-1756. Meck~iig, H. (1971). In Work Hardening IN Tension and Fatigue (ed. A. W. Thompson), The Meta~lurgicalSociety of AIME, Cincinnati, Ohio, pp. 67-87 Mecking, H. (198 I). In ~ e f o r ~ a t i o onf Polycry.stals: ~ e c h a n i ~ ~and i ~ sMi~rostructures (eds N. Hansen, A. Horsewell, T. Leffers, and H. Lilholt), RISO National Laboratory, Roskilde, Denmark, pp. 73-86. Meckmg, H., and Estrm, Y. (1 987). In 8'" RISO Intcrnationnl Symposium OFZMetallurgy and Materiuls Science (eds S. I. Andersen, J. U. Bilde-Sorensen, N.Hansen, T. Leffers, H. Lilholt, 0. B. Pedersen, and B. Ralph), pp. 123-145. Mecking, EI., and Kocks, U. (1981). Acta Met., 29, 18651875. Mendiratta, M. G., Kim Y.-W., and Dimiduk, D. M. (1993). In ~ i g h ~ ~ c m p e r a tOrdered ure Intermetallic Alloys V (eds aker, R. Darolia, J. D. Whittcnbcrger, aiid M. W. Yoo), Materials Rcsearch Socicty, Pittsburgh, PA, pp. 543-548. Messerscliniidt, U., Haushalter, R., and Bartsch, M. (1997). Mater. Sci. Eng., A 2 3 ~ 2 822. 3~~ Miracle, D. B. (1993). Acta Metall. Mater., 41, 649. Mitehell, T. E,, Foxall, R. A., and Hirsch, P B. (1963). Philos. Mughrabi, H. (1975). In Comfitutive Equations in Plasticity (ed. A. S, Argon), MIT Press, Cambridge, MA, p. 199. Nabarro, F. R. N. (1989). Actn Metall., 37, 1521. Nabarro, F. R. N., Basinski, Z. S., and Holt, D. L. (1964). Aclv. Phys., 13, 193. Nabarro, F. R. N., and de Villiers, H. L. (1995). The Physrcs of Creep. Taylor & Francis, Amsterdam. Nandy, T. I<., and Baiierjec, D. (1997). In Strrkctural ~~ZternzetaZlj~s 1997 (eds NI. V Nathal, R. Darolia, C . T. Liu, P L. Martin, D, B, Miracle, R. Wagner, and M. Yamaguchi), TMS. Warrendale, PA, pp. 777-785. Noebe, R. D., Bowman, R. R., and Nathal, M. V. (1993). Intl. Noebc, R. D., Cullcrs, C. L., and Bowman, R. R. (1992). J Mutls. Res., 7,405-6 12. Oh, M. H., Inui, H., Misaki, M., Kobayashi, M., and Yamaguchi, M. (1993). In High-Temperature Ordered Intermetallic Alloys V (eds 1. Baker, R. Darolia, J. D. Whittenbcr~er, and M. H. Yoo), Materials Research Society, Pittsb~irgh.PA, pp. 1001-1006. Qtt, E. A. (1998). Mechanis~sof Creep Deformation in Gamma-based Titanium Aluminides, PhD Thesis, Carnegie-Mellon University, Pittsburgh, PA.
Palm, M., Corzel, A., Letzig, D., and SauthoE, G. (1 997). In Structural ~ n t e r m e t a ~1997 ~ ~ ~(eds , ~ M. V, Nathal. R. Darolia, C. T. Lm, P. L. Martin, D. B. Miracle, R. Wagner, and M. Ya~aguchi), TMS, Warrendale, PA, pp. 885-593. Parks, D. M.. and Ahzi, S. (3990). J. Mech. P/zys. Solids, 38, 701-724 Pope, D. P., and Ezz, S. S. (1984). Intl. Meealk. Rev., 29, 136. Pnnz, F., and Argon, A. S. (1980). Phys. Sfaf. Sol., 57, 741. Rollett, A. D., and Kocks, U. F. (1994). Solid State Pheno~ena,35-36, 1-18. Sauthoff, G. (1989). 2. ~ e ~ a l ~ ~ ~ . , Schoeck, G. (1969). Acta Mefall., Schoenfeld, S. E., Ahzi, S., and Asaro, R. J. (199-7). J Mech, Seeger, A., Diehl, J., Mader, S., and Rebstock, H. (1957). Plzil. Mug., 2, 323. Sevillano, J. G. (1993). In 34uterial.s Science aPzd Technology: Plastic Dejormation qf Mutcrids (ed. H. Mughrabi), VCH Publishers, p. 21. Shi, X., Mahajan, S,, Pollock, T. M., and Arunachalam, V. S. (1999a). ~ n t e r m e ~ a l ~7,i c1255. ~~, Shi, X., Mahajan, S., Pollock, T. M., and Arunachalam, V. S. (1999b). Philos. Mug., B A , 15-75. Shi, X., Saada, G., and Veyssiire, P. (1996). Philos, Mag., "PA, 1159. Shili, D. S., Schwartz, D. S., and Q'Neal. J. E. (1993). In High- Temperatwe Ordered Intermetallic Alloys V (eds 1. Baker, R. Darolia, 5. D. Wbittenbcrger, and M , W.Yoo), Materials Research Society, Pittsbmgh, PA, pp. 579-584. Siltka, V. I<., Viswanatban, S., and McKamey, C. G. (1993). s R, Darolia, J. J. In Str~ctura1 I n t e r ~ e t u l l ~ c(eds Lewandowski, C. T. Liu, P. L. Martin, D. B. Miracle, and M. V Nathal), TMS, Warrendalc, PA, pp. 483-491. Sizek, H. W., and Gray 111, G. T. (1992). In S ~ o c ~ " W and ~ve H i g ~ - S ~ r u i n - ~ aPt he e ~ o ~ e ntna Materials (eds M, A. Meyers, L. E. Murr, and IS.P. Staudhammer), Marcel Dekker. New York, pp. 117-125. Sizek, H. W., and Gray 111, G. T. (1993). Acta Metall. g t Mater., 41, 1855-1860. Srinivasan, R., Savage, M. F.. Daw, M. S., Noebe, R. D,, Garg, A., and Mills, M. J. (1997). In Structural ~ ~ ~ e ~ ~ e1997 ~ a(eds l l M. ~ c V. s Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B Miracle, R. Wagner, and M. Yamaguchi), TMS, Warrendale, PA, p. 665. Srinivasan, R., Savage, M. I?., Daw, NI, S., Noebe, R. D., and Mills, M. J. (1998). Scripta Mater., 39, 457. StoloE, N. S., and Davies, R. G. (1966). Prog. Mat. Sci., 13, 1. Sun, Y. Q. (1995). I ~ ~ t e r ~ e Comnpounds, t a l ~ ~ ~ , F'ol. I. Principles (cds J. H. Westbrook, and R. L. Fleischer), Wiley, p.4 Suzuki, T., Mishima, Y., and Miura, S. (1989). ISIJ Int., 1- 23. da, M. (1995). Mater.
., and Ono, S. (1990). Pliilos. Mag., 61A, 785. Taylor, G. 1. (1938). J. Inst. Metals ( U K ) ,
Strain ~ u r d ~ n i n g
377
Yamaguchi, M., aiid Inur, H. (1993). In Structurul Inter~etallics (eds R. Darolia, J. J. L ~ w a n d o ~ s C. k ~T. , Veyssiere, P., and Douin, J. (1995). Inter~eta$licC o ~ p o ~ n d s , Liu, P. L. Martin, D. B. Miracle, and Vol. I : Prmc@les, (cds J. H. Westbrook, and R. L. TMS, Seven Springs, PA, pp. 127- 142. Fleischer), Wiley, p. 519. Yarnaguchi, M., aiid Inui, H. (1995). In I ~ t e r ~ z e t a ~ l i c Vidoz, A. E., and Brown, L. M. (1962). P h i l ~Mug., ~ ~ . 7, 1167. C o ~ p o ~Vol. ~ ~2:s Practice , (eds J. N.Westbrool~,and Vidoz, A. E., Lazarevic, D. P., and Cahn, R. W. (1963). Acta R. L. Fleischer), Wiley, p. 147. &!etaK, 11, 17. guchi, M., and Umakoshi, Y. (1990). Prog. Matls. Sci., Viguier, B., Bonneville, J., Hemker, K. J., and Martin, J. L. 1-148. re Interrnetallic (199%). In ~ i g h - T e ~ ~ e r a t u Ordered M. N., Horton, J . A., and Liu, C. T. (1988). Acta Alloys VI (eds J. A. Horton, I. Baker, S. Hanada, R. D. Metall., 34, 2935. Noebe, and D. S. Schwartz), Materials Research Society, Yoo, M. H., Takasugi, T., Hanada, S., and Izumi, 0. (1990). Pittsburgh, PA, pp. 629-634. Mater. Trans. JIM 31, 6, 435. Viguier, B., Hemker, E;. J., Bonneville, J., Louchet, F., and Yoo, M. €3. (1995). Acta Martin, J. L. (1995b). Philos. Mag. A, 71, 1295-1312. Viguier, B., and Hernker, K. J. (1996). Philos. Mug., 73 Yoshimi, K., and Hanada, S. (1993). In Str~ctural 575-599 Inter~etal~ics (eds R. Darolia, J. J. Lewandowski, C, T. Westbrook, J. 13. (1957). Trans. AIME, 209, 898. Liu, P. L. Martin, D. B. Miracle, and M. V. Nathal), ng, S. El., Feng, Q., and Cao, Y.-Q. (1998). Acta Metall., 4,6485. TMS, Seven Springs, PA, pp. 475-482, Temes, .f. I<., Farltas, D., ancl Kriz, R. (1995). Philos. Mug.,
This Page Intentionally Left Blank
Dislocations exist in both conventional intermetallic crystals aiid iiitermetallic quasicrystals (QCs), especially in deformed specimens. Figure l shows densely distributed dislocation lines in a Al-Pd-Mn single icosahedral quasicrystal C) bent at 720°C. Studies of dislocations in QCs h drawn extensive attention because of their importance, not only to structure investigation, but also to understanding many physical and mechanical properties of QCs. Since plastic deforination of single QCs occurs mainly via dislocation glide, studies of Burgers vectors of dislocations and dislocation arrangements in differently pretreated aiid deformed QCs are very iinportant for understanding the microi~eclianis~ of plastic deformation of QCs. As described by Keltoii (1995) in Chapter 20, Vol. 1 of this series, a quasiperiodic lattice may be described as a projection froin a N-dimeiisional (ND) liypercubic lattice onto a 3D physical subspace that has an irrational orientation with respect to the original lattice. N = 6 for IQCs, and N = 5 for decagonal quasicrystals (DQCs), octagonal quasicrystals, dodecagonal quasicrystals and pentagonal quasicrystals. Therefore, Burgers vectors = [BI, B2, . . ., B N ] of dislocations, displacement vectors W and reciprocal vectors G = [Gl, G2, . . ., GN] in a QC are all vectors with N components. This unique character of QCs complicates the experimental determination of rs vectors and elasticity theory of dislocations in compared with those of dislocations in convenIn the present chapter, we will describe the Volterra process of a dis~ocationin Cs, methods of experi-
mental determination of Burgers vectors of dislocations in QCs, linear elasticity theory of calculation of the elastic field arouiid a dislocation in a QC, as well as some experimental results about Burgers vectors of dislocations, slip systeins and dislocation reactions in QCs. Finally, we will explore dislocation niechaiiisms for high-temperature plastic deforination of QCs.
and Trebin (1989), Klernan and leinan (1995), Yu et al. (1997), Y (1998) and Diiig et d.(1998) discussed the Volterra process, a constr~ction method for dislocations in QCs. The Volterra process for dislocations in coiiventional crystals consists of four steps: (1) Cut the crystal along an arbitrary surface (which i s bounded by the dislocation line L, forming two lips C1 and Cz. (2) Displace these two lips relatively one to the other by a urgers vector of the dislocation line. (3) Fill the space between 61 and C2 with perfect crystal matter or re ove extra matter; (4) Glue the lips together. Because is a lattice vector, this process restores the lattice everywhere outside the dislocation line L except for the continuously field around the dislocation line L. stress fields around the pletely defined by the line and the elastic properties of the crystal, but are
/nfPrmefallic Cornpocdv: Vol. 3, Principles Lmd Practice. Edited by J. II.Westbroolc arid R. L. Fleischer. @ZOO2 John Wiley 6t Sons, Ltd.
380
~ e c h a n i ~Properties ~L
d re 1 Densely d i s t r i ~ u t edislocation ~ lines in Al-Pd-Mn icosahedral single quasicrystal bcnt at 720 "C
independent of the special choice of 2'2. Such a Volterra process can not be performed in QCs themselves because tr~~nslation s y ~ ~ e t isr ylost. Since a Burgers vector of a dislocation in Cs is a translatioii vector in
haded strip S is the projection window that has an irrational orientation with respect to the the lattice points lying within this d onto the physical subspace PI1 to form the QC. The Volterra process constructing a dislocation line Lil with a consists of the following steps (see Figure 2): dislocation line Lit, described by a in physical subspace Pll into a -location byperline L in ND hyperspace. L is a direct sum of 5 ,11 and the perpendicular subspace PL:
Schematic d i a g r a ~describing the Volterra process in ND hyperspace constructing a dislocation line Lll in 3 0 physical (parallel) subspace PI1 with R ND Burgers vector The vertical line represents the (N - 3)D pe~endicular subspace PL. The l D dislocation line Lll in physical subspace, designated by a filled circle i s lifted into ND space forming a (N - 2)D hyperliiie L, The 2D cutting surface Cl1 and the corresponding two lips CI and Cl are lifted into (N-l)D hypersurface t: and two hyperlips CI and Cz. The shaded strip S is the projection window, that has an irrational orientation with respect to the ND hyperspace. All the lattice points lying within this window are projected onto the physical subspace PI! to form the QC
vector of the dislocation line ,511. This ND Burgers sists of a p~ionon-type physical and a phason-type ~erpendicular
(2) Feuerbacher et al. (1997) introduced a strain {=I a c c o ~ ~ n o d a t parameter io~
L = L" @ P-'
hyperspace along an arbitrary surface C which is bounded by n hyperline L, forming two (N- 1)D and C2. The projections of the rsurface t: and two (N- l)D hyperlips 221 and 2'22 onto tlie physical subspace
vector, this process restores the N D lattice everywhere outside the dislocation byperline L except for the continuously di~tributedstrain field around the dislocation line L. Notice the displacement vectors are ND vectors (11) consisting of phonontype physical compone~tsu(d) and phason-type perpendicular components vv(rl1):
(3)
Dislocations ir2 Quasicrystnls
38 1
where the phason type dis~lacementsw(r11) lie in the perpendicular subspace PL and cause some lattice points to move into the strip S and some other lattice points to move out from the strip Project all the lattice points lying within the strip S onto physical subspace Plf, forming a quasiperiodic lattice co~itainingthe dislocation line LII. must be emphasized that: The displacement field is a function of the position vector d in the physical subspace Pi' around the dislocation line defined by the line I;//position, and the elastic properties of the QC, but is independent of the special choice of C .
cutting surface Ell, and this dislocation is not a perfect, but a partial o erfect dislocation. he displacement field , after traversal of a c C surrounding l;ll in Pll:
(P")
=
(4)
Figure 3 Schematic diagram of defocused convergeiit-bea~ electron diffraction. The crossover C of the ~onvcrgentincident beam is deviated from the specimen (Sp) by a distaiice Af (defocus). Thus the diffraction pattern (EDP) at tlie back focal plane of the objective leiise (Obj) contains both diffraction and real-space information of the illuminated region
2.2.1 Defocused CBED D e t e ~ ~ ~ n a qf t~on Vectors of D ~ s l o ~ a t in i~n~~
Figure 3 shows a schematic diagram of a defocused CBED. A conical incident electron beam with a half angle 0.4"-1 .O" converges at the crossover C which deviates from the specimen (Sp) by a distance Af (defocus). Thus a circular region of the specimen is illuminated by the incident beam. If the illuminated There are three methods of experimental detennination of Burgers vectors of dislocations in quasicrystals: (1) Defocused convergent-beam electron diffraction (CBED) method. By this method all the inforination about a Burgers vector, including its direction, sense and magnitude, can be determined. This method is appropriate for isolated inclined dislocation lines. (2) Diffraction contrast method. Usually only the direcurgers vector may be determined by this method. If coiiiplemented by computer simulation of the contrast images of a dislocation, full information about a Burgers vector inay be obtained, This method is appropriate for inclined dislocation lines whose density niay be higher. (3) High-resolution lattice fringe method. This method is appropriate for nearly end-on dislocation lines. When there are two neighboring zone axes with a small angle for high-resolution imaging, full infoimatioi-z about a Burgers vector may be obtained.
ion of the displacement field around the dislocation DlD2. For diffe nated points, the displacement vectors incident directions are different, resulting diffracted and transmitted intensities, and one observes a correspoiiding contrast In (l-zkl) diffracted and (000) transmitted disks as shown in Figure 3 where the points D1 and D2 in the specimen have their corresponding points D1' and D2' in the (000) transmitted disk and points D1" and diffracted disk. Therefore, a defocused contains both diffraction and real-space the illuminated region. As described and reviewed recently by tioii line along tlie direction U with a Burgers it and twist of the briglit diffracted disk and their
correspond in^ dark de~cientfringes in the t r ~ ~ s ~ ~ i t t e d
physical components and
~ e t e r i ~ i i i iby ~ dthe inner er-order reflection.
if G is a weak
ED pattern taken from a perfect an A170,Pd,, 2Mn,, IQC. Indices G, o f rcflection numbered as I = 1, 2, . . ., 17 are listed i n Table 1
modulus of the value the criterion may be expressed as follows: If a weak higher-order reflection fringe corresponding to a ~ ~ $ i o ~ ~line t i oofn a will split into In1 -+ 1 brandies with In\ nodes wheii
of reflection fringes G , 111 Figures 4 and values for a dislocation line shown in Figure S(a) ineasurcd froin thc &focus CBED pattern sliown in Figure 5(b) (111: not i n t ~ r ~ ~ ~ t ~ d ~ M,
= rl.
linearly i n d ~ p e ~ ~ N e nrec~procal t vectors GE2,. . ., G 2 ~(1E= 1, 2, . . ., N ) whose fringes interct the dislocation line. The indices G, of the fringes may be obtained by computer simulation of the '
1 2 3
4
s
6
a
7
6 1 1 -3 -4 -3 3
11
and the values
32 13 14 J5
17
from a perferct region ol' an A170,4Pd2i~zMn, IQC. The bright. circular disk of 0.6" radius is the transmitted disk and each bright fringe ~ ~ ~ ase 1 ~2, e. . d., 17 1s a niain reflection fringe in the respective diffraction disk.
2 2 -4 -6 -4
8 -4
-8 -4 4 6 I I) 5
-3 -5 -2 3 7
--I
-3
-12 6 -2 -3 5 1
-6
-4
-4 4 2 -6 4 2 2 4 -2 8 -4 -4 -6 2 -3 3 1 -3 3 -1 2 2
-5 5
-7
-1
5
1
-1
-3
1
2
2 4 -6 -10 -8 12 8 1
3
-6 8 2 -6
1
0 nl -2 ni -2 (1
nr
E 111
high temperature. hen the d e f o ~ u s ~incident d beam illuminates this dislocation liiie, one obtains ii defo-
side of the dislo~ationline twists opposite direction and hence we ~ i ~ ~ l a toifothis n ~ ~ twe~deduced ~ r all n the indices of the reflections appearing in Figure 4 as listed in Table 1. age of a clislocation line deformed plastically at
not intersect the dislocation line, hence they do not split and we cannot btain the correspondiiig experimental values of Gi,= nj for i = 4, 6, 12 and 14. AIL
Ddocutions in euasicrystuh
383
1994b; Feng et al., 1995; selecting ( N - 1) linearly al., 1994b; Wang, 2000). (i = 1, 2 . . . N - 1) under independent reflections ast of the dislocat~onis which the diffraction c iiivisible or weak, then we have (6)
e2
LGN-1.1
Figwe 5 (a) Contrast magc of c?. dislocatmi linc i n an Al,, Pd,, Mn, IQC deformed plastically at lt~gh-teiii1-’el-ature. (b) Defocused CBED pattern takcn front the dislocation line shown in Figure 5(a). Indices Gtl of reflection fringes the G, . B = n, valucs are listed in Tablc 1
the information is listed in the last columii of Table I. There are many possibilities of selecting six linearly from the nine experimental values lving for any select arne Burgers vector =(1/2)[~3~~2~].
2.2.2 Burgers Vectors qf ~ i s l o ~ a t in i ~ QCs ~~s Determined by a Contrast Experiment ~ e c h n i ~ ~ e
ED technique described above is appropriate for isolated dislocatioii lines. Iii case of densely distributed dislocations, especially dislocation dipoles and small dislocation loops, one may use a contrast experiment technique (Feng and Wang,
GN-1.2
...
eN
* * *
GN-1,N
where fi is a coefficient dependent on the niodulus and , G, thejth component o f sign of the Burgers v the reciprocal vector eJ tlieBth unit vector of an orthonormal system. sing properties of the deterniinaiit one caii confirm that the vec equation (7) satisfies equation (6) 1994b). The technique of matching the c o i ~ p u t e ~ ~ s i m u l a t e ~ dislocation coiitrast images to e~perimentalelectron micrographs (Head et al., 1 vectors of dislocations was al. (1994b) identified the dislocation by using both d ing techiiiq ues. By matching c o n i p ~ t ~ r - s i m ~ ~ l a t e ~ contrast images to experimeiita loops were identified in Al-Si1991) atld Al-Pd-Mn IQC (W By coin bining defocused techniques we successfully identi~eda dissociation of exteiided dislocations of the forin 1/2[ 1?OO? I] --3. i,qi31111] 1/4[iiii31] with a sLacking f a ~ l t between these two partial dislocations (Feng et aE., 1995).
+
2.2.3 Burgers Vector Determination by the High- Resoht ion LLIt tice Frmge Method As pointed out by Yang et aE. (1998), in a highresolution lattice friiige iinage obtained from an inverse Fourier transform by selecting the reciprocal E , there are Inz,/extra dislocation with a Burgers vec Thus we can es from high-resolution lattice fringe images of a disloca Fourier transform by selecting a w we will discuss how to deter ni. Figure 6 shows s c ~ e ~ ~ t an i c edge a ~ ~ ~
384
~ e c ~ i a i t i Properties ca~
image there are only four linearly indepenTherefore, we need at least two HRTEM images with the dislocation line being nearly end-on. Owing to the high symmetry of the icosahedral point group, we can choose a dislocation with a line nearly along a three-fold axis and get its images at this three-fold axis and also at a neighboring pseudo-two-fol~ axis. Since the angle between these two axes i s only 10.8", the dislocation may remain almost end-on for both eases. At last, selecting any six linearly independent j, and the experimentally determiiied i t i into equation (9,one can solv . ., BG of the 6D Burgers vector the dislocation line. ecently, Yang et al. (2000) determined experimen== [-2, 0, 3, - 2, 3, 01 in a n quasicrystal by the highresolution, lattice-fringe technique described above.
...-.....
........*
I
ic diagram showing the definition of the an edge dislocation with a line direction U
By using the defocused CBED technique and other
dislocation with an extra half-plane. According to the University (Wang and Dai, 1993; Wang ez al., 1994a; Feiig and Wang, 19948; Ding et ul., 1998; Wang, 2000) d e t e ~ i n e da series vectors of dislocation Cu25,5Fe, lines in and in Al,oCo,sNi,s s c ~ u n g s z e n t ~ Ju u~ enfeld et al., 1995; ~ e u e r b a c ~ eetr al., 1997; Urban ex al., 1999) determined a series of Burgers vectors of dislocation lines in annealed and h i g l ~ - t e ~ p ~ r a t uplastical re AITOPdzl Mn, IQCs. The deter~ined tors are listed in Table 2 where the superlattice constants a % 1.3 nni for the fkce-ceiitered AI-Cu-Fe and AI-Pdrelated to the periodic of dislocations in Al-Co-Ni DQC. In Table 2 we also ) images. In their lattice
d 5 in Yang et al., 1998), obtained by inverse Fourier transform in each figure by selecting 15 difTerent reciprocal vectors Gl with i = 1, 2, , . ., 15, extra fringes are clearly seen. We find the 30 lattice fringe images are all in excellent reement with the sign criterion expressed above.
1997) defined as the ratio of the modulus o f the perpendicular phason to the modulus of the p of a high dimensional a dislocation in a QC. Notice that usually the urgers vectors in IQCs possess rather large 5 values ranging from z3 to 2' (z = (1 + &)/2) indicating a large phason component of the strain field around a dislocatlon line. According
385
Dislocations in Quasicrys tals ~ x ~ e r i n i ~ n t adete~mined lly Burgers vectors
32
DQC DQC DQC DQC DQC DQC
= (~00000) 4 = (-00000) = ~00000)
I I 1
0.4 = (10000_0) D,1 = ( 0 1 0 ~ ~ 0 ) 0.2 = (0111 10)
I1 II
0,2 0,3
BJ, 234, B5, B,5]of dislocations in ~uasicrystals
Al-Pd-Mn, Al-Cu-Fe Al-Pd-Mn, AI-Cu-Fe AI-Pd-MIi Al-Pd-Mn Al-Cu-Fe Al-Pd-Mn AI-Pd-Mn Al-Pd-Mn Al-Cu-Fe Al-Pd-Mn
0.296 0.183 0.1 13 0.070 0.215 0.10% 0.174 0.067 0.513 0.098
= (1/4)(515151) 0.1
= [ B I ,3 2 ,
I
to the statistics provided by osenfeld et al. (3995), Feuerbacher ef al. (1997) a Urban et al. (1999), about 90% of the dislocations investigated are perfect dislocations and show two-fold urgers vectors, i.e. their directions in the physical su pace are parallel to a two-fold axis of the IQC. Slip-plane normals PI of an
five-fold axis. Less frequently, two-fold and three-fold slip planes were also observed. Thes densely packed planes of an A1-Pdincreasing plastic strain from 0% to 1% the frequency of occurrence of dislocations with higher accommodation ratio ( increases. For example, while the frequencies of occurrence of dislocations with I; == 2 3 , z5 and 27 are 34%, 60% and 6% respectively for the specimens annealed at 760'6, they become 094, 64% and 36% respectively for the specimens compressed at '760 'C to a strain of 170.They remain almost constant when plastic strain increases still fbrther.
0.20 0.40 0.60 0.80 0.19 0.30
0 0 0
Al-Co-Ni Al-Go-Ni AI-Co-Ni
0 1iz. 1 p
correspond to trans~ations,distortions and rearrangements of unit cell (Levine et al., 1985; 1986). In the higher-dime ND displaceme~tvector (equation (3)). Then we have
where V = ejV, with V, = a/ax, = ai being a differential operator relative to the position vector IP// = (XI, x2, x3). can be decomposed into symmetric and antlsymmetric parts. The antisymnietric part describes a rigid rotation that does not change the elastic energy. Since there is an energy cost associated with the a n t i s y ~ e t r i c derivative of w, the elastic en components of the gradient of Thus, the elastic energy depends only on the following strm1s:
It is easy to verify the com~atibi~ity e ~ u ~ t i o( n al., 1993, 1994):
s have two types of elastic hydroone is the phonon variable u and the other is the phason variable w. Moreover, the excitations analogous to dislocations do top0 as well. Phonons, phasons and dislocations exist
where ejJlcis the alternator symbol. ~ u r t h e ~ m ~the re? elastic energy density can be expanded in terms of the Taylor series in the vicinity of Ezj= 0 and wlJ= 0 to the second order:
MechanicaI Properties
384
T, there is another stress tensor H with components Hz, along the x, direction in PI- acting on the surface orthogonai to the xJ direction in PI/. If n is the outward unit vector normal to an element surface, then we have
(17)
t, = Tlpj h, = H@,
As Lubensky et aE. (1985) pointed out, in the case of QCs the theorem of‘ momentLlm possesses the form
-1 d
p(ir+w)dV=
df.v
J .v
where V i s an arbitrary volume in PI! and S is the boundary surface of V. Application of Gauss’ theorem to equation (18) leads to two equations of motion
where
+ g , = pi+,
C$TzJ+J1 = pii,
(19)
and two static equilibrium equations are quadratic elastic constants in the classical elasticity theory with C I J k l= c k l I J = Cji,cr c i j l k . We Call denote all Crilrfby a symmetric matrix [C]. Similarly, =z1
(13) are elastic constants of the phason field with fir!&/ = which can also be decoded by a symmetric matrix
are the elastic constants associated with the phonoiiiously, Rllkl = RJlfcl,RijkE= R/iiik3 f fi)kEEjI R b k l $ Riilj-w e denote and ER’] with [R’]= [RIT.These four matrices compose a matrix [C, I(,RI:
i.),T,,+JI = 0
(20)
The theorem of angular momeiituni associated with the phonon field has the forin dY=
s,.
rli x f d Y +
Hence, by Gauss’ theorem and equation (19) we obtain T,\ = 1;i
(22)
This means that the phonon stress tensor is symmetric. ) transform differently under different representations of the point group (niore precisely, the former transforms like a vector, but the latter d the product represenlations, contain any vector represent~tions.This implies that for the phason field there is no equation aiialogous to (21), from which it follows that, generally, HIJ
ing et al., 1994: Yaiig et d.,1993). With these notations, we can write the elastic energy in a compact form:
+
h’&YiJ g, = 0
# HJI
(23)
By an argument like that given in classical elasticity theory we obtain
Substituting equation (1 1) into the above equations gives the generalized The m o v e ~ ~ of n t a t o ~ through s a barrier, ~eading to a rearrangement of the unit cell, needs some forces. Thus, besides the conventional body force density f and surface force density i n Pll, a generalized body force density g and surface force density h in P-Lshould be introduced in calculating the elasticity of QCs. Similarly, in addition to the conventional stress tensor
Tzj
HLJ
= c i j k ~ ~ k+l Rijkl =~
~
+~
“fJ/Cl
wkl
“h~ I
which can be expressed in matrix forni
(25) ,
~
/
~
i
387
Dislocations in Quasicrystals Substituting equation (25) into (20) gives the nonhomogeneous partial differential equations satisfied by U and w: Czjk/@%uk
Rklya&%uk
+ Rzlkla,alWk +.6 = + KzJk/qalwk f gz =
(27)
3.2 Elastic Constants of Various Laue Classes of QCs The number of independent components and their explicit forms in the elastic constant tensors can be calculated with the aid of group representation theory. From group theory one knows that the number of independent tensor components is equal to the number of times that the identity representation is contained in this tensor representation, i.e.
r = rs+ r, + r7
(30)
+
It follows that a vector in PI1 transforms under r5 rl, whereas a vector in PL transforms under r7. The tensor characters for the elastic constants Cjjkl, KijkI and Ro/clcan be derived as follows:
x(e)
x(a)
x(a2)
21 21
3+2& 3+28 -6-4&
1 1
36
x(a3)
3-2d 3-2& -6+4&
0
x(m4) x(P) x(@) 5 5 4
5
3 0
5 3 0 (31)
From equation (28) it follows that the numbers nc, nK, and n~ of independent elastic constants Cijkl, Kijkj and Rijkl are
nK=4,
n,=5,
nR= 1
(32)
To determine their explicit forms we should notice that the transformation properties of Cijkl, Kijkr and Rijki follow directly from those of Eij (T'j) and Wij (Hij). If we find the precise components of Ei, (Tij) and Wjj (Hij) that transform under the same constituent representation, we can construct all the invariants formed by their c.ombinations and then establish the independent components of Cijkl, Kukl and Rijkr. For the phonon field six symmetric components of El, transform under
w, + r,)x (r,+ r,)is= 2r1+ rs+ rs
(33)
where {. . .}s denotes the symmetric product representation. This means that E11 E22 and E33 form two ID subspaces (the identity representation 1 ) . They are two linear invariants. E11 - E22 and 2El2 form one 2D subspace (r6) giving one quadratic invariant (scalar product)
+
- 0 1 0 0 0 0 0 1 0 0
o o o
r(E)=
1
o
- 1 0 0 0 0 - 0 0 0 0 10 0 00 - 1 0 o o -1 o o 0 - 1 0 0 0 0 0 0 0 1
-1 0
r(p)=
0
0
(E11 - E22HEll = (E,, -E&
- EZ?)+ (2E12)W12)
+ 4E:2
and E23 form another 2 0 subspace another quadratic invariant
.E13
(34) (r5) giving
E73 + E:3
(35)
Thus, there are a total of five quadratic invariants due to the phonon field (Ell
+ E2d2>E L , (El1 + E22)E33, E:, + Et3,
EllE22
- E:2
(36)
Among them the first three are products of two linear invariants: Ell E22 and E33. From equation (36) it can be see that the non-vamshing components are
+
388
~ e c I ~ u EProperties i~~1
K= IK1 K. K2 0 0 0
which can be written in the matrix form
\o
Kl 0 0 0 0
0 0 0 K4 0 0 K1 + & + K , 0 0 0 rc, 0
0
(44) where the double indices labeling the phason strains are arranged in the order 1I, 22,23,12,13,21. Finally iiotice that the pairs (E11 - E22, ~ E Q and ) (Wl1 W22, W ~-I W12)transform acco~dingto the same representation (r6).This means that there exists an invariaiit
+
(Ell
where the subscript 5 stands for the number of independent components. The correspondences between iiidex pairs and single indices are, as usual, ( j j ) = 11 22
i = l
2
33 23 31 3 4 5
12 6
(39)
iniilarly, for the phason field six coniponents of Wij transforni under
I - ?T22 and W21 t FVl2 form two different ID subspaces (r3 and r4) giving two quadratic iiivariants
w21 - 'c.1/12) and (T'T/139 W2.3) pairs (kV11 form two different 2 subspaces (r'6 and ]r7) giving two other quadratic iiivariants W k 9
Thus, there are a total o f four quadratic invariants due to the phason field. They are listed in equations (41) on-vanishing components are
The correspo~di~g matrix form i s
- E22)(Wll
+ W22) + 2EIdW21 - w;,)
(45)
coupling El, to Wif.Non-vanishing components are Rl,,,
= RI122 = -R2211 = 4 2 2 2 2 = RI221 = R212r - - R m = -R2112 = R
(W
with the matrix form
R=
0 0 0 0
0 0 0 0
0 0 0 0 0 0 0 - R
0 0 0 O
0 0 0 R ,
I
(47)
Therefore, it can be seen that there are 10 quadratic iiivariaiits and hence 10 independent second-order elastic constants for 8~~ s y ~ m e t r yAmong . them five components (nc = 5) are due to the pbonon field, four constants (IZK = 4 ) due to the pfnas~nfield, aiid one constant ( E R = 1) associated with the ~honon-phason coupling. The results coincide with those given in equation (32). Table 3 gives all the point groups of2D QCs. They are divided into 18 Laue classes which belong to 10 systems. 111 the same way, we can determine the numbers of the independent components and the explicit forms of the elastic consta~tsof various Laue classes of QCs. All the results are listed in Tables 4 aiid 5.
In this subsection we apply the generalized elasticity theory to the case of icosahedral QCs. Elastic constants of icosaliedral QCs have been discussed by several authors. The elastic constants depend 011 the choice of the coordinate system. Unfortunately, a
389
Dislocations in Quasicrystals Table 3 Systems, Law classes, point groups and numbers of independent elastic constants €or 2D QCs
__
No. of Laue classes
Systems
-
Orthor~omblc Tetragonal Tr1gonal Hexagonal Pentagonal Decagonal Octagonal Dodecagonal
groups
- - -
Nc
NK
21 13 13 9 7 6 7 6 5 5 5 5 5 5 5 5 5 5
21
--
1,
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Triclinic Monoclinic
POlllt
NR
____
._ -.
i
2, in, 2/m 12, 1111, 12/m 2rnn1, 222,mmrn, mm2 4, 4/1n 4mm, 422, 4%2, 4/mmm 3, 3 3rn,-32, 3m 6, 62 6/m (Iiiim, 622, 6752, 6/inmm 5, 5 5m,52, 5m 10, 10,10/m lOmm, 1022,-10m2, 10/rnmm 8, 8, 8jni Smrn, 822L8m2, 8 / ~ m m 12, 12,12/m 12111m, 1222, 12rn2, 12/mmm
13 12 8
7 5 7 5 5 4 5 4 3 3 5
4 5 4
Sun?
36 20 18 10 10 5 12 6 8 4 6 3
78 46 43
27 24 16 26 17 18
1 0
33 16 12 10 9 12 10 10
0
9
2 1 2
particular choice for the coordinate system 1s somewhat arbitrary, and at least two choices appear in the literature. A coordinate system can be chosen with the z axis pointing to a vertex of an icosahedron (Levine ak, 1985; Ding et al., 1992). Another coordinate system has the x, y and z axes parallel to the three twofold axes of an icosahedron (Lubensky et al., 1985; Jaric, 1985). Here we use the former coordinate system to express elasticity tensors, i.e. its y and z axes are parallel to two- and five-fold axes, respectively. The corresponding elastic conslants ;we given in Table 5. In this case the generalized Hooke’s law is given as follows: T,, = 1 2 0 T’22
= A8
T’,, = A8
+ 2 , ~ E , l+ R(31~1+ 82~2+ + 2pE27 + Piz(Ol~1
+ + 83~1)
8 3 ~ 3 &3tl.’1)
8 2 ~ 2 8 3 ~ 3
+ 2pE3, - 2R8,w3
+ +
T73 = 2,~E23 R(82w3 - 8 2 ~ ~ 8 11 ~ 2 =) T32 = 2@31 R ( & w ~- $w, 8 1 ~ 3 = ) ‘Ti,
+
+
Tlz = 2,~1E12 R(alw2 - 83 W? - 8 2 ~ 1 = ) T21 = R(E11 -E22 +2E31)+251;alw1 +K2(dlw3 + a , ~ l ) H22 = R ( E 1 1 - ~ 2 , - 2 ~ , , ) + K ] a ~ ~ ~ + K 2 ( a-a,w,) ~W1 f& = X(E1, 4- E22 - 2&) (K; K2)03w3 ff23 = -2RE12 (Kj - K2)83~~2K2(82~l- i 3 j ~ 2 ) ff3l = 2RE31 &(a1 w1 - a,w,> (Kl 1Y3 f f 1 2 = -2R(E23 E , ~ ) rqa2w, ~ ~ ( 0 ~ 1 . 1 )a21.1J3) ~ .ff,2 = 2RE23 ( K , - K~)&IV,- K2(82~1 31~2) ff13 = R(E,1 - E22) K2(a11~1 8 2 ~ 2 ) ( K , - K 2 ) a 3 ~ , ff2, = 2R(E12 - E23) - K 2 ( 8 3 ~ 7 82~3) KID,PV~ (48)
+ + + + + +
+
+
+
+
+
lasrtic
&)a,
+ + + + +
In order to find out the elastic fields induced by dislocations in QCs, we nmst solve the inhomogeneous partial differential equations (27) with given boundary conditions as people have done in the elasticity theory of dislocations in crystals. In the derivation. that
Mech ~ n i Pvuprties c ~ ~
390
Elastic ~ o n s ~ a nof t s 2D QCs with noncrystailographic sy~~metries
All 2D QCs with nonc~ystai~ograplllc syrnrnetries have the same phonon elastic constants (CtJk()listed below:
Laue class 12
All the same as above except for Kc, = Rz = R3 = R5 = 0 ( 2 jlX l, 1X1). All the same as above except for rCr, = R2 = R4 = Rc,-- 0 (211x2, ypz d_ x2). Laue class 13 K6 = Kl = R3 = .& = Rs = S6 = 0 Laue class 14 Kfj = K7 = R2 = R3 = R4 = 115 = jP6 = 0
Laue class 15 Phason field Ktjlc/
Laue class 11 (Thc numbers of Laue classes i s the same as in Table 3) Phason field K,j/cl 11
22
23
12
13
21
11 22 23 12 13 21
11
22
23
12
13
KI
0 0
K5 -K5
0 0 0 0
0
K2 K1 0
K5
-K5
0
0 -K5
K2
K5
where C = K1
K
4 0
0
0
C 0
0
K
4
21
Kj
-& 0 K3
0
K3
0
c
12
13
21
112
0 0 0 0
-R2
+ Kz -I-K3
Phonoa-phason coupling R i j k l
11 22 33 23 31 12
Phonon-p~asoncoupling Rljk/ 11
23
22
13
12
21
11
22
23
RI
RI -RI 0 0 0
0 0 0 0 0 0
-R1 0 0 0 R2
R2
-R2
0 0 0
-R1
0 0
R2
0 0 0
RI
Laue class 16 Kj = R2 = 0 Laue class 17 The same phason elastic constant as Laue class 15, but no constants associated with the pl~onon-phasoncoupling Law class 18 The same phason elastic constant as Laue class 16, but no constants associated with the phonon-phason coupling
follows we assume a QC (body 7) to be an infinite, homogeneous and free body. Thus, the boundary condition T,,11, = .ElzIn, = 0
(50)
According to the eigenstrain method (Mura, 1987), a subdomain 7’ in the body z i s subjected to eigenstrains E ; and Wc during the Volterra process. The actual strains in domain z, E,J arid WzI,are the sum o f the eigenstrains E: and W; and the elastic strains E;j and W’iI, i.e. --j.
00.
Et/ I= Eij +El:’
Wit
Wi,
and EllJ by the generalized Wooke’s law equation (25), we obtain
+ “v;
(51)
Since the elastic strain fields E;i and W ; induced by the dislocation are related to the internal stress field Tij
Tij
= cyklEkl
$- R i j k t w k l
€$zj
= RI,It$&
+ Kijkl wkf - ~kitjEk*I- K j k l wz,
- CijkEE?l
.+. Rzjklw?E
(52)
Substituting equation (52) into equation (20) and considering ji = gL= 0, we have ~
i
j
Rt,I,+%’k
k
3-~R ~ ~l k l~~ l ~~ l= w~k~~~~~~1~~~ k $- ~ ~ wfl, k (53) -I- ~ ~ k = 4c/,ja/-Gl ~ ~ 4-~ K,lx ,14w;, ~ ~
By comparing equation (53) with equation (27) one can see that the contribution of E$Land W & to the equilibrium equations is similar to that o f two body forces X,and YE:
~ ~
~ k
39 1
Dislocutions in Quasicrystals
Table 5 Elastic constants of lQCs (the coordinate system is chosen with the y and z axes parallel to two- and five-fold axes of the icosahedron) Phonon-phason coupling
22
33
23
31
12
~ t - 2 ~a
a a
0 0
0 0
0 0
p o o
o p o
o o p
11 11 22 33 23 31 12
a a
i+2p
i
a+q
0 0 0
0 0 0
o
o
0 0
o
33
23
31
12
32
13
21
R R R - R - R R
O
O
O
O
R
O
11
11 22 33 23 31 12
o
22
Rijkt
0 0 0 0 - R O 0 0 - 2 R 0 0 0 0 0 0 0 0 0 0 0 - R R O - R R - R O 0 R 0 0 0 0 0 0 0 - R O - K O 0 R
Phason field K&
11
22 33 23 31
12 32 13 21
11
22
33
23
31
12
32
13
21
K] 0 0 0 K2 0 0 IG 0
0 plii
0 0 KI +K2 0 0 0 0 0 0
0 0 0
K2
--kfi
0 0 0 Ic, 0
0 0 0 0 0
K2 Ic,
0 0 0
0 0 0 0 I(1 - Ic, 0
-K2
0 0 -K2 0 0 K2
0
K1
- K2
0 K2 0 0 --K2
0 0 KI - K2 0 0 0 0
-112
-K2 0 0
K1 - K2 0 -K2
o
0 0 -K2
0 K1
if we define A@@, Vcr and
Upon Fourier transformation equation (53) beconies
M,kgk
RLkG/c
R& 3. N t k W k =
2,
r,
(55)
and
where an overbar represents the Fourier transfori of a function, i.e.
.m> =
where
+f(k)exp(ik.x)d
where
Equation (59) is a set of inliomogeneous algebraic equations, the solution of which can be expressed in terms of Green’s functions as
and
where
2, = -iCi,,lk,&:L F, = -iRk,ljk&l
- iRIJklk, Wll - iKrJklkj@f,
163) (58)
Equation (55) can be written in a compact forin (59)
Mecltan ical Properties
392
When we transform equation (62) from Fourier space back to coordinate space, we find the general expressions for the elastic ~~splacement fields induced by Cs (Ding et al., 1994, 1995~): w
)Zfl (x’)dx’
(65 )
-#
42.1 The ~ e n e r ~ l i z a t iof’ o ~ ~ ~ h e l b ~ 3~ ’ ~s t h o d (Diigg et al., 1995a)
From the generalized elasticity theory of described in the preceding section, we know tha linear partial differential equatio phonon and phason displace~~ents follows (we assume J;: and gi in equation (27) to be zero):
with Green functions
~ u b s t i t u t i n equations ~ (54) and (60) into (65) and ~ntegratingby parts, we have
where we choose a dislocation coordinate system in which the .U;? axis is along the straight dislocation line in physical space. it should be noted that the elastic constant tensors ClJkl, KIJkland R,/,I in equation (68) are all referred to the dislocation coordinate system and the subscripts with small letters (i, k ) take the values 1, 2 or 3, while the capital letters (J, L ) take tlie values 1 or 2. Using
vqx) = 6 ~ ~ ~+~~ ~( -x3)w , ~ ( x )
(69)
(see equation (60)) and setting
q’
6r($cTj,c,
+
(’i-3Ri/kl)
+ CSP-~(~~)R/~*~, +
@-34jkl)
(70) The following are noteworthy: (1) The body forces X, and YL(equation (54)) both act at the point x’ in Pi’, but the directions of X,and Y, are along the axes in PI1 and P’, respectively. (2) Gap(. - x’) represents the a component of the displacernents at the position x in PI1 produced by a m i t point body force along the fl direction actiiig at the point x: in PI. (3) The elastic constants in the dislocation coordinate system are not, in general, the same as those in the conventional coordinate system of QCs, and accordingly a coordinate transformation must be performed.
Equation (68) can be written in a compact form (7 1) According to Eshelby et al. (1953), equation (71) has solutions of the type
vqx) = A’If(q)
q = XI +pxz
(72)
where p is an undeteriiiined parameter which we will find below. If we substitute equation (72) into (71) and introduce a inatrix with the elenients
then we obtain a set of linear algebraic equations for A as given by
As we know besides the Green’s function method given in section 4.1, there are other methods in the case of ordinary crystals, such as Eshelby’s method 953; Hirth and Lother, 1968) and troll, 1958, 1962; Bacon et al., 1978), which may have an advantage over the Green’s method for some actual calculations, in the following, e will generalize these two methods to the case of Cs and give some general expressioiis for the elastic fields of straight dislocation lines.
a@AP
0
(74)
It follows that the parameter p in equation (72) i s determined by the condition de;ltla”Pl = 0
(75)
which is an algebraic equation for p . energy is always positive and the coefficients are all real, the roots p(n) of equation (75) are necessarily coinplex occurring in conjugate pairs. Following the analogous
Disloca Eiopzs in &uasicrystals procedure proposed by Eshelby et al. (1953), the displaceinents induced by an infimte straight dislocation line parallel to the x3 axis can be expressed in the form AI-'(n)D(n)In
,4
(76)
where Re means that only the real part is to be taken, the sign of 2ni is taken to be the same as the sign of the imaginary part of p(n). The constants D(M)are determined by the following equations:
r
6
393
Since the elastic energy (equation (16)) is positive, i.e.
the symmetric matrix
is always positive-definite. The elements of [C, K, RI are none other than the terns in ~ q ~ a t i o(70). n It follows that there exists a real, non-zero vector a in PI1 and a real, non-zero vector P = + P~ such that
1
where b@are the components of the Burgers vector the dislocation. Furthermore, if we introduce the symbol
M; = "Fl,+ 6:-"Hg,
(78)
the generalized Hoake's law (equation (27)) can be written as
Ad; = ~ ~ ~ ~ , ~ ' ( x ) (79) By the same method as Foreman (1955) used, the elastic energy per unit length of the dislocation can be easily obtained as
4.2.2 The Generalization of Stroh's Method (Ding et al., 1995b)
The anisotropic elasticity theory of dislocations in crystals developed by Stroh (1958, 1962) has played ail i ~ p o r t a npart t in the theory of defects in solids. Bacon et al. provided an overview on the Stroh theory in their review article (1 978) in which the orthogonality, completeness and invariance relations for the Stroh eigenvectors, and the integral formalism were derived in considerable detail. In what follows, we shall extend the 6D formalism o f Stroh to a higher-dimensional one and then give an integral representation of the elastic fields induced by dislocations in QCs, which may be more suitable for numerical calculations. For definiteness we consider icosahedral QCs.
This means that the matrix
is positive-definite. This result guarantees the existence of an inverse matrix ( a a - ' . Moreover, if we again define another non-zero vector in PI,then it is easy to prove that the permutation s y ~ ~ e t ~ofi the e s elastic tensor indices (see equations (12--14}) assure tliat
(ad)"I-'= ( d ~ ) B a
(85)
We assume that a QC is infinite, hoiiiogeneous, anisotropic and a free body. Let mutually orthogonal unit vector parallel to a straight dislocation line. Let x denote the position vector from the origin to a point in N.In the present case, U and vv, induced by i n ~ n i t estraight dislocation lines parallel to t , are independent of the
displacement field solution of the form V q X ) = AI-'f((n)= APf(
where AB is a 6D constant complex vector, p i s a complex constant andf is an analytical function of its a r ~ i ~ e n t s . Using the expressions ~n equations (69-71), and substituting equation (86) into equation (69), we obtain
+
+
+
{ ( m n ~ ) " ~ ~ ( m ~ ) (tzm)"'] ~I-' p 2 ( p z ~ ~ ~=P0~ A(87) '~ The set of equations (87) has non-zero solutions for AB only if the determinant o f the coefficient matrix vanislies:
ce the above e q ~ a t ~ oi sn a polynomial with real ~ c i e i i t sthe , 12 roots (p = 1, 2, , . ., 12) occur in coniplex conjugate pairs uppose p p ( p = 1, 2, . . ., 6) is equal to the roots with a positive imaginary part, and then Ygf6 = p;, I f we define another vector as
LE
p= 1
y= 1
(34)
+
L; = --[inm)"P ~ l l ( n n ~ aPP J A y where 8 is the aiigle between the basis plane ( ') in the plane normal to a new basis plane ( Additionally,
or equivalently
the n ~ r ~ ~ l ~ z acondition tion
(95)
should be satisfied, which ensures uniqueness for A; and ~ i r ~ ~ e r ~theo fo~lowing re, conjugate relations hold: IJ/L+b
I_
1';
=
;'+6
L;+6
( p = 1, 2 . . . 6 )
$.
Furlhei-niore, we can define three 6 x 6 real matrices and give integral representations of them:
= (LE)*
(91)
t shodd be noted that the sum rule is not available any subscript in equations now on sunii~ationover repeated
the position vector x, then Therefore, in a polar coordinate system we have
+
taken for ,LL = 1, 2 . . .6, - for y using the definition equation (89), equation (87) can be r ~ ~ r i t t easn where
is
with s",L
=
[;*.I
1. Using equations (89) and (96), equation (92) can be expressed in ternis o f Sap and
~ l t i ~ l y equ~tion in~ (98) by mP and nP, respectively, we have
and ~ntegratingthis equation yields the r~presentationof the displacement field 8 1 v q r , 0) = -b"(-Sfl" In r + 4 0 ' " Su[(n~)-~]~idO 2n; ~ r t h ~ ) g o ~ ~ c~o~~l ~i t~y~, e t e iand i e s sinvariance are still and pkL,i.e. valid for A;,
(1009
An uncomplicated calculation on the Fourier transeaiiwhile9inserting equation (98) into equation (199), ) of the Green's functions gives the the resulting stress field is bp following results for k3 = 0: M," = B"P { - ~ ~+s~ ~ ~ [ "( ~ ~ ~ ) - ' ~ ~ ~ ' [ 4 ~ ~ ~ " J I 2nr ~
+ (.m)'Asi"]}
(101)
Consider a straight dislocation line with Burgers vector = (bl , bi, bi, hf , b i ) parallel to the periodic direction. In this case the dislocati~ncoordinate system is with the conventional QC coordinate system. along the periodic direction (x3 axis) without corresponding phason components, the phonon displacement field corresponding to b[ is the same as that in any transversely isotropic co~itinu~m. Hence we need only consider the elastic displacement field induced by the (hl, b!, 0, h f , h i ) . In this case V"(x)(in equation (60)) is dependent of x3, i.e. V" = vol(.xl, xz). For = (bl, 0, 0, bf, 0) the subdomain z' consists of an infinite s e ~ i p l a ~(x'~ e < 0, xk = 0, - 00 e x i < 00). The eigenstrains can be obtained as *
irac delta function and H( -x), the Heaviside step function. Substituti~gequations (66) and (103) into (67), we have
9
exp[i(klx,
+ k2x2)]dkldk2
k3-0
(lo4)
the other G a b = 0 where k2 = k: ki C = (C66K1 - R2#C,1K, - R 2 ) (106)
+
Substituting equation (105) into (104) and carryiii the Fourier integrals (Yao et al., 19971, we can find the
396
Mechanical Proper ties
expressions of the displacement field for h f , 0). Moreover, the expressions corres = (0, b!, 0, 0, b i ) can be obtained by rotating our coordinate system by n/2 co~nterclockwise in the physical space and by 3774'2 m the complenieiitary space. Finally, the expressions for the elastic displacement fields duced by the dislocations with Burgers vector = (h!, b!, 0, hi". h i ) (Ding et al., 1994, 1995c) are *
+
where r2 = .xy xz and ro is the radius of the dislocation core. If a straight dislocation line t lies in the quasiperiodic plane, say, it i s perpendicular to a two-fold symmetry axis, we can choose a dislocation coordinate system (q,x2, x3) such that the x3 axis is along the positive direction of and the x2 axis is along the periodic axis. This system can be obtained from the c o ~ v e n t i o ~system ~l (xi, xi, xi) by the subscript transformation: 1' -+ 1, 2' --+ --3? 3' + 2. After this transformation we find that non-vanis~ing elastic co~stantsin the dislocation ~oordinatesystem are
From equations (71), (72) and (108) it can be seen that equation (74) in this case consists of five equations which are divided into two sets ofequations: one set for A ' , A2 and A4 given by (C1,
+ C44p2)A' + (C13 + C44)pA2+ RA4 = 0 + C44)pAt + ( C M + C33p2)AZ= 0 (109) RA1 + (Kl + K4p2)A4 = 0
(Cl3
wit11 P(4 = P ( l ) , P(2>, p(4), p(7), p(8) and P ( W , and the other set for A' and A6 given by
(C,, RA3
+ C44p2)A3+ RA6 = 0 + (K; + K4p2)AB= 0
(1 10)
with p(n) = p(3),p(6),p(9) and p( 12). ~orrespondin~ly, equation (75) consists of two independent algebraic equations for p :
397
D is10 cat ions in Q uusicrystals
+ 2K4Rh$]tan-“/-”>
2C44K4 X I
and
For simplicity, we consider a dislocation with 0, b!!,0, h i ) , In this case the only equations we have to solve are (110) and (112). Equation (112) has two pairs ate roots (p(3), p(9)) and ( ~ ( 6 ) ~ p(12)). Suppose the roots with a positive imaginary part to be, say, p(3) and p(6) given as (1 16) Elastic displacenlent fields induced by dislocations in QCs with other symmetries have been discussed by a number of authors (Yang et al., 1995; Li and Fan, 1998). where Of
~ubstitutingequation (1 13) into equation (110) and putting ~ ~ ( =3 ~) ~ ( =6 1,) we have 2C44R
A6(3)= c44 K1
- c66K4
- s2
2CMR
A4(6) = c44 Kl
- cfi6K4
+ s2
Substituting equations (113) and (114) into (77) and noting h3 = hi, b6 = b i , we have
Finally, combining equations (60), (72) (76), (1 13), (1 14) and (115) all together, we find the displacement fields induced by a dislocation with (Ding et al., 199%) are as follows:
Quasicrystals are all brittle with a high Vickers hardness between 800 and 1000 at room temperature. Upon heating, the hardness decreases and the plasticity increases. It was found that above about 600°C Al-Cu-Fe IQC becomes ductile (Bresson and Gratias, 1993), and above about 700°C AI-Pd-Mn 1QC does also (Inoue et al., 1994). In the fo~1owingwe will cite some experimental results of compression deformation of Pd,,,, Mn,,, icosahedral single ~uasicrys~a1s, carried out by the Juelich QC research group, summarized by Feuerbacher et aE. (1997) and Urban et nE. (1999). Figure 7 shows true-stress versus true-strain curves at 760 “C and 800°C and at a strain rate of lowss-l. Both curves show the same features. After the elastic deformation stage there is an initial hardening stage YM with a very high hardening rate. Then a maximum stress is reached at M, followed by a pronounced yield drop and a continuous softening. Neither saturation nor secondary hardening observed up to strain h the maximum stress values of more than 20% c~~~~~~ and the flow stress cflow decrease with increasing the temperature T, as s u ~ ~ a r i z in e dFigure 8. Figure 8
described in section 3, the most frequently occurring dislocations in defor ed Al-Pd-Nn TQC are those with ers vector of c = r5 corresponding to .183 nm. The correspond in^ activation areas = AlAx) range from about 0.87nm2 to , which are about two orders of magnitude larger than the area per atom. As discussed by Urban et al. (19991, AI-Pd- n IQC can be described by an arrangement of Mackay-type clusters of a diameter 0.9 nm. Thus the obstacles, which are thernially circumvented or cut by the gliding dislocations, are not singular atoms, but ackay-type clusters.
b
0
igure 7 True-stress versus true-strain curves of AI,, Pd,, Mn, IQC at 760°C and 800°C and at a strain rate of j*-9 ;-I
shows the t e ~ ~ p e r a t udependence r~ of the maximum stress a-max, which decreases from about '750 activation volume V', which is defined as
modulus of the physical component of the Burgers vector of the ~lidingdi~location,and A x i s the distance swept by a dislocation segment of length Al during thermal activation. Hence V'/lbll I = AlAx represents area of the obstacles therma~lycircumt by the gliding dislocations and is called the activation area. The temperature dependence of Y can in good a ~ ~ r o x i ~ a t i be o n described by Pa). As determined by Rosenfeld ct al. (1995) and Feuerbacher et al. (1997) and
most cases as isolated lines, while in ~ i g ~ ~ t e m ~ e r a t u r e deformed AI-Pd-Mn IQCs usually more or less dense dislocation networks are observed. This i s the result of various dislocation reaction processes taking place in the course of plastic deformation. Therefore, by investigatmg one of the most lrequently observed ~ r r a n g e ~ ~of n tthe s networks - a triple node formed by three dislocation line segments - information on the reaction processes a n be obtained. Wang et al. (1998a) dis d possible dislocation reactions in face-centered s from the geometric and energetic points of view. Feng et al. (1995), W'nng et al. (1998a), and Wang (2000) provided some experimental examples of the reactioi~sby i~entifyingthe Burgers vectors of dislocations in triple node arrangement. These are 1/2[OiOlZ2](f = 2 7
+ 1/2[022iOl]~<= 2 )
= 1/2[032023](<= T 7 )
1/2[03?023](C == 2 7 )
+ 1/~[0Z2iOl](C= 6)
= 1/2[OlOiZZ](~ = T 5 )
i / 2 [ i o i i i o ] ( ~= =
of the stress-stram curves of At,, Pd,, ,Mn, icosaliedral single yuasicrystal compressed at a strain rate 10-5 s-* (after Feuerbacher et al., 1997)
(irnax
(120)
+ 1/2[121Zoo](<= 2 )
= i / z ~ o Z i o i z l (--~ ?)
and
(1 19)
+ ~ / ~ [ ~ i i i o =o ~T (~ f )
= 1/2[i0iiio](f = 2)
i / ~ [ i 0 0 2 i ~ ]= (f
(1 18)
+ i / ~ ~ i i i i o o ]=( fr3)
i/2[0122101(< = 2 )
1/2[01Z2io~(f= 2 5 )
(1 17)
(121)
in Q u a s i c ~ ~ ~ t u / ~
Among these dislocation reactions, reactions (1 17)(120) and (122) lead to a reacted product with different { ratios coinpared with the reacting disloc~tions. Therefore, it i s reasonable to suppose that the dislocation reactions give rise to an increase in C at an early stage of plastic deformation, and subsequently a dynamic e~uilibrium between the increase and decrease in the strain-accommodation parameter < is reached. The reaction (122) forms an extended dislocation consisting of two partial dislocations with a stacking fault in between. ere the partial dislocations with Burgers vectors of 1/4[1iil-?I] and 1/4[131iil] a five-fold type with I III = 0 . 1 7 4 1 ~and ~ I / = T ~ see , Table 2. es of'
In order to understand the iniportant role played by dislocations in plastic d e f o ~ ~ t i o nit, is crucial to observe the microstructures in differently deformed QCs. 111 the following we summarize the results observed by transmission electron microscopy.
and Feuerbacher et al. (1997) observed dislocation motion directly by an m-situ straining experiment in a high-voltage transinission electron niicroscope. The motion was viscous with a velocity ran 0.12 prn s-l to 1.2 pins-' , i.e. neither jumps nor local pinning of the dislocation line were observed. The moving dislocations left behind a contrast of the slip trace, which in some cases was observed to disappe~r after some time. The dislocations liad a proiiouiiced tendency to follow each other on the sanie slip trace. No examples of demobilization of ~ n ~ v i nd gis locations by interaction processes of any kind were observed. QLiantitative measurements of carried out by Wollgarten et al. (19 (19951, Feuerbacher et nE. (1 997), a differently deformed and aiineale IQCs revealed that: Dislocation density is d i ~ e r e n t for annealed specimens compressed to different plastic deformations epipl. For example, while the dislocatio~ density in an annealed specinieii is 8 x 107cm-', the dislocation densities of spec~~nens deformed at 730°C at a strain rate of 1W5s-' to 0.5%, 1.5%, 3.8%, 5.2%0, 8.2%, and 11% are 3 x 109~ m - ~ 8.8, x 109~ m - ~ 11,x 109~ m - ~12,x log 9.4 x 109 and 6.3 x 109~ m - res~, pectively. Namely, dislocation deiisity increases at the beginning of the deformatio~and increases further when the stress drops steeply after the maximum stress. The dislocation density reaches a maximum value at a plastic strain tpl 5-6%, and then decreases conti~uouslyuntil it reaches half of the inaximuin value (for the specimen 2i 11%). Wlth Dislocation density decreases with i n c ~ e a s i n ~ deformation temperature. For example, when specimens were deformed to a plastic strain 1% at a strain rate of 1 0 - ~s-l at various ~~1 temperatures, dislocation densities in the specimen deformed at 695"C, 730"C, 790"C, arid 820°C were 9 x 109 6 x log cmP2, 0 . 7 5 ~ 109 and 0.3 x 109 respectively. Dislocation density decreases during annealing. For example, when annealing a specimen with a dislocation density of 12 x log at 730 ('C for 13 rnin, 45 min, and 90 min, the dislocation density decreases to 7.7 x 109~ i i i - ~4.9 , x 109 and 4.8 x 109 respectively. This experiment indicates that the measured dislocation density in the unloaded specimen is lower than that in the stressed specimen, because the time period from I
Wang et u2. (1993) and Yan and Wang (1993) deformed an A17@ CO,,Ni, decagonal polyquasicrystal by hot impact and observed the microstructures of the deformed samples. In most of the regions they observed either densely distributed stacking faults or densely distr~buteddislocations. The fault plane was perpendicular to the 10-fold axis and the displacement vector lay in the fault plane and was parallel to a two-fold axis any dislocations possessed a Burgers vector parallel to the 10-fold axis. On the basis of this observation they proposed tlie following mic ism of ~i~h-temperature plastic deformation whose 10-fold ams is a periodic direction. In t ed by dislocation motion. the moving dislocation is d axis, the situation is the same as in coiiventional crystals, and densely distributed dislocations are observed, When the Burgers vector of the gliding dislocation i s a higher dimensional vector with its physical comp parallel to a quasiperiodic direction, for example, axis, the region slipped by a moving dislocation becomes a stacking fault region. The Juelich QC research group systematically studied the dislocation nisni o f plastic deformation in A17@., Fdzl.oMn, Wollgarte~et al. (1995)
,
399
400
Mechanical Properties
unloading to quenching a specimen is usually longer than 2min. As summarized by Urban et a/. (1999), ~~lthough a number of plastic d e f o ~ a t i o nexperiments were carried out on Al-Cu-Fe IQCs, no evidence had been found for a dislocation mechanism in this inaterid. However, Wang et a/. (2000) have recently observed densely distributed dislocation lines in hot-impacted Al,, Gu,, Fe,, IQC. All the experimental observations described above indicate that dislocation motion is the predominant process mediating the liigh-temperature plastic deformation of quasicrystals, at least for Al,, CU,~,,Fe,, Cs and Al,,Co,5Ni,, DQC.
,
5.3.2 tacki in^ ~ a ~ lTt ~, ~and i n C ~ y ~ t a l lPhases i~e in Dejormed QCs
ensely distributed stacking faults were observed in hot-impacted and rapidly cooled Al,, CO,,Nil, DQC ang et id., 1993 and Van and Wang, 1993), A170,4Pd21.2Mn8.4 IQC (Wang et id., 1998b; Urban et al., 1999; ~ a n etg al., 2000) and Alb2Cuz5.5 XQC ( ~ a n get al., 2000). Contrast and trace analysis revealed that the fault planes are ail parallel to one of the slip planes, are usually dense atomic . The displacement vector in the fault plane and is the physical component of one of the Burgers vectors of the dislocations. However, stacking faults in ~~igh-temperat~~re deformed and slowly cooled Al,o,4PdZ1.,Mns ZQC were only occasionally observed ( ~ ~ etnal.,g ~998b).They are not conventional planar defects but extended into wavily bounded walls of about 2.5-10 nm thickness. The apparent displacement vector o f each stacking fault, determined by a contrast extinction experiment, is parallel to the fault plane nornial. In order to clarify why stacking faults are only occasionally observed in high~~emperatL~re deformed and slowly cooled Al-Pd-Mn amples, Yang et al. (2000) carried out an in-situ observation of the hot-impacted and rapidly quenched sample, in which densely distributed stacking fault contrasts were observed. Up to a temperature of about 650°C, no remarkable change of the contrast could be observed. Further increasing the temperature to 700°C leads to an increased blurring of the fringe contrast. Within approximatel~5 min the fringe contrast completely disappeared. The positions of the bounding dislocations o f these staclciiig faults did not change compared to the initial state, and the contrast of the bounding
dislocations became sharper. This ir?-situ observation revealed that the stacking F~ulksformed during hot impact of Al-Pd-Mn IQC may be annealed out without any motion of the bounding partial dislocations. All these experimental observations reveal that a inoviiig dislocation in Cs leaves behind a stacking fault with a displaceme vector, which is equal to the physical component of the urgers vector of the gliding dislocation. This stacking fault is not a conventional planar defect but a wavy wall. Owing to the intense diffusion at high temperature, it broadens along the lateral direction, inducing an apparent displacement along the normal direction. of the stacking fault, and ~ n a l l ydisappears. The situation for AI-Cu-Fe IQG is complicated. In addition to stacking faults and dislocation lines, twins (Shield et a/,, 1993; Shield and Kramer, 1994) and shear-induced, one-dimensionally periodic lamellae (Shield and Kramer, 19973) were observed. Based on these observations, Shield and Kramer (1994, 1997a) suggested mechanical twinning and martensite-type shear inechaiiisnis in the high-temperature deformation. In addition, Shield and XCramer (1997b) suggested also a grain boundary mechanism, and Bresson and Gratias (1993) emphasized that atomic diffusion is the major cause of deformation. 5.3.3 Pha.son Strain Fields in Plas~ica//yD
Franz et ul. (1999) demonstrated that phason strain fields in plastically deformed quasicrystals can be quantitat~velycharacterized by means of selected area electron diffraction in a transmission electroil microscope by using iinaging plates. They found that niovlng dislocations may introduce phason strain into the regions free of dislocations in quasicrystals. The inteiisity of the phason strain varies with position in a given specimen and increases with increasing amount of plastic deforniation.
Based on the experimental obsewations reviewed above and discussions described by Feuerbaclier et aE. (1997), Wang et al. (1998a, 1998b) and Urban et aE. (1999), we conclude thdt the elevated temperature plastic deformaFe,,, and tion of quasicrystals, at least for A~,,CU~~., IQCs, and Al,, Go ,,Nil DQC is PdZ1 mediated by thermally-activated dislocation motion. The phonon component of a dislocation on a densely
,.
,
401
~ ~ s l ~ c ~in t Quasicrystals i o ~ s packed plane of a QC may glide under applied stress leaving behind a stacking fault. Owing to the strong intermetallic bonds in clusters in a QC and large stacking fault energies, the critical stress for dislocation movement i s very high. Hence ahnost all the QCs are brittle at room temperature. At high temperature, thermal activation helps moving dislocations to circumvent or cut these clusters in the course of plastic deformation. The stacking fault left by a moving dislocation broadens and then disappears due to intense di~usionat high temperature. There are three points that must be emphasized: (1) Dislocation motion in QCs is not an equilibrium niotioii, in which the static equilibriuni structure of the dislocation is maintained at all times. (2)The obstacles to dislocation motion are clusters with a high density. Thus a large activation area and a viscous motion are observed expe~mental~y. (3) Dislocation motion at high temperature is dependent on atoinic diffusion, which leads to the recovery of the phason strain. However, it is controlled, not by difhsion, but by localized obstacles (Feuerb~cher et al., 1997; Urban. et al., 1999). There are two competing processes in the course of the plastic deformation of QCs: strain hardening and disorder softening. Owing to the high structural symlnetry, plastic deforlna~-ion in f QCs is usually enhances the ‘lip carried Out by strai~-hardenin~ process, so that we observed an initial hardening stage VM with a very high hardening rate. On the other hand, the strain accomn~odationpara11 I of the moving dislocations become larger during the early stage of deformation and reach the values of 2 5 - r7. Hence the phason-type strain left behind by the moving dislocations is rather strong and its recovery is insufficient. This causes a disordering aiid hence a softeiiing of the obstacles to dislocation motion, leading to a decrease of the flow stress with increasing plastic strain, as observed experimentally. Based on these concepts, Schall (1998) tried to calculate the stress-strain curve and obtained a good consisten.cy with experimental curves. In the preseiit work we stressed dislocation motion as one importa~tmicromechanism for high-temperature plastic d e f o ~ a t i o nof quasicrystals. Of course other mechanisms, such as diffusional creep, grain boundary sliding, twinning, and marteiisite-type shear may also contribute to the hi~h-temperatureplastic deformation of quasicrystals.
s was supported by the National Natural undation of China.
ces Bacon, D. J., Barnett, D. M., and Scattergood, R. 0. (1978). Proc. Matter Sci., 23, 5 1. Bak, P. (1985). Phys. Rw., Barnett, D. M., and Lothe 105. ~ the c ~ ~ u n Bohsung, J., and Trebiii, H. R. (1989). 111 ~ ~ ~ r o d to ~ a ~ ~ ~ e m ac?f‘t Quasacrystols ics (ed. M. V Janc). Acadeinrc Press, London. L., and Gratias, D. (1993). J. ~ o ~ - C r y sSolids, t. 1 468. . H., Yang, W. G., and Hu, C. Z. (1992). J . Wuhan Univ. (Nat. Sci. Edition) 323-330 (in Chinese). Dmg, D. H., Yang, W. G., Hu, C. Z., and Waiig, R. H. (1993). Generalized clasticity thcory of quasicrystals, Phys. Rev.. 7003. . H., Yang, W. G., Hu, G. Z., and Wang, R. H. s and defects (1994). Linear elasticity theory of C/LI 111 quasicrystals. Mater. SCE, Forz;lm, 345. and Yang, Ding, I).H., Qin, Y. L., Wang, R. H., W. G. (1995a). Actu Phevs. Siiz. (Overseas Editlon) 824. D. Wallg, R*H.9 Yang, W* 6 . 7 Hu, C* and Qin, Y. L. (199Sb). PhiE. Mug. Lett., 7 Ding, L), H., Wang, R. H., Yang, (199Sc). J phys., Condens, Mutter 7, 5423-5436. Diiig, D. H., Wang, R. H., Yang, W G., and Hu, C. Z. (1998). Elasticity, plasticity and dislocations of quasicrystals. .?
“-3
z.*
Eshelbi, J. D., Read, W. T., and Shockley, W. (1953). Acta metal^., 1, 251. Feng, J. L., and Wang, R, H. (1994a). Convergcnt-beam electron diffraction study of the Burgers vectors of dislocations in icosahedral quasicrystals. Phil. Mag., A69, 98 1-994. Feng, J. L., and Wang, R. H. (1994b). Burgers vcctor determination for quasi-crystalline dislocations by the diffraction contrast technique. J. Phys.. C~ndens.~ ~ f t e ~ , 6, 6437-6446. Feiig, J. L., Wang, R. H., and Dai, M. X. (1995). Observation and analysis of extended dislocations in ail Al-Pd-Mn icosahedral quasicrystal by trasmission electron mcroscopy. J. Mater Res., 10, 2742-2748. Feuerbacher, M., Metznnacher, C., Wollgartcn, M., Urban, K., Baufeld, B., Bartsch, M., and ~ e sse r sc h i ~ i d tU. , (1997). The plasticity of icosaliedral quasicrystals. Mater.
Foreman, A. J. E. (1955). A m et all., 3, 322. Head, A. K., Humble, P., Clarebrough, L. M., Morton, A. J., and Forwood, C. T. (1973). CompLited electron inicw~ r a p hand ~ ~ defect i d e n t ~ c a t i o ~~.o r t h - H o l l ~ ~Publ., nd Amsterdam. Hirscli, P., Howic, A., Nicholson, R.B.. Pashley, D. W., and Whelan, M. J. (1977). Electron immscop.y ojthirz crystals, Robert E. Krieger, Huntington.
Hirth J. P., and Eother J. (1968), Theory o j Dtslocations, McGraw-Hill, New York. Hu, C. Z., Ding, D. H., and Yang, W. 6 .(1993). rlcfa I'hys. Sin. (Overseas Edition) 2, 42-47 Hu, C . Z., Dirag, D. H., and Yang, W. G. (1993). 3. Wzkhan Umv.(Nat. Sci. Edition) 3, 21-28 (in Chinese). . H.. Yang, W. C., and Ding, D. H. (1996). Actu Cryst., 8 5 2 , 251-256. ,, and Masumoto, T, (1994). Mater. , 850. Cryst., 842, 261 -271. Jam, M. V. (1985). Pkrys, Rev. Lett., 55, 407-610. Kelton, K . F. (1995). In ~ n t e r ~ e t ~Cl ~oi ~c p ~ u Vol. ~ ~ s1,: P ~ z ~ e z ~(eds l e . ~J. H. Westbrook, and R. L. Fleischer). Wiley, pp. 453-491. IClenian, M. (1995). Czech. J. Phys., 45, 935, omniers~C. (1991). Actn Met. Mat., 39, 287. Levine. D., Lubensly, T. C., Ostlund, S., Ramaswamy. Steinbardt, P. J., and Toner, J. (1985). P h y . Rev, Lett., 1520-1523. Levi~ie,D., and Steinliitrdt, P. 1. (1984). Pl7.y~.Rev. Left., 53, 2477-2480. Li, X, F., and Fan, T. Y. (1998). Chin. Phys. Lett., 15. 278. Lubensky, T. C. (1 988). Introduction to Quasrcrystuls (ed M. V. Jaric). Academic Press, Boston, p p 199-280. amy, S., and Toner, J. (1985). Plzys. clzutiics ofdejeects in solids,Martinus Mijhoff Publ., Dordrecht. . H., Ding, D. H., and Lei, 3 . L. (1997). J. Plys.. Condcns. Mufter. 9, 859 872. Rosenfeld, R., Feuerbacher, M., Baufeld, B., Bartsch, M., anke, G., Beyss, M., Messerschm~dt, (1995). Study of plastically deformed Mn cpsicrystals by transmission electron microscopy. PhiE. Mag. Lett., 72, 37.5-384. Schall, P. (1998). Diploma thesis, RWTH Aachen. S1iield, J. E., Krarner. M. J., and McCallum, R. W. (1993). J. Mater. Res., 8, 1199. Shield, J. E., and Kramer, M. J. (1994). Plzil. Mag. Lett., 69, 115. Shield, J. E., and ICrmer, M. J. (1997a). J . Mater. Res.. 12, 300. Shield, 3. E., and Kramer, M. J. (1997b). J. Mater. Res., 1 2043.
Stroh, A. N. (1958). Phi/. Mag., 3, 6 Urban, K., Feuerbacher, M., Wollgarten, M., Bartsch, M., and Messerschniidt, U. (I 999). In Physical Properties if Quus.sicr~vstal,r(ed Z. M. Stadriik). Springer, Berlin, pp. 361-401,
Wang, R. H. (2000). Micron, 31. 475. Wang, R. H., and Dai, M. X. (1993). Burgers vector of dislocations in an icosahedral Cu,, Fe,, quasicrystal determined by means of conver~ent-bea~electron digraction . Phys . Rev ,, Wang, R. H., Feng, J. L., Yan, U. F., and Dai, M. X. (1994~). Convergent-beam electron diffraction determination of the ors of dislocations in quasicrystals. Mrxfcr. Sci. 151, 323-334. Wang, euerbacher, M., Wollgarten, M., and Urban, K. (I 998a) Dislocation reactioiis in icosahedral AI-Pd-Mn quasicrystals. P M . Mag., APf7, 523-540. Wang, R. H., Feuerbacher, M., Yang, W. G., and Urban, K . (1 998b). Stacking faults in high-temperature-deforrned AlPd-Mn icosahedral quasicrystals. Yhil. Mug., A?$, 273284. Wang, R. H., Yang, W. C., Gui, J . N., and Urban, K. (2000). Dislocation mcchanrscn of hi defor~ation of Al-Cu-Fe and quasicrystals. Maler. Sri. E~zg.A , Wang, 2. G., Wang, R. H., and Transmission-eleceron-i~icr~~scopystudies of small dislocat~on loops in Al,, Si, Mn,,, icosahedral phase. Phys. Rev. Lett., 66, 2124-2127. Wang, Z. G., Wang, R. M., and Feng, J. L. (1994b). Dynamical siinulations of dif'fraction contrast images of straight dislocations in icosahedral quasicrystals. Phil. Mo.~.,A70, 577-590. Wang, 2;. G., Dar, M. X., and Wang, R. H. (1994~).Phil. Mag. Lett., 69, 291-296. Wen, J. Q., Wang, R. H., and Lu, G. H. (1989). Distortion of the zeroth-order Laue-zo in a silicon crystal. Acta Wollgarten, M., Beyss, M. Koster, U. (1993). Phys. Rev.Lett., 71, 549. Yan, Y. F.>and Wang, R. (1993). Phil. Mug. Lett., 67, 51. Yang, W C., Wang, R. El., Ding, D. ET. and Hu, C. Z., y theory of cubic qwsicrystals.
Yang, W. G., Lei, J. L., Ding, D. H., Wang, R. H., and Hu, C. Z. (1995). Phys. Lett., A ~ 177-183. ~ ~ , Yang, W G., Feuerbacher, M., Tamura, N., Ding, D-H., Wang, R., and Urban, K. (1998). Phi!,M L ~ 879, . , 14811497. Yang, W. G., Wang, R. El., Feuerbacher, M., Schall, P., and Urban, K. (2000). D e t e r ~ i n ~ t of ~ othe n Burgers vector of dislocations in icosahedral qu~sicry~talsby a hi resolution Iatticc-hge technique. Phil. M q . Lett., 281. , D. H., and HLI,C. Z. (1997). Yu, D. P., Baluc, N., and Kleman, M. (1997). Dcfkct tmd
The phenomenon of twinniiig in crystalline solids has been a subject of scientific research thr 20th century, mostly by mineralogists crystallographers in the first-half and and electron inicroscopists in the second-half. the past few decades, it has been recognized by ~ a t e r i a l s scientists and solid-state physicists that twiiining plays important roles in high-temperature intermeta~licc o ~ p o u n d s ferroelas, tic sliape memory alloys, ferroelectric sensors and actuators, ferromagnetic high-eiier~y~densitymaterials, high T, superconducting oxide compounds, as well as in a niultitude of ele c materials such as compou~idse~iconductors. n the past decade, two symposium proceedings and Wuttig, 1994; ande, 1999) were ~ublished" whicli cover all types of twins in the so-called advanced materials including intermetallic compounds. Whe recent review articles by ~hristian and (1995) and by ~andershaeve(1998) are on defor~ation twinning, and mechanical twinning in semiconductors, respectively. ~ollowingFriedel(1926), a twin may be defined as a portion o f a crystalline body which has undergone a 'tra~isformation' in such a way that the crystal structure of the resulting product is identical with that of the parent, but oriented diKerently according to well-defined relationships. In accordance with their twins may be classified into three different urgers, 1945): (a) growth twins, (b) transformt-tlioii twins, and (c) defbrmation twins. Growth twins are those formed by a solidification process from a liquid or vapor phase, and f o r i ~ a l ~these y belong to a
special kind of grain b o u ~ d ~ ~ ~ ri ae ~ s .s ~ o r n i ~ ttwin^ io~i ate tran$~orniatio~, an may result during they are further div of a niarteiis~t~c tr within a new crystal struct~~re ~ u r i nthe~ t r ~ n s f o r ~ l a eformation twin~ingor ~ e ~ ~ a n i c a l a shear defon11ation process caused applied stress, and i s one of the two (slip is the otlier) o f plastic d e ~ o r ~ ao t i ~ ~ solids at an ambient t e i ~ ~ e i - araiige t ~ r ~where defor-
cold-work and aIinealin~of polycrystals. s far as the r ~ ~ a t i o nbet s~~~ t w ~ n n i n and ~ i ~ e c ~ a bn e~~ ~ vl i oinr i n t e r ~ e t ~ l l i ~ compounds is concerned, xt i s worth ~ e ~ i t isomc o ~ ~ ~ i ~ general notions often stated in e~ei~entary text~ooks twinning occurs p i - e d ~ ~ ~ i n a n t i y
404
Mechanical Properlies
need to be modified. t will also be shown that transformation twins and growth twins play important bchavior of intermetallic alloys. ves (1952) has been often cited as pred~cting that loiig-range ordered alloys of cubic be incapable of ‘true-twinning’ o be d e ~ n e din the next section) Its in the literature are in support of this prediction, e.g. ahn and Col1 (1941) and ki and Fisher (1943). In some ordered s of bcc~basedstructures, ine~asticdeformation can occur due to the proftise ‘pseudo-
are readily available, such as heterophase interfaces and/or grain boundaries, and (b) the degree of chemical ordering (7’ as conipared to fccy) and/or solute segregation to the interfaces can lower the intrinsic stacking-fault energy of an alloy and hence provide a surmountable energy-barrier for twinning. The tension/co~pression asymmetry observed in a creep-deformed Ni-based superalloy is attrib~tedto the unidirectio~ialityof itie twin system ( which may be an important factor in plastic instability and creep-fracture initiation of this important engineering allay system based on Ni& (Anton, Ch. 1 in Vol. 2). en, 1979) and NiTi (Goo er al., The i ~ p o r t a n c eof the role twi~ning plays in physical properties and mechanical beliavior of interng of these intermetallics, the restoring forces stemming from the ‘pseudo-twins’ nietallics has been further demonstrated in those of cause them to untwin or detwin and give rise to the sonon-cubic fcc-based str~ctures,viz. the Ll, (tP4) and r and the related shape(tIS) types. In CuAu-I (El,)-type alloys, such as Inoue, Ch. 35 iii Vol. 1; 8, Fe-Pt, and CO-Pt alloys, the A1 (cubic) +Ll, ragonal) transformation is accoi~panied by the suka and lien, 1999). formation of a po~ysyntheticallytwinned (P bic structure, L1, (cP4) ture (a repeated twi d below, Clialcraborty ttendant to the at the so-called ‘superresult of the strain atomic ordering re inode occurred in fully 1942; Vlasova et al., 1972). A significant enhancement ordered alloys, whilst the usual fcc twinning mode was of magnetic properties of polytwinned ferroma~netsis observed in disordered and partially ordered alloys. In cons id ere^ to be possible through better under~tand~ng Ti) y’ phdse of the Ll, s t r u c t ~ rat~ elevated of the relationship between the PST i~icrostructu~e atures, ~ i c r o t w i n n i nof ~ the true-twin mode and the mechanism of coercivity (Zhang and was observed to initiate from the y/y’ interfaces under a 1992). In the Ti-A1 system, the fact that a variety of c r e ~ ~ ~ d e f o r m a t icondition on uimier and Strudel, two-phase ~icrostructu s consisti~gof y (Ll,) phase ar er al., 1970). The experimenta~results and a2 phase of the 0,9 (hP8) structure can be ~emonstratetwo important points: (a) contrary to the conventio~alw i s d o ~ twinning , in ordered i n t e r ~ ~ t a l - obtained because of the presence of 01 phase (hcp solidsolution) of the A3 (hP2) structure at high temperalics can occur at ~ i t e~~ ~ he r a t u r and e s low strain rates tures was discussed in the review by Kuaiig and if hetero~~neous ~ucleationsites for twin formation
~ r y ~ ~ a l l o ~ r aelements p h i c and parameters of ~
~
i
~
n
i
~
~
Twirt n irzg and ~ The roles of trans~ormation ic~ostructure-propertyrelationshi~sand of deformation twinning in i~eclianical
discussed recent
eical Be ~ havio r~
~
a
~
~
405
(or K i ) planes. U! is the ~iiterplanarspaein
ucture have been review articles,
e pLi~ose of this chapter is to present a comprehensive review of research highlights aiid depicted in Figure 1, the result 1s a s~ipercellof the uiiresolved issues on the roles of twinning in deformatwinned lattice with the base tion and fracture of intermetallic c o ~ p o u n ~Special s. liile -rC, and q r are fixe emphasis is placed on material that was not covered in by (180"2211/) to the earlier volumes (1995). pecific terminolo~yused in respectively. The twinning shear IS g = 2 cot$ = this chapter regard in^ crystal structure, aiid genesis of twins is ev I/@* The four orientation relationships of the classical In Section 3, experii~entaldata reflection in theory are: (i) are s u ~ m a r i ~ eaccordin d to the crystal classes. The ql, (iii) reflection in the plane normal to q l , and (iv) current status of our understand in^ o 180" rotation about the direction norinal to K,. The iiieclianisnis of twin formation i s reviewed require~entfor a crystal superlattice to be uncli~nged 4. The next two sections, 5 and 6, discuss th and ductility of i n t e ~ ~ e t a l ~ i cby the shear leads to the condition that either K , and q2 both have rational, low-index elements (type I twin), or ts of teiiiperature, microome of the key issues are l o ~ - i n e~l ~ m x e i ~(type ts which suggests specific s are rational, the twin is directions of future research. twins observed in metals aiid alloys are of the coi~poundtype. ~ ~ ~ e ~ ~ando t - r ~~ ~e ~~ tin ~ ~ai i ~~ ~ ii ~~ ~ g pound of the L1, structure are shown in and 2(b), respectively, and In general, a true twin niay be defined as a twinned portion of a crystalline body which has a crystal structure (including the state of chemical order) ~denticalwith that of the parent matrix. The orieiitaq values should be lialved i lion relatio~shipsbetween the true twin and the matrix representing the primary twin s can be specified using the classical theory of the CompreEicnsive ~ r ~ oC the ~ c t~ y ~ s ~ ~e~ ~ ~ of o g r~incompleke ~ ~ ~p ~ y~ twin unless atomic s h u ~ i n g( qr 2) of those atoms indicated by the strokes on the circular symbols possible t w i n ~ i nmodes ~ in sup~rlattice structures based on fcc and bcc structures were given by follows. In the co~plementarytwin s Aruiiachalam and Sargent (197 1 Cliristian and also called a .wper-twiR (Chalcrabo g i s twice as large as in the pseu ~ a L i ~ h l (1989), in and 6 0 0 (1994). atomic s ~ i L ~ is ~ inecessary n~ (q = 2). ments and parameters for other cubic L2, types), tetragonal (L1, and O,, type) structures 1 and 2. We are concerned only with compound twin modes which give small twiniiiiig~shearvalues, g < 2/2. the acute angle between q l and q2 (or q i ) or between In this chapter, unless ~ e n t i o n e d oth~rwise, the )?
406
ure 2 Crystallographic elcnients of (a) pseudo-twinning, $ = 70.53", and (b) complementary anti-twinning, $?, = 54.74", in an A,B compound of the L1, structure
able 1 Crystallo~raphicelements and parameters of twin modes in cubic ordered structures Structure type
Twin type
P (Primary); C (Complenientary)
K,
KZ
yI1
yI2
S
(I
g
Twinning und ~ e c h u n i c a Behavior l
407
Crystallographic elements and parameters of twin modes in tetragonal and hexagonal ordered structures. based on disordered unit-cell dimensions
A=
c/a 1s
P (Primary); C (Complem~ntary)
Miller-Bravais indices in an ordered crystal structure are referred to the crystal structure of the corresponding disordered state (fcc for Ll,, Ll,, and DO,,; bcc for B2, DO3, and L2,; and hcp for DOl9). The nuiiierical value of e is simply 2/3 for the twin modes in cubic str~ctures(Table 1). In contrast, the nLi~~erica1 value of c! in the primary (K,, ql) twin systenis of non-cubic crystal structures depends on the axial ratio, 1,= c/a, as given in Table 2. On a primary twin plane of K, = (1 11) in the tetragonal cases (L1, and O,, structure types in Table ), one of the three twinning shear vectors, = [112]/6, creates true-twinning or order-twinning and the others, [?11]/6 and [121]/6, lead to pseudotwinning (e.g. see Figures 8 and 12 by Yamaguchi and Umakoshi, 1990; or Figure 1 in Yoo et al., 1994b). It is this true-twinning mode with relatively low values of'g and q that ont tributes signi~cantlyto the plastic deformation of Ll, and DO,, alloys. In the cubic and tetragonal structures (Tables 1 and 2), primary twin systems and conjugate (reciprocal) twin systems are crystallographically equivalent, i.e. {Ki,> = {K'J and (qJ = (q2).In all other twin modes, they are not. Furtheri~ore,the conjugate deformation mode, (K,, q,), to the co~plementary twin (antitwinning) mode is found to be the secondary slip system at low and intermediate temperatures, which becomes the primary slip system at high t~mperatures, in all the ordered alloys for which the information on deformation modes xs available (Yoo, 1989). Table 3 summarizes these conjugate re~atio~sliips between the complementary twin (anti-twinning) systems and the slip systems at elevated temperatures in various superlattice structures.
Table 3 Slip-twin conjug~te re ~ ~ t ~ o n s h i m p s ordered superlattice structures between c o mp l e i ~ e n ~ ~twin ry (antitwin~iiig)system of (K,,- q,) and the slip system of (K2,q2)
Structure type
Complementary twin
i'c;
rll
Active slip (hkl)
fuvw] g
A simple criterioii for twinning results from geonietrical consideration^ that a twinni~gmode needs to have a small shear strain, g, and either no atomic sliufling (qg 2) or a simple shuffling mechanism represented by a small value of q when q> 2. Formally, these two parameters, g and q, are related, respectively, to energetic and kinetic aspects of deformation twinning. For a given c l ~ s e - p a c ~plane ~ d in a non-cubic crystal structure, there exist a unique set of translation and rotation operations for crysta~lograpliicalydistinct interfaces. For instance, y / y interfac~swithin a ? + a z lamellar doinain of a y-TiAl based alloy exhibit three orientation relatioiiships (Feng ei al., 1990; Inui et al., 1992; Yang and Wu, 1992). The f o ~ r t hkind of
orientation relationship mentioned above 6' %Ivesa truetwin with the rotation angle of 0 = 180", a pseudo-twin (8 = 60"), and a rotational boundary (0 = 120"). Interfacial energies of these three types in TiAl were given ao et nl. (1994), and Yoo and Fu (1~98b). In the classical nucleation theory of twin formation, the most important physical parameter is the specific inter~~cial ener rlattice intrinsic/extr~nsic
orn fir~t-prii~ci~les calcuntermetallic compounds
scopy ( techiiique~.A couple of examples are shown es 3 and 4. Figure 3 shows a lattice image picture of a true-twin taken in a deformed Ti54at.%Al (hereafter, ~ ~ - 5 4 Aspeci l) al. (1994). Figure 4(a) shows a typical a y/y lamellar ~ o u n d a r y of the ( B = 180") taken along the [1T0] direction of a TiAl PST crystal by Inui et al. (1993), and a true-twin boundary in anothe ift by the transla, 1997) js shown in
There are two types of dislocation describe a twin embedded in the matrix
or atomic structures of twin/matrix interfaces can be electron microchar~cterizedby using hjg~-resolutio~
Lattice image ~R~~ picture of a twin in Ti-54A1 (from Couret et al., 1994, reproduced with p ~ r ~ i s s i ofrom n The Minerals, Metals and Materials Society). M is the matrix and T 1s the twin
09
Schematic illustratioiis aiid HREM images of y/y lainellar boundaries in a TiAl PST crystal, (a) the true-twin type (from Inui et al., 1991) and (b) the true-twin type with an APB type translation (from Sicgl rt al., 1997)
twin formed by dis~ocationswith (Frank aiid Stroll, 1952; allel to q2 gliding on lough, 1957). Alternat ly, disclination models for winning are sometimes useful to describe certain aspects of twinning phenomena (~rmstromi~, 1965; Romanov and Vladirov, 1992), such as a case of twin(M~llneret al., 1994; M~llnerand ta of ~ransmission electron microscopy (TEM) indicate that an incoherent twin boundary consists of steps and/or ledges on the CT interface, where each step is a twin or glissile-type twinning dislocation as Frank (1951) originally proposed. Figures 5 and 6 are shown as typical exam~les. Low- and high-i~agnification H ~ E Mimages from y-TiAl shown iii Figures 5(a) and 5(b) confirm that tapering of a twin l a ~ e l l ais due to the presence of steps (Singh and Howe, 1992). Figure 6 shows a bright field in situ TEM image of a twin tip in Ti-54AI under g (from right to left) at a very low ouret et et al., 1994). Each sequence (I, 2, and 3) indicates a set of three Shockley partial dislocations, and the gradual decrease in the interspacings of these partials (especially of the first 15 partials at the left in a thcker region of the foil) confirms the validity of a dislocation pile-up model of deformation twiiis in y-TiAl. This dislocation pile-up ler (1992%) to alnalyze the dual twin dislocations of a in TiAl. The results of
n u m e ~ ~analyses ~l indicate that twin nuc~eation is difficult, but twin propagation is rehtively easy. Using an in situ strain in^ TE al. (1997) investigated deform migration in a fully lamellar two-phase TiAl alloy at room temperature. The results showed that only the interfacial dislocations in tw for slip along the mobile, suggesting that the a& interface is higher than (0 = lSOO) or pseudo-twin IS = 60") interface at room temperature. The junction between a coherent ii~coherent{121} twin boundary in a yinvestigated using (1996). The straight iiicoherent (122 ) boundary perpendicular to the coherent (111) ~oundarycan be produced either by a shear process in the y phase or as a result of the a--+? t ~ a n s f o i ~ i a t ~ oThe n . niobility o f such an incoherent twiii boundary is believed to be intrinsically low, but can also be hampered by solute segregation. The topological theory of interfac~aldefects was proposed by Pond (1989) who showed that ad~issible interfacial defects are characterized by combinations of symmetry operations froin each of the two adjacent crystals. In their recent H interfaces in T i - 4 4 ~ l - ~ Z r (1999) found no lattice misfit dislocations, but a variety of interfacial steps. The Burgers vectors deter~inedfor
image d e ~ ~ o n s t r ~the t i i tapering i~ of a twin lamella due to the presence of steps 111 yarea showing multiple-pl~nesteps at the twin interface marked by arrows (from Singh and Howe. 1992, reproduced with permission of Taylor & Francis)
image of a twin tip iri Ti-54A1(from Gouret et al., 1994, reproduced with ~ ~ ~ i ~ i sfrom s i oThe n ocrcty). The letter X points out an iniperfectio~iin the sequence (1, 2, and 3) or partial
clislocati ons
aisaz e2: al., 1997) were t mapping metliod by
step and di~locationc ~ a r a c t ~The r . spIi and dislocation components yields a s tion of interface flux for d i ~ u ~ i o n a ~ transport ~ s s o c i a t ewith ~ the motion and interaction defects has been treated formally nd (1998). That the presence of' a ( I 11) / 3 Frank partial dislocation at a twin bo~~ndary in a l u ~ ~ i np~oduces ~m a step that can move by c ~ ~ was concluded from th dlin et al. (1997) and
b
(1999). This is an important finding in that it provides a dislocation i~echaiiism whereby twin growth can occur by non-conservati~e thermally activated processes, which is consistent with the role of Frank partials in twin nucleation in y-TiAl scussed recently by oo (1997) and Yoo and ishinuma (I 997, 1999). se
In accordance with their genesis, twins niay be classi~ed into four different kinds: growtli twins, annealing (recrystall~~ation)twins, transformation twins, and deforination (i~echanical)twins. All of these kinds have been observed in intermetallic compounds. Interfaces created by growth twins in an ch the order-disorder han the melting point boundaries in it. A 113) type in Ni,Al is , Ch. 24 in Vol. 1). In the compounds (T,K T1J, e.g. in Vol. l), annealing twins are also possible because ‘cold working’ in the disordered state ( T > T,) i s easier than in the ordered state (T
atom-by-atom correspondence between positions in the parent phase and the twin, not only for those cases with 4>2, but also for the cases of intrinsic anti-site defects and/or vacancies, sL~bstitLitioiia1solutes, and interstitials trapped in different positions. Whilst we d~stinguish t r u e - t ~ i n n i n and ~ pseudo-twinn~ng for ideal superlattice structures in the belief that this distinction inay also be important in real materials (Christian and Laughlin, 1989), one must keep in mind possible roles of intr~nsicand extrinsic point defects in deformation twinning of ~ u l t i c ~ n i p o n eintermetall~c nt alloys at elevated te~peratures.
atio
3.1 fcc-
res
3.1.1 L1, (cP4) ~ 0 ~ ~ 0 ~ n ~ s
~ e f o r ~ a t i otwin~ing n was reported to occur under shock loading in Cu,Au ( ikkola and Gohen, 1966) and Ni3A1 (Gray, 1994). In the latter, the operative twinning mode is assumed to be of the complei~entary anti-twinning type (Albert and Gray, 1995). This true twin .in the Llz structure can be regarded as an array of SISFs on consecutive K , = (111) twin pia TEM microstructures of shock-deformed Ni,Al (e.g. Figure 2 of Ilcebuchi et al., Ni,Al+B (Albert and Gray, 1994) show a dispersed distribution of stac~ing-faultfringes that are indicative Transformation-induced twinning in interinetallic o f individual S than ‘microtwins’.A similar superlattice compouiids was reviewed by Yamaguchi et distribution of observed by TEM investigaal. (1994) and a general theory of transformation tion of defornie e (Takeuchi et al., 1973). twinning kinetics during ordering reactions was given Emission of extended SISFs from a crack tip has by Khachaturyan et al. (1994). Since pseudo-twinning been reported in Ni,Al (Yoo et al., 1994a) and in intermetallic compounds, as (Co,Ni),Ti (Liu et al., 1988), where the former was q ~ a s i - t ~ i i i n i n(Cahn, g ~ 1992) or of stoichio~etriccoinposition and the latter had a Goaxtoii (1994, 1995), is a special 23Ti-3Ni composition. case of stress-iiiduced mart~nsitic transformation, The term ‘microtwins’ may be reserved for those ongoing debates on unresolved issues (e.g. mixed twins whose thickness (D)is on the order o f 7000 characteristics of ‘diffusional-disp~acive traiisforniainterplanar lattice spacings ( D w 7 x 1O3 d, where eneous nL~cleatioiisites, etc.) recorded 3d.;;r;0.4nm for Ni,Al) and the term ‘macrotwins’ in a number of‘ inonographs on displacive t r a n s f o ~ a - inay be used otherwise. Accordingly, twins of t~~ickness tions apply here as well (Tanner and Wuttig, 1990; on the order of several lattice spacings (Da7d) may be Cahn, 1992; Christian et al., 1995). called ‘nanotwins’, but these must be thicker than 3d so ~ ~ c o r d i ntog the strict de~nitionby Cahn (1977), that the twin orientation based on the stacking true-twi~ningas defined in Section A B ~ can ) be ascertained in the deformation in most loys except completely random structure. A geonzetrical method of det substitutional alloys. y a homogeneous lattice strain, thickness of a nanotwin under an in situ T real intermetallic alloys with long-range order lose an was given by Gomjou et al. (1988).
awasaki (1966) showed actual examples of TEM icrographs in which overlapping of distinct stacking faults sometimes shows contrast similar to those of thin twins or nanotwin~. Whether the twins observed in Gu3Au ( ~ i k k o l and a Cohen, 1966; Chakraborty and Stark, 1975) are of the ~seudo-twin type, or the true-twin type was not resolved unanibiguoLisly. The rotation (or tilt) of the angles of $ = 70.5" and =[i01] or [101] can be determined by measuring the deviation of marker lines drawn on pre-polished surfaces, which gives an estimate of the magnitude of a twin shear after deformation, but this is obviously very difficult to obtain when microtwins are very thin.
The relationshi~sbetween twinning and mechanical and other L1, alloys were discussed ew papers (Yoo, 1998a; Yoo and ~ a ~ a g u c h2~00). i , The twinning shear associated with system shown in Figure 7(a) was in y-TiAl (Shechtuian et af., 1974). ar as a f u n c ~ i oof~ the axial ratio (A = c/a), as given in Table 1 is plotted in Figure 8 for the primary and ~ Q m p l e ~ e n t atwin r y systems and also f o r ~ a t i o ntwin system with
unit twin dislocation for the primary twinning is ]/6, and that of the a~ti~twinning case is = [ii2]/3 as shown in Figure 7(b). Wher ent twin boundary energies for the two cases are the same and calculated to be rather low, rT = 5 0 - 6 ~ i ~ ~for / ~ TiAl 2 (Fu and Yoo, 1991; ~ o o d ~ a et r daZ., 1992), tlie twinning shear oi" the former is about one-half of that for the latter (gp= g&) wlien it zz 1. Figure 7(c) shows that the primary and its conjugate twin systems are crystallographically equivalent. Under an a ~ ~ r o p r i a internal te stress field, therefore, te pair niay give rise to a pure shear strain, on1 an algebraic average of two simple cated by the dashed parallelograms in Figure 7(c). In contrast, the slip-twin conjugate relationship of (OOl)[l 2 and (111) [ii2]/3 is s ~ e ~ c h eind ~ i g u r e cause of the large niagnitude of twinnin (gc> 1.2 for it< Ll)? deformation by the complementary or anti-twinning mode is not 1i~elyto occur unless it is needed at a site
U2
Ib I
ldl
ure 7 Twinnin~ in the Ll, s t r u c ~ ~ r e(a) : primary t ~ w i ~ ~ nsystem, i n ~ ~ (b) co~plementary ~ an t i ~ t w i n n i n ~ ) system, (c) crystaliograp~i~ally equivalent conjugate twin systems, and (d) non-equivalent twin-slip conjugate system
where a large amount of strain i n ~ o ~ ~ a t i bmust i ~ tbe y relieved, such as at a twin-twin intersection or a crack tip, In addition to finding the direct role of twinning partial dislocations in plastic deforniation of y-TiAl, Singh and Howe (1992) attr~butedthe formation of a three-plane (9R) deformatio~-induce~ structure parallel to pre-existing twin interfaces to an ~ ~ t i - t w i n n ~ n g shear operation an every third plane CAB). This result, together with the oh accommodation shear by basal (00 1)[11 twin-twin intersections of type-1 (don cozonal direction) at high t e ~ ~ e r a t ~ r e s al., 1993; Sun et al., 1993), gives an indirect evidence of the slip-twin conjugate re~ationshi~ (Table 3). However, Abe et al. (1937) coiiciuded from their analyses with tlie aid of image simulation that the ~~~
AuCd MnPt
Nblr
AgTi
1.6
I.2
0.8 0.4 0
95
1
1.05
7.1
re 8 Variation of thc twiiining shear wlth the axial ratio in LI, c o m ~ o ~ n d s
structure was caused by overaiid not by the hen an (001) pi01 1~ode-Icrack is s ~ b ~ ~ ctot ean d uniaxial tension along the [0011 direction, crack-tip blunting by two symmetric tension twins of the co~plemeiitary (antitwinnin~) type is ~eometrica~lyand energetically
the so-called 'c-domains') is formed are known to deform plastically by twinning. investigation of deformed CuAn-I, Syutlrina ovleva (196'7) observed prominent bands of deformat~on which changed direction at the domain boundaries and Pashley ei al. (1969) interpreted then1 as s concluded from their inant deforn~ation U-]I at room temperature was { 111) deformation twinning and not uAu-I alloy o f coarse grains ( 5 to
10 pm) without the lamellar structure (colonies of cdomains), Greeiiberg et al. (1 999) observed microtwins in the temperature range -96°C to 3 primary { I l l ) (113) twinning i s reported to be the major deforination mode in FePt from room temperature to 600 "C (Gao and ~ h a 1994). n ~ In~ FeFd, while defor~ationtwinning was the secondary mode of deformation under tension, profuse twinning was observed after cold rolling (Rao and SoEa, 1997). In two-phase TiAl alloys, deformation twins originate froin various interfacial boundaries. Figure 9(a) represents the twinned microstructure typically observed after high tei~peratured e f ~ r ~ a t i o(Appel n and Wagner, 1994). The propagation of the twins i s essentially limited by the widths of y-lamellae. The twins are generally very thin (D< 50 nin), as documented in Figure 9(b) by the frequency distri~utionof their thickness. Tlie mean thickness of deformation twins lies in the 15-20nrn range. There have been many reports of ~ ~ ~ t -3: ~ o r ~ observations of twin intersections in y-TiAlalloys, e.g. Singli and Howe (1992), Wa (1993), Appel et al. (1993), Chaturvedi (1993), Sriram et al. (1994), Loubradou er d.(19951, Jin et G I ~(1995), . Hsi
0,16 0,12 II h
9 0,08 2 0,04 0 0
20
40
60
igure 9 Deformation twins in Ti-4881-2Cr alloy (a) deformed at 800 "C to failure at sf = 10.2%; foil orie~tat~on [loll, and (b) frequency d~stribL~tionof twin thickness. N, = 884 is the total numbers of twins investigated (from Appel and Wagner, 1994, rcproduced with ~ e r ~ i s s i ofrom ii The Minerals, Metals and Materials Society)
414
~ e c ~ ~ a n Properties ~cal
Figure 1 Q Temperature dependence of the apparent CRSS for twinning observed in y-TiAI single crystals (after Kisbida et al., 1996)
Inui et al. (1997), and hang et a/. (1999). As ~ e n t i o n e dabove, there are two types of twin-twin type-TI in reference to the = [lOT] directions, respectively. These studies revealed a variety of localized deformation processes (subsidiary twins, basal or cube slip, nanotwins, formation of 9R structure, etc.), but no clear evidence of microcrack initiation. In single crystals of Ti-54A1 and Ti-56A1 under co~pressive~ ~ ~ ~ i aloading, x i a l twinning was observed at the ~ 0 0 1orientation ~ and high temperatures ida et al., 1996; Inui et al., 1997). A plot of the vs. temperature (Figure 10) shows a transition of the defor~ationmode from superlattice slip to defoi~ationtwinning at about 750°C. At the [OOl] orientation and room temperature, Mahapatra et al. (1995) reported that d e ~ o r ~ ~ t iino n23-51AI single crystals occurred ~ r i ~ d r i l by y twinning, but an increase in the A2 content to 54 at.% resulted in deformation primarily by slip.
The primary (1 1I) [112] twinning of the true-twin type in Ni,V was first observed by Vanderschaeve and Deformation in this alloy arrazin (1977) using proceeds mainly by p lion of stacking faults and twinning due to the rela tiveiy low stacking-fault energy about 22 i n ~ ~ m (~a~derschaeve 2 and Escaig, 1978). cause of the Al--+DO,, transformation at T,= 1045 "C, domains in the shape of elongated prisins b o u n d e ~by ( 110) transfo~ation-twinplanes inay be
formed. Thus, inechanical properties of this alloy are affected by interactions of { 111) slip dislocations and deformation twins with the (1 10) twin boundaries. Zigzag propagation of deformation ~ ~ c r o t w i across ns the domain boun~aries observed by TEM (Vanderschaeve and Escaig, 1983) was attribBted to crossslip of partial ~ i s l o c a t i o ~via s the stair-rod mode (Fleischer, 1959). eformation twinning in A13Tj was reviewed by Uaniaguchi and Umakoshi (1990), and Yoo et al. (1991). The Fact that (111) (112) true twinning is the major deformation mode in A1,Ti at elevated temperatures (20-800 "C) was established following the compression study of single crystals by ~ a m a ~ u c h i (1986) and Uaniaguchi et al. (1987) and the TEM analyses of deformed polycrystals by Vasudevan et al. (19898). Twins were often observed to propagate from one side of the sing1e-cryst~~lsample to the other accompanied by audible clicks, and serrations were recorded on the stress-strain curves which were thought to be due to the discontinuous propagation of twins. Four variants of the (1 11) ( 1 l2> twin system were observed to be augmented by (OOl)[llO] slip at high temperatures, and this observation is consistent with the slip-twin conjugate relationship in the structure (Table 3). Twinning plays an ~ m ~ o r t a nrole t in the plastic deformation of A1,N b (Shechtman and Jacobson, 1975). Conipressive ductility of AI,V was found to increase with replace~entof vanadium by titanium, e.g. Al,(.95V-.O5Ti) showed 7 % compressive fracture strain at room tenipemture (Umakoshi et al., 1988). The i ~ p ~ o v e in ~ ~~n to ~ ~ ducti1ity ~ ~ of ~ the s s ~
~
T ~ ~ i nand ~ i ~~e gc E ~ a ~ iBehavior cu1
ternary alloy was attributed to a substantial increase in the activity of ( 1I I ) ( I I 2) true twinning which was not operative in the binary Al,V compound.
415
Crystallographic elements of twins observed in Ni,Nb single crystals (Hagihara et al., 2000) in tcims o f the Miller-~ravais indices based on the DO, structure
YI 3.1.4 DO, (oP8) C o ~ ~ o u ~ ~ s Tensile strains in excess of 11TOwere reported in a NiNi,Nb composite at room te~peratLire~ which were attributed to extensive twinning on { 112) planes under tensioii along the [loo] growth direction and (011) planes under compression of the ertzberg, 1971). Afte of the directionally solidified r(Ni)/y'(Ni,Al)-G(Ni,Nb) wal et al. (1978) reported that no (011) observed, whereas the hysteresis loop and TEM characteristics could be fully interpreted by means of twinning and untwinning from two specific variants of the four (21 I > twinning planes. Grossiord et al. (1972) determined the (21 1) twinning elements in Ni,Nb to be K , = (21 1) and q2 = ( I1 1) in terms of the Miller-~ravais indices based on the g = 0.384. A systematic study of plastic deformation of Ni,Nb single crystals was performed by Hagihara et al. (1999), and the crystallographic elements of three twin systems were obtained as given in Table 4. The temperature~dependentCRSS plot (Hagihara et al., 2000) showed that while (012) (021) twinning occurred at low temperatures (0 to 200 "C), (01 1) ( o i l ) twinning and (21 1)'(m7 13)' twinn~ng(quotes indicate irrahigh temperatures (200" to for the former mode was about 70 MPa, whereas that for the (01 1) (011) twinning was
3.1.5 C15 (cF24) C o ~ p o u ~ ~ ~ ~
The Laves phases (sometimes designated as FriaufLaves phases), with an AB, composition in the binary case, form a very large group of intermetallics which crystallize into the hexagonal C14 (hP12) structure, the cubic C 15 (cF24) structure, or the dihexagonal C36 (hp24) structure (Laves, 1967; Sauthoff, 1996). These t h e e structures belong to the class of topologically closepacked (TCP) structures (Gladyshevskiiand Bodak, Ch. 17 in Vol. 1). Any one of these structures niay be transformed into any other by shear, and, in this sense, twinning of the cubic C15 phase is just an example of polytypic transformation (IComura and Kitano, 1977). In the C15 prototype MgCu, at 600 "C, deformation twins were observed by Moran (1965). The deformation microstructures of NbCr, observed ming TEM
(211) (011) (012)
'(40nB)' (oil) (010)
'[m7 131' [Oil] [021]
v2
[111] [OI 17 [OOl]
g
Kind
0.40 Type I 0.13 Compound 1.07 Compound
'(hkl)' or '[uvw]' means irrational indices.
were dominated by twins up to 1400"6,whereas at higher temperatures the deformation appeared to proceed by a mixture of twinning and slip et aE., 1997). In all cases the twins appeared composition plane within each grain and this could be attributed to an autocatalytic nucleation effect. Livingston and Hall (1990) showed in plastically deformed (Hf,Nb,Ti)V, alloys that { 111) (1 12) twinning was the major deformation inode at room temperature, and they linked the twinning mechanism to a synchroshear process (Kronberg, 1957). In HLf-22Nb-64V alloy, Chu and Pope that at elevated temperatures (25-300 " fairly ductile due to ( 11 1) ( 112) type defori~ation twinning without dislocation slip. Twinning at low temperatures (below 25 "C) in ternary alloys of composition Hf-25Nb-62V and Hf-26~b-64Vwas investigated by conventional and high-re 1998). A large concentration homogenized alloy confirme been reduced in the ternary deformation twin clusters after compressive deformation was similar to that of stacking faults in the undeforrned alloy The twin plane was deteri~inedto be the net of vanadium atoms which comprises the highest density atomic plane in the Laves phase. The reason for the intrinsic brittleness of Cl5 Laves phase at room temperature is believed to be that slipI twinning, and stress-induced polytypic transformation are all accomplished by synchroshear rather than by ordinary shear processes. In each case, tlie rate-lin~iting step could be the motion of synchro-Shock tions wh~chis d i ~ c u l at t low temperature. (I 994) described the crystallography and geometry of the defects responsible for twinning in the Cl5 structure and proposed a twin nucleation model based on the partial source m e c h a n i s ~ (Pirouz, 1987). He also discussed niechanisnis by which synchro-Shockley dislocations may become mobile, such as the possible role of vacancies in the cooperative atomic shuHing associated with the motion of a kink.
res
Following the original remarks by Laves (1952), a s were made to produce twins in the en /?-CuZn single crystals were aded at several temperatures after various thermal treatments, pseudo-twinning of the (11?)(111) system (Table 1) was observed to occur when some disorder arose (~ouchardet al., 1970). It was concluded that the absence of ‘twinning’ in the ordered stale was due to a ‘detwinning’ ineclianisrn rather than i~capabilityof ‘twinning’. i-50Ti-3Fe alloys, deformation twins of two pseudo-twin types with the {112} and 14) twin planes were identifie~in the TEM study by 00 et aE. (1985). T ~ e ~ cycling a l under a constant load produced a high density of these twins. The first
structure. An apparent (1142 twin ‘double twinning’ rcus, 1967), but this was not believed to be the case since the observed composition plane was the twin plane 1985). ~ o b ~ r liJty nl. (1990) further t ~ ~ e r ~ o - ~ e c h ~snt ri ~ c a~ l~ t h e n oi nf gthe alloy and f o u that ~ ~ defor~~ation twinning on { 114) planes, together with concurrent dislocation slip o f the (0 10) type, enabled the polyc~ys~~liine ~ a t e r i to a ~yield greater than 50 % r o o m ~ ~ e ~ ~ e r a~ucti1ity. ture ~ndividual twins did not grow thicker than 50-150nm, tzrid the density of twins i ~ i ~ r e a swith ~ d the extent of cold working.
3.2.2 DO3 (cFI6) ~
~
~
~
As was discussed in the Introduction, the profuse
1,/&
and 4 5= 8 as listed in
Table 1.
atomic s h u ~ i n g(q > 2) acco~panied the twinnin process. Therefore, the (114) twinning does result in
ati ion twin bands were ~ e ~ o r t eind Fe3 y a n ~ ~ b of e ri~vesti~ators (e.g. Cahn a Coll, 1961; Guedo et al., 1978). Using the ~ o r i ~ u l a t i o n evis and Crocker ~ 1 9 ~ ~ , 1 9 ~ 9 ) ~ (199 1) derived the c r y s t ~ ~ l l o ~ r ~ pelements hic of
Variation of the twinning shear ( g ) with the axial ratio in hcp metals and DO,, c o ~ p o u n d s
~
T w i n n i ~and ~ ~ e c ~ a n i Belzavior ca~ deformation twins in Fe,Al to be I(i = ( 2 i i ) , q l = [ i l l ] , K, = (211), r2= [TTI].The density of twin bands in specimens compressed at liquid-nitrogen temperature was higher than that in the deformed specimens at room temperature. res
3.3.1 DO19 (hP8) C O ~ ~ O ~ ~ ~ ‘ ~ in hcp metals was reviewed by Yoo (1981) who showed a plot of twinning shear vs. the axial ratio. This g vs. ;1 plot was extended to include A, compounds of the DOl9 structure as shown in Figure 1 1. The largest and the smallest axial ratios correspond to A = 1.70 for Pt,U and /I= 1.43 for A1,Ce. There are many compounds which have axial ratios in the range to those five (Mn,Ge, In) shown in Figure Fe,Mn, Ni3Sn, Ti, tion twinning has been observed in only one of them, viz., Ti,Al, The {ioiz} (ioii) twin system (Table 2), which is most comnion in hcp metals and alloys, gives a socalled ‘tension twin’ (in reference to an uxiiaxial applied stress along the c axis) because ;1< 1/3 for all known DO,, compounds. That atomic shuffling is necessary in the disordered hcp alloys ( q> 2) is demonstrated with the aid of Figure 12. Th divided conceptually into shear of atomic motif shuffling. Figure 12(a) shows the homogeneous shear the unit cell for the (10i2)[TOll] 19 structure. In the fully disordered n four layers (q = 4) must shuffle by small additional displacements, as indicated by short arrows in Figure 12(b). In the case of the fully ordered state, on the other hand, all the a t o m in eight layers ( q = 8 ) must shuffle, but the i ~ p o r t a n t difference between the two cases is the fact that in the ordered DOl9 structure the atomic motif pairs situated at tlie four face-centered positions of the unit cell must shuffle by much larger displacements as shown by long arrows in Figure 12(c). Christian and Laughlin (1988) suggested the term ‘interchange shuffling’ or ‘order shuffling’ for this process. Inspecting the shLifI-ling displacements of those atoms labeled A in Figure 12(b), one notices coordinated combinations of a ~ t e r n a t i n ~x- and ymechanisms (Jaswon and ove, 1960) which together appear like a ‘ring mechanisni3 in the clockwise direction, This is consistent with the synchroshear of
417
atoms in the core of an edge zonal twin dis~ocationin the A3 structure ( senbaum, 1964). In an A compound of the D structure, such a coordinat motion of atoms is disrupted because of the required interchange of A and atoms over relatively large distances as shown in Figure 12(c). This is the likely twinning 1s intrinsically In the tensile deformation study of Ti,Al by Lipsitl et al. (1980), no twins were observed over the temperature range of 25-900 “C; however, some extended Pdults, or micro-twins, were observed at 900 “C. The occurrence of twin-like def~rmation bands in Ti-24Al alloys q d from observa temperatures was ted from by Morris and Morris (1991). but no confirmation of the twinned orientation by select~d-areadiffraction (SAD) patterns was given. Qhtsuka (2991) reported that twin-like deformatioii bands were observed during creep tests of Ti,Al polycrystals, and the occLirrence of
i ~ l? ~ Atomic ~ econfigurations associated with the process of (1012)[70!1] twinning in the DO^, structure: (a) homogeneous shear t r a n s f o r ~ ~ t i o n(b) , atomic shuffling, and (c) i ~ t e r c ~ a n gshuBing e
418
Meclzcanicul Properties
these bands became pronounced with a deviation from stoichionietry to an Al-rich composition. eformation twinning was reported in a liyperstoichiometric Ti-34A1 polycrystalline alloy in the temperature range of 800-1 100 "C (Lee et al., 1995). The doininant twin system was identified to have K1= (IOU),tlr = [ioiZ], li=j = ,(mi3), y ~ = ? [3032],and g = 0.122; and { 1122) and { 1012) twins were also observed with a relatively low density. The { 10T1) twins showed fine stru~tureswhich were characterized as having { 1012) habit planes, suggesting that the fine plates were secondary twins. Most deforlnatlon twins were niacroscopically lens-shaped and their thickness appeared to increase with increasing temperature. Also, the volume fraction of deformation twins was observed to increase with temperature in the range from 800 "C to 1000 "Cand with increasing strain at 1000 "C. 3.3.2 DR,y (
~~
~~
1~
~~
o~
~
n
A compressive deformation study of Ti,Si, single crystals at very high te~~peratures (1300-1 500 "C) was conducted by Umakoshi and Nakashima (I 994). Numerous twins were observed in deformed samples, and these were characterized to be the {10i2){10ii) type twin in terms of ravais indices based on the D8, structure. Th tion dependence of the SS suggested that the plastic deforination of Ti,Si, might be controlled by this twin system.
Deformation twins were observed in Cu,Sb which is the prototype of the C38 (tP6) structure (Paxton and Entwisle, 1985). Twins in five separate crystals were analyzed by two-surface inethods iii order to estimate the composition plane and the angle of shear. The twinning elenients were deter ned to be KI = (T12), rll =[ni 11, K = (I 1 3 , qt= [i111 =[liO], aiidg=0-152. Deformation twinniiig has also been reported in other non-cubic superlattice structures. Nesbit and investigated slip and defornzation 1, (1110) structure of a fixlly ordered i,Mo alloy and found that wide SZSFs and deformation twins were pronounced features of deformation. Tawancy (1 98 1) reported that the ordered Pt,Mo prototype phase in certain Ni-MO-Cr alloys (commerastelloy) deformed predominantly by twinning. In aged samples of Ni-25Mo-8Cr alloy, Kuinstr and Vasudev 4) observed that the ,er) precipitates of the deformation mode of
o structure changed from crystallogra~hicslip to microtwinning of the true-twin type at large strains. I
The homogeneous twin nucleation suggested by rowan (1954) is based on a model that a l e n ~ i c u ~ ~ r uiided by twin dislocation loops of lying on the K , habit plane. The of a twin and the stress field of a calculated using a dislocation rth, 1991). According to nucleation of a deformation twin by homogeneous lattice shear can take place with the aid of a local stress concentratio~iin the absence of any thermal activation. In. the analysis of a competitive relationship between twinning and crack initiation in hcp metals, Yoo (19'79a,b) treated the twin lamella as if it were a slit crack, being ~ c t i v a t ~by d a ~ ~ , ~ mode-ZI shear slate. Extending this concept further, Lebensohii and Toiiii (I 993) investigated the stress state associate^ with the activati elastically anisotropy materials. inclusion theory (Eshelby, 1961 bifurcation theory (Johnson and and Yoo (1990) analyzed the effects of applied stress on the elastic strain energy of an ellipsoidal twin and the stress state around the twin tip in tetra~onal crystals, and this work was later extended to hexagonal crystals (Yoo and Lee, 1991). Using a modified Monte Carlo technique, Lee and Yoo (1994) made an attempt to atomistically simulate heterogeneous twin nucleation in a bcc crystal under stress. The i n ~ ~ ~ ofactors r ~ ~ nthat t control the twin formation at heterogeneous nucleation sites depend on the niicrostructure and the internal stress state of a given material. The following statement made by Frank arid Stroh (1952) still applies today for iiiter~e~ailic compounds: 'Bounded slip-bands, ~ s ~ r i k - b a ~plates ds, of deforiiiation, twin or inarteiisitic transformation product, and thin cavities, cracks or notches under tangentid stress, all produce stress fields of the same kind: in various circuinstances any one of these may serve as the source of any other.' In addition to these bouiided lattice inhoinogeneities, we also include surfaces, grain boundaries, and heteroyhase interfaces as possible sites for twin initiation, Once a twin i s nucleated, the increase in twinned volume takes place by its lengthening in the yl direction, widening in the S direction, and tliiclseii~ng in the direction normal to the K , plane. The former
419
~ w z n ~ iund n ~~ e ~ h a i ~Behavior i~al two results from propagation of twin dislocations at the incohereiit twin boundary, and the latter from acquiring additional twin dislocations, either by the incorporation of slip dislocations into the twin or by the nucleation of partial dislocation loop(s) with the aid of an internal stress concentration.
Stress-indLiced ‘microtw~nning’in the y-phase of a twophase TiAl alloy was observed during an in situ field ion microscopy (FIM) study by Larson and Miller (1998). The thickness of a typical twinned region was estimated to be between 0.5 and 3.5 nm, so these may be called nanotwitis. These twins differed from preexisting y / y lamellae in that they were created in response to the internal stress associated with the lattice expansion due to field-induced strains in the sample. This twinning process is similar to the in situ TEM observation of twin nucl~ationat a r e - e ~ t r a ~ t corner of the surface of a nearly perfect zinc whisker (Price, 1960). Two types of twin formation at a crack tip are sclie~iiaticallyillustrated in Figure 13, as observed by han-Couson (1993) and Appel et al. (1995). Tlie former type of anti-twinn a mode-1 crack (Figure 13(a)) was discussed in on 3.1 (b). When the mode-IT or mode-111 corn eiit of an external loading is mixed with a (1 11) [ 1TO] mode-1 crack as depicted in Figure 13(b) or 13(c), coplanar deformation twinning or ordinary slip at the crack front can occur so as to shield the crack-tip shear stress field (Yoo et al., 1995). crack-tip twinning was directly the leading process of translagation in Ti-rich TiAl alloys by Appel et al. (1995). Grain boundaries provide the most potent sites for twin nucleation in polycryst Iloys, e.g. in TiAl (Sriram et al., 1994; Jiii and 299%; Seo et al., 1997; Morris and Leboeuf, , FePt (Gao and Whang, 1994), Ti,Al (Lee et 1995), and NbCr, zantzis et al., 1997). At a = 9, 1221) type tilt ndary in the Ll, structure, in formation in one grain by a pile-up of slip dislocations in the adjacent rain is crystallograpliically possible (Yoo and King, 1988). In the case of C15 Laves phases, such as NbCr, and (Hf,Nb)V, alloys (Luzzi et al., 1998). parallel twin bands of tliickness ranging from lattice dimensions to app~oximately8 nm were observed to form collectively in an a~tocatalyticmanner ( azantzis et al., 1997). The occurrence of a group of parallel twins, rather than a single twin, is a simple nia~ifestationof free
energy i~inimizationfor the system ( 1994). In two-phase alloys, it i s now well establish~dthat deformation twins are readily nucleated at a variety of homo- and hetero-phase interfaces. Twin initiation in the y-phase at yl/y2 and a2/y lainella phase TiAl-Ti3AI alloys IS widely conventional TEM, REM, and in situ straining investigations. For example, the Figure 14 shows (a) a typical example of the formation of a SISF at an a,/y interface and two ~istinctly different types of twin nucleation at (b) a true-twin
initiation of a true twin in a y 2 lamella at the head of a pileup of ordinary dislocations in a y, lamella was analyzed to be energetically feasible (Yoo c Hsiung and Nieh (1997) reported direct TE for twin nucleation as a result of i n t e r ~ ~ cd~i ~ a locatio~ pileup against ledges along the a2/y lamellar boundaries
m I
[m]
mode I
r
mode I
mode I
mode II
mode 111
~ i 13 ~Crack-tip ~ twiiining r ~ and slip in the L1, structure: (a) two symmetric tension twins, (~b)coplanar primary twinning (edge type), and (c) coplanar ordinary slip (screw type)
Fi Slacking fault initiation and twin nucleation in 1'TiAI deformed about 3% at room t e ~ ~ e r a t u r(a) e : formation of a stacking fault at a misfit d i ~ ~ o c a tof i oan ~ a,ly interface in Ti-47Al-lCr with additions of Nb, Mn, Si, and B, (b) generation of deformation twins (I and 2) and an embryonic twin (3) at an interface of the ~rue"twin boundary contain in^ misfit dislocations (arrowed), and (c) nucleation of a deformation twin at a small-angle grain boundary in the samc alloy as in (a) (from Appel and Wagner, 1998, reproduced with permission of Taylor & Francis)
42 I
2
2
Twin formation by a pileup of niterfacial dislocations on an boundary in Ti-47Al- 2Cr-2Nb: (a) a bright-field TEM image showing several (11I)[ 1121 d~formatioiitwins formed in lamellae, and (b) a s c ~ e ~ a t i c representation of twin formation showing mterfacral ( stair-rod (bJ, and twinning (b,) partial dislocations (from Wsiung and Nieh, 1997, reproduced with permission of Taylor & Fraiicis)
in a two-phase TiAl alloy as shown in Figure 15. When T i - 4 7 A l - ~ ~ r - 2alloys ~ b were stressed at hard orientations (i.e. either parallel or normal to the a,/y i~terface),a defori~at~on twin in a y lamella was observed to pile up against an aZ/y boundary. The response of the adjacent cxZ lamella was the generation urgers vector on a of dislocation§ with the e + pyrainidal plane in a ~ i - 4 7 ~ 1 - 2 ~ r alloy - 2 ~ bat room temperature (Codfkey et ul., 1 observed to occur in a binary Ti
but not at room temperature ( iezorek et al., 1997). It is conceivable, under an appropriate applied stress of the opposite sense, that this process could be reversed to facilitate a twin ~ucleationmechanis~in TiAl as a result of slip transfer from Ti,Al.
Many types of dislocation sources for twinning partial dislocations in cubic and ~ e ~ a g o ncrystals al have been
proposed, as given in several reviews (Christian and Crocker, 1980; Yoo, 1981; Narita and Takamura, 1992; C h r ~ s t and ~ ~ ~~ na h a ; a n ,1999, and some of these may be applicable to deforniation twinning in cubic and non-cubic superlattice structures. Dislocation models for twinning may be divided into two main types: (a) pile-up mechanisms (Cohen and Weertman, 1963; Fujita and Mori, 1975) and (b) pole mechanisms (Cottrell and Rilby, 1951; Venables, 1961). The pileup inechanisms are based on a ‘glide’ source, and the pole mechani~msoperate from a ‘prismatic’ source (Venables, 1961). In addition, Maliajan and Chin (1973) proposed a model for the formation of twin embryos stacking faults in close-packed crystal Many dislocation models for twinning in the L1, structure were proposed, and these have been discussed in recent review papers (Yoo, 199817; Yoo and Fu, 1998a). First of all, using a (111) APB energy of 200 mJ/m2 ~ ~ a r c i n k o 1963) ~ v ~and ~ ~ an , SISF energy of less than 10mJ/m2 ( A ~ r i a ~ o v set~ ial., i 1971) for reeiiberg (1970) investigated many plitting configurations of both glissile and sessile types aiid identified one of them (tlie socalled C-configuration) to be a twin source. The model based on this twin source was found to be a distinctly glissile type in that imperfect twins niay result, i.e. not all STSFs necessarily lie in adjacent planes. On the other hand, tlie TEM study of an ordered CuAu-I alloy by A~rianovskii et al. (1971) proved that deformation twins origimte within the domains due to the action of a spiral source (i.e. a prismatic source) and not at their interfaces. A nuniber of possible mechanisms for the formation of twin sources in y-TiAl have been suggested on the basis of dislocation dissociations or transformations in their core structure: (a) expansion of an SISF loop resulting from a [101] superdislocation (Fu and Yoo, ite dissociation’ of a [112]/2 et al., 1986; Girshik and Vitek, e niechanism anchored by a 110 >. /2 ordinary dislocation (Farenc et al., 1993; Voo, 1997). Using an i~ situ straining TEM technique, Couret et al. (1994) found direct evidence for a single pole at advancing twin tips in Ti-54A1. In a post analysis, Zghal (1997) observed a ‘pole d by a Shockley partial when it met with a forest dislocation of the ordiiiary type. To explain the experimental results of Figure 10 that the apparent SS for twinning in Ti-56A1 is in the range of 200 MPa at 1000 K and about 100 MPa at 1200 K (Inui et al., 1997), Yoo (1997) proposed a dislocation pole
mechanism 011 the basis of dissociation of a jog on an ordinary [ 1 l0]/2 dislocation. The prisniatic source in this case is schei~aticallyillustrated in Figure 16. An important finding of this work is that a repulsive interaction between Frank and Shocltley partials exists in the near-edge orientation due to elastic anisotropy, and this promotes the spontaneous expansion of the Shocltley partial loop. Based on their ~ o s ~ ~ 7 ~weak-beam z ~ ) ~ t ~ TEM ~ i rr-doped Ti-54A1, Hug and Veyssikre (1989) suggested a twin nucleation mechanism involving a [112]/2 screw ~uperdis~o~ation, but no direct evidence for pole dislocations was given. In the TEM investigation of the Ti-57Al-SCr alloy after compression along the near [001]-[011] orientation, Morris et [ 1 12]/2 superdisloal. (1994) observed p~edo~niiiantl~ cations and a microtwin created by the movement of [T12]partials on a (1 1I ) plane. Singh and Howe (199 I ) observed that nucleation of nanotwins occurred by superposition (overlapping) of extended SESFs on alternate (111) planes in Ta-doped Ti-48A1 alloy. Direct observation of twin formation by electron niicroscopy techniques is inherently difficult because of the dynamics of’ twin nucleation and propagation. While unambiguous characterization of the pole e ~roposinga pole dislocation(s) is a p r e r e ~ ~ i s i tfor mechanism (Farenc et al., 1993; Couret et al., 1994; Zghal, 1997), identification of an array of stair-rod dislocations is often used to support a pile-up mechanism by interfacial dislocations (Hsiung and Nieh, 1997).
In the past decade, a significant advance has been made in our understanding of point defects and diEusion in intermetallic compounds ( Point-defect properties and diffusion coefficients of titanium aluminides were reviewed recently by Yoo and Fu (1998a). In TiAl, no structural vacancies were predicted from first-principles totnl-energy calculations (Fu and Uoo, 19931, and the dominant point-defect types are therefore aistisite defects on both sublattices, as in the case of intermetallics having a close-packed structure such as Ni,A1 (Fu and Yoo, 1994). Thermal vacancy ~onceiitrationsin TiAl are predicted to be very low, even at high temperatures (e.g. co= 10A7-10-6 at 1500K). In explaining twin formation in TiS6A1 under low stresses and high tem~eratures(Figure lO), excess thernlal vacancies that can expand Frank partial segments are thought to play an important role in the early stage of the pole mechanism (Yoo, 1997).
423
.
-
.
ure 16 Schematic illustrations of a pair of Frank and Sliockley partials resulting from an ordinary [I 10]/2 dislocation of the edge type in the Ll, structure: (a) atomic configuration showing the primary (K,, yl) and conjugate (K2,yz) t ~ i n i ~elements, i i ~ ~ and (b) the in-plane and out-of-plane c o ~ p o n e ~(FX, t s Fy)of the elastic interaction force
Tn the absence of stress concentration, the sufficient condition for a chemical stress that comes solely from a vacancy sui~ersaturatioii is c/co>, 13. In an actual case, the climb stress can be overcome by a combination of the stress concentration and a more reasonable vacancy supersaturation (c/co < 13). In addition to the effect of vacancy condensation, clustering of supersaturated self-i~terstitial atoms could also assist the nucleation of deformation twins, The so-called radiation-~nduced ductility (RI reported in Ti-47Al alloys at 600°C ( ~ i s h i n u ~eta al., 1996) was attributed to the effective formation of twin embryos (prismatic sources) in the presence of interstitial-type Frank loops and the subsequent nLicleation and growth o f twins during plastic deformation (Yoo and Hishinunia, 1997). In austenitic stainless steels of the fcc structure, Frank loops of interstitial type are also formed during neutron or heavy-ion irradiation, and twinning is the predominant deformation mode in irradiated steels at room
temperature (references in Yoo and 1999). Two reasons, of energetic and kinetic origins, were offered to explain the st~bilityof large faulted Frank loops that act as het~rogeneoussites for twin.
the mobility of a composite Slisckley partial (Hug et al., 1986), which was originally called a I)-type Shockley by ~ e ~ r t m a and n ~ e e r t m a n (1964), nucleated on an interstitial-type loop is considered to be much lower than that of a hockley partial on a vacancy-type loop. Formation of Shockley partial dislocations by intersection of glide dislocations with Frank loops in fcc lattices was analyzed by Song twin nucleation model proposed by and the model by Yoo and Hishinim essentially the same end result: A t w i ~nucleus can
424
~ e ~ h aProperties ~ i ~ a ~
emerge from a Frank loop as a result of the slip-loop interactions. The main difference between the two models is that the former model gives an imperfect twin nucleus with the thickness limited to the number of slip dislocations that intersect the loop, and the latter provides a three-dimensional description of twin nucleation and growth beyond the twin thickness of six atomic layers. While the profuse activity of deforniation twinning in y-TiAl in two-phase TiAl alloys of fully l a ~ e l l ~ r microstructure is well establis~ed,it i s also known that d e f o i ~ a t i otwinning i~ becomes more difficult as the microstructure clianges from fine lamellar, to duplex, and then to fully singl~-phasey-TiAl over the compoto 54 at.% Al. This variance in be rationalized in terms of two factors: (a) the decrease in area1 density of heterogeneous nucleation sit interfaces and (b) the increase in SISF and s as the composition of y-TiAl varies fr chiometry (4 at.% Ti-rich) to hyperstoichioi~etry(4 at. % AI-rich). Investiga -y T~-(48-52)AIpolycrystalline alloys, Hall and 1989) reported that the ~icrostructureof the lloy after room-tei~~perature deformation consisted primarily of deformation twins and ordinary slip dislocations. In the same composition range, Si-iram et aE. (1994) noted that twin activity in all alloys increased with increasing temperatures, and at aiiy given tei~perature,the nuniber density of deformation twins was higher for the Ti-rich alloys. This has been ~ttribLitedto mi~rostructuraleffects, e.g. the fine grain size, rather than to intrinsic effects of A1 content. From the viewpoint of dislocation core s t ~ c t ~ r and e s their behavior under t of an applied stress (Cirshik tra et al., 1995), Vitek and his both SISF and CTB energies increase sharply with increasing A1 content, and thus, twinning stress is also increased. o far, this conjecture on the composition depency of twinning in y-TiAl has not received support from the available experimental data and theoretical results. For instance, the TEM *
ry weak composition energy, and the experimental showed an opposite ossible roles of extrinsic substitutional and interstitia~ imp~rities in twinning in intermeta)lic coniyounds are not fully understood. The absence of twinning in the a2 phase of two-phase TiAl alloys may
be attributed either to the d i ~ c u l t yas~ociatedwith the interchange shufffling neces (Christian and Lauglilin, 19 is an intrinsic eflect, as men to an extrinsic effect due to the presence of interstitial iinpurities. The extrinsic-effect the oxyge~-scavengingeffect ( due to tbe fact that interstitial elements have a higher solubility in Ti,Al than in TiAI. It was proposed by ~ ~ s u d e v aetn al. (1989b) that ~ ~ t e r ~ te~ ~t ~ea l~ ~ ~ t enhance the degree of directionality of the Ti-Ti bonds, and so their removal reduces this directionality and therefore the anisotropy of the Peierls stress in y-TiAl. Since the oxygen content in a2 phase can be as high as 20 OOOppm even when the average oxygen con the two-phase Ti-A1 alloy is only 40Oppm Partaix and ~ e n n a n d 1996), , this level of oxygen may be expected to hinder the twinning process in a,-Ti,Al.
Once nucleated, the possible roles of twins in rnechanical behavior may be generally divided into two opposite ways: (a) wlzile slip-twin andlar twin-twin interactions contribute to work hardening, they can also cause the nucleation of a ~ i c r o c r a c ~and , (b) extension of pre-existing cracks can be suppressed by slip and/or twinning at the crack tip. In bcc metals and alloys, the available experi~entaldata show a dichotomy in that deformation twinning can eKectively strengthen a material under some circumstances and weaken it under others ( eid, 1981). In intermetallic compounds of a low-syrnmetry crystal structure, on the other hand, twinning is believed to play beneficial roles in contrib~ting to both strengthening and toughening. The case of 11-TiAl, which has been investigated extensively in the recent decade, is an excellent example. cts
The roles of deforniation twinning in the strength of monolithic Ti, Ti,Al, TiAl, and TiAl, single crystals were discussed in the review papers (Yoo et ul., 1991; Yoo and ~ a m a g u ~ h2000), i, Figure 17 shows schematically the temperature dependence of compressive yield the c axis at a nominal strain rate of Ti,Al inon on is hi and Yoo, l990), TiAl (Kawabata et al., 1985, 1990), and TiA1, (Uamaguchi, 1986). An anomalous (positive) t~mperaturedependence of fly
Twinning and Me1:hanical Behavior
COMPRESSION ALONG THE C-AXIS
*-
.A
600
TEMPERATURE, T (K)
FSgure 17 Temperature dependence of the coinpressive yield strength of titanium and Ti-aluminide single crystals
occurs in all except TiAl,, and deformation twinning plays an important role in all except Ti,A1. In Ti, the deformation mode at temperatures below the peak strength (Tpx650 K) was almost entirely (> 90%) by { 1122}( 1123) twinning. On the other hand, at temperatures above Tp, the major mode was (c + a) slip and the minor mode (10-30%) was {lOil}(lOT2) twinning. Therefore, the rise and fall of the yield strength with increasing temperature may be related to an apparent transition from twinning to slip involving the displacement vector along the same (c+a) directions (Paton and Backofen, 1970). Whereas, in Ti3Al, no twinning was observed, and the anomalous yield behavior at a relatively high level was attributed to the temperature dependence of the mobility of edge superdislocations (Minonishi and Yoo, 1990). In Ti-54A1 single crystals, no specific role of twinning in the yield strength anomaly was elucidated after the early experiments of Kawabata et al. (1985, 1990). The recent experimental data of Ti-54Al and Ti56A1 single crystals (Figure 10) by Kishita et al. (1996) indicated that the apparent CRSS for .( 1 1I } (112) truetwinning had a strong temperature dependence so that a transition of the active deformation mode from { 11l}(lOi) superlattice slip to deformation twinning occurred at about 750°C. The latter result, which is
425
only a part of the extensive experimental data (Inm et al., 1997), demonstrates also how much more difficult twinning nucleation in y-TiAl becomes in the absence of heterogeneous nucleation sites. such as interfaces and extended defects. The TiAl, compound shows the weakest yield strength among the four,
The von Mises criterion is a necessary condition for ductility in homogeneously-deformed polycrystalline solids, and methods for determining whether a set of slip systems satisfies the criterion are well established and presented in many textbooks (Groves and Kelly, 1963; Hirth and Lothe, 1982). In alpha-brass (Cu30Zn) polycrystals, the numbers of active { l l 1) slip planes were counted by etching freshly cut surfaces in deformed samples (Fleischer. 1987). For all strains and grain sizes tested, fewer than five independent slip systems were active in about 40% or the grains. It was therefore suggested that the demand of the von Mises criterion for five slip systems is not obeyed because the assumption of uniform deformation is not correct. As for the role of twinning m plasticity, Kocks and
424
Mec han ical Properties
estlake (1967) pointed out that ductility of polycrystalline hcp materials may not require five tlidependent slip systems, and that the internal stresses rain boundaries when inelastic strains are not compatible may be relieved by localized deformation by twinning. These statements are coiisistent with the experii~entalresults of active deformation modes observed in the case of y-TiAl alloys of fully lamellar ishida et al., 1998). A eth hod was intr~d~iced to determine if a set of twin systems satisfy the von Mises criterion by taking into account the midirectionality of twinning shear ark, 1989). The results are, for example, ing observed in the B2 structure criterion, but tlie basal and bined with the twin system in h.c.p. ~ e t a do ~ snot. 6.00 (1998) examined which of the slip and twin s y s t e ~ sobserved in TiAl and Ti3A1 are needed to satisfy the von Mises criterion in the cases of single (y or a2) and dual (a2/y)phases. In the y phase, the “easy’ { 111) ( 1TO) 12 slip system combined 2) true-twinning system do not herefore, the ductility some of the ‘hard’ ms must have been activated. In be met by slip with er with ~ l o i T } ( i o i ~ )
cube-type texture could be s i ~ ~ u l a t equite d satis~actoSS of equal magnitude to the rily by ascribing a ordinary slip and t modes and a value 2.5 times higher to tlie superdislocation mode. Typically, cast material of y-TiAl is made up of Pimellar grains with a strong morphological texture, since the lamellae are preferentially aligned perpen~icularto the direction of solidi~cati~n, The anisotropy is large and can be calculated rather accurately with the tal plasticity codes by employ in^ the (~ebensohnet al., 1998).
Fracture behavior of an intermetallic compound depends strongly on tts microstruct~rewhich is, tn turn, dependent on its chemical coi~position.The fracture behavior of binary Ti-A1 alloys at room temperature can be discussed with the aid of Figure 18 (Mitano et al., 1991), which shows s ~ ~ e ~ a t i c afour lly typical microstructures: (a) fully lamellar structure at =:44Al leading to intergranular fracture, (b) fine lamellar structure at x 48A1 leading to interla~ellar
40
orie~itation relationship (Willia 197Q),i.e. (1 1l ) ~ , / / ( O O O l ) ~and [lTO]y//[l 120],, the von es criterion may be satis~edby the easy ordinary and t r ~ ~ e ~ t ~ i i isystems n i i i ~ (without the hard s u ~ ~ r l a t t i cslip e system) provided that all three variants (six with directions) of the orientation relationships are present. Texture analysis f o l ~ o ~ i nthe g compression experiment of TiAl by Cheong et al. (1996) indicated that ~ l e ~ o r t i i ~ ttwinning ion occurred at very low strains and 850 “C,where thermal activation makes
30
20
!.E 8
w
10
t
0
0
44 t.%A!
rmation twi~ning.The topography of the yield surfaces depends on the relative values of the CRSS of odes (Mecking et al., 1996). In r m~crostructure,the hot-rolled
Composition dependence of the fracture toughness of as-cast and isothermally-forged-and-annealed (IF-A) binary Ti-A1 alloys and schematic drawings of typical crack paths (from Mitao et al., 1991, reproduced with permission of Taylor & Francis)
Twinning and Mechanical Behawor and/or iiitergranular fracture, (c) duplex structure at x5OAl causing cleavage fracture in the y phase, and (d) y-phase polycrystals at x 50Al exhibiting transgranular fracture. In addition to these, Ti-54/11 single crystals may be classified as a fifth type of structtire (e), for which fracture toughness was obtained to be KO (provisional .KIc value) > 7 ~ P a in ~ Mn-d~ped m crystals (Phan-Courson, 1993) and K1,* 1 MPal/m in 1997). The latter cal stress intensity ) = 0.89 and 1.04 MPaJm for the (100) and (001) surlaces at 0 I< (Yoo et al., 1995), respectively. This indicates that y-TiAl is intrinsically brittlc without aay appreciable crack-tip plasticity. The seven-fold increase in fracture tough-doped single crystals may be attributed, in enhanced crack-tip twinning. The isothermally forged and annealed (IF-A) Ti50.2A1 alloy which consisted of single phase y grains the lowest fracture toughness (KQ = in Figure 18, but this is still much higher Ti-54Ai single crystals by a factor of about 10. This difference may be due to the role of grain boundaries as sites for slip and twin initiation as the crack propagates through a polycrystal. In two-phase m u l t i ~ o ~ ~ oTiAl ~ i ealloys, ~ t fracture toughness is measured to be reasonably good, e.g. Ktc = 16-23 MPa t/ni from a fully lamellar structure (Chan and Kim, 1995), KO= 10-32 variety of microstructures (Figure MPaJm from a lamellar structure (Chan and Kim, 1994, and KO > 50 MPaL/m from fine (0.1-0.4 pm) laniellar iiiicrostructures (Liu et al., 1996). Deve and Evaiis (1991) reported that the fracture toughness o f TiAl alloys containing equiaxed y grains was greatly influenced by defori~ationtwinning, which resulted in it twinzone, a source for the crack-shielding effect. r, the work by Mitxo et d. (1991) indicated that the occurrence of cleavage cracking in the y grains was also an important controlling factor. Deve et al. (1992) reported that twinning occurred within both a process zone and crack bridging liga~eiits, with. about equal contri b ~ t i o narising ~ from the process and bridgiiig zones. Obviously, in two-phase TiAl alloys, the presence of various lamellar interfaces influences the crack-tip plasticity, which is strongly orientation dependent (Chan, 1992; Mi tao et al., 1992; Naltano et al., 1995; Yoltosliirna and ~ a ~ a g u c h1996), i, and therefore affects translamellar and interfacial microcracking in the crack-tip region in a complex manner (ChPtn, 1993). One area not examined in detail i s whether synergistic toughening
427
effects can be obtained by comb~ningmechanisl~s, which might include dLictile-p~~s~ blunting, ductilephase bridging, ligament ~oughening,and twin toughening (Chan, 1992). A quantitative assessment for the specific role of twinning in the crack-tip plasticity is yet to be made.
Deformation twinning is an i r n ~ o r ~ dmode n t of plastic defornia tioii not only during strain-controlled, shortterm deformation at ambient temperatures, either by monotonic loading or by cyclic loading (Sastry and Lipsitt, 1977; Yilsuda et al., 1996; but also during long-term deformiltion at elevated temperatures and very slow strain rates, such as creep conditions (Loiseau and Lasalmonie, 1984; Oikawa, 1990). In a Ti-53.4A1 sample crept in compression at 950 K under 317 MPa up to 0.3 strain, observed many deformation twins in optical microstructures. The iiuinber of grains that contained twins increased with increasing creep strain. As the total strain increased, bending of twins was observed, which indicates a significant slip deforrnatioii iii the matrix. Sinall recrystalli~edgrains were found locally at an early stage of deformation, and the recrystallized area increased as the creep strain increased. In as-cast binary TiAl alloys in the co~positionrange of 44-51 at.%Ai, Mitao et al. (1991) observed tkat, as aluminum content increased, dynamic re~rystalli~ation became more difficult at a given temperature, result in^ in an increase in creep resistance, Most of the experimental data in the literature, pertaining to the present subject, are on i~u~ticomponent alloy systems, part~cularly those based on titanium aluniinides. It is therefore difficult to discern the general mechanistic role of twinning in creep flow and fracture behavior of i n t e ~ e t a l l i ccompounds. In Ti-rich TiAl, alloying (Tsujimoto and Hashi Hall, 1990) increase However, these transition-metal al~oyingelements do not affect the ductility of Al-rich single-phase y-TiAl. An efYect of Mii addition to e ~ h a n c the e ~ropeiisityfor twinning in two-phase TiAl has been reported by several investigators (Hananiura et al. 1988; Hug arid Veyssiere, 1989; Tsijimoto and ~ a s h i m o t o 1989). , The reasons for this t w ~ n ~ i nen~ancement g were proposed to be as follows: (a) the pinning effect of the trailing twin dislocations that is caused by Mn segregation to transformation-twiii boundaries (
1 9 8 ~ (b) ) ~ the lowering of the SI F energy (Hug and yssikre, 1989), and hence the CTB energy anamura et al., 1988), and (c) the weakening of 1-A1 covalency bonds n replacing A1 in the Atsites (Tsujimoto and imoto, 1989), which is by electronic charge density calculations el! al., 1990). Also, additions of Ga (Mad r (Vasudevan et al., 1989c) were reported to promote a solid solution softening and an increased ~ o b i ~ i of t ytwin dislocations. In the latter case, it was concluded that the Er gettering of interstitial elements from the TiAl lainellae was
2Nb alloys of the duplex microinicrostructures at 700 "C under low stress of 280 MPa was observed to be the extensive twiiining activity which was already present after only 1% strain ( ~ o r and ~ s Leboeuf, 1997). Figure 19(a) shows an example, taken at it low magnification in the TEM, of one of the large y grains that was completely traversed by several twins, This twinned microstructure was confirmed to be produced by emission and glide of twin dislocations on {I I 1) planes. The extensive twinning activity was also responsib~efor the activation of a large number of ordinary dislocations at intersections with some of the 171 y interfaces. Details from some of these intersections ) are shown in Figures 19(b) and 19(c). It is noticed that some of the ordinary dislocations remain on the { 111) planes of the twin interface, as was situg TEM studies of twin also e v i ~ e n c ed~ ~ ~ inr ~ propagation in y-TA.1 by Couret ef al. (1894). Some of the twins do not propag the entire grain, as is seen w i t h the area labe igure 19(c). It appears was iiot sufficieiit1y that the local stress ca high to provide the driving force for twinning across the entire grain, as suggested by Jin and contribution of deformation twinning to the total creep strain was about 10 and 50% in the low and high stress regmes, respectively. A two-phase alloy of Ti-47.5AI-2.5Cr composjtion was studied under two different heat-treatment conditions in order to obtain different microstructures orris, 1995). While deformation twinning was observed in 90Y0 of the equiaxed y-grains, it was only observed in 50% of the coarse lamellar grains. This observation led to the conclusion that the
internal stress distri~utionin the equiaxed microstructure was more favorable for the initiation of twins than in the lamellar grains. In another two-phase Ti-48 alloy with equiaxed, fine-grain microstructure, deformation twinning was observed to occur in less than 50% of grains, while ordinary were active within all the y grains ( Leboeuf, 1998). The twins were responsi a sniall fraction of strain, but they led to a subdivision of the ~icrostructure and ~ e t e r ~ i n e(directly d or indirectly) the hardening process observed during the pninary stage of creep deformation.
peratures (650, 700, and 750°C) under tensile stresses varying between 150 and 300 M 1998). In this study, deformation twins were observed already after a small strain (0.3%) in y-grains as well as in lamellar grains. The lamellar spacing was reduced by a factor of two at strain of 0.3%0,which was not much affected by higher creep strains or applied stresses. is in good agreement with the results obtained by et al. (1997) who studied the twinning behavior o laniellar grains in a duplex alloy. They showed that twinning parallel to the lamellar in was other saturated after 0.2 to 0.5% creep strain. hand, in the y-grains, the mean interface spacing was found to decrease with increasing strain and stress, whereas it decreased with decreasing temperature. These results suggest that twins contribute to the deformation process of the y-phase in Mn~bearing alloys during the whole creep life. Instability of twin structures during creep flow at a low strain rate is an important factor for deformation ~crostruct~ires developed at elevated temperatures. The investigation by TEM mid HREM of Ti-48A1-2Cr deforiiied in tension at 1073 K (Appel, 1999; Appel et al., 1999) revealed that the twin-matrk niterfaces often contain local networks o f dislocations which are similar to subboundaries. These networks acted as the soiirces for dislocation loops which often caused f r a ~ i ~ ~ t a t i o n of twin lamellae. This result suggests that the competition between the generation and reconstruction of twins depends obviously on the applied stress conditions, and, consequently, a broad variation of the creep microstructures becomes plausible. A fine-grained ~ ~ - 4 7 A l - 2 C r -alloy ~ ~ b of fully lamellar structure (the same alloy used [or KQ measurements by Liu et al. (1996)) was creep tested at 760 "C under stresses between 69-723 MPa (Wan al., 1995). It was found that this alloy exhibited higher
Twinning und ~ ~ ~ ~ aBehavior n i ~ a l
429
Deformed microstructure o f the duplex Ti-4SA1-2Mn-2Nb alloy: (a) example of extensive twinmng traversing equiaxed y grains after 1% creep strain at 700 "C under 280 MPa, (b) and (c) details from areas A and B in (a) (from Morris and Leboeuf, 1997, reproduced wtth permission from Elsevter Science)
creep resistance than other TiAl alloys with s i ~ ~ l a r needed, particularl~in the kinetics of twinnin compositions. In this case, the evolution of deformaorder to develop 'constitutive laws for twinning'. tion substructure revealed that the deformation at the Some highlights and unresolved issues are noteprimary creep stage was mainly accommodated by soft worthy. The i~echanist~c role of Shockley partial lamellar grains tlirough the movement of interfacial dislocations in the inotion of incoherent twin bounddislocations in the a,/y and y / y boundaries, and aries is well established in the case of the El, structure. defor~ation twins were found to form within 1' They are instrume~tal not only in the growth of lamellae only at the secondary creep stage (Hsiung deformation twins, but also in the stability of and Nieli, 1997). If the blocking of the sliding motion traiisEorniatio1.l-induced true-twin and pseudo-twin of interfacial dislocations i s a controlling mechanism ~oundaries. Whereas, the possible role of rank of twin nucleation, the creep rate at the secondary partials in twin nucleation and growth i s essentially stage should increase with the number of interfaces. unresolved. The experimental finding of the radiationThe question as to whether twinning must be induced ductility in Ti- 7Al alloy at elevated temperaconsidered as contributing to a work hardening ox" to tures is remarkable in that a nearly two-fold increase in a niicrostructural softening process is difficult to tensile ductility was recorded without any appreciable answer. Several investigators (Seo et al., 1997; Hsiung changes in the yield and flow strengths. Additional orris and Leboeuf, 1998) concluded experiinents of this kind to validate the ra~iationfrom their studies that twin interfaces binder dislocainduced ductility result (which was attributed to an 'on motion and, therefore, harden the mici-ostructure. enhanced twinning activity) are needed for TiAl and ut another possibility is that twinning reorients the yother intex"~eta11ics~ In Ti,Si3 single crystals at elevated phase into a more Cavorable orientation for deformatemperatures, compressive twinning was reported to be tion, i.e., the formation and migration of dislocations the major mode of plastic deforniation. This is very become easier in the twinned crystal than in the parent encouragin~as we respond to the demand by materials y-crystal, which would result in a microstructural technology for tougher and stronger materials which softening process (Skrotzki et nl, 1998). Further work have very high melting temperatures and often possess on microstructu~alc~aracterizationis needed to clarify limited operative slip systems. There have been the answer to this question, as well as investigati~nsof increased rgsearch activities on twinning in Laves the interaction of twins with other eleinentary prophases, particularly of the Cl5 type. The motivation cesses like dislocation slip (Skrotzki, ~ 0 0 0 ) and for this research effort stems from the fact that there dynamic recrystallization. are hundreds of binary Laves phases and thousands of ternary Laves phases, mizny of tlneinn sharing the same family of crystal structures and deformation inechanisms. This i s a case in which a little unders~andin~ would go a long way ( In 1991 a couple of workshops on 'Scientific Issues on I n t e ~ e t a l l i cC o ~ p o u ~ -d Status s and Forecast for ~ u n d a i n e n t Research' ~l and 'Coniputztional Issues in avior of Metals and Intermetallics' The author would like to thank C. L. Fu and J. IS.Lee y reports of which address unrefor collaborations over the years and S. solved issues on the role of twinning in mechanical Appel, T. R. Bieler, . W. Cahn, J. W. Christian, E. behavior (Uoo et al., 1993; askes et al., 1992). Since COO,and R. C. Pond for helpful comments on the then, s i g n i ~ c a nadvances ~ have been made in our manuscript. This work was sponsored by the Divisi understanding of these issues owing to the extensive Materials Sciences and Engineering, Office of use of advanced techniques for microstructural and Energy Sciences, US t of Energy, under analytical cliar~cterizatioiiand the interpretive and Contract ~ E - A C 0 5 ~ 0 0 predictive capability of computational materials science. For example, the question 'is there a CRSS for twinning'!' has been partially answered in the case of y-TiAl, and interfacial structures and energies of twin boundaries are reasonably well understood. But, Anneuting t w i m those twins formed by a thermomore effort in experimental and theoretical interdiscimechanical process of cold-work and anneal plinary research at multi-levels of length scale IS
Twinning and ~peclzanicalBehavior
43 1
Type II twin: when Ka and q t both have rational, ~ n t i - t w ~ n n i twiniiing n~: on the primary K1 plane but low-index elements in the opposite direction, --ql C o ~ ~ o utwin: n ~ when ~ all four crystallographic elements (K,, K2, yl, and ya) are rational C o ? ~ ~ l p e ~ etwin: n t f ~that r ~ results from anti-twinning (Kb -4,) C o m ~ ~ ~~ , ~~formed s ~~ o ~ under n : compression applied along the c-axis of hexagonal or tetragonal crystals Conjugate twzn: (K2, +) twin as contrasted to the ., and Naltaniura, M. primary (K,, q 1 )twin A., Syutkina, V. I., ~ e ~ t w ~~~ n ~arn shear g : ~ d e f~o ~ ~ a~t i oprocess n~ z ~ ~ Shashkov, 0. D., aiid Yaltovlcva, E. S. (1971). P/?ys. caused by an exteixally applied stress Stat. Sol. ( U ) 6, 323. Double ~ w ~ ~ na~second-order ng: twinning within the Albert, D. E,. and Gray, G. T. (1994). Phil. Mag. primary twins Albcrt, D. E., and Gray, G. T. (1995). Phil. &fog. ~ ~ r twifznhg: ~ i the~same ~ as ~ pseL~do-twinnin~ n P. A., and Wagner, R. (1993). A G r o ~ ttwins: ~ those formed by a solidification 1. process from a liquid or vapor phase ph, U., aiid Wagner, M ~ c r o t ~ ~when i ~ , s twin : thickness is larger than that A, 72, 342. defined as a microtwin Appel, Fa,Qehring, M., and Ennis, P. J. (1999). In G f f ~ ~ u ~ e c ~ ~ n i c twinnifzg: aZ the same as deformation T i ~ a A~l ui~ ~ i n~i ~~(eds ~ e sV.-W. , Kim, D. M. Dirniduk, twinning and M. €3. Loretto). TMS, Warrendale, PA, p. 603. Appel, F., aiid Wagner, R. (1994). In Twirzning in Advanced Microtwins: those twins with thickness on. the order ~ U t e ~ i (eds a l ~ M. ~ El. Yoo, and M. W u t t ~ ~ TMS, ). of seven thousand interplanar lattice spacings Warrendale, PA, p. 3 17 ~ ~ ~ ~ o t twins ~ i ? zof~ thickness s: on the order of the Appcl, F., and Wagner, R. (1998). M m r . Sci.Eng. R, 22, 187 lattice spacing Armstrong, R. W. (1968). Scimce 16 ~ r ~ p e r - t ~ ~ i n nthe i n gsame : as tr~~e-twi~ning V S., and Sargent, C. M. (1971).Scr. ~ ~ e t U 1 1 . Ar~i~iachalaiii, P o Z ~ s y n t ~ e t i ~ ~ Zt l~y i ~ n( Pe S~T ) . ~ t r ~ ~ t uar ~ : 949. repeated twinning structure that emerges as a result ., and Needleman, A. (1992). of the strain energy relaxation attendant to the atomic ordering reaction 1968). Proc. Roy. Soc. A 3 P~l-vtwinnpe~~ structure: the same as PST structure 123. Primary twin: the particular ( K l ,yl) twin system that Bevis, M., and Crocker, A. G. (1969). Proc. Roy. Soc. A 313, 509. receives the largest driving force for twinning Bhowal, P a., Prewo, K. M., and McEvily, A. J. (1978). P ~ e u ~ o - t ~a ~special ~ i ~ scase : (twinned orientation) of M e t d . Tram. A 9, 747. stress-induced niartensitic transforination Bilby, B. A., and Crocker, A. G. (1965). Proc. Xoy. Soc. A ~ $ i ~ ~ i - t w i n n the i n g same : as pseudo-twinning 288, 240. ~ e c ~ r o c atwin: Z the same as complementary twin Bolling, G . F., and Richman, R. H. (1965). Acta MetufE. Xecrystdlizatiovz twins: essentially the same as 709, 723, and 745. annealing twins Booth, A. 5.. and Roberts, S. 6.(1997). Acta M a ~ e ~ . ~ e c o fwins: ~ ~ those ~ r formed ~ by double twinning Bouchard, M., Claisse, F., Tardif, H. P., and Dub&,A. (1970). Szipper-twinning: true-twinning occ-tlrriiig in fullyCm. Metall. Quarter. 9, 395. ordered alloys Braisaz, T., Ruterana, P., and Nouet, G. (1997). Plziil. Mag. A 76, 63. T e ~ ~ i otwin: n formed under uniaxially applied tension along the c-axis of hexagonal or tetragonal crystals T r a ~ ~ ~ o r ~twins: a t i ~those n formed during a solidstate ~ransforniation ~ r ~ e - t w i n n i ndefined ~: as that when the twinned Cahn, R. W., and Coll, J. A. (1961). Ac portion of a crystalline body has the identical crystal Calbrick, C. J., and Marcus, R. B. (1967 structure with that of the parent matrix, including the Chakraborty, S. B., and Stark, E. A. (19 , state of chemical order 63. Tvpe I twin: wheii X I and y2 both have rational, lowChan, M. S. (1992). J . oif'Metals, May, 30. index elements Ghan, M. S. (1993). Metall. Trans. A
32
Mechanical Properties
Chan, K. S., and Kim, Y.-W. (1994). Metall. Tram. A 1217. Clzaii, K. S., and Kim, Y.-W. (1995). Acta Metall. Mafer. 439. owitz, J. D., and Bieler, T. R. (1996). Scr.
Gao, Y. Q.. and Whang, S. €3.(199 ). Scr. iMetall. Muter. 31, 1583. Girshik, A., and Vitek, V. (1995). In ~ i ~ i z ~ T e ~ ~ p e ~ a t ~ Ordered ~ n t e r ~ e t a l l A i cl / o y ~VI, (eds J. A. Horton, I. Baker, S. Hanada, R. D. Noebe, and D. S. Schwartz). MRS, Pittsburgh, PA, Christian, J. W., Olson, G. B., and Cohen, M. (1995). J . de Godfrey, A., Hu, D., and . (1998). ~ h ~Mag. l . A 77, 287. Christittn, J. W., and Croclter, A. G. (1980). In D~slocations~n Goo, E. (1994). IT] Twinning in Advanced Materials, (eds N. Nabarro). North-Holl~~n~, M. El. Uoo, and M. Wuttig). TMS, Warrendale, PA, p. 305. glilin, D. E. (1989). Acta Metall. 36, Coo, E. (1998). SCP Mater. 1617. Goo, E., Duerig, T., Melton, inclair, R. (1985). Act61 Christian, J. W., and Mahajan, S. (1995). Prog. Mater. SCL Goo, E., and Park, K. T. (1989) Scr. Metall. Chu, F., and Pope, D. P (1994). In Twinnzng in Advanced Gray, G. T. (1 994). In Twrnrzrng 7n Advanced Mmerids, (eds '"400, and M. Wuttig). TMS, M. H. Yoo, and M. Wuttig), TMS, Warrendale, PA, p. 337 Green, M. L., and Cohen, M. (1979). d c f a A ~ e ~ ~27, ~ l1523. l. Gottrell, A. H., and Rilby, Greenbcrg, B. A. (1970). f'hys, Stat. Sol. 42, 459. Goujou, A., Beneteau, A, Creenberg, B. A., Antonova, 0. V., and Volkov, A. Yu. (1999). ~ ~ i r ~ r ~ 7, e t121 ~ 9. ~ l ~ ~ c s Couret, A., Farenc, S., Caillard, D., and Coujou, A. (1994). Grossiord, C., Lesoult, G., and Turpin, M. (1972). Electron I n Twimting in Advanced Malerials, (eds M. H. Y o o , and Microscopy and Structure of Mufcnals, (eds 6. Thomas, M. Wuttig). TMS, Warrendale, PA, p. 361. R. M. Fulratb, and R. M, Fisher). ~ n i v ~ r s i t of y 962). Pkil. Mag. 7, 1901. California Press, Berkeky, CA, p. 678. K., and Asaro, R. J. (1996). Yhil. Mag. A Groves, 6 . W., and Kclly, A. (1963), Phil. Mqq. Guedo, J. V.. Paliard, M., and Rieu, J. (1978). Scr Deve, H, E., Evans, A. 6.. and Shih, D. S . (1992). Acta 927. Guiniier, A., and Strudel, J. L. (1970). In Second Int. C o n . on eve. H. E., and Evans, A. G. (1991). Aclu Me~all.Mater. 3 ~ ~ r ~ ?o ~~g ~t he tand ~ Alloys ~ l s (ASM, Metals Park, OH), 1171. 1145, Hagihara, K., Nakano, T., and Uniakoshi, Y (1999). In Eshelby, 3 . D. (1961). Prog. SolidMech. 2, 89. l ~ ~ ~ l ~ - T i ~ ~Ordered ~ e r u Intermetallic ~ure AItoys V I Z (eds Farenc, S., Coujou, A., and Couret, A. (1993). Phil. Mag. A E. P. George, M. J. Mills, and M. Yaiiiaguchi). MRS, 67, 127. Warrendale, PA, 552, KM10.7.1. Feng, C. R., Michel, D. J., Crowe, 6 . R. (1990). Phi/. Mag. Hagihara, K., Nakano, T., and Umakosht, U. (2000). Acta Lett. 611, 95. Mater., 48, 1469. Fleischer, R. L. (1959). Acta b4&~//.,7, 134. Hall, E. L., and Huang, S.-C. (1989). J . Mater. Res. Fleischer, R. L. (1987). Hanamura, T., Uernori, R., and Tanmo, M. (1988). J. Mater. Frank, F. C. (1951). Ph Res. 3, 656. Frank, F. C., and Stroh. Hazzledine, P. M. (1994). In T w i ~ ~ z nmgAdvffneed~ ~ t e r i a l s , 81 1. (eds. M. H. Yoo, and M, Wuttig). TMS, Warrendale, PA, Frcnkel, J., and Kontorova, T. (1939). J, Phys. A4oscow 1, p. 403. 137 P. M., ICad, Ea. I<., and Mendiratta, M. G. (1993). Friedel, 6. (I 926). Ledcons de ~ r i s t a l ~ ~ g (Bergerr ~ ~ ~ ~ ~ i Hazzlcdine, e In Thin Ei'lms: Stresses and ~ e ~ ~ a n iProperties cal IV, (eds Levrault, Parts). Townseiid, T. P. Weihs, Fu, 6 . L., and Yoo, M. H. (1991). Phil. &!age Lett. sen). MRS, Pittsburgh, PA, Fu, C. L,, and Uoo, M. H. (1993). Inter~etffllics I, hi, M., and Weissnian, S. ( Fu, C. L., and Yoo, M. H. (1994 ). In Alloy M ~ ~ e l i nmgd 25. Design, ( C ~ S G. M. Stoclts, and P.E.A. Turchi). TMS, Hirth, J. P., and Lothe, J. (1982). Theory of location,^, Warrendale, PA, p. 225. Sccond Edition (Wiley, New York), 293. FLi-jita, H., and Kawasalti, Y. (1966). J q . J . Appl. Phys. 5, Nirth, J. P,. and Pond, R. G. (1996). Actu Mater. 788. ., Sawai, T., and Wakata, N. (1996). Fujita, H., and Mori, T. (1975). Ser. Metall., Fujiwara, T., Nakamura, A., Hosomi, M., Nishitani, N.R., Hoover, W. R., and Hertzberg, R. W. (1971). Metall. Trans. Shim, Y., and Yainaguchi, M . (1990). Phil. Mag. A, 61, 2, 1283. 591.
Kisliida, IC., Yokoshima, S., Johnson, Hsiung, L. M., Schwartz, A. J., Nich, T. G. (1997). Scr. Yarnaguchi, M. (1996). In Pmc. of NATO-AS1 on 1017, Stal~ilit}~ of Materials, (eds A. Conis, P.E.A. Turchi, and ., and Nieh, T. C. (1997). Mafer. Sci. Erg. A J. Ku~rnovsky).Plenum, New Yos 38. and Hall, E, L. (1989). In ~igh-Temperatz~re ICishida, K., Inui, H., aiid Yamaguchi, Ordered Iktcrinetulk Alloys III, (eds 62. T. Liu, A. I. Taub, . ~ Kocks, U. F., and Westlake, D. 6. (1967). ~ r a ~ ~~ s M N. S. Stolofl, and C. C . Koch). RS, Pittsburgh, PA, 1 A I M E 2 ~ 9 1107. , 373. Komura, Y., and Kitano, Y (197 Huang, S.-C., and Hall, E. L. (1990). In Alloy Phase Stability ICroiiberg, M. L. (1957). Acta M e and Design, (eds G. M. Stocks, Kuinar, M., and Vasudevan, V. Giarnei). MRS, ~ittsbusgh,PA, Advanced Material~s,(eds M. ., arid Lasalrnoiiie, A. (1986). Phik. Mag. TMS, Warsendale, PA, p. 423. n, D. J., and Miller, M. K. (1998). ~ ~ tSci.e Eng. ~ .A Hug, G., and VeyssiGre, P (1989). In E~ectroFZ~ i c r o ~ s c o pmy 0, 72. plastic it^ a d Fracture ~ e ~ ~ e a rofc hM ~ i e r ~ a l ~(eds s , U. , F. (1952). Naturw2,s~sensc.hafte~~ Messerschrnidt, F. Appel, J. Heidenrich, and V. Schini~~t), Laves, F. (1967). In I n t e r ~ ~ t a l lC~ c Westbrook). Wiley, New York, p. 129 i, Y ., ~ a i i i a ~ c hM., i , Fujita, S., Lebensohn, R. A., Uhlcnhut, H.. I-rartig, C . . aiid Mecking, H. aiid Nishisako, T (1995). Mater. Sci. Eng. A 19 Inkson, B. J., and ~-Iumphreys,C. J. (1996). Phil. Mag. A 73, Lebcnsohn, R. A., and TomC, C. N. (1993). Phi/. Mag. A 67, 1647. ., and Y a m a ~ c h i .M. 187. Lee, J. K., aiid M. H. Yoo. (1990). M ~ t u l lT. r a m A 22, 2521. Y ~ n ~ ~ gM.~ (1992). c ~ i , Lee, J. K., and M. H. Yoo. (1994). In Twinning in A d v ~ ~ n c ~ d Mate~ials,(eds M. H. Yoo, and M. Wuttig). TMS, bayashi. M., Shirai, Y., Warrendale, PA, p. 51. a i d Yarnaguchi, M. (1995). Phil, Mag. A 72, 1609. Lee, J. W., Hmada, S., and Yoo, M. H. (1995). Scr. A ~ e ~ a l l . D.-H., and Y a ~ a g u c h M. ~, er. 33, 509. Lipsilt, El. A., Sheclitrnan, D., aiid Schafrik, R. E. (1980). (1960). Acta Cryst. 13, 232. Mt?tnll. Trims. A 11, 1369. er, T. R. (1995). In Gamma Liu, C . T., Schneibel, J. H., MaLiasz, P. J., Wright, J. L., and ~ i ~ uA n~ a~i ~~i n& ~(eds ~~e Y s . -W. Kim, R. Wagn~r,and M. Easton, D. S. (1996). l n t e r ~ ~ t a l l i ~ s Yarnaguchi). TMS, Warrendale, PA, p. 975. Liu, Y., Taksugi, T., Izunii, O., and Takahashi, T. (1988). Jin, Z., and Bieler, T. R. (1992). Scr Mecull. 27, 1301. Acta h.Petal1. 36, 2959 Jin, Z., and Bielcr, T. R. (1995a). Phil. Mag. A ~ivingston,J. D., and Hall, E. L. (1990). J . Mater. Re,s. Jin. Z., and Bieler, T. R. (1995b). Phil. Mag. A 7 Loiseau, A., and Lasalmonie, A. (1984). Mater. Sri. E E ~ Johiison, W. C.. and Cahn, J. W (1984). ilcta 163. 1925. Loubradou, M., Bonnet, R., and PCnisson, J. Kad, B. K,, and Asaro, R. J. (1997). Phil. Mag. A 7 Kad, B. K., and Olivcr, B. F (1989). In ~ i ~ ~ h - T ~ i ~ p e ~ a t z ~ r e s, M. A. (1995). ~ e f a l lMater. , Trans. A O r ~ e r e J 1 n t e r ~ e tAlloys c ~ ~ ~ 111, ~ c (cds G. T. Liu, A. 1. T N. S. Stoloff, and C. C. Koch). MRS, Pittsburgh, PA, s, T. A., and Pope, D. P. 237 Kaltchi, K. (1999), Scr. Mater. Mahajan, S., aiid Chin, GaY . (1973). Acta ~ ~ ~ t 21, u l l1353. . ICawabata, T., Kaiiai, T., and Izunil, 0. (1985). Acta Metall. Mahapatra, R., Gisschick, A., . 3 (1995). Scr. ~ e t a ~ ler niya, T., ICanai, T., and Izumt, 0. (1990). Marcinkowski, M. J. (1963). €381. Strength of Crystals (Intersctcnce, New York), 333. indow, M., and Jones, 1. P (1997). 2kfrter. nkowski, M. J., and Fisher, R. M. (1963). J. Appl. Phys. 44. , 2135. k, J. M., and Giamei, A. F. (1970). In Markctz, W. T., Fislicr, F, D., and Clcments, Meta[l., 90, 588. ocks, U. F, (1996). Acia Khachaturyan. A., Scmcnovskaya, S., and Chen, L.-Q. (1994). In T ~ z n n i ~i n2 ~ Advanced Materials, (eds M. H. Medlin, D. L. (1999). In Advancm 111 T~)iFZnin~, (eds S. Yoo, and M. Wuttig). TMS, Warrendale, PA, E>. 193. Ankeni, and C. S. Pandc). TMS, ~ a r s c n ~ a lPA, e , p. 29
434
Mechanical Proper tics
Medlin, D. L., Illartcr, C. R., Angelo, J. E., and Mil~s,M. J. (1997). Plzil. Mag. A "7, 733. Melirer, H. (1 996). M a ~ e r Truns. JIM Mikkola, D. E., and Cohen. J. B. (1966) Minonishi, Y., and Yoo, M. EX. (1990) 203. .. and Minakawa, K. (1991). Mater.
Paton, N. E,, and Backofen, VV. A. (1970). Metall. T r a ~ s . 1,2839. ~ n ~ M a t e ~ i u l (eds ~, Paxton, A. T, (1994). In T ~ i n in~Advatzced M. H. Uoo, and M. Wuttig). TMS, Warrendale, PA, p. 27. ~. Paxton, A. T. (1995). Arta M e t ~ i lMuter Paxton, A. T., and Entwisle, A. R. (1985) 573. ~ ~ a n - C o u r ~I.o (1993). n, Doctoral Thesis, Univ. of Paris VI, 140. Mitao, S., Isawa, T., and Tsuyaina, S. (1992). Scr. Metall. Pirouz, P. (1987). Sm. MetalE. Mater. 21, 1463. Mater. 26, 1405. Pond, R. C (1989). In D i s l u c u ~ ~ (m~ ?Sol ~s Mitchell. T. E., and Hirtk, J . P. (1991). Acta MetaEZ. Mater. Nabarro). North-Holland, Amsterdam, 39, 1711. Pond, R. C. (1995). Inferface Sri. 2, 1. Moberly, W. J., Proft, J. L., Duerig, T. W., and Sinclair, R. Price, P. B. (1960). Plzil. Mag. (1990). Actn Metall. Mater., 33, 2601. Rao, M., and SofTa, W. A. (3997). Scr. Mater. 36, 735. Moran, J. B. (1965). Tram. TMS-AIME 233, 1473. Rao. S., Woodwarcl, C., aiid I-Iazzlcdine, P. M. (1994). In ,Yukawa, N., and Adaclii, H. (1 990). Y efcct- In f e $ m In teractio , (eds E. P, Kvam, A. H. King, M. J. Mills, T. D. ands, and V. Vitek). MRS, S., Leboeuf, M. (1994). Phil, Mag. A Pittsburgh, PA, 319, 285. 69, 527, Rao. S . , Woodward, C., Simmons, J., and Diniiduk, D, M, Morris, M. A. (1993). Phil. Mug. A 68, 259. (1 995). In ~ ~ ~ ~ i - ~ e n ~Ordered p e r a Irn ~~ ~ ~e ~ ~ ~Alloys ieta~lic Morris, M. A., and Leboeuf, M. (1997). Interrnetallics 5, 339. VI, (cds J. A. Horton, I. Baker, S. Hanada, R. D. Noebe, Morris, M. A., and Leboeuf, M. (1998). J . Mater lies. 13, and D. S. Schwartli}. MRS, Pittsb~rgh.PA, 3 Reid, C. N. (1981). Metall. Trclns. A 11 orris, D. G. (1991). Phil. Mag. A 63, Romanov, A. E., and Vladimirov, V. I. (1 9x9, In isl location^ in Solids, (ect. F. R. N. Nabarro), North-Holland, Mull~ier,P., Solenthaler. C., and Speidcl, M. Q. (1994). In Amsterdam, 9, 191. Twinnikg in Advmmd Materials, (eds M. H. Yoo, and M. Rosenbaum, H. S. (1964). In ~ ~ f ~ ~ r ~~~ uz ~t ?~n(eds ai nngR. ~ Wuttig). TMS, Warrendale, PA, p. 483. E. Reed-Hill, J. P. Hirth, and H. C. Rogers). Gosdon and Mul~ner,I?., and Pirouz, P (1997). Mater Sci. Eng. A 233, Breach. New York, p. 43 139. Sastry, S.M.L., and Lipsitt, PI. A. (1977). Merall. Tram. A 8, Nakano, T., Yasuda, W. Y., Higashitanaka, N., Uniakoshi, 229. Y. (1997). Aeta at er. Sauthoff, G. (1996). ~ ~ a ~ e r iScience als and ~echnalogy,(eds Naknno, T., Kawanaka, A., Hosomi, M., Yasuda, H. Y., and R. W. Cahn, P. Haasen, and E. J. Kramer). VCH Uniakoshi, Y. (1995). Mater. Sri. Eizg. A 194, 43. Verlagsgesellschaft, Weiiiheini, p, 643. Nanta, N., and Takamura, J. (I 992). In Dislocations in Solids. Seliogl, S . M., and Fisher, F. 53. (1997). Phil. Mug. A, 3 (ed. F R. N. Mabarro). North-Holland, Amsterdam, 3, Seo, D. Y., Bieler, T. R., and Larsen, D. E. (1997). In 164. Sfructural Z n ~ e r ~ ~ ~ a (eds l l ~ cM. ~ s V. , Nathal, R. Darolia, Nerc-Partaix. A., and Mennand, A. (19%). In ~ i t a r ~ i ~ ~ , C . T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. ~ i g ~World t h Congress (Inst. of Metals, London), 121. Y a m a g ~ c ~ iTMS, ). Warrenda~e,PA, 137. Nesbit, L. A., and Laughlin, D. E. (19x0). Acla Metall. 28, Shang, P., Cheng, T. T., and Aindow, M. (1999). Phi/. Mag. A 989. 49, 2553. Nixon, T., and Pond, R. C. (1998). In Inter~aces and Shechtman, D., Blackburn, M. J., and Lipsitt, €3. A. (1974). ~ l a s t i c j t (eds ~ , N. Clement, and J. Douin). Scitec Publ., Ziirich; Solid Sfatr' Plienomma 59-60, 201. Shechtman, D., and Jacobson, L. A. (1975). Mefall. Trnrzs. A Ohtsuka, M. (1991) Doctoral Thesis, Tohoku University , 1325. (Sendai, Jay an). Siegl. R., Vitek, V., Inui, da, IS.,and Yamaguchi, M. Qikawa, €3. (1990). In High T e n ~ ~ e r a t ~ArEe u ~ i ~ i t ~and e~s (1997). Phil. Mag. A Intermetullics, (eds S. H. Whang, C. T. Liu, D. P Pope, Singh, S. R., and Wowe, 91). Scr. Metall. Mater and J. 0. Stiegler). TNS, Warrendale, PA, p. 699. 485. Orowan, E. (1954). D i s l a c a ~ iin~ ~Metals ~~ (AIME, New Singh, S. R., and Howe, J. M. (1992). PRil, Mag. Lett., York), 116. 233. Skrotzki. B. (2000). Acta Mafer., Otsuka, K., and Reii, X. (1999). I ~ t e r ~ e t u E I3, ~ c51 s 1-528. Skrotzki, B., anal, M., Eggler, 6. (1998). Scr. Mafer. 3 Park, K. T., and Goo, E. (1991). Acta Metall. Mater. 1 1023. 3027. J. L., and Bruemmer, S. M. (1997). Acltr Pashley, D. W., Robortson, J. L., and Stowell, M. J. (1969). Phil. Mug. 19, 83.
Twinning and ~ ~ c ~ a nBehavior i c ~ l Srirarn, S., Viswanatlian, G. B., and Vasudevan, V. K. (1994). In T ~ ~ n n I ni ng Advanced Materials (eds M. H. Yoo, and M. Wuttig). TMS, Warrendale, PA, p. 383. Sun, Y. Q., ~ a ~ z l e d i nP. e , M., and Christian, J. W (1993). Syutkma, V I., and Yakovleva, E. S. (1967). PhevLs.Stat. Sol., , 465. Takeuchi, S., Kuramoto amoto, T., and Taoka, T. (1973). Jap. .J. Appl. 1486. Tawancy, H. M. (1981). Sci., 16, 2883. Tsujimoto, T., and Hashimoto, K. (1989). In NighTem~erutureOrdered I ~ t e r ~ e t a l lAlloys ~ c III (eds C. T. Liu, A. I. Taub, N. S. Stolof’f, and C. C. KO&). MRS,
.
43 5
Yamag~lclii, M. (1986). In ~ ~ ~ h ~ T e ~ p e r a~ tr ud re re e f ~ I ~ ~ e r m e t ~Alloys ~ l l ~ cIf (eds N. S. StoloK 6 , C. Koch, G. T. Liu, and 0. Izumi). MRS, Pittsburgh, FA, Yarnaguchi, M.*U ~ a ~ o s hY., i , and Yarnane, T. ( Mag. A 55, 301. Yamaguchi, M., Shirai, Y . , and Umaltoshi, V (1988). 111 ~isperszonStrengt~ze~ed A l u ~ i n Alloys u ~ (eds Y.-W Kim, and H. M. Burte). TMS, Warrendale, PA, p. 721. Yamaguchi, M., Inm, H., and Shirai, U. (1994). In Twrnning in Advanced Materials (eds M. H. Yoo, and M. Wuttig). TMS, Warrendale, PA, p. 177. Yamaguchs, M., and Umakoshs, Y. (1990). Prog. Mater. Sci.
Yang, Y. S., and WLI,S. K. (1992). Phil. Mug. A Yasuda, H. Y., Nakasio, T., and Uii~d~oshi, Y. (1996). Phil. Mag. A 73, 1035. Yokoshima, S,, and Yarnaguchi, M. (1996). Actu Mater. 873. Yoo, M. H.(1979a). Scr. et all. Yoo, M. H. (1979b). In Proc. 5th ~ e t uand l ~ Allo-vs, ~ Vol. 2, (eds P. Haasen, V. Gerold, and G, Kostorz). Pergarnon, Oxfor Yoo, M. H. (1981). Metall, Tram. Yoo, M. H, (1989). .I. Mater. Xes. Yoo, M. H. (1997). Phil. Mug. Le Yoo, M, H. (1998a). I n t e r ~ e t ~ l l ~ c s I<., Wheelcr, R., and Fraser, H. L. (1989a). In Yoo, M. H. (199%). Scr. M u m . 3 ~ i ~ h - T e ~ ~ e rOrdered a t ~ r eIntern~etallicAlloys I11 (eds C. Yoo, M. H.,Fu, C. L., and Lee, J, K. (1989). In HighT. Liu, A. I. Taub, N. S. StoloE, and C. C. Koch). MRS, Te~peratureOrdered I n t e r ~ ~ t ~ lAlloys l i c III (eds G. 71. Pittsburgh, PA, 133, 70.5. Liu, A. I. Taub, N. S. StoloE, and C. C. Koch). MRS, Vasudevan, V. K., Stucke, M. A., Court, S. A., and Fraser, H. Pittsburgh, PA, 133, 189. L. (1989b). Phil. Mag. Lett. 59, 299. Yoo, M. H., Fu, C. L., and Lee, J. K. (1991). J . Phys. I11 Vasudevan, V K., Court, S. A., Kurath, P., and Fraser, €3.L. 1065. (1989~).,!er Metall. 59, 907. Yoo, M. H., Sass, S. L., Fu, C. L., Mills, M. J., Dimiduk, D. M., Venables, J. A. (1961). Phil. Mag. 6, 379. and George, E. P. (1993). Acta M e ~ a lMater. ~. Vlasava, E. N., Gorbunov, V. L., Kosnarova, I. A., and Yoo, M. H., Fu, C. L., and Horton, J. A. (1994a). M u m . Sci. Sapozhkova, T. P. (1972). F k . Metall. Metallov. 3 Eng. A 176, 431. Wang, J. N., Scliwartz, A. J . , Nieh, T. G., Liu, C. T., Siltka, Yoo, M. H., Fu, C. L., and Lee, J. I(.(19941~).In Twirznrreg in V. K., and Clernens, D. (1995). In Gu~znzaTitanium A~vancedMaterials (eds M. H. Yoo, and M. Wuttig). Alu~inides (eds Y.-W Kim, R. Wagner, and M. TMS, Warrendale, PA, p. 97. Yamaguchi). TMS, Warrendale, PA, p. 975. Yoo, M. H., Zou, J., and Fu, C. L. (1995). Mater. Sci. Eng. R Wardle, S., Plian, I., and Hug, G. (1993). Phil. Mag. A 67, 1921193, 14. 497. Uoo, M. W., and Fu, C. L. (1998a). ~ e t a l Mater. ~. Tram. Weertrnan, J., and Weertrnan, J. R. (1964). ~ l e ~ e i z t f f r ~ 9 4 49. Distocnfion Theory (Macniillan, New York), 99. Yoo, M. H., and Fu, C. L. (199%). In Mirroscoptc Simiulati Wiezorek, J .M .I<., DeLuca, ills, M. J., Fraser, €3. L. of Inte~acialPhe?io~enain Soli~sand ~ ~ (eds ~S. u (1997). Phil. Mag. Left. Phil~pot,P. D. Bristowe Wilbrandt, P. (1994). In T ~ ~ ~ ~innAclvanced ing M a t ~ r ~ a(eds ls MRS, Warrendale, PA, M. H. Yoo, and M. Wuttig). TMS, Warrendale, PA, Yoo, M. H . , and Risliinuma, p. 15.5. Yoo, M. M., and M ~ s h i ~ u r nA. a ~ (1999). In Advunces in Wil~iarns,J. C., and ~lackburnyM. J. (1970). In Ordered T w z n ~ ~(eds ~ g S. Ankem, and C. S. Pande). TMS. Alloys: Structural Applicsxtions and Ph-vsical ~ ~ ~ t a l l ~ r g y Warrendale, FA., p.225. (eds B. H. Kear, G. T. Sirns, N. S. Stoloff, and J. II. Yoo, M. H., and King, A. H. (1988). J . Mater. Ses. 3, 848. Westbrook). Claitor’s P~~blishing,Baton Rouge, LA), Yoo, M. E., and Lee, J. K. (1991). Phil. Mag. A 63, 1001. p. 425. Yoo, M. H., and Loh, B.T.M. (1970). In F ~ i z ~ a ~ z e nAspects tal Woodward, C., MacLaren, J. M., and Rao. S. (1992). J. o ~ ~ ~ . ~ l ~ cTheorji a t i o n(eds J. A. Sim Mater. Res. 7, 1735. R. Bullough). NBS Spec. Publ. 317,
Umakoshs, Y., Yaniaguchi, M., Yarnane, T., and Hirano, T. (1988). Phil. Mag. A 58, 651. Umakoshi, Y., and Naltashnwa, T. (1994). Scr. Metall. Mater. 30, 1431. Vanderschaeve, G. (1998). In Solid State P h e ~ o ~ e n5a9 ~ 0 , 145. Vanderschaeve, C., and Escaig, B. (1978). J. Phys. Lett., 39, L74. Vanderschaeve, G., and Escaig, B. (1983). Phil. Mcig. A. 265. , G., and Sarrazin, T. (1977). Ph-vs. Stat. Sol.
~
Yoo, M. H., and Famnguchi, M. (2000). In ~ i c ~ o s ~ r u c ~ u rHirth, e J. P., and Lothe, J. (1982). Ddorination Twmning, and P r o p e r t ~ eo~f ~~a t e r i a l sVol. , 2 (ed. J. C. M. Li). World Chapter 23 in Theory of ~islocations,Second Edition, SCLPubl. Co., S ~ n g a p ~p. r e79. ~ Wiley, New York ghal, S. (1997). Doctoral Thesis, L'Universitt Paul Sabatier Khachaturyaii, A. G. (1983). The Theory (?f' Structural de Toulouse, 136. T r a n s ~ o r ~ a t ~ in o nSolids, s Wiley & Sons, New York. Zhang, B., and SoRa, W. A. (1992). Phys. Stut. Sol. (a) 131, ~lassen"Neklyudova,N. V. (1964). ~ e ~ h a n i c aTw~~ning l of 707. Crystals, Plenum Press, New York. Zliang, L. C., Chen, G. L., Wang, J. G., and Y e , H. Q.(1999). Kosevich, A. M., and Boiko, B. S. (1971). Dislocation theory ~ n ~ e ~ ~ e 7, t a1241. ll~~s of the elastic twiiining of crystals, Soviet Phys. ~ s ~ 1 e ~ ~ ~ hang, Y C., and Chaturvedi, M. C, (1993). Phil. Mug. A 68, 286. 915. Mahajaxl, S. (ed.) (1981). The role of twin metals and alloys, Metall. Trans. A , papers). Mahajan, S., and Will~~nis, D. F (1973). Deformation .) (1999). Advances in twinntng in metals and alloys, Int. ~ e ~ Rev., ~ l l . s, Adv. EB Phys., 3, 363, Reed-Hill, R. E., Hirth, J. P., and Rogers, H. C. (eds Christian, J. W. (1965). Thcory of' T r a n ~ ~ f o r ~ a t iino nMetals .~ Dcforination Twinning, Cordon and Breach, New York. and Alloys, Pergarnon Press, Oxford. Tanner, L. E., and Wuttig, M. (eds.) (1990). First-order Christian, J. W., Olson, C , B., and Cohen, M, (1995). displacive phase traiisfor~ation: review and recomC ~ a s s i ~ c a t iofo ~displacive ~ tr~nsformation: what is a mendations, Mater. Sci. Eng. A 127, 137. (15 other J. de Pizys. IV, Colloque C8, papers). C8-3. (3 other papers). ~andersch~eve,G. (1998). Mechani Christian, J. W., and ~ a h a j a n ,S. (1995). Deformation semiconductors, Solid State P h e n o ~ e ~ Twinning, Prog. Mater. Sci.39, 1 . Yoo, M. H., and Wuttig, M. (eds.) (1994). Twinning IIZ Hall, E. O., (1954). Twinnirig and ~ z ~ u s i o ~ lTrans~or~ution es,~ advance^ material^^, TMS, Warrendale, PA. in M ~ t a l . ~Butterwort~s, , London.
t
s Czech ~ e p ~ Czech ~ l i ~ ~ stiizate of Physics, c a ~ e ~~ ~y S c i ~ nofc ethe
ion
The stacking-fault-type interfaces we explore in this chapter play a very important role in dislocation configurations and thus in plastic behavior of materials. The existence of inetastable APBs and stacking faults is of major significance for the structure of dislocations in a given intermetallic compound. In general, a vital characteristic of dislocations is their dissociation into partials connected by ribbons of metastable planar faults. Such dissociations, when energetically favorable, frequently determine the slip planes and thus control the operative slip systems. A simple example is the well-defined slip systems in fcc materials with dissociation into Shockley partials, in contrast to an uncertain plethora of slip systems in bcc metals in which no metastab e stacking faults exist (Christian, 1983; Duesbery, 1989). Nevertheless, understanding of dislocation dissociation involving metastable planar faults i s only the first step in comprehension of the atomic aspects of dislocation behavior. The next is the core structure of either uiidissociated lattice dislocations or partials. Here y-surfaces, defined and utilized in this review, may be an important source of physical information. In the case where the cores are planar, i.e. spread in the slip plane, the Peierls model, combined with the knowledge of the y-surface may be used to
determine the details of the core structure (~hristian and Vitek, 1970; Vitek, 1992; Schoeck, 1994; Duesbery and Vitek, 1998; Mryasov et aE., 1998). most significant core phenomena occur when the dislocation cores are non-planar, and spread into more than one crystallographic plane. This is well known for screw dislocations in bcc metals (~hristian, 1983; Duesbery, 1989; Duesbery and Vitek, 1998), but it may be at least equally important in intermetallic compounds. Indeed, the non-planar core of superpartials in L12 Ni3Al is, presumably, responsible for the anomalous increase of the yield stress with temperature ('Vitek et al., 1996), and non-planar cores were also found in TiAl (Girshick and Vitek, 1995; Simmons et al., 19971, NiAl (Schroll e significant in MoSi ng these studies is b An all-inclusive definition of an internal interface in crystalline materials is that it is the region of contact of two crystallites which are in some respect different from each other. These regions are domains of distinct atomic structure and composition that are, in general, substantially different from those of the bulk of either of the adjoining crystallites. As a consequence, the physical and chemical properties of interfacial re differ significantly from those of the bulk, and this i s
~ n ~ ~ r rC n~ ~~ t~ ~p o~Vol. ~~ ni 3, d~sPrincipkes ~ and Practice. Edited by J. H. Westbrook and 0 2 0 0 2 John Wiley & Sons, Ltd.
438
~ e & h u ~ iProperties &a~
the reason why interfaces phy a prominent role in a broad variety of physical, chemical and mechanical properties of materials. The character of an interface is determined by tlie nature of the difference between adjoining crystallites. The stacking-fault-like interfaces are formed by joining two identically oriented crystallites that are relatively displaced with respect to each other by a vector which is not a lattice vector of their crystal structure. In ordered alloys these faults involve various types of topological and/or chemical disorder in the interfacial region. Special cases are antiphase domain boundaries (APB) for which the displacement vectors are identified with one of the vectors connecting the sublattices of different species. Aiitiphase boundaries (APBs) were discussed in detail by Sun in Volume 1, Part IV, Chapter 21 (1995) and all tlie basic concepts can be fouiid in that review. In the present chapter we include some new data on energies of APBs that appeared since the publication of Volume 1, and thus update Tables 1-6 un. We then discuss other stackingfault-like defects and concentrate 01%aspects that have not been considered previously. An ii~portantnotion, employed throughout this chapter, is the concept of y-surfaces which are theoretical constructs that have been used extensively in studies of stackin~~fault-like defects and dislocations. An exact definition of y-surfaces is presented at the beginning of Section 2. For each specific crystal structure it is its s y m ~ e t r ywhich determines the most general features of these surfaces, while distinctions found for different materials with the same structure are controlled by the specific character of atomic interactions. This concept is then utilized when discussing possible stacking"~~ult-like defects in various types of crystal structures encountered in intermetallic compounds. The specific structures discussed in this chapter were selected to cover most c o ~ ~ o nstudied ly intermetallic compounds. After surveying the various crystal structures, we consider briefly effects of temperature on the structure and properties of stacking-fault-like defects. These effects mainly relate to segregation and disordering, which in turn may lead to thickening of ion. Since energies o f the faults s u ~ m a r i z ein ~ this review have been de~ermined from the measured widths of dislocation dissociations, we also discuss problems associated with such meas u r ~ ~ e n t These s. include the effects related to the formation of the contrast in the electron microscope, as well as effects of the core structure o f the dislocations. Finallv. we summarize. for all the structures
coilsidered in this chapter, the crystallographic planes on which the staclcing-fadt-like defects can be formed, the displacements leading to such faults, and discuss briefly some implications for dislocation behavior.
Interfaces discussed here are metastable planar translational faults confined to certain crystallographic planes. They encompass antipbase boundaries (AP and various types of stacking faults (SF) and play a prominen t role in dislocation dissociation, Such stackin~~fau~t-like defects can be very conveniently analyzed using the notion of y-surfaces. This concept was first introduced when searching for possible stacking faults in bcc metals (Vitek, 1968, 1992; Duesbery 1989); but it becomes even more useful in the case of more complex crystal structures siiice the variety o f planar defects increases with the structural complexity. To introduce the idea of y-surfaces, we first define a generalized stacking fault: Imagine that the crystal is cut along a given crystallographic plane displaced with respect to the lower parallel to the plane of the cut, as . The fault created in this way is ~~ j ~and~ it is not in called the g e ~ e ~ u s~ ~~ za ec .~~~~~ general metastable. The energy of such fault, y ( be evaluated, at least in principle, when an appropriate description of atomic interactions is available. In such calculations relaxation perpendicu~ar to the fault must be allowed but no relaxations parallel to the ed. Repeating this procedure for within the repeat cell of the given
Definition o f the generalized stacking Eau1t
crystal plane, an energy-displacement surface can be constructed, comnionly called the y-surf‘ace. The local minima on this surface determine the displacement vectors of all possible metastable stac~ing-fault~like defects, and the values of y at these ininiiiia are the energies of these faults. These faults play an important role in dislocation splitting, and vectors characterizing these faults are the urgers vectors of the correspoiiding partial dislocations. Symmetry arguments can be utilized to assess the general shape of these surfaces. Neuniann’s principle (Neumann, 1885), which states that ‘Any kind of s y ~ m e t r ywhich is possessed by the crystal structure of the material is also possessed by any physical property of this material’ has the followiiig iniplicatiorn for the y-surfaces: If a mirror plane of the perfect lattice perpendicular to the plane of the generalized stacking h the point corresponding to a first derivative of the y-surface this mirror plane vanishes owing to the mirror symmetry. This implies that the y-surfkce will possess extrema ( m i n i ~ a maxima , or inflexioiis) for those dis~lacementsfor which there are at least two non-parallel mirror planes of the perfect lattice perpeiidicular to the fault. Whether any of these extrerna correspond to minima, and thus to metastable faults, can often be ascertained by coiisidering the change in the nearest neighbor coiifigui-ation produced by the corresponding displace~ent. tigating the s y ~ m e t r yof the plane for which the ysurface is studied, the existence of inetastable stackiiigfault-like defects can be assessed without calculations. Since such faults occur as a result of crystal symmetry, they are common to all materials with a given crystal structure. However, other minima than those associated with symmetry-dictated extrerna may exist in any particular material. These cannot be anticipated from crystallographic grounds, and their existence depends on the details of atomic interactions.
Complete calculation of y-surfaces for a given material can only be done by einploying reliable descriptions of interatomic in tions have been made in a various empirical sclienies ran (e.g. Vitek, 1968; Basinslti et 1999), t h ~ o u ~niany~body h (e.g. Cserti F) al., 1992; Khaiitha et al., 1992; Farkas and Vailhe, 1993; Girshick and Vitek, 1995; Simmoiis et al., 1997) to tight binding schemes (e.g. Xu and Moriarty, 1998) and recently even by electronic structure methods based on the density functioiial theory (DFT) (e.g. Kaxiras and uesbery, 1993; Juan and Kaxiras, 1996; Ehmann and Fahnle, 1998; Mryasov et al., 1998; LWryasov and Freeman, 1999; Waghmare et al., 1998, 1999a,b). An i ~ p o ~ t a n t advantage of y-surface calculations i s that they can be done relatively easily compared, for example, to atomistic calculations of dislocations, so that ah initio studies of y-surfaces can be made even for complex structures.
Now we coiisider stackin~-fault-li~e planar defects in intei-iiietallic compounds possessing diverse crystal structures. When defining these stru~tureswe use the structural types, Pearson symbols and the notation of the International Tables for Crystallogr~phy(Wahn, 1996) including the space group number. Table 1 summari~~ Pearson s symbols and space structures discussed in this chapter together with the two-di~e~isional(2D) plane groups aiid sectioi~al-layer groups for the crystallograp~cplanes on which the faults are located. Zn the aiialysis of Stacking-~aLilt~likedefects we employ figures that present atomic con~gurationsin
able 1 Pearson symbols and space groups of intermetallic structures discussed, and the correspond in^ plane groups and sectional-layer groups o f the planes of stacki~g-fault-l~ke defects ~~
Structure
L1,
Pearson symbol Space group Space group iiumber Fault plane Plane group Layer group
CP4 Pm3m 221 (111) p6mm p3ml
Layer group nuniber
L72
L*o
DO,,
tP4
t18
DO,,
139 (111) pmm2 p2/mll
L14
L14
L2,, DO,
C11,
cF16 t16 CP_2 Pm3m Fm3n-1 14/iiimm I94 22 1 225 I39 (0001) (110) (110) (110) p ~ ~ pmm2 ~ m pm~2 cmm2 p2ml pinrnin prnnini c~mmm p61n2 pinmb pmmn cmiiie L72, L78 L37, L41 L37, L46 L47, L48
hP8
~ 4 / ~~ 4m ~ r n~l n Pl ~6 , / m c 123 (111) pmm2 p2/mll
B2
C40 hP9 P6,22 180 {OOOl) ~ 6 c211 Ll0
c54
oF24 Fddd 70 (001) ~ ermm2 m C222 L22
440
~ e c l ~ ~Properties ~ i c ~ l
perfect crystals in projections perpendicular to the plane of the fault, and the following conventions are observed:
(i) The atomic positions in different planes located in the direction perpendicular to the plane of the fault are distinguished by circles of different sizes; the largest circles correspond to the top and the smallest circles to the bottom layer, respectively. (ii) When i ~ t r o d u c i n ~the generalized stacking faults, the cut is always made just below the plane of atoms depicted by the largest circles and these atoms belong to the upper part which is being shifted. (iii) The rectangles drawn in these pictures represent two-dimensional repeat cells of the plane considered, and dashed lines denote mirror planes perpendicular to this plane. The two-dimensional planar groups and three-dimensional sectionallayer groups, i.e. the largest subgroups of the perfect crystal space group preserving the orientation of the chosen plane of the fault, are also given. (iv) Since we can expect that y-surfaces will possess maxima for displacements which place the atoms of the upper part exactly over the atoms in the first layer of the lower part the domains of displacements for which the generalized stackingy is expected to be very high are indicated by large dotted circles centered at atoms of the first layer of the lower part. Local minima at y-surfaces, corresponding to metastable stacking-fault-like defects, are then anticipated for displaceinents leading to locations in between the dotted circles. The same concept is used when employing a hard-sphere model. One type of these displacements brings the atoms of the upper part to positions of other atoms of the same part (largest circles). Such displacements are lattice vectors in monoatomic structures and define possible APBs in ordered alloys. Other displacements to positions of expected minima at y-surfaces define possible stacking faults. In some structures, for example Llz and Llo, they place the atoms of the top layer (largest circles) exactly above the atoms in the bottom layer (smallest circles).
We start our survey with the fcc-based cubic Ll2 (cP4) any i ~ t e r ~ e t a compounds l~c with this
structure have been investigated during thes last 30 years, and we present a comprehensive review of measurements of energies of APBs and stacking faults. The most frequently studied Ll2 compound is Ni3Al which is the major phase in Ni-based superalloys (Pope, 1992; Nabarro, 1994; Anton, 1995). In Llz compounds the slacking-fault-like defects play a crucial role in the important phenomenon of the anomalously increasing yield stress with increasing temperature (Westbrook, 1959) (for reviews see Pope and Ezz, 1984; Liu and Pope, 1996; Vitek et al., 1996; Veyssikre, 1998)+ The L12 structure can be considered as derived from a parent fcc lattice in which one simple cubic sublattice is occupied by the minority atoms and the other three sublattices by the majority atoms. The symmetry of this structure is cubic, and Figure 2 represents the atomic positions in different (111) planes using the notation defined above. The elementary twodimension~lcell of the (1 11) plane is centered, and the corresponding 2D-plane group has the highest symmetry of ~ 6 The ~ direction ~ of ~the projection ~ . is parallel to the three-fold rotational axis [ I l l ] . Only one plane of each of the three sets of mirror planes is drawn in Figure 2. Other equivalent mirror planes are parallel planes passing through the dark circles. Similar planes passing only through the open circles are not the mirror planes because of the two different atomic species in the ordered alloy. However, they are mirror planes in the parent fcc lattice, The highest symmetry of a generalized stacking-fault is then given by the sectional-layer group When searching for stacking-fault~ty~e defects we follow the procedure defined above. The following are the three displacements for which such faults are anticipated (see Figure 2): (i) The displacement of the type 1/2 { 110) from the large dark circle to a large open circle defines the APB; it interchanges the simple cubic sublattice occupied by the dark circles into that occupied by the open circles. (ii) The displacement of the 1/3 (112) type from the large dark circle to a small dark circle produces the superlattice intrinsic stacking fault (SZSF). (iii) The displacement of the 116 ( 112) type from the large dark circle to a small open circle, that leads to the well-known intrinsic stacking Fault in the parent fcc lattice, defines the complex stacking fault (CSF). This fault disturbs both the lattice stacking and the location of the dark-atom sublattice.
S t a c ~ ~ ~ g - F aT uy p~et Interfaces -
44 I
I
igure 2 Three (1 1 1) planes of the LI 2 structure. The atoms belonging to three adjacent atomic planes are dj$ting~ishedby the sizes of the circles, the largest being from the top plane. The rectangle defines the two-di~ensionalcentered repeat cell of the (111) planc, and the dashed lines represent the planes of mirror symmetry perpendicular to (1 1 1). The two different atomic species are depicted by open and filled circles, respectively
The displacement yielding the SISF leads to a position where three mirror planes intersect and thus the symmetry of the sectional-layer group served. This implies an extrernuiii at the y-surface and since for this displacement the separations of the first and second nearest neighbors remain the same as in the perfect lattice, it is most likely a minimum. Hence, the existence of the SISF is in Llz structures guaranteed by symmetry. The only other displacements leading to positions where three mirror planes intersect are shifts restoring the perfect crystal agd shifts to positions marked by ~ ~ d i u ~ - sdark i z e circles, corresponding to maxima of the y-surface. However, the symmetry does not guarantee extrema on the y-surface for displacements corresponding to the APB and CSF since only one mirror plane passes through these points. Hence, these faults may, but need not be, metastable, depending on the specific ~at~riaF 1 .u ~ t ~ e r m if~ rthe e ~mini mu^ is present it may be displaced away from the geometrically defined position along the line of the intersection of the mirror plane with the (111) plane (~amaguchiet al., 1981b).
Notice that in the parent fcc lattice the 1/6(112) displaceiiient leads to the position where three mirror planes intersect, and thus the existence of a meta~table intrinsic stacking fault is guaranteed by symmetry. While the SZSF and CSF are faults specific for { 1 I 1} octahedral planes, the plane of the crystallographic plane. Another which plays a role in the formation of Wilsdorf locks, is on (001) cube pl corresponding to the 1.12{ 110) dispiacenient is at the point of intersection of four mirror planes (two of the { 100) and two of the { 101) type), and the y-surface must, therefore, possess an extremum at this point. This is usually a minirnuni since this displacement disturbs neither the separation nor the chemi of the first nearest neighbors, and thus the (001) planes is metastable in all Ll2 ordered alloys. The de~nition of the generalized stacking fault introduced above assumes that the fault is localized in a single plane. However, this concept can be generalized to faults spread into several parallel a t o ~ i c
~ e c ~ i a ~ iProperties cal
t can be created by making cuts ent planes and displacing the part e by a vector 1, above the second e corresponding y-surface will then in the five-, seven-, or higherAn example of a two-layer stacking sic stacking fault in fcc crystals isp~ace~nents of the I /6 (1 12) type (111) planes (Wirth and Lotlie, situation can arise in the Llz mple, displacements in two adjacent e same vector which leads to the ce a superlattice extrinsic
there may also be noiithat are not formed by a d are, therefore, not connected with perdislocations. These faults may n~gurationswhen the structure of lel to the boundary is different. We hese faults in detail but instead situation that may arise in the Llz two types of atomic planes in this 1987). The planes of the first type, es odd, contain atoms from all four ices. The planes of the second type, mbinatioiis of Miller indices, comrom two sublattices, and adjacent contain different sublattices. In an r type formed on the planes of the s (OOl}, only the atomic planes with lattices can be in conttact at the ained below. In an X3Y ordered lattice i s occupied by Y atoms, and the second type can (each letter stands ). Consequently, three APB concur that differ in local chemical
~ concentrations ~ ~ r of different cornained by segregation. The XX-XX h correspon~sto a layer of pure X, the lowest energy while XY-XV, a very significant devkation from rds , is likely to have the highest a non-shear c o n ~ g u r ~ t i owith n the second type that possesses a ely to occur (Paidar, 1985).
The energies of stacking-fault-like defects, determiiied from measurements of the widths of dislocation splittings, are summarized in Table 2 for various materials with the L12 structure. ( Suii (1995) for earlier data). The methods employed in these measurements and related qu precision are discussed in Section 4. energies may depend on temperature, in particular in materials with some structural instability near the critical temperature of a phase transformation; this is discussed in the Section 3.10. In general, the energy of stacking-fimlt-like defects inay depend sensitively on alloying, as seen in a number of cases summarized in Table 2. For example, the energy in Ni3Al increases by 50% when only 3 HE is added to the alloy (ICrurnl et al., 2000). In this context the SISF is an interesting case. Its energy is generally low in nickelbased Llz alloys, as expected from the fact that the separation of the first and second nearest neighbors remains the same as in the perfect crystal. However, in titanium trialuminides that have the L12 structure stabilized by alloying with transition metal Mn, Ni), the energies of the SISF and AP c o ~ p a r a b ~Furthermore, e. the SISF energy also varies wi tii traiisitioii-metal additions and appears to be related to the phase stability (Morris et al., 1995). These effects suggest that directional bonding is of importance in A13Ti-based alloys.
The Llo (tP4) structure can also be derived, at least formally, from a parent fcc lattice, but now two of the simple cubic sublattices are occupied by atoms of one type and the other two sublattices by atoins of the other type. The most notable compound crystallizing in this structure is TiAl that is considered as a very promising high-temperature material I automotive industrie ~ i m i d u k 1998.1999; , et a/., 2000). This structure is tetragoiial with different species alternating on the atomic planes parallel to the basal plane. Nevertheless, it is cL~stomaryto use the cubic notation, with mixed parentheses { hkl) and (uvw], to denote planes and directions in this structure. In general the c / a f 1; if c = a the parent fcc lattice is uiidistorted while the structure remains, of course, tetragonal. igure 3 represents the atomic positions in three different (1 11) planes using again the notation defined earlier. The elementary two-dimensional cell of this plane is not centered since its center is occupied by an
443
~ t ~ ~ ~ ~ n Type g - ~Interfaces ~ u l t -
Energies ( n ~ J m - ~o)f APBs on (1 11} and (001) planes, CSF and SISF in Ll, c o ~ p o u n d measured s mostly at room temperature and m several cases after plastic deformation at higher temperatures Com~osition (at.%) Ni,Al 21.9 A1 22.9 A1 22.9 A1 23.5 A1 23.5 A1 24 AI 24 AI 24 AI 24.2 A1 24.4 AI deformed at 823K deformed at 1073I< 24.6 AI 25 A1 25.5 AI 25.5 A1 25.9 AI deformed at 623 K defoimed at 973 K 25.9 A1 26 A1 26.5 A1 23.5 AI, 0.3 B 23.8 AI, 0.7 B 22.7 A1, 0.3 Hf 22.7 Al, 1.6 Hf 21.9 Al, 3.3 Hf deformed at 683 K 23.5 AI, 2 Pd 24.1 Al, 0.9 Sn 21.3 Al, 3.7 Sn 24.7 Al, 1.0 Ta 17.4 AI, 6.2 Ti 23.4 Al, 1.0 V 20.8 Al. 4.0 V
Ni,Fe
APB (111)
APB lOl0)
CSF
SISF
Method
Referciice
1753. 15 206 122 1403.21
1 0 4 1 15
2353.40
6k0.5
WBIS WBC WBC" KREM WB WBIS WBIS WB WBC
(Karnthaler et al., 1996) (Diiiiiduk et al., 1993) (Dimiduk et al., 1993) (Crimp, 1989) (Yu et al., 1994) (Lo Piccolo, 1999) (Hemkcr and Mills, 2993) (Hemker and Mills, 1993) (Dimiduk et al., 1993)
183If 12 1761-11 1803.20 14'9115 1943.22
1793.15 1953.13
lilt-15 180130 175rf:13
1 4 9 1 12 1573-8 1 3 5 1 13
206 1- 27 206 130 1773.20
WB WB WB WBIS WB WBh WB
(Douin aiid Veyssiere, 1991) (Douin and Veyssiere, 1991) (Yu et al., 1994) (Lo Piccolo, 1999) (Douin erf al., 1986) (Douin et al., 1986) (Yu et al., 1994)
WB WB WBC WBIS WB WB WBIS WB WB WBIS WBIS WB WBb WB WBC WBC WBIS WB WBIS WB WBC WBC
(Douin and Vcyssierc, 1991) (Douin and Veyssiere, 1991) (Dimiduk et al., 1993) (Lo Piccolo, 1999) (Yu et al., 1994) (Yu et al., 1994) (Hemker and Mills, 1993) (Neveu, 1991) (Neveu, 1991) (Kruml et al., 2000) (Kruml et al., 2000) (Krum1 et al., 1997) (Kruml et al., 1997) (Sun et al., 1999) (Dimiduk et al., 1993) (Dimiduk et al., 1993) (Baluc and Schaublin, 1996) (Baluc and Schaublin, 1996) (Korner et al., 1993) (Korner, 1988) (Dimidulr et al., 1993) (Dirniduk et al., 1993)
553.5
WB WB"
(Korner et al., 1987) (Korncr and Sclioeclr, 1990)
1503-10
WBIS
(Balk et al., 1997)
2 9 0 1 15 2783. 18 148
WBIS WB WB
(Balk et al., 1997) (Kumar and Hemker, 1997) (Fang et al., 1994)
751-25 2 5 1 15 1433- 7 1603.16 9 0 15 140 1343.8
236 129
1203.20 I05 3. 10 250 3.29 2 1 9 1 17 1783. 12 1'903. 13 173k 15 1503.20 190$.20 300 k25
1443.20 201 1 2 2 2153.27 237 t- 30 1903.20 250-420 250 3. 30 254531 276 3. 37 93k6 163.5
286 3.22 163.9
1773- 16 113110 124$8
27'9 3.49 335 3.60
1203. 20 170 3.20 250 k 25 220 I 2 5 205 1 2 5 1653. 15 1021- 11 200 k25 1863. 15
3 460
15k3
12 352 3. 50 227 3.30 > 370
15k5 1515
250 3. 30
215 3. 35
15-30
WB 124 77
WB WB
(Yoshida aml Taltasugt, (Yoshida and Takasugi, (Yoshida and Takasugr, (Yoshida and Takasugi,
1994) 1994) 1992) 1992)
Coinposition (at.%) Cu,NiZii
AI& Al,Ti in L1, 25 Ti, 7.5 Cr defoi-med at 773K 25 Ti, 9 Cr 25 Ti, 7.5 Fe 25 Ti, 8 Fe 28 Ti, 8 Fe 28 Ti, 8 Fe deformed at 773K deformed at 973K 23 Ti, 6 Fe, 5 V
Zr,Al deformed at 67313: deformed at 873-1073 K
APB ill11
APB (0101
CSF
SISF
Method
Reference
9 7 5 17 88 & 29 6256 4357
WBd W Be WB’ WBg
(Van Der Wegen et (Van Dcr Wegeii et (Van Der Wegen et (Van Der Wegen et
313
WBh
(Gcorge et al., 1990)
175 130 305 1 3 5 95
WBh WBh WBIS “REMh
80510 100
305 140 72 274
100 340160
45
2 30--45
al., al., al., al.,
1983) 1983) 1983) 1983)
WBh WBh
(Miura and Watanabe, 1993) (Miura and Watanabe, 1993) ( K m a r and Hemker, 1998) (Inoue et al., 1991) (fnul et al., 1992a) (Morris and Guilther, 1993) (Lerf and Morris, 1991)
WBh W B”
(Lerf and Morris, 1991) (Ceorge et at., 1990)
WB
(Holdway and Staton-Bevan, 1986) (Holdway ancl Staton-Bevan, 1986)
w €3”
Meaning of acronyms: WB - weak beam; WBlS - weak beam combined with the image simulation; WBC - weak beam with anisotropic correction; WREM - high resolution electron microscopy. “Annealed at 973 K for I hour and water quenched. ’IR situ observation at 983 K. ‘Iri situ observation at 753 IS near the order-disorder transition temperature of 776 K. for 164 hours above the transition at 598R to modified Ll,, not aged and deformed. ‘Aged at 623 I< for 75 min. ‘Annealed at 723 IS for 112 hours and subsequently deformed. gAnnealed at 723 IS for 72 hours and at 763 IS for 2 hours then deformed; disordering temperature is 774 K. hIsotropic elasticity used.
atom of a di~erenttype than at the corners. The 2Dplane group of the (111) atomic plane has the and the respective 3D-sectional-layer group, which determines the highest symmetry of a ed stacking fault on a given fault plane, is Thus there is only one set of mirror planes ular to the (1 11) plane, the planes (170). The direction (1101 is not equivalent to { 1011 and, therefore, while the displacement 1/2 (lib] restores the other two 1/2 (1011 isp placements analogously to the L12 structure. Similarly, two types of stacking faults can be formed on the { 111) plane. The SISF is created when the 116( 1121 d~~placementvector is perpendi~ular to 1/2 [ 1701 while the other two 1/6 { 1211 displacements The notation of the layer groups follows (Vainshtein, 1981) (Figure 2.63 on pages 120-121) and Voluine E of the ~ ~ ~ ~ ~ v iTables a ~ if o~r nCa lr ~ ~ ~(Kopsky ~ ~ and l Litvin, l ~ ~ 2000).
lead to the CSF. Since there is only one set of mirror planes, the symmetry does not guarantee that the ysurface will possess extrema for these displacei~ents and, consequently, the APB, CSF and SISF may be metastable, but need not be in this structure. Furthermore, when they are metastable, the fault vectors may be somewhat d i ~ ~ r e nthan t those deduced purely crystallographically, depending on the given materiaL2 Similarly as in the L12 structure, the APB can be introduced on other crystallograp~icplanes than { 111) and the other most important APB may again be on { 001) planes. 21t should be noted that the decoinposition of the superdislocation with the Burgers vector (1011 into two superpartials 1/2 (1011 is not symmetric. Both supe~partials can further split into two 1/6(121] Shockley partials, but these bound the CSF in one case and the SISF in the other ~case.a An~ analogous ~ ~ dcconiposition in Llz IS symmetric since the Shockley partials always bound the CSF.
445
ure 3 Three (1 11) planes of the Llo structure. The notation and the meaning of the rectangle and the dashed line are the same as in Figure 2
An example of the recently calculated (1 11) y-surface in TiAl with the Llo structure is shown in Figure 4 (Znam, 2001). The calculation was performed using the tight-binding-based bond-order potentials, which explicitly include non-central p- and d-electron bonding (Znam et al., 2000). The basic features of the ysurface are as anticipated geonietrically (see Figure 3). There are three minima corresponding to the APB, CSF and SISF; but their positions deviate, albeit iiot drastically, from those envisioned ~eometrically. The ~ i n i m u mcorresponding to the APB is very shallow aiid the energy is 542 mJ in-2; tlie energies o f the SISF and CSF are 141mJ mV2and 412 mJ m-2, respectively. A similar calculation employing the DFT-based LAPW method was made by Ehinan and Fahnle (1998). The overall form of the y-surface is tlie same, but no ~ i n i m uwas ~ found in the vi~inity of the presumed APB and thus this fault is unstable. The calculated energies of the SISF, CSF and the energy at position are 172mJ m-2, 362 mJ m-2 and 667 m J m-2, respectively. Several other calculations of the (111) y-surface in TiA1 have been made using central-force, many-body ~otentials (Girshick and Vitek, 1995; Simmons et al., 1997). In these calculations all three types of faults are stable, but the energy of the SISF is rather low owing to the central-force
approximation. Furthermore, the superlattice extrinsic stacking fault (SESF) formed by displacement in two adjacent { 111) planes by the same vector 1/6 (1121 was found in atomistic studies of d~s~ocationswith 1/2 (1 321 Burgers vector (Girshick and Vitek, 1995). These dislocations dissociate into partials separated by a combination of SISF and SESF and may serve as nuclei for twinning. Such dissociations have, indeed, been observed in TiAl (Inkson and Wumpbreys, 1995; Inkson, 1998). The energies of APBs on { 111) and { 001) planes aiid the SISF iii TiAl, determined from measurements of the widths of dislocation s~littings,are summarized in Table 3. A relatively large variation in the measured energies cannot be related to diRerences in composition. Nevertheless, the measured energies of the SISF are within the scatter in a reasonable a~reementwith the above calculated values. On the other hand, energies of the APB on { 111) are signi~cantlylower than the ca~culated ones. In fact, the calculations suggest that this AP is either unstable or possesses so high an energy that plitting with APB cannot occur. This is, indeed, implied by the results of Stucke et al. (1995) who observed only SISF in the dissociation of the (1011 superdislocat~ons.In (1989) observed splitting with an
446
Meclzanicul Properties
y-surface for the (I 11) plane in TiAl calculated using the ~ight-bIndingbased bond-order ~ o t e ~ t iand a ~ repre~ented s by contours of constant energy (Znam, 2000)
the dis~arityof these two observations is not clear; but it is possible, for example, that segregation stabilized and decreased the energy in the latter case - if a long time after the deformation of tlie samples. In addition, when tlie width of dislocation splitting is very small, it is even difficult to c o n ~ rthat i ~ the ~orrespoiidings t a c ~faults i ~ ~ or antiphase boundaries exist 3s w e l ~ - d e ~ ~defects. ed While the structure of the dislocation core may be described cations, and two distinct erved, the ribbon of the veloped. It was shown in idas et al. (199~),in which the calculatioii of the of dislocation splitting in is discussed, that the error in determiaatioii of energy can be exceedingly large in this case.
We now proceed to other lose-p~c~ed structures, tetragonal DO22 (tE) and hexagonal I3019 (hP8). In the latter case the inaterial which is of principal interest is erjee, 1995) which is an importa~t llar TiAl-based alloys (Kirn, 1989; Tnui et al., 1992b; Kirn, 1994); and in the former case, it is Ni3V and A13Ti, the deforination properties of which were investig~tedby several authors (~ande~schaeve and Escaig, 1978; Fraqois et 1992; Vasudevan et al., 1989; Hug et al., 1989; rris and Lerf, 1991; ~ a ~ a ~ uand c hInui, i 1995). The DO22 structure is tetragonal and can be considered as fo from the Llz structure by a s on every second (001) plane.
447
S t a c ~ ~ ~ ~ - Type ~ a u liz l tter - fcices
of APBs on { 111) and (001) planes and SISF in TiAl with L1, structure, measured mostly at room le 3 Energies (~n.Tm-~) temperature and in several cases after plastic defomatlon at higher temperatures Composition (at.%)
APB (111)
APB (00 1)
SISF
Method
100 210
140 73*3 60 140 185 116k8
WBIS WB" WB" WBb WBIS WB WBIS WB
Reference
~
52 AI 54 A1 deformed at 873 K 54 A1 56 AI deformed at 873 K
> 250 1 4 5 1 15 I20 250
386 1983.25
(Wiezorck and Huinphreys, 1995) (Hug et al., 1989) (Hug et al., 1989) (Woodward et a?., 1992) (Stuck et al., 1995) (Stucke et al., 1995) (Stucke et al., 1995) (Stucke et al., 1995)
Meaning of acronyms is the same as in Table 2. "Calculated from the measured width of splittiiig using cubic elastic constants. bCalculated from the measwed width of splitting measured m (Hug et al., 1989) using the tetragonal ekastic constants.
Nevertheless, similar to the case of the Llo structure, it is customary to use the cubic notation for crystallographic planes and directions. However, in general c / a # 1, and thus the parent fcc lattice is undistorted only if c = a. Figure 5 represents the atomic positions in three different (1 11) planes, similar to Figures 2 aiid 3; note that there are six (1 11) planes within the period. The elementary repeat cell is again rectangular. The 2D-plaiie group of the { i l l ) atomic plane has the syrnmetry as xn the Llo structure, and the respective tional-layer group, which determines the highest symmetry of a generalized stacking fault on . Again there is only one set of mirror cular to the (111) plane, the planes (170); but only every second one of these planes is a mirror plane, as seen in Figure 5. Although the f a u l t ~ l i ~defects e are of the same structure; because there is a greater structural complexity, two types of APB and SISF uished. The correspondiiig displacement vectors chosen as described in the introduction to Section 3, are shown in Figure 5. They are 1/2 ( l i O ] type for APBI, 112 (lO?] type for the APBIl, 1/3 (?21] type for the SISFI, 1/6 ( 1121 type for the SISFII and 1/6 (1311 type for the CSF. Once more, the symmetry does not guarantee that the y-surface will possess extrema for the displacements associated with these faults. Thus, the APBs, GSF and STSFs may be metastable but need not be in the DO22 structure; and, when they are, the fault vectors may again be somewhat different than those deduced purely crystalliically, depending on the given alloy. ilar to the L12 structure, the AP other crystallogra~hicplanes. Again, the most important inay be (001) planes, where stability is guaranteed by symmetry, since the 1/2 ( 1101 displacement vector points
to the intersection of four xnirror planes (two { 100) and two { 110)).This APB represents a two-layer slab of Ll2 structure, aiid its energy will be relatively low if the difference between the cohesive energies of the same alloy in the DO22 and Llz structures, respectively, is sInall. This can be expected in A13Ti which transforms into L12 structure upon alloying with transition metals (Cr, Fe, Mn, Ni) (Ceorge et al., 1990; lnoue et al., 1991; Lerf and Morris, 1991; s et al., 1991; Tnui et al., 1992a; Miura and Wat Gunther, 1993; Yamaguchi and Inui, Heniker, 1998). The y-surfaces for both (1 11) calculated in Khantha et al. (1 9 using an e~pirical iiiany-body central-force pote Sinclair type (Finnis and Sinclair, 1984; Aclsland and Vitek, 1990), which assure the stabi~ityof the with respect to alternative structures. In these culations all five faults on { 111) planes were found to be metastable, albeit with displace~entssomewhat deviating from the geometrical ones. As envisaged, the lowest energy i'ault was found to b the SISFI has the highest energy. AP on {OOl) planes, as well as the GSF energies in the same range within 20%. However, the diEerences rnay be more p r o n o u n c ~if~ directio~a~ bonding is taken into consideration. ~urthermore,the SESF formed by displacement in two adjacent ( 111) planes by the same vector that leads to the §ISFll, 116 ( 1121, was found in atomistic studies ( ~ h a n t h aet al., 1992); the energies of SISFII and SESF are similar. Such faults were, indeed, observed in (Vanderschaeve and Escaig, 1978; Franqois et al., 1992) and rnay serve as nuclei for twinning which inain deformation mode in intermetallics with structcire.
448
Three (I 11) planes of the DO22 structure. The notation and the meaning of the rectangle and the dashed line are the same as in Figure 2
The energies o f SISF, ESF and APBs on ( 11 1) aiid (001) planes in two CO pounds with DO22 structure, Al3Ti and ~ i 3 estimated ~ , from measuremeiits of t1ie widths of dislocation splittings, are summarized in Table 4. As expected, in Al3Ti the energy of the APB on (001) is niuch lower than that of the APB on ( 111) planes. However, the same is not the case in Ni3V, which is, presumably, mzzch niore stable in the DO22 2 structure. On the other hand, energies I and SESF are in this compound low and very similar. 112c
No. 1
~
~
octahedral plane in L12. However,- in DO19 -two different ~~-sectional-layergroups and can be distinguished with respect to the position of the cut along the normal to (0001). Comparison of Figure 6 with Figure 2 suggests that the y-surface for (0001) in DO19 is very similar to that for (111) in Llz and the following are the anticipated stacking-fault-type defects.
)
This is a hexagonal structure related to the cubic L12 structure in the same way as the hcp lattice is related to the fcc lattice. In this section we einploy the notation for crystallo~~aphic planes and directions commonly used for hexagonal structures. Figure 6 represents the atomic positions in two diEerent (0001) basal planes using the notation defined at the beginning of Section 3. The elementary t~~-dimensional cell of the basal plane is ceiitered, and it has the same 2D-plane group of the highest possible symmetry, ~ 6as the~ ( I 11)~
(i) APB produced by displaceme~t of the type 1/3[’i-210] from the large dark circle to a large open circle. (ii) SISF produced by di~placement of the type 213 [TOlO] from the large dark circle to a position between three inaxima surrounded only by open circles, i.e. atoms of a different type than the atom displaced. (iii)CSF produced by displacement of the type 113 [ lOTO] from the large dark circle to a position between three maxima surrounded by two open and one dark circles; note that the latter is an atom of the same type as the atom displaced. Three mirror planes intersect at the position corresponding to the SISF which implies an extremum at the y-surface. The extremum is most likely a minimum ,
449
S ~ a c ~ ~ ~ Type ~ - ~I ~ut e~ ~l f ta -c e ~
Energies (mJrn-,) of APBs on (111) and (001) planes, SISF and SESF in two compounds with DO,, structure, determined from obscrved width of dislocation splitting Composition (at.%) A1,Ti unde formed deformed at 1023K deformed at 873 I< deformed at 873 IS deformed at 673 K Ni,V undefo~ed deformed at 1073 K deformed at 873 K
APB, (111)
APB (001) 28 13+3 32 25 25
168k 13 190 200
100k20
100k15 32
SISF,,
SESF
99k8
21 k 3 21k5
41 2 5 28k8
Method
Reference
HREM WB WBa WBd WB
(Francois, 1992)
WB" WB WB"
(Vasudevan et al., 1989) (Morris and Lerf, 1991) (Hug et al., 1989) (Vandcrschaeve and Escaig, 1978) (Vasudevan et al., 1989)
Meaiiing of acronyms is the same as in Table 2. "Calculated froin the measured width of splitting using isotropic elasticity
since, similar to the Llz structure, the separations of the first and second nearest neighbors remain almost the same as in the perfect lattice if the c / a ratio does not diger very significantly from the ideal one. Hence, existence of the SISF i s guaranteed by symmetry. Other displace~entsleading to positions where three mirror planes intersect are either shifts restoring the
perfect crystal or shifts to positions of envisaged maxima of the y-surface. C o n s e ~ ~ i e n tthe l ~ ,s y ~ ~ e t r y does not guarantee extrema on the y-surface for displacements corresponding to the AP and, therefore, these faults may be but are not necessarily metastable, depending on the specific ~ present, u m material. Again, as in L12, if the ~ i ~ i is
\
I
cl' Figure 6 Two (0001) planes of the same as in Figure 2
DO19
------'I'
0
structure. The notation and the nneaiimg of the rectangle and the dashed lines are the
450
~ e c h a ~ iProperties ~al
it may be displaced away from the geometrically defined position along the line of the intersection of the mirror plane with the (0001) plane. The y-surface for the (0001) basal plane in a DO19 structure was calculated by Cserti et al. (1992) using empirical many-body, central-force potentials of the Finnis- inc cl air (1 984) type. Three non-e~uivalent minima of metastable faults corresponding to APB, SISF arid CSF were found very close to the positions shown in Figure 6. No other metastable faults have been found on the basal plane. Similar to other structures, APBs can essentially lie on any plane. In the DO19 structure, besides the basal plane, the other most important crystallographic plane is the prism plane (l0TO). The significance of these two types of planes may differ from material to material, similar to hcp metals (Hirtli and Lothe, 1982). Indeed, two well-known alloys crystallizing in the DO19 structure, Ti3A.1 and Mn3Sn, display a different preference for the slip planes. Both alloys show a coniiglon slip direction, ['7210],but in the latter case the slip occurs preferentially on basal planes (Takeuchi ramoto, 1974) while in the former case on anes (Minonishi, 1991). When considering prism planes, it is seen froin 019 structure can be constructed by stacking eight different (1070) planes. Half of this motif is repeated with the displacement 1/3 [T2x0], and there are two distinct spacings of neighboring (1070) planes. When analyzing the planar faults, we consider only cuts between widely spaced (1010) planes, since we anticipate that possible stac~ing-fault-typedefects formed between narrowly spaced planes have much higher energies. Still, two distinct types of cuts between widely spaced planes need to be distinguished. In the first case the pair of neighboring planes has two diflerent atomic species on opposite sides of the interFace, and in the second case all atoms are of the majority species. Figures 7a and b represent the atomic positions in three consecutive (ioio) planes for these two distinct cuts, respectively, using the notation defined in the introduction to Section 3. However, tlie dashed circles nnarking high-energy regions are centered at the atomic position of both planes that are positioned below the top plane because these two planes are very close to each other. Two types of anticipated stacking-fault-type defect, APBs and a stacking-faults (SF), are labeled in Figures 7a and b. The displacement vectors are the same in the two cases; but since the cuts are different, these faults niay posses different energies. As pomted out by Umakoshi and Yarnaguchi (1981), both APBs corres-
pond to the symmetry-dictated extrerna of the ysurface and may thus be expected to occur in any DO19 material. However, for APB, the separations and stoichiometry of the first and second nearest neighbors reinain the same as in the perfect lattice and its energy is, therefore, expected to be lower than that of the APBII for which the stoichiometry changes. This inference has, indeed, been corroborated by calculations (Cserti et al., 1992) which give ene by one order of magnitude. The po faults do not correspond to any symmetry-dictated extrema 011 the y-surface, but calculations in Cserti et al. (1992) revealed another ni~nimumthat corresponds to a metastable stacking fault with the fault vector u/6 [12ix], close to SF,, for the cut leading to the low energy APB. Here x determines the [OOOl] c o ~ ~ o n e n t of this fault, and both a and x depend on the details of the atoinic interactions. Since this fault vector is not determined by the symmetry of the DO19 structure, it will be different in different materials; and in some ay not be metastable. in TijAl has been determined ex~erimen~dl~y by ~ e a s u r i nthe ~ width of dislocation splittings usiiig weak-beam electron microscopy (Legros et al., 199Ga,b). Since in Ti3Al the glide of 213 /12iO] super dislocation^ is preferred on prism planes, the observation of dislocations on these planes IS more coilamon. Nevertheless, splitting of dislocations was observed on both prism and basal planes, and in the latter case even splitting involving the SISF was found. The average values of the energies of these stacking-fault-li~edefects are summarized in Table 5. The difference between APBI and APE311 on the prism plane could be determined; because in the case of the high-energy APB the width or splitting depended only on the orientation of the dislocation line, as predicted by elasticity, while in the case of the low energy APB the width of splitting varied somewhat randomly by about one order of magnit~~de (Legros et al., 1996a). Similarly, a large scatter of APB widths was found for dislocation splitting on the basal plane. This variation of the width of dissociation can be exphilied by independe~tlocking of the two partials, and it is consistent with the sessile dislocation core structure revealed in these two cases by calculations in Cserti et al. (1992). The distributlon of observed APB widths IS well described by a statistical model of nonequilibrium dislocation dissociation (Paidar et al., 1994; Paidar, 1996). It should be emphasized that this efTect has to be taken into account when determining the fault energies from individual observations of dislocation splittings that are composed of sessile
igure 7 Three (XOlO) planes of the DO19 structure. (a> The fau,, plane IS located between the planes composed of the same atomic species. (b) The fault plane is located between the planes composed of different atomic species. The notat~onand the meaning of the rectangle and the dashed lines are the same as in Figure 2
partials. The mean value of the observed partial dislocation separations can be significantly different from the separation that is found with the highest probability, ~ o r r e s p o n ~ to i n the ~ equilibrium separation.
The structures of the next group of intemetallics discussed bere are ic, based on the bcc lattice. These are binary (cP2) and DO3 (cF16) and ternary L21 (cF16) ctures. Important conipounds which crystal~i~e in these structures are NiAl (B2), another candidate for high-temperature applications authofX: 1989; Darolia et al., 1992; M i r d e and arolia, 1995), and Fe3Al (DO3) considered as possible
replacement for certain steels (Vedula, 3995; 1998; Neumann and Sauthoff, 19 The parent bcc lattice of the decomposed into two simple cubic lattices, each of them occupied by different atomic species. Since there are no nietastable stacking faults on either ( 1l O } or (112) planes in bcc metals (Vitek, 1968, 1992; Duesbery, 1989), we will colisider only the antipbase boundaries as possible stacl~ing-fault-like defects in ordered structures derived from the bcc lattice. The most densely packed atomic planes are { llO}, and we discuss faults on these planes. However, occurrence of APBs on { 112) planes was also considered (Ta~euchi, 1980). Figure 8 shows the atomic positions in two different (110) planes using the notation defined at the beginning of Section 3. The s y ~ ~ ~ eoft rthis y plane is
452
~ ~ ~ ~ h Properties a ~ i ~ f f l Energies (rnJn~-~) of APBs and SISF on basal and prism planes in Ti,Al ~
Composition
APB (000 1)
APB,
(1 oio)
APBU (I oio)
SISF
Method
63
42
84
69
WB
_
_
Reference (Legros et al., 1996a, b) -~~
~~
Meaning of acronyms is the same as in Table 2.
,, ,,
I
I
,
i
_I--*
, - - * ” - - - - -
Two ( 1 10) planes of the B2 structures. The notation and the meaning of the rectangle and dashed lines are the same as in Figure 2 , and its elementary two-dimensional cell is
with the sectional-layer groups . All the atomic planes (001) and ar to the fault plane, are planes of mirror symmetry. The displacement 1/2 [ 1111, which defines on (110), points to the intersection of two mirror planes. Owing to this symmetry, the corresponding y-surface will possess an extremum Tor the 1/2 [ 11 11 displacement in any B2 alloy. number o f atomistic calculations of y-surfaces have been made using pair potentials and the embedded atom method (EAM). In a study using pair-potentials for a model I32 lattice (Yarnaguchi et al., 1981a), the 1/2[111] splacernent was found to lead to a eta stable A On the other hand, in calculations employing EAM type potentials for NiAl (Farkas and Vailhe, 1993; Parthasarathy et al., 1993;
Farkas e f al., 1995; Vailhe and Farkas, 1997a, b; Schroll et nl., 1998) a saddle point was found for this displacement. However, two symi~~e~ry-rel~ted ininima were detected for an additional non~crystallographic displacement in the [I101 direction (see Figure 8); the magnitude of this displacement depends on the interatomic potentials used. The possibility that instead of a single m i n i n i u ~corresponding to a displacement for which an extremum on the y-surface is guaranteed by symmetry, several symmetry-related, ~nergeti~ally degenerate minima occur, is a general feature that may appear in any structure. However, to our knowledge, this type of degeneracy of stacking-fault-like defects has not yet been established ex~erimentally, Energies of APBs were detemined from observed widths of spliltings o f superdislocations 111 several B2 compounds using electron microscopy . Earlier data
Table 6 Measured energies (mJm-') of APBs in B2 compounds Coinposition (at.%)
APB Ill01
APB r2111
Method
> 500
> 750
WB
(Veyssiere and Noebe 1992)
WB
(Shyue et al., 1993)
150 15
References
WB
Meaning of acronyms 1s the same as iii Table 2.
can be fouiid in Table 1 of Sun (1995). Several inore recent measurements are summarized in Table 6. (~~~~ t Ti)
The binary DO3 and ternary L21 structures are both derived froiii a parent bcc lattice that can be decomposed into four fcc sublattices (two simple cubic sublattices can each be subdivided into two fcc sublattices). Both structures possess cubic syninietry. When two fcc sublattices are occupied by the same
atoiiis but the other two sublattices by atoms of two different types, a ternary L21 alloy is formed. In the binary DO3 structure, only one out of four fcc sublattices is occupied by atoms of a different type. Figure 9 represents the atomic positions in two different (li0) planes in. either L21 or the notation is similar to that defined in the introduction to Section 3, but figure captions explain the distinction between the two structures depicted. The 2D-plane group of the (110) atomic planes has the symmetry ~ ~the same ~ as2for the , The elementary two~dimensio~al cell is again rectan~ular,
--
Figure 9 Two (110) planes of the L21 and/or DO3 structures. In this figure components Y and 2; of an X2YZ ternary alloy with the L21 structure are represented by dark circles and squares, respectively. This figure can also be used to represent an X,U binary alloy with the DO3 structure, but in this case the dark squares should be replaced by open circles of the same sizes since they now denote the X component. The notation and the meaning of the rectangle and dashed lines are the same as in Figure 2. The points A, €3, C, D. E and F correspond to the positions inarked by the same letters on the y-surface shown iii Figure 10
54
~ e ~ h a n iProperties ~al
but its size in the [001] direction is twice that in the e largest sectional-layer groups The mirror planes include all cular to the [ l l O ] direction, but only every other atomic plane perpendicular to the LOO1 J direction (planes passing though the atomic positions drawn as filled symbols in Figure 9). Two types of APBs niust be distinguished in L21 and in DO3 structures, as shown in Figure 9. The APB,, corresponds in both structures to the position at which two mirror planes intersect and thus the y-surface will possess an extremmm, most likely a minimum, for this n the other hand, the APB, position lies on only one mirror plane, and so this possible t determined by the symmetry. structure the y-surface was calculated by Paidar (1976) using a set of pair potentials describing the atomic interactions in Fe3Si. This was one of the first y-surfaces calculated for an. ordered loy, and its contour map is presented in Figure 10. o calculatioiis o f y-surfaces have been made for the L21 structure. Letters A-F denote corresponding points in Figures 9 and 10; points A, B and C coincide with minima (the perfect crystal and two APBs), and poiiits D, E and F with rnaxiina on the y-surface. Notice that the disp~acementvectors pointing to A, C, aphic while the remaining vectors nd E (energy maxima) can differ tors. The non-crystallo~apl~c related to the lower symmetry
urc 10 y-surface for the (1 10) plane in the DO3 structure calculated using pair potential8 for F;e,Si and represented by contours of coiistant energy (reproduced by permission of Institute of Physics, Academy of Sciences, Czech Republic, from Paidar (1 976))
of the DO3 lattice when coinpared with the €32 lattice, and the corresponding displacenient depends on the particular type of atomic interactions employed in the calculation of the y-surface, i.e. on the ~ a t e r i a ~ considered. However, this effect is not the saiiie as the degeneracy of the APB discussed in the previous section. In recent years the most investigated materials with DO3 structure were Fe-AI alloys, coinpositionally close to Fe3A1, since they are Considered as possible replacement for certain rather expensive steels. The AP energies were measured in Fe;roAlso alloys under full load using in situ deformation in the electron microscope and in stoichiometric Fe& after deforma~ion at 573 K (Rosner et al., 1996). Such post mortem measurements were also conducted for two alloys with the same content of transition metals after deformation at different temperatures (Krhl et al., 1997). The APB energies obtained from these measureiiients are summarized in Table 7. In this table we also include very systematic and careful nieasurements reported in Crawfoi-d and Ray (1977) as well as one measurement of the APB energy in Fe75.2Si24.8.
In the follow in^ sections we discuss tetragonal Cllb (tI6), hexagonal C40 (hP9) and orthorombic CS4 (oF24) structures in which a number of transitionmetal silicides crystallize. These materials, which combine properties of metals and ceramics, have been extensively studied in recent years aiid are regarded as a very promising basis for a new ~enerationof hightemperature structural materials (Petrovic, 199s; Petrovic and Vasudevan, 1999; lso the chapter by Lipsitt et al. in this volume and umar in volume 2). One of the most studied silicides is MoSi2 (It0 et al., 1995, 1997b, 1999; Guder et al., 1999). It crystallizes in the C l l b structure which IS, therefore, disc~ssedin more detail. Some more complex structures are also briefly discussed. Of these the most important are Laves phases (Liu et al., 1993; Livingston, 1994; ~ o n K e i t zet al., 19981, but these merit a separate more extensive review. Cllb is the crystal structure of several transition metal (TM) di-silicides, in particular ReSi2. The physical and mechanical properties of MoSi2 have been studied in most detail, since this t~mperature,high materlal combines very high creep strength and corrosion resistance with at least moderate ductility at room temperature (Tto et al., 1995, 1997b, 1999; Guder et al., 1999; chapter by
455
~ t a c ~ ~ ~ g - Type F a ~ Interfaces lt-
able 7 Energies ( ~ n J m - ~of) APBs in DO, compounds as measured mostly at room temperature and, in indicated cases, after plastic deformation at a given ternpcrature
Fe@ def.573 K Fe,, 7 4 6 3 Fe,, ,A116 I) Fe72 ,AI,, 8 %2*J28
Fe,,AI,, def.773 K Fe,,Al,, def.873 Fe,, 7A118.1 I”e,o*l,, Fe7,Al,, def. 573 K Fe,,Al,, def. 693 K Fe,,Al,, def. 753 K Fe6,A12,Cr6 Fe,,Al,,Cr, def.773 K Fe,,Al,,Cr, def.873K
723-5S4 783- 10 693-510 753-9 803-7 67rt8 653-55 79+9 76+ 12
WB
391- I8 85+ 14 71 12 52&9 73&7
WB WB WB
+
77k5 763-6
643-9
2siZ4 8
WB VVB
44+7
WB WB WB WB
78k7 763-5 75+6 I22
WB WB WB
50+8 583-59 100t-6 893-6 81 rt7
WB WB
WB
(RBsner et at., 1996) (Crawford and Ray 1977) (Kral et al., 1997) (Crawfoord and Ray 1977) (Rosncr et al., 1996)
(Kral et al., 1997)
~Poschmannet al., 1995)
Meaning of acronynis i s the same as in Table 2 .
extremum on the y-surface where two mirror planes intersect. This extremum is located at the small dark circle in Figure 12 and it can be expected to be a ~ a x i m u m .Since the o ~ u r r e n c eof ~ i n i mcorre~ sponding to metastable stacking-fault-like defects is not assured by symmetry, the existence of the three above-mentioned faults is not guaranteed a priorz. ~ e v e r t ~ e ~ erecent s s , calculations employiiig ~ensityfunctional-theory-based ab initio methods (Wagli~are et al., 1998, 1999a,b, 2 suggest that a metastable 1/4 ( 1111, corresponding planes and has a relative n t 114[ 33 1). This displace~ent114[ 111] is e ~ u i v a l ~ to fault apparent~yparticipates in the dissociation of 1/2 ( 1111 dislocatioiis according to the reaction 1/2(111) = 1/4(111] l/4(111] that was observed using weak beam TE (Evans et al., 1993; Ito ef al., 1995; Maloy ct al., 1995; Evans c*t ul., 1997; lnui er al., 2000). On the basis of these observatiQns the energy o f in previous sections, based on considerations this fault was determined to be in between 260 mJ m-2 introduced at the beginning of Section 3, t h e e minima (Evans et al., 1993) and 360 mJ m-2 (Ita cf nl., 1995); on the y-surface can be expected. They are marked by this energy appears to be aAFected by alloying (Evans et correspond to APB, SISF and aE., 1997; Inui et al., 2000). The calculations give the et al., 1993). However, the energy 370 mJ rn-’ in good agreement with experisymmetry guarantees the existence of only one mental observations. While { 110) (1111 slip has been observed in the most important slip system that appears to contro~ ’The mixed notation { hkl) and (hltl] is used to dif~erenti~te the ductility of this material at ambient temperatures is the first two indices from the third one which plays a different the (013)(331] system (lto et al., 1995, 1 9 9 6 ~ ,1999; ~, role, owing to the tetragonality of the Cllb structure.
Lipsitt et al. in this volume and Kumar in volume 2). Most other silicides, such as di-silicides crys~allizingin structures discussed in the following section, aiid/or Mo5Si3 with the body-centered tetragonal D8,, structure that was also investigated recently (Petrovic, 1995; Chu et al., 1999a,b; Petrovic and Vasudevan, 1999), display either very low ductility or none. The unit cell of the C11b structure, shown in Figure 1la, is composed o f three bcc-like elementary cells and it can be coiisidered as an . . , A {lIO) atomic planes.3 Figure 12 shows the atomic positions in two consecutive {ITo) planes using the notation defined at the beginning of Section 3. In this plane the two-dimensional elementary cell is rectan-plane group would have the highest symmetry of ~~~~~~2 for the ideal ratio c / u = &. s froni the ideal one plane group is only emm2. The sectioiid-layer groups are
456
~ e c h ~ n iProperties c~l
igure 11 Structures of three transition metal di-silicides that can be considered as various stackings of the same type of atomic planes, indicated by shading. Dark circles: transition metal (e.g. MO), open circles: Si. (a) C11k> structurc; shaded plane is of the (110) type. (b) C40 structure; shaded plane is of (0001) type. (c) C54 structure; shaded plaiic is of (001) type
Inui et al., 1999; Mitchell and Misra, 1999). For this reason we shall discuss possible stacking-fault-type defects on { 01 3) planes. Figure 13 shows the atomic positions in three consecutive { 073) planes using the not~tiondefined in the introduction to Section 3;
however, note that there are five (013) planes in one period. The elementary 2D repeat cell of this plane is again rectangular and centered. Owing to the greater complexity, there are two types of APBs (Umakoshi et al., 1990): APBI with the displacement vector 1/6[331],
457
Stac~j~g-FaultType Jnterjbces
I
Two { l i O ) planes of the Cl 11, structure. The notation and the meaning of the rectangle and dashed lines are the same as in Figure 2
and APBII with the displacement vectors 116[73T]. In addition, in analogy with the ( 110) plane, we can consider three distinct stacking faults, as marked in Figure 13; these are not the same as those discussed by Rao et al. (1993). Ab jnj~jocalculations of the y-surface cross-section along the [331]direction have been recently performed by two groups ( ~ a g l i m a r e et al., 1998, 1999a,b; Mitchell et al., 2001). Both reveal metastable faults akin to APBI and APBi1 but the corresponding displacernents are smaller, approximately 1/8 [ 331J' and 1/8 13313, respectively. These faults have also been found in calculations employing the modified embedded atom method (MEAM) (Baskes, 1999), presented in Mitchell et al. (2001); but, in addition, ~ i n i m ahave also been found exactly at positions corresponding to APBI and APBIl. The energy of the APBl is then the lowest fault energy detected. Inspection of Figure 13 suggests that the minima
corresponding to 1/8 [371] and 118I3371 are actually related to faults SF1 and SF3 rather than to APBl and APBll. Thus, it is possible that the true minima of the y-surface are close to the displacernents jc 1/8 [321]but deviating somewhat iii the direction pei-pendicular to the 13511, a surmise that has not been tested in the calculations. The weak-beam TEM and of 1/2 (3311 dislocations suggest (It0 et al., 1996a,b) that they dissociate on the (013) plane according to the reaction l/2(331] = 1/6(331] 1/3(331], which involves APB1. Still, slightly different lengths of Burgers vectors of the partials would not be easily detected, and thus 1/8 (331) and 3/8 (3311 vectors are a possibility. However, the two ab initio calculations differ significa~tlynear the position corresponding to the displacement 1/4 [ 3311. Mitchell et al. find a ~ i n i m u m , indicating a metastable fault with tliis displacement, albeit with energy higher than that with the
+
458
Three 1013) planes of the Cl 11, structure. The notation and the meaning of the rectangle and dashed lines are the same as in Figure 2
displacement 1/8[3517, while Waglimare et al. do not find any extreniui~at the y-surface near this point. A metastable fault with the displacement 1/4 [33t] was also found when using A N itchel ell et al., 2001). It i s seen by inspec of Figure 13 that the displacement 1 /4/331] is very close to the position rapliically equivalent to SF2. Again, it the true niinimum of the y-surface is close to the displacement 11413% ] but deviates soiiiewhat in the direction perpendicular to the [331]. o ~ e as~pointed ~ ~ out , in ikbell et al. ( 2 ~ ~ 1 ) , di~erencesbetween different calculations may arise ing to difrerent types and extents of relaxations. ly the relaxation perpendicular to the fault plane aghmare et al. (1998, 1999a,b),while relaxation including the atomic displacements parallel to the fault plane was carried out by Mitchell et al. (~00~). Besides the faults discussed above, another type of plan s observed oii the (001) plane in MoSi2 and e i al., 1995; Ito et al., 1997a). This fault i s of the Frank-type, with the displacement vector ~~rpendicular to the plane of the fault, and it arises by removal of silicon atomic layers parallel to (001).
3. As mentioned earlier, the C11b structure can be coiisidered as an . . . A . . . stacking of { 110) atomic
planes, the structure of which is shown in Figure 12, Similarly, the hexagonal C40 (e.g. NbSiz, TaSi2, CrSiz, VSiz and the ternary di-silicide Mo(Si,A1)2) and the orthorhonibic C54 (e.g. TiSi2) structures can both be regarded as stackings of the same type of planes. The C40 structure is formed by an . . . ABCABC . . . stacking, and these planes become basal planes of the (0001) type (see Figure Ilb). The 654 structure is formed by a n . . . A . . . stacking, and these planes become base e (001) type (see Figure 1Ic).4 Atomic positions in three consecutive (0001) planes of the C40 structure are shown in Figure 14 using the notation defined at the beginning of Section 3, and, similarly, four consecutive (001) planes of the C54 The 2D-plane group and the s~mnietryis n the case of the ideal ratio b/a = &* However, 111 the latter case the 2 0 plane group i s only sectional~l~yer grou C54, respectively. Again, based on co~siderations introduced at the beginning of Section 3, five types 4Another group of transition metal di-silbdes (CoSi2, NiSiz) has the C1 (cF12) structure which 1s similar to the fcc lattice. Stackiii~-fault-li~e defects can occur m this structure on all three highly symmetric planes and the energy was evaluated for SF on { 001 } from high-resolution observations in the electron microscope to be only 17 3~4 mJ m-* (Suzuki and Takeuchi, 1993).
stack^^^- Fault- Type ~ n t e ~ f a c e s
Three (0001) planes of the G40 structure. This hexagonal structure is fornied by the ...ABCABC... stacking (largc, medium and small circles). The notation and the nieaning of the rectangle and dashed lines are the same as in Figure 2
of metastable planar faults can be anticipated on (0001) planes in the C40 structure. These are one AP two types of SISF and two types of CSF, as marked in faults of the same type can be structure, as shown in Figure 15. irror symmetry planes in either the C40 or the C54 structure, there are no extrema of the corresponding y-surfaces dictated by try. There15 may be fore, planar faults defined in Figures metastable but need not be, and different faults may exist in different materials crystallizing in the C40 or C54 structures. In C40 di-silicides the slip system is (0001) (1210) (Umakoshi et al., 1994; Inui et al., 1997, 1998b), and dislocations have been found to dissociate according to the reaction 1/3 (1210) = 116 (1210) + 1/6 (1210) with an SISF between the partials in a number of disilicides. The energy of the SISF was evaluated in
several cases from weak beam observations. It was found to be 360 mJm-2 in VSi2, 246 ~ n J r n ' ~in TaSi2 (Inui et aE., 1998b) and 397 m J n ~ -in~ et al., 1997). In the ternary alloys ( M o * . ~ ~ ~ b ~ . * ~ ) ~ i 2 ? (~0O.~7CrO.*3)§i2 and ~ o ( S i * . ~ 7 ~ 1 * , *the 3 ~ 2energies respectively were found to be 315,297 and 281 ~nJm--~, 7V~ (Ishilsawa et al., 1998), and in ( ~ 0 * . ~ 03)§i2 321 rnJrnF2 (Inui et al., 2000). These data suggest that the SISF energies in ~ o ~ i 2 - b ternary a ~ ~ d alloys with the C40 structure are lower than the energy of the SISF on (1 10) in MoSi2 with the C11b structure (It0 et aE., 1995). A similar dissociation was observed in which also crystallizes in the C40 structure, but the basal slip is associated witli a synchroshear mec~anism spanning more than one (0001) plane ( h i et al., 1998a); similar behavior was observed for CrSiz (Iiiui et al., 1997). In short, the dislocation s p l i t t i ~then ~
460
Mechanical Proper ties
Four (001) planes of the CS4 structure, This structure is fornied by the ...ABCDABCD... stacking (very large, large. medium and small circles). The notation and the meaning of the rectangle and dashed lilies are the same as in Figure 2
involves formation of multi-layer faults on (0001) planes. The only more detailed investigation of the deformation of a C54 silicide was made for TiSi2 (Takeuchi et al., 1994). The slip systems vary for different orientations and the observed systems are (001) [110], (510) 11301 and (071) [Oll]. However, no infor~ationis available on the structure of dislocations, and the reason for the occurrence of different slip systems is not known. It is likely to be related to complex dislocation core structures.
s are also found in ii~termetalliccompounds with more complex structures than those discussed so far, for example, in A15Ti3 (tP32), which is a constituent phase in alu~inum-richy-TiAl (Nakano et al., 1996;
Nakano et al., 1998; Nakano et al., 1999b). Since sy ~ m e tyr) elementary cells of r-Al?Ti, h-Al2Ti ( and A15Ti3 ( symmetry) phases are multiples of the eleiiientnry cell of y-TiAl, the lattice dislocatioiis ted in the Llo structure become partia~s~ o n ~ ~ c by APBs in the structures with larger elementary cells (Nakano et al., 1999a). It was observed that the width of paired dislocations on the (001) cross-slip plane of the parent fcc lattice is much wider than on the (111) glide plane in Al5Ti3, indicating a strong anisotropy of the APB energy in this structure (Nakano et al., 1999a). In analogy with the hexagonal D019 structure (Umakoshi and Yamaguchi, 1981), various stackingfault-like defects were proposed for the orthorhombic of the Ti?Al~b-type(0structure of ternar phase, space group interactions up to
S t u c ~ iFault~ ~ ~Type - Interfuces
1994). Owing to the low symmetry, the existence of all geometrically possible faults derived froin this model is unlikely in real materials. everth he less, this analysis, which discusses 29 possible APES, two SISFs and two CSFs on various atomic planes, is a useful starting point for investigation of stacking-fault-type defects in this complex structure. issociated dislocations were, indeed, observed in titanium aluminides alloyed with Mb and the energy of the APB on the (001) base plane was estimated as 3 5 0 f 100 mJm-2 and the energy of intrinsic and/or extrinsic stacking-~~ults as 14 & 5 mJm-* (Douiii et al., 1993). Using a similar model, possible APBs and their energies have also been analyzed in the DO, orthorhombic (oP8) structure II symmetry) of Ni3Mb (Hagihara et al., 2000). Stacking sequences similar to those of close-packed atomic planes in the fcc and hcp lattices are found in the C15 (cF24) and C14 (hP12) Laves phases, and thus siinilar stacking faults can be expected in these more complicated structures. The 112 (110) dislocations are split into two 116 (1 12> Shockley partials, as in a fcc lattice, and from the analysis of triple nodes the stacking-fault energy in the C 15 Laves phase of NbCr2 was determined to be 8 i n J n ~ by - ~Uosbida et al. (1995) and 25 mJmW2 by Kazantzis et al. {1996). Still, although crystallography of binary and ternary Laves phases has been investigated extensively, very little is known about planar defects and dislocations, and in general, about the mechanical behavior of these materials (Livingston 1994). Grown-in planar fiiults and dissociated dislocations containing a fault with the 112 (100) displace~ent vector were also observed in the A15 (cP8) structure of the binary Nb3A.1 and ternary Nb55V20A125 alloys (Smith et al., 1993, 1994).
In previous sections the structure and energy of stacking-fault-li~edefects have been treated as essentially independent of temperature. However, changes of temperature may affect signi~cantlythe structure and properties of these defects. The influence of temperature originates in entropic effects and it may be manifested by segregation o f certain species to the planar faults and by local structural changes andlor disordering that may lead to spatial de-localization of the fault and, effectively, to ail increase of the thickness of the faulted region.
46 1
While the APB energy can be lowered by local atomic rearrangemeiits, the most important thermal effect i s often chemical. position can change at t impurities to stacking faults in fcc metals i s the wellknown Suzuki (1952) effect. Naturally, such segregation can also occur at all types of stacking-faul~-like defects in intermetallics. These effects have been already discussed by Sun (1995) in Volume 1. New experi have been presented where they are comp observations on N&Fe (Korner and S and with tlie theoretical calculations of cxl. (1992), who also investigated ch configurational entropy associated with disordering. These results are presented in decrease of the energy of the FeligA12gCr4 can be expected approaches the order-disorder transition temperature T,. However, the decrease is still steeper in cubic trialuminides. A similar effect of temperature on dislocation dissociation has been observed in in situ annealing experiments in the electron microscope (Morris, 1991). At room temperature the superdislocations in A15Ti2Fe (Llz structure) were just visibly dissociated on { 11I } planes into supei-partials linked by the APB. However, upon annealing for minutes at temperatures as low as 5'73
Figure 16 Variation of the APB energy with teniperature in Ni,Fe (Korner and Schoeck 19901, trialurninides (Morris 1992) with the L12 structure, Fe,,AI,,Cr, (Morris and Leboeuf, 1994) with the D03 structure; data for the €32 structure were obtained by theoretical calculations (Bcauchamp et al., 1992). (Reproduced by permission of Taylor & Francis Ltd, from Morris and Leboeuf, 1994)
462
~ ~ ~ h aProperties ~ i ~ a l
increased visibly to about 4 nin, and at increasing temperatures the width of the APB ribbon expanded steadily to 7 nm at 773 K and up to 20-30 nm at 973 . The increase o f the width of the dislocation splitting obviously results from a significant decrease of the APB energy, which most likely is the conse~uenceof local disordering. However, the change of the width of splitting was not recovered upon cooling. Furthermore, edge and screw dislocations behaved differently, The width of splitting of edges increased gradually with annea~ingteniperaLure, while that o f screws started to expand only above 773 K. Thereafter the spacing of supe~partialswas practica~lyin~ependentof dislocation character, This difference between edge and screw dislocations is most likely related to a dependence of dislocation mobility on dislocation character due to signi~cantdifferences in their core structures. The variation of the thickness o f grown-in A with teinperature in the rapidly solidified Ni,Al been recently investigated by erez et al. (1999). Sainples of Ni75A123Cu2 were annealed for 30 minutes range 700-900 "C. The observed thickness (domain wall thic~ness) 011 the annealing temperature is shown in Figure 17. Since this alloy remains ordered practically up to the melting temperature, the chosen annealing temperature range is about 0.6-0.75 of the homologous tem~erature,i.e. far below the disordering temperature. ~nnealingat 600 "G did not lead to a measurable
change of the AP thickness even after 24 h. In contrast, no APBs ere visible after only 10 minute anneals at temp~ratures higher than 950 "C. The reason is that the d o ~ a i ngrowth was so fast that the domain size that of the grains. The increase of the AP accompaiiied by v behavior reflects a change from the original structure described by a shear vector 1/2(110) to a inore complex form, indicating structural and conipositional changes a~sociatedwith disordering and/or segregation plzeiiomena that have not yet been ideiitified.
ies of
and stacking faults, summarized in e all been determined from the measured separations of the partial or super~partial dislocations that participate in the dislocation dissociations coinprising these faults. However, already the deter~~ination of the separation of the partials from observed images is not a str~ig~tforward task. As first et al. (1969) the positions of weak o f partials are generally different tions. An isotropic correction for the dislocation image shift away froin the position of the partials was derived by Cockayne et al. (1969). This correction, modified by the use of anisotropic ed to measurements of dislocation iduk et al. (1993); this method is evert~ie~ess, in a ~ u m b e ~ of studies this effect was not taken into account and positions of images were identified wit partials; this approach i s d e ~ i g ~ a as te~ 2-7. The most reliable method that links the position of images to the true positions of partials is simulation of the contrast formed in the trans~issionelectron microscope. This i s usually performed using a manybeam approximation in the dyiiainical theory of electron diffraction, which takes into account the elastic anisotropy of the crystal, but the ~ a r ~ i a are ls considered as singular dislocations (Schliublin and ~ t more Stadelinann, 3993). For a critical a s s e s ~ m eand details of the com~utationaiprocedure see Schiiublin (1996). The combination of w observations with image siiiiulation is deiioted as ~
3
Variation of the APB thickness with annealing solidified Ni,AI, (Reproduced by Francis Ltd from Perez et al., 1999)
The true separations of partials are usually smaller than the separation of images and, therefore, Fault energies evaluated directly from the observed images
463
Siackmg-F~~uEtT v ~ Interfaces c
can be systematically uiiderestii~ated (cf. values in Table 2 for Ni74.3Al24.7Ta (Balmc and Sch~ublin,1996) or for Mi76A124 (Hernker and Mills, 1993)). For example, for the dislocation dissociation on the p r i s ~ a t i cplane in I3019 941, the observed width of APB’s was found to be 0% larger than tlie actual separation of the partials (Wiezorek et al., 1995). This diflerence increases with decreasing partial separation and is thus the largest for faults with high energies, such as CSF in the L12 structure or APB in the L10 structure. When the true separation of partials is known, the corresponding fault energy is determined from the equilibriu~condition which implies that the energy is to the separation of partials inversely proportion~~l (Hirtli and Lothe, 1982). In earlier studies the proportionality constant was derived using isotropic elasticity, but the majority of data included in Tables 2 -7 were obtained using anisotropic elasticity. In observations of both split dislocations at high temperatures in sitei, and quenched-in dislocations after high-temperature plastic deformation, the temperature dependence of elastic constants should be taken into account. Moreover, in the elastic calculations relating the separation of the partials aiid the fault energy, the partials are usually regarded as singular Voltera dislocations and the finite width of their cores is ignored. However, 11 was shown by Cockayne and Vitek (1974) and Schoeck (1997), using a Peierls-type model, that the separation of partials with wide cores may be significantly larger than if singular dislocation lines are assumed. Schoeck applied his model to the four-fold dissociation of ( 110) superdislo L12 structure, which involves a ribbon of ribbons of CSF bounded by Shockley partials. This
analysis shows that for a given separation of partials the energy of the CSF may be up to 10% higher than tlie value deduced when singular dislocation lilies are assumed. However, owing to the use of the Peierls model, this treatment of the core eRects is appropriate oiily for planar dislocation cores spread in the slip plane. When the partial dislocations possess nonplanar cores, their separation may deviate from that corresponding to the lowest energy configuration. The reason is that metastable con~gurationswith higher energy may not easily relax to the lowest energy state since the partials are sessile and a large stress is needed to move them from oiie Peierls valley to another. Since a large number of such metastable ~onfigu~ations is possible, the fault energy can only be determined with a high confidence from a statistical distribution of observed widths of the fault (cf. the measurements for the DO19 structure of Ti3Al (Paidar et al., 1994; Legros et al., 199th; Paidar, 1996)). The non-planar cores are a prominent feature even in relatively simple structures such as bcc metals ( uesbery, 1989; Vitek, 1992) and are expected to be common in intermetallic compounds, in particular as the complexity o f the structure increases.
We have analyzed possible stackiiig-fault-type defects, APBs and various forms of stacking faults, in 10 distinct intermetallic structures. The most important findings of this exaiiiination are suinniarized in Table 8. Here we present for each structure studied the planes of the faults and the displace~entvectors that lead to metastable APBs or stacking-Faults.An effective device for such study has been the concept of y-surfaces. If
T Displacement vectors of stac~ing-falilt-li~e defects on selected plaiies in structures discussed 111 Sectlon 3; 1z.c.: noncrystallograph~cd i s p l a c e ~ e vector ~t Structum
C40 (2.54
IjZ(li0) (111) 1/2(101] (111) 1/2(110] [ l l l ) , 1/2(101] (111) 1/3(izio] (0001) 1/3(T2io] (ioio) 1/2(llI) (101) 1/2(111) (IOl), (010) (101) 1/3(001] (110) 1/6(331] (013), 1/6(331] (013) I /3 (i0 101 (0001) 1/3[100] (001)
CSF
SISF
1/6(112) ( I l l } 1/6(121] (111) X/6(121] (111) 1/3(ioio] (_oooi) 11.c. (1010)
1/3(112) (111). 1/6(112] (111) 1/3(121] {til), 1/6(112] (111) 2/3(1010] (0001)
1/12(331] (110) tbree n.c. (013) 1/6(iOIO](OOOI), 1/6(0110] (0001) 1/6[1007 (ooi), i/12(i30] (001)
1/4(111] (110) i/6(i2io] (oooi), 1/6(2110] (0001) 1/2[010] (OOI), 1/4[110] (001)
464
MechaE icaE Proper ties
they can be calculated using reliable descriptions of atomic interactions all possible stacking-fault-type defects and their energies are found. However, even if such a complete study cannot be made, symmetry considerations and assessments of interactions of first nearest neiglibors, similar to hard-sphere models, allow us to anticipate con~gurationsof metastable faults. While the displacement vectors presented in Table 8 have all been deduced from these approximate ts, in cases where calculations of y-surfaces n made, these estimates have been confirmed. This statement is true even for relatively complex structures, such as, for example, Cllb, which was discussed in Section 3.7. Important stacking-fault-type defects that have not been discussed in this chapter and which may also play a very important role in mechanical behavior of materials, and intermetallics in particular, are multilayer faults. Such faults are characterized by welld e ~ n e ddisp~acements in more than one layer and should not be confused with faults of finite thickness discussed in Section 3.10. Calculations of tlie core of dislocations in DO22 and L10 structures et al., 1992; Girshick and Vitek, 1995) hat s~iclrfaults may play a similar role in dislocation dissociations as do sin~le-layer faults. oreoves, these faults may play an important role in the nucleation of twins. Such multilayer faults associated with dislocation splitting have, indeed, been observed in TiAl (Inkson and Humphreys, 1995; Inkson, 1998) and in I3022 Ni3V (Vanderschaeve and Escaig, 1978; Vanderscliaeve et al., 1979). Investigation of multilayer stacking-fault-li~edefects in complex intermetallic compounds will be a further important step in studies of their niechaiiical behavior that will lead to understanding not only of slip but also of twinning which is often an equally important cieforniation mode.
support during his stay at the Kyoto University, where part o f this study was performed.
The authors wo~ild like to thank Professor M. rs. H. lnui and I(. Ito for many valuable discussions of the experii~entaldata presented in this chapter. This research was supported by the US ~ e p a r t ~ ~ of eEnergy, n~ BES Grant. no. DEand by the NSF - International Programs Grant no. ~~T"96-05232.One of the authors (VP) would like to acknowledge the support of the Japan Society for the Promotion of Science for
hamp, P. (1986). Phil.
Ackland, G. J., and Vitck, V (1990). Phys. Rev. B, Anton, D. L. (1995). In rnter~ie~allir ~ ~ ~ ~ p (eds o u J.~ €3.d s Westbrook, and R. L. Fleischer). Wiley, Vol. 2, p. 3. Balk, T. J., Kuniar, M., and Henikcr, IS. 3. (1997). Tn High Tempera~ureOrdered I r i ~ e r ~ n e ~ aAlloys l l ~ c VII (eds C . C. Koch, N. S. Stoloff, C. T. Liu, and A. Waiincr). Pittsburgh, Mater. Res. Soc. Synip. Proc., Vol. p. 641. Baltic, N., and Schaubliii, R,(1996). P l d . Mag. A, '7 Banerjec, D. (1995). In ~ r i t e r ~ e t i ~Compounds /li~ (eds J. H. Westbrook, and R. L. ~ l e i s ~ h e rWiley, j. Vol. 2, p. 91. Basinskt, Z. S., Duesbery, M. S., and Taylor, R. (1971). Cmmd. J. Phys., 49, 21 60. Baskes, M. I. (1999). Mater. Scz. Eng. A, 26 ~eauchamp,P., Dirras, G., and Veyssiere, P. (1992). P h ~ / . Mag. A , 65, 477 Christian, J. W. (1983). Mt;ta/l. Tram. A , 14, 1237. Christian, J. W., and Vitek, V. (1970). Reports Prog. Phys., 31, 307. Chu, F., Thorna, D. J., McClcllan, K., and Peralta, P. (199921). Mater. Sci. Eng. A , 261, 44. Chu, F., Thoma, D. J., McClellan, K., Peralta, P., and He, Y. (1 999b). I n t ~ r ~ e t a l l7, i r 6~1~1.~ 1. L. F.. and Whelan, M. J. (1969). itek, V. (1974). Phys. Stat. Sol. ( b ) ,
Crimp, M. A. (1989). Phil. Gserti, J., Khantha, M., Vitck, V., and Pope, D. P. (1992). Mater. Sri. Eng. A , 352, 95. Darolia, R., Lahrman, D. F., Field, R. D., Dobbs, J. R., Cbang, I!. M., oldm man, E. H., and Konitzer, D. G. (1992). 111 Ordered I n ~ t e r ~ ~ e t a l l i Pc .~~, y ~ ~~ie~taa ll l u r g y and ~ e c h ~ ~B ze h~aav il~ (eds ~ r C. T. Lm, R. W. Cahn, and G. Sauthoff). Dordrecbt, Mluwer, p. 679. Diiniduk, D. M. (1998). r ~ t e r ~ e t ~ l l i6, c .613. ~, Dimiduk, D. M. (1999). Mater. Sci. Eng. A, 263, 281. ., and Williams, L. C .
M. (1993). In NighT e ~ ~ ~ r aOrdered ~ u r e r ~ ~ e r ~ eAlloys ~ ~ lVl (eds ~ c A. Baker, R. Darolia, J. D. ~ h i t t e ~ i b e r and ~ c ~ M. , H. Yoo). Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. p. 317. Douin, .I., and Veyssiere, P (1991). Phil. Mug. A, 6 Dowling, W., Donlon, W., and Allison, J. (1999). In HighTeinperu~ur~ Ordered I ~ t e r ~ ~ e t Alloys u ~ ~ i cVi (eds J. A.
Horton, I. Baker, S. Hanada. R. D. Noebe, and D. S. Schwartz). Pittsburgh, Mater. Res. Soc. Symp. Proc.,
Inui, H., Moriwaki, M., Ando, S., and Yarnaguchi, M. (1997).
Yamaguchi, M. (1998a). Phil. Mag. A, 77, 375. 89). In Dislocations in Solids Inui, H., Moriwaki, M., and Yamaguchi, M. (1998b). Nabarro). Amsterdam, North-Holland, Vol. Intermetallics. A, 723. Duesbery, M. S., and Vitek, V (1998). Actu Mat Inui, H., Nakamoto, T., Ishikawa, K., and Yarnaguchi, M. Ehrnann, J., and Fahnle, M. (1998). Phil. Mag. A, 77, 701. (1999). Mater. Sci.Eng. A , Evans, D. J., Court, S. A., Hazledine, P M., and Fraser, H. L. Inui, H., Ishikawa, K., and Yamaguclii, M. (2000). (1993). Phil. Mag. Lett., 67, 331. In~ermetallics,8, in press. Evaiis, D. J., Schelten Ishikawa, K., Inui, H., and Yarnaguchi, M. (1998). In N.L. (1997). Phil. Advance M~terial~s and Processing (eds M. A. Iinarn, Fang, J., Schulson, E. R. DeNale, S. Hanada, Z. Zhong, and D. W. Lee). TMS, 70, 1013. Vol. 2, p. 455. Farkas, D., Mutasa, B., Vailhe, C., and Ternes, K. (1995). Ito, K., Inui, H., Shirai, Y., and Y a ~ a g u c hM. ~ , (1995). Phil. Modelling Simul. Mater. Sci. Eng., 3, 201. Mag. A , 72, 1075. Farkas, D,?and VaiUle, C. (1993). J . Mafer. Res., Ito, K., Moriwaki, M., Ntlkanioto, T., Finnis, M. W., and Sinclair, J. E. (1984). Phil. Mag. Yamaguchi, M. (1996a). M ~ ~ eSci. r . Ertg. FranGois, A. (1 992). PhD Thesis, Universitk Pierre et Mane Ito, JS., Yano, T., Nakamoto, T., Inui, H., a Curie, Paris. M. (1996b). Intermetullics, 4, S119. Frangois, A., Hug, G., and Vcyssi&re,P. (1992). Pid. Mug. A, Ito, K., Nakarnoto, T., Inui, H., and Yarnaguchi, M. (1997a). 66, 269. In High ~ e ~ p e r a t u Or~€ered re I ~ t e r ~ e t a Alloys l l i ~ VII (eds George, E. P., Horton, J. A., Porter, W. D.? and Schiieibel, C. C. Koch, N. S. StolofT, C. T. Liu. and A. Wan J. H. (1990). J. Mater. lies., 5, 1639. Girsliick, A., and Vitek, V. (1995). In ~ i g h - ~ e ~ ~ e r a t u r e Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. p. 599. Ordered I n t e r ~ e ~ a l l Alloys ic VI (eds J. Horton, I. Baker, R. D. Noebe, and D. S. Schwartz). Pittsburgh, Mater. Ito, K., Yano, T., Nakanioto, T., Moriwaki, M. Res. Soc. Synip. Proc., Vol. 364, p. 145. Yamaguchi, M. (1997b). Prog. M L m . Sci., Guder, S., Bartsch, M., Yarnaguclii, M., and Messerschrnidt, Ito, K., Masuda, K., Shirai, Y., Inui, H., and Yamaguchi, M. U. (1999). Mater. Sci. Eng. A, 261, 139. (1999). Mater. Sci. Eng. A, no, T., and Umakoshi, Y. (2000). Acta Juan, Y. M., and Kaxiras, E. (1996). Phil.Mag. A, 7 Kad, B. K., Veccbio, K. S., and Asaro, R. J. (1995). Phil. Hahn, T, (1996). ~nternational Tables .for Cry~~tallography* Mag. A, 72, 1. Vol. A: Space-~roupS y m n ~ ~ t rDordrecht, y, Kluwer. E. T., and Renteiiberger, C. Hernker, K. J., and Mills, M. J. (1993). PFzil. Mug. A , 68, 305. Hirtli, J. P., and Lothe, J. (1982). Theory of Di.sZocaiions, . (1993). Phys. Rev. Lptt., 7 Oxford, Pergainon. 3752. Holdway, P., and Staton-Bevan, A. E. (1986). J. Mater. Set., Kazantzis, A. V. Aindow, M., and Jones, I. P. (1996). Phil. 21, 2843. Huang, S.-C., and Chesnutt, J. C. (1995). In Intermetallic , and Pope, D. P. (1992). M ~ t e rSci. C o ~ p o ~ (eds ~ d sJ. H. Westbrook, and R. L. Fleischer). Eng. A , 152, 89. Wiley, Vol. 2, p. 73. Kim, Y.-W (1989). J . M e t ~ l ,41, ~ , 24. Hug, G,, Douin, J., and VeyssiGre, P, (1989). In HighKim, Y.-W. (1994). J . Metals, 46, 30. T e ~ p e r a t ~Ordered ~e I n t e r ~ e t a l l iAlloys ~ 111 (eds C. T. Kopsky, V., and Litvin, D. B. (2000). Int~rnation&~l Tablesfor LIU,A. I. Taub, N. S. Stoloff, and C , C. Koch). Pittsburgh, Cry.s~ulE~graphy, Vol. E: S a ~ b ~ e r i Groups, o ~ i ~ Dordrecht, Mater. Res. Soc. Syiiip. Proc., Vol. 133, p. 125. Kl uwer. Inkson, B. J. (1998). Phil.Mag. A , 77, 715. Korner, A. (1988). Phil. Mag. A, Inkson, B. J., and Hurnphreys (199.5). Phil. Mag. Lett., 71, Korner, A., Karnthaler, W. P., a 307. PhiE. Mug. A , 56, 73. Inoue, H. R., Cooper, C. V,, Farrow, L. H., Hanada, Y., and Korner, A., Cockayne, D. J. H., and Sun, Y Q. (1993). Phil. re Wayrnan, C. M. (1991). In ~ i g h - T e ~ p ~ r a t uOrdered Mag. A, 68, 993. ~ n t e r ~ e t a l l Alloys ic IV (eds L. A. Johnson, D. P. Pope, Korner, A., and Schoeck, G. (1990). Phik. Mag. A, 61, 909. and J. 0. Stieglcr). Pittsburgh, Mater. Res. Soc. Syn~p. Kr2, F., Schwaader, P., and Kostorz, G. (1997). Acta Mater., Proc., Vol. 213, p. 493. 5, 67.5. Inui, H., LUZZI,D. E., Porter, W. D., Pope, D. P., Vitek, V., Kruml, T., Vigmer, B., Bonn and Yamaguchi, M. (1992a). PhiE. Mag. A , 65, 245, (1997). Mater. Sci. Eng. A , Inui, H., Oh, M. H., Nakamura, A., and Yamaguclii, M. Krurnl, T., Martin, J. L., and Bonneville, J. (2000). Phil. N a g . (1992b). Phil. Mag. A , 66, 539. A, 80, 1545.
Kumar, K. S. (11995). In ~ n t e r ~ e t u l C l ~ac~ p ( ~ ~(eds n d ~J.s H. Westbrook, and R. L. Fleischer). Wiley, Kumar, M., aiid Hemker, K. J. (1997). Phit 399. umar, M., and ~ e m k c r K. , J. (1998). J . Muter. Res., 13, 610. Legros, M., Courcl, A., and Caillard, D. (1996a). Phil. Mag. A. 73, 61. Legros, M., Couret, A., and Caillard, D. (1996b). Phi/.Mag. A. 73, 81. Lerf, R., and Morris, D. G. (1991). Acta Metull. Mater., 39, 2419. Llu, Y., Allen, S. M., and Livmgston, J. D. (1993). In HighT e ~ ~ p e r ~urt d~~~r ~I~iter~etnllic ed Allays I.' (cds I. Baker, R. Darolia, J. D. ~hittenberger, and M. H. Yoo). Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. 288, p. 203. LIU, C. T., and Pope, D. P (1995). In ~zter~etullic Co~~pounds (eds J. H. Westbrook, and R. L. Fleischer). Wiley, Vol. 2, p. 17. Livmgston, J. D. (1994). In ~ i g h - T e ~ p e r a t ~Sil~cides re and ~ e ~ r u c t aAlloys ~ y (eds C . L. Brzant, J. J. Petrovic, B. P. Bewlay, A. K. Vasuclevan, and H. A. Lipsitt). Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. 322, p. 395. Lo Piccolo, B. (1999). PhD Thesis, Ecole Polytechnique FCdCrale dc Lausanne. Maloy, S. A., Mitchell, T. E., and Heuer, A. H. (1995). Acta
Nabarro, F R. N . (1994). Mater. Scz. Eng. A , Nakano, T.. Matsumota, K., Seiio, T., Q Umakoshi, Y. (1996). Phil. Mug. A, '74, 251. Nakano, T., Negtshi, A., Seno, T., Surnida, M., Y a ~ a ~ o t o , M., and Umakoshi, Y. (1998). Phi/. Mug. Lerf., 7 Naltaiio, T., Hayashi, IS., Ashida, K., and Uniakoshi, U. ( 1999a). In High- Te?~perature Or~eredI ~ i t e r ~ e t a /Alloys /~c VIII (eds E, P George, M. J , Mills, and M, Ya Pittsburgh, Mater. Res. Soc. Symp. Proc., p. KK5.9.1. Nakano, T., Negishi, A. I, K., and Umaksshi, Y. (1999b). Actu Mater., Neuiixznn, F (1885). Vodesungen Cber die Theone der Ela~~ticit~t, Leipzig. Neuniann, P., and Saulho~,G. (1999). I n ~ e r ~ e t u / I i c7,s ,233. Ncveu, G. (1991). PhD Thesis, Unive Paidar, V. (1976). Czech. J. P l ~ y s B, . Paidar, V. (1985). Acta Metall., Paidar, V. (1987). Acta Melall., Paidar, V. (1996). In S ~ ~ i ~ i l i t ~ ~ I?. E. A. Turchi, and J. Kudrnovsky). New York, Plenum, p. 645. Paidar, V., Legros, M., Couret, A. and Gaillard, D. (1994). In P1asticit.y of M c fuls und Al1o.v~(ed. Switzerland, Trans. Tech., Key Eng. Mater., V p. 201. Paidar, V., Inui, H., ICishida, K., and i, M. (1997). Mater Sci. Eng. A , 233, 111. Parthasarathy, T. A., Rao, S. I., and Dimiduk, D. M. (1993). ~ i ~ u lR., ~ aSchaaf, . G., and Trcbin, H. R. (1999). Phi/. Mug. Phil.Mag. A, 67, 643. Perez, J. F., Shang, P., Morris, D. G., and Joncs, I. P. (1999). Minonislir, U. (1991). Phil. Mag. A, 63, 1085. Phil. Mag, A, 79, 179. Miracle, D. B., and Darolia, R. (1995). In I ~ t e r ~ e t a l l ~ cPetrovic, J. J. (1995). Mater. Sci. Eng. A, 89 Petrovic, J. J., and Vasudevan, A. IS. (1999). C o ~ p o ~ n d(eds ' s J. H. Westbrook, and R. L. Fleischer). Wiley, Vol. 2, p. 53. A, 261, 1. Pope, D. P. (1992). In Ordered Inter~eta//ics Physics/ Mitchcll, T. E., Baskes, M. I., Chen, S. P., Hirth, J. P., and Metall~rgyand ~ $ c l ~ a n i eBehavio~r ul (eds C. T. Lsu, R. W. . 6. (2001). Phi/.Mag. A, 81, 1079. Cahn, aiid G. Sauthoff). Dordrecht, Kluwcr, p. 143. Mitchell, T. E., and Misra, A. (1999). Mater Scz. Eng. A, Pope, D. P., and Ezz, S. S. (1984). fnt. Met. Rev., 25, 233. 106. Poschmann, I., Ortel, C.-G., Bcrgmann, 3., and Tobisch, J. Y., and ~ a t a n a bN. ~ ,(1993). Scri~)taMerall. Mater., (1995). Phil.Mug, A, 71, 1083. 139. Rao, S. I., Dimiduk, D. M., and Mendiratta, M. G. (1993). Y amaguchi, M . (1997). Phil. Mug, A, 68, 1295. Rosner, H., Molenat, G., and Nembach, E. (1996). Muter. Morris, D. C. (1991). Scriptu Metal/. Muter., 25, 713. Morris, D. C.(1992). Scripta Metall. Muter., 26, 733. Sauthoff, G. (1989). 2. ~ ~ t a / l k u ~ z d ~ , Morris, D. C., and Cunther, S. (1993). Phil. Mag. A, 68, 81. and Stadclmanii,P. A. (1993). Muter. Sci. Eng. Sch Morris, D. G., and L e b o d , M. (1994). Phi/.Mug. Lett., 70, 29. Morris, D. G. and Lerf, R. (1991). Plzil. Mag. A , 63, 1195. Morris, D. G., Lerf, R., and Leboeuf, M. (1995). In HighSchroll, R., Vitek, V., and Gumbsch, P. (1998). Acta ~ ~ t e r . , ~ e ~ p e r a t u rO e r d e r e ~Intermetall~c ~ Alloys V l (eds J. 6 , 903. aker, R. D. Noebe, and D. S. Schwartz). Shyuc, J., Hou, D. H., Aindow, M., and Fraser, H. (1993). Pittsburgh, Matcr. Res. Soc. Symp. Proc., Vol. 364, ~ a ~ eSci. r . Eng. A , 1 7 ~ 1. , p. 1215. Sirnmons, J. P., Rao, S. I., and DimidL~k,D. M. (1997). P~il. Mryasov, 0. N., Gorno N., and Freeinan, A. J. Mag. A , 75, 1299. (1998). Phys. Rev. B, Singh, A. K., Nandy, T. E=,, Mukerjee, D., and Baneijee, D. Mryasov, 0. N., and Fre . (1999). Mater Sci.Eng. (1994). Phil. Mag. A, 69, 701. ~
Smith, L. S., Cheng, T. T., and Amdow, M. (1993). In HighTemperature Ordered Intermetallic Alloys V (eds A. Baker, R. Darolia, J. D. Whittenberger, and M. H. Yoo). Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. p. 263. Smith, L. S., Aindow, M., and Loretto, M. M. (1994). In High- Temperat~re Silici~l~s and R~fractoryAlloys (eds C . L. Briant, J. J. Petrovic, B. P. Bewlay, A. I(. Vasudevan, and tsburgh, Mater. Res. Soc. Syrnp. Proc. Sparks, C. J., Porter J. H., Oliver, W. C., and Golec, C. 6. (1991). In Alloy Phase Stabi1it”v avzd Des7gn (eds G . M. Stocks, D. P Pope, and A. G. Giamei). Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. X p. 175. Stoloff, N.S. (1998). Mater. Scr. Eng. A, Stucke, M. A., Vasudevan, V. K., and Dimiduk, D. M. (1995). Mater. SCE.Eng. A, ~ 9 2 / ~ 9 111. 3, Sun, J., Lee, C. S., Lai, J. K. L., and Wu, J. S. (1999). Intermetallics, 7,1329. Sun, Y. Q. (1995). In ~Pzter~~etallic ~ o ~ p o ~ (eds 7 ~ dJ.s H. Westbrook, and R. L. Fleischer). Wiley, Vol. 1, p. 495. Suzuki, H. (1952). Sci. Rep. Rcs. Inst. Tahoku Univ
Takeuchi, S., H
Vedula, K. (1995). In Intermetulli~ Comp~?u~d~s (eds J. H. Westbrook, and R. L. F1 Veyssikre, P. (1998). IPzterme Veyssikre, P,, and Noebe, R. Vitek, V (1968). Phil. Mag. Vitek, V. (1992). Prog. Mater. Vitek, V., Pope, D. P., and Bassani, J. L. (1996). In ~islocutiovz~s in Solids (eds F R. N.Nabarro, and M. S. Duesbery). Amstcrdani: North-Holland, VoI. G., and Neurnann, I?. as, E., Rulatov, V., and Duesbery, M. S. (1998). ~ ~ ~ ~ ~ eSirnul. l I ~ P mate^. z g SCE.Erg., 4,493. Waghinare, U. V., Bulatov, V., Kaxiras, E., and Duesbery, M. S. (1999a). Phil. Mag. A , 79, 655.
Wiezorek, J. M. K., and Wuniphrcys, C. J. (1995). Scrijstu Metall. Mater,, 33, 451. Woodward, C . , MacLaren, J. M., and Rao, S. (1992). J . M a t ~ r Res., . 7, 1735. Xu, W., and Moriarty, J. A. (1998). Cotnp. Mater. Sci., Yarnaguchi, M., Pope, D. P., Vitek, V., and Urnako (1981a). Phil. Mag. A, Yamaguchi, M., Vitek, V., and Pope, D. P, (1981b). Phil.
~rnakoshi,Y., Sakagarni. Y., Hirano, T., and Yarnane, T. (1990). Actu Metall., 38, 909. Urnakoshi, Y., Nakashirna, T., Nakano, T., and Yanagisawa, E. (1994). In H i g ~ - T e ~ ~ ~S~licides ~ a t ~ ravzd e R ~ ~ r a c ~ o r y Y a ~ ~ g u c h M., i , and Inui, H. (1995). In I ~ t e r ~ ~ e t ~ i l l i c Alloys (eds C , L. Briant, J. J. Petrovic, B. P. Bewlay, A. K. C o m ~ o ~ (eds n ~ sJ. H. Westbrook, and R. L. Fleischer). Vasudevan, and H. A. Pittsburgh, Mater. Res. Wiley, Vol. 2, p. 147. Soc. Synlp. Proc., Vol. Yamaguchi, M., Inui, H., and Ito, E;.(2000). Acta Mate Urnakoshi, Y., and Yarna (1 98 1). Phys. Stat. Sol. 307. Yoshida, M., and Takasugi, T. (1992). I J 1 d M ~A , ~4 , Vailhe, C., and Farkas, D. (1997a). Acta Mutcr., Yoshida, M., and Takasug, T. (1994). In S t r e n ~ t ~of’ z Vailhe, C., and Farkas, D. (1997b). J . Mater. Res., 12, 2559. Materi~ls(eds €3. Oikawa, K. Maruyarna, S. Takeuchi, Vain~htein,B. K. (1981). Modern crysta~lograph~y I, Symmetry and M. Yamaguchi). Sendai, Japan Institute of Metals, qf c~?/sta~,r, i!!f@fhods of’ s t ~ i ~ tCi”~tsl.llogr~l/7,, ~ r ~ l Berlin, p. 357. Springer. Yoshida, M., Takasugi, T., and Hanada, S. (1995). In ~ i g l z ” Van Der Wegen, 6.J. L., Bronsve T e ~ p e ~ u t ~Ordered re Inter~etallic Alloys VI (eds J. J. T. M. (1983). Phil. Mag. A, Horton, I. Baker, R. D. Noebe, and D. S. Schwartz)., Vanderschaeve, G., and Escaig, B. (1978). f. Phys. France, 39, Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. 36 L74. p. 1395. Vaiiderschaeve, G., Sarrazin, T., and Escaig, B. (1979). Acia Yu, H. F., Jones, 1. P., and ~ i n a l ~ i ~R. a n E,. , (1994). Phil. Metall., 27, 1251. Mug. A, 70, 951. Vasudevan, V. K., Wheeler, R., and Fraser, H. L. (1989). In Znarn, S. (2000). PhD Thesis, University of Pennsylvania. Higl~-Temp~ratureOrdered ~~termetullic Alloys III (eds Znam, S., Nguyen- an^, D., Pettifor, D. G., and Vitek, V. C . T. Lm, A. I. Taub, N. S. Staloff, and C. C. Koch). (2000). In Properties of Conzplex Ivzorgunic Solids-II (eds Pittsburgh, Mater. Res. Soc. Symp. Proc., Vol. 133, A.-M. Meike, A. Gonis, P. E. A. Turchi, and K. Rajan). p. 705. New York, Kluwer Acadeinic/Plenu~,p. 479.
This Page Intentionally Left Blank
ractice is everything.
~ n e e r isi ~the ~ art of doing that well with one dollar, h any bungler can do with two after a fashion.
v ~ r y t h i nthat ~ can be i ~ v e n t has e ~ been inv~nted.
S
ht State ~ ~ i v e ~ s iayt t y ton, , OH, ~S~~
uctio Why has the commercialization of intermetallic alloys been so slow and difficult? This question will be examined in this chapter, concentrating on the potential use of such alloys in high performance systems, specifically propulsion and power generation. Much development activity has centered on such applications, because the greatest potential benefits are available in these systems; therefore, they should offer the greatest incentive for use. Future growth of turbine and automotive engines depends upon the cont~nued development of high-temperature, lightweight materials. For example, the operating efficiency of a gas-turbine engine will increase by over 1% for every 10°C increase in the turbine-inlet gas temperature. Substantial fuel savings in comrnercial aircraft and power generation can be achieved by using new materials which can provide this temperature increase - a major incentive in these times of increasing fuel prices. As electric power demand is also growing rapidly, it would be advantageous to design steam turbines larger than those on-line today, but to do so requires considerably stronger and lighter disk, blade and vane materials. An example of the influence of materials on the design of highly efficient, smaller automotive engines, is a need for lighter-weight valve materials that can operate at higher rotating speeds
without efficiency loss due to valve float. In all of these examples it has been suggested that intermetallic alloys could be part of the solution. The reason is illustrated in Figure 1, which shows the strength limits of the metallic alloys currently used for construction. These are relatively mature materials, and it i s generally agreed that further improvements will not come easily. Also in the figure are the strengths of a number of intermetallic alloys, and from these data it can be seen that there does appear to be growth potential. Two things must be remembered in making such comparisons. The current materials of construction exhibit a balanced set of properties, and strength is only one of these; but, in addition, inter~etallicalloys are not simple extensions of conventional alloys. In particular, especially at lower temperatures, tensile ductility tends to be marginal and toughness and fracture characteristics are not attractive. So while each class o f intermetallic alloys evaluated over the past years has some attractive properties, it has proved very d i ~ c u l t to achieve all the engineering requirements of current design practices. All devices that operate at high t e i ~ ~ e r a t udemand re rather special materials with an extensive set of design requirements. Strength and stiffness are just two of these, other characteristics include creep and fatigue resistance coupled with good oxidation and corrosion properties. Materials with low density are especially
*See List of Contributors €or current address. ~ n ~ ~ ~ r nCompounds: e ~ ~ l l iVol. ~ ~ 3, Pnnc@les and Practice. Edited by J. H. Westbrook and R. L. Fleisclier. 0 2 0 0 2 John Wiley & Sons, Ltd.
472
Applications
F i 1 ~§peci~c"strength ~ ~ ~ limits of structural alloys: (a) alloys used today; and (b) with. selected iiitermetallic alloys superimposed. Data for wrought TiAI from Kin1 and Dimiduk (1997b) and Chen et aE. (1999)
A frequently used measure of aircraft turbine-engine efficiency, especially in military systems, is the thrust-to-weight ratio. Thus, research and development strategies often include material elements that lead to increased thrust (through higher-te~peratLirealloys) and to decreased engine weight (with low-density materials). In a number of cases, described in Section 3 , alloys based on TiAl appear very attractive based on these criteria, but design uncertainties coupled with projected cost increases cloud the application picture. Overall affordability is tied to a myriad of Factors including acquisition cost, which will include manufacturing issues such as part yield and quality control expense, together with in-service outlays for repairs and field maintenance. These issues, as we shall see, present some special challenges for intermetallic compounds. Life-cycle cost and reliability are key char~cteristics for power-generation equipment. generators operate for years without shutdown, system reliability is essential and higher initial cost to achieve this may be acceptable. The substitution of c o ~ v e n t ~ o ntitanium a~ alloys for steel to improve corrosion fatigue performance is an example. The introduction of intermetallic alloys into such wellestablished systems may be difficult, but the emerging trend toward localized power generation using microturbines ( < 1 MW) could be a better opportunity. Using today's technologies, mi~roturbineshave d i ~ c u l t ycompeting economically with the utility grid. The industry goal is to increase the overall efficiency to 40%; here it is possible that intermetallic alloys could play a role. The timing is right for such new systems, but again, reliability together with long-term stability and cost will be key issues. Advanced automotive performance goals show some similarities. Here agdin there is emphasis on lighterweight materials development, exemplified by the rapid alloys, but material cost increasein the use of a l u ~ i n u m is of paramount importance. So with the possible exception of government-mandated regulatory changes, higher-cost materials will not be incorporated. In addition to the intrinsic part cost, manu~actu~ng yield or assembly problems cannot be tolerated as these could jeopardize high output and stringent delivery schedulerequirements. In spite of this there has been more progress in applying ~ n t ~ ~ i i e t a i l i c alloys in this industry than in either of the others.
attractive as system weight and inertial effects can be innportant, and this makes some of the alLii~ii~um-rich intermetal~ccompoun~sespecially attractive. In-service conditions can also provide additional challenges including the ingestion of debris, etc. Thus, the ability to absorb damage without compromising safety is another crucial characteristic. The introduction of a new material for use in structural components can be a complex process. In the present business climate there must be a quantifiable benefit from the outset. Often this is associated with a performance improvement based 011 increased properties, but cost reduction and reliability increases are becoming equally important drivers. In practice, developmeill o f a new material usually begins in earnest because a property improvement, often based on preliminary data, and a system requirement that can exploit this advantage, lead to a product 'pull'. Part of the explanation for the automotive lead Some illustrative examples are as follows: lies in the different reliability~ris~ relationships and
473 associated development practice in the three industrial sectors. A new material cannot be introduced into aero-propulsion systems until a very stringent series of tests and analyses have been completed includ~ng This section will examine the requirements for the specific ~eq~Lirements set by regulatory agencies. This certification of new structural materials in the selected is obviously because any significant level of risk is systems, designated in the Introduction, and in doing u ~ a c c e p ~ a b land e ~ the liability implications of a crash so identify some of the special c ~ ~ l l e n g efacin s and loss of life due to equipment malfunction can be intermetallic alloys. It is not easy to strike the correct inhibiting. The risk to human life is small for the balance between the complexity and expense of the failure of a power-generating system; however, society qualification process. There is no doubt that the alloy is so dependent upon a constant supply of electricity certification process i s long, difficult, and in many cases that repeated shutdowns could cause severe economic incomplete for intermetallic alloys. On the other haiid, disruption. Thus, from this performance perspective, the fact that intermetallic components have been the properties and long-term stability of a new successfully tested in engines is an indication of material, under conditions that simulate the extreiiies progress. Another complicatin~ issue is that the of turbine operation must be known precisely; and as a sequence of events that lead to certification are result, long and extensive testing is required. In interconnected, and therefore separation into discrete automotive systems the risk to human life is small if steps is at least an oversimplification and can lead to an engine Fails, and this is reflected in much shorter problems. The approach here is to start with material testing programs and considerable in-service testing of ~evelopmentstrate~iesfor intermetallic systems, examcandidate materials. Here, since many thousands of iiie process options, look at properties and units are manufactured, the major concern is a recall, manufacturii~gissues, and then touch on some design to repair or replace defective components, which can factors. In the current business climate it i s not be extremely expensive and time consuming. possible to escape the a~ordabilitycriteria which can Williams (1997) has discussed business and cultural override the most compelling engineering benefit case. issues that have certainly in~uencedthe introduction of int~rmetallicalloys, especially in aerospace systems. There is no doubt the adventurous atmosphere of the 1960s and 1970s, has been superseded by the more Over the years structural development of materials has conservative environment of the 1990s. Thirty years been semi-empirical, serially execu ago, resources seemed unlimited and the drive to occurred in a series af logical steps. ‘lighter, higher, faster’ to a great extent overwhelmed of compromises are made over time in the context of other considerations. Now the pendulum has swung to emerging data and inventions - a process that passes the other extreme. Programs today have to be justified for optimization. There are signs of change, for in excruciating detail, the iinpact assessed every step of example, alloys developed under the now defunct the way, and overcome extremely challenging financial High Speed Civil Transport program combined hurdles. However, the situation is not intractable; but a statistical methods with modeling to achieve quite more disciplined and focused approach than the current substantial progress, ~igli-strengtlisteel developiiient fra~mentedefforts is needed, If the case for intermetallic has taken modeling and o~timizationeven further in alloys can be expressed in quantitative, credible terms the design of alloys, with impressive results (Olson, and remains attractive when compared with conven1997). (See also the chapter by Nalca and Kahn in this tional materials using the applicable trade-off factors, volume.) These efforts illustrate the significance of a progress can be made. To do so will require the current, §ystems-engineering approach to new materials, one more limited, resources to be focused on the key areas to which recognizes the full spectrum of i~terdepen~ent move as rapidly as possible to the appli~ation/produc- requirements in the process of Optimization. As a tion stage; after 20 or more years of effort, it becomes general rule, the materials science profession teaches a ever more difficult to maintain momentum. Any new linear relationship between synthesis, structure, propthrust must be more co-operative than in the past; the erties, and use. This relationship is not well suited to resources of both gaver~mentand industry must be the need for an interactive, linked approach that combined to advantage. ~e will expand on the design, demands compromises in mechanical design and engineering and econoimc factors needed to establish material optimization. The problem is exacerbated in credibility in the next section. that structure-property relationsh~~s are often given
474
A ~ p l iions ~~t
too great an e~pliasisrelative to processing-structure, and especially property-use relationships. Such linear thinking is also not compatible with the world of busi~esswhere cost and reliability domi~iatedecision making. Intermetallic alloy development has tended to follow the standard process in which the first focus is on problem properties. Thus, the search for solutions to iniprove tlie typically poor ductility and fracture characteristics has dominated alloy development efforts, and has tended to obscure the search for optimized or useful balances of properties. Further, the ever-present ~ ~ o m i sofe improvement has distracted desi~nengin~ersfrom interpreting non-coiiventional property balances and defining the compromises that may lead to successful engineeriiig applications. ince ductility and fracture are still major design concerns, it is worthwhi~eto examine some general features of the current situation. There is a sinall n u i ~ b e rof i~it~rmeta~lics that are ductile at ambient ples include Zr,Al (cP4, LIJ. aiid TI,AlNb (oC16, Ela), and brittle-to-ductile transitions at interniediate ~e~peratLires. ~nfortunately,many also show pronounced strain-rate sensitivity that can delay ility to unacceptably high temperaoading rates. Single crystals of some ductility in certain orientations, but behavior is often anisotropic; early hopes of exploiting specific orientations aluminides seem to have been abandone tropy does open the potential for utilizin red material in thin sectioiis; this has been demonstrated in Ti3A1 (bP8, alloys, where ductilities of 30% have been ed in an otherwise marginally ductile material. In general, there appear to be two discernible trends in the ductility of iiiter~etallicsystems. The alloys cited above are single-phase materials and are too weak to be considered for structural application; therefore, more complex alloying strategies have to be employed to achieve strengths needed for engineering exploitation. Alloying additions to base compounds, within the solid-solutioi~range, can improve ductility and also increase strength; Nb in Ti,AI, several transition 1 (tP2, LI,) aiid Zr in Ni,A1 (cP4, J are exam~les,but adding $ufficientconcentratioii form a second phase has proven the most fruitful approacli. Thus, the formation of the beta phase in 1 alloys and the two-phase Ti,Al and TiAl mixtures could be coiisidered the most “successful’ intermetallic alloys. Of course, the distribution of the pha~esis extre~elyi ~ ~ o r t aand n t we sliall cover some
of the specifics in the following sections for specific compounds. In general, the fracture characteristics of this class of alloys follow the ductility trends, although there is at least one instance of fracture and ductility showing different trends; see Section 3. A number of points should be made in respect to fracture c~aracteristics.As noted above, under high strain rates, i.e. impact conditions, virtually all intermetallic alloys have rather poor properties even at intermediate (service) temperatures; impact strength of a joule or two in a Gharpy test does not breed design confidence. This can be an important chmacteristic if containment capability is required should a problem failure) be encoLintered during system operation and part fragments liberated. Aircraft systems are especially ~ l n e r a b l to e such events since an engiiie-case penetration can be dangerous, but such requirements exist at some level for all dynamic environments. Environmental effects can also override attractive intniisic properties; the rapid decay of fracture strength in -doped Ni,Al at temperatures of 550°C in humid air is a well-documented example. The bottom line is that, although extensive investigations have resulted in some improve~entsin ductility and fracture cliaracteristics, it has proven vii-lually impossible to match those of co~veiitionalmetallic alloys. In many cases improvements come at the expense of the attractive high-teiilperature properties of thc intermetallic alloys, a result wlich only re~emphasi~es If one the need for effective o p t i ~ ~ ~ a t procedures. ion looks at tlie current state of the art, an inesca~able conclusion is that, although it is possible to sift through all the data and id~ntifysome quite attractive general characteristics, to ~ ~ p all ~ the ~ r desired e en~i~eer~~~ propertips in oyle d o y has proven elusive, This explains to some extent the continued search for thc ‘best alloy’, and continues to delay scale-up decisions. It has been stated in a recent review ~ i l l i a i n s ~ 1997) that ftirtlier improvements may not be possible, and a better approacli might be to attempt to exploit what we have in hand - and in the absence of o p t ~ i ~ ~ a tmethods, ion this suggestion may have some merit.
The evolution and control of microst~cturet h r o ~ g hthe specific process cycle chosen is key to the properties and perforniance of a component. In addition, the specific features of this cycle also play a major role in determining the cost of parts. In general, the methods used to produce and shape internietallic alloys are the same 11s those used to produce conventional alloys, although special
precautions are sometinies needed to guard against cracking, Therefore, an encouraging aspect of the processing situation is that existing infrastructure can be used to make forgings, mill products and castings, as can be seen from the following examples. Melting of intermetallic alloys has progressed substantially in the past 20 years: early ingots spoiitaneously shattered during cooling. Today, sound ingots weighing over 500 kg (and larger) can be produced by a variety of production processes. solidification is not extreme and can usually be handled by relatively standard, thermal homogeni~ationprocedures; but as larger ingots are produced, the situation should be reviewed. Subsequent conversion to billet i s most readily accomplished by high-temperature extrusion, nearly always performed on encapsulated, i.e. canned, ingots to minimize surfiice distress and thermal shock. Use of the more conventional forging methods to produce billets usually leads to unacceptable cracking, but such methods have been used successfully on Ni,AI alloys. Xsotherrnal forging of components has been possible for a number of years, but the desire to forge at higher temperature does represent a major challenge to both eqiripmeiit capability aiid die life. ill-product development has also shown impressive progress; and the present ability to produce uniform thin TiAl sheet, which scemed an impossible dream 10 years ago, is a tribute to the ingenuity, expertise, and investments of the Plansee ~ o m p a n y (Clernens et al., 1999 and chapter by Appel t7t al. in this volume). Xiitermetallic alloy castings have a longer history and have been the most widely (See also the chapter by volume.) Xn general, cas although inold niodification is often required to ininimize stresses during cooling; and elimination of challenging design details, such as sharp notches and rapid thickness changes, is desirable. But there are also a number of other di~culties.Although investment castiiigs are usually coiisidered to be ‘iiet-shape’ products, deformation sizing ensure blueprint compliance. usually performed at room te pose real problems for low-ductility materials. The substitution of hot sizing adds both process compl~xity and cost to the product. In addition, castings are often repaired by welding, again not a straightforward process for most inter~etallicalloys. ~owder-metallurgical ( /M) processing of intermetallic alloys eliminates some of the melting and casting problems and can also improve niaterial-
utilization Factors. ( ee also the chapter by man and Seiniatin in this volume.) This process has been used to produce sniall parts, and there is some evidence that improved properties can be attained in P/M products. However the lack of a robust production base and associated cost uncertainties, coupled with quality concerns, have slowed development. The understanding and control of structural factors such as grain size, phase distributions and overall ho~ogeneitythrough melting and wo~kingoperations have shown progress. Process simulation and mapping have begun for some intermetallic systems, most notably TiAl, where features of the recrystallization behavior and strain limits for fracture during hot working have been documcnted. These are in a form that could be integrated into a broader set of models for component production, including the prediction of design-property levels. (See also the chapter by and Khan.) Another interesting issue is the pot for designing alloys for process capability. For example, if forging in the alpha-phase field is a path to superior properties for TiAl-base alloys, then serious consideration should be given to designing an alloy to lowcr the required forging temperature. The situation for secondary processing is not as encouraging, especially for low-ductility ~ystei~is. Standard machining practices used for typical engineering alloys will often result in chipping and sometimes cracking of parts. odified methods have been developed, but tend to be slower and more expensive, and insufficient data have been developed to confirm total reliability. Fusion welding, used extensively for repairs and in the building of subassemblies, can be accomplished, but special equipment to incorporate high levcls of preheat iii an inert atmosphere, with complex han~ling systenis, is usually required and adds to the processing costs. Two messages should be extracted from the above discussion. First, complete integration of the processing parameters into c o ~ ~ o s i t i o n a l and microstructural design is needed, iiot only to maximize the utility of the material, but also framework for production control. non-standard processing leads to pric contributes negatively to the eneral a~ordability equation, another issue to ~ h i c hwe will return.
2.3
es
What are the factors that determine if viable components can be designed from a new material’? Foremost is the need for data on test samples of a single alloy
476
~ p ~ l ~ions cat
composition ~ a n u ~ a c t u r eind s u ~ c i e n quantity t by a controlled process cycle to give information on quality standards, and from which enough test information has been generated to establish estimated minimuni properties. In addition, a credible engineering benefit case is required that is coupled with a rigorous cost an established econoniic criterion. The subsequent steps are dependent on application and industry: we shall use the procedures that prevail for aerospace applications as examp~es.Scale-up and product certification include a number of interconnected activities. Setting material-specification limits is usually performed on ~~termediate-size heats in which compositiona~and processiiig parameters are varied systeinatically (~t~tistically) to define limits. Full-scale heats are then produced to confirm processability and establish product variability; such information is su~sequentlyused to build process-control documentation. The material from these heats is used to set the stati~tically-basedproperty minima (design allowables) that can be used for final designs. Material perforniaiice i s fLirther calibrated by rig testing of selected coi~ponents,for example, the demonstration of cyclic and overspeed capability. It is important to emphasize that the tests must be conducted on material taken
destructive inspection procedures are usually finalized at this stage. Note that no ~ n t e r ~ e t a l lalloy i ~ has achieved this level qf ~ u t ~ r i t ~ . As current design practice is based on the characteristics of conventional metallic alloys, the ductility/ fracture characteristics of interinetallic alloys set some special design challenges. These have not been totally r e s o ~ v eand ~ roba ably will not be, until there is consi~erablymore component- and engine-test experience. Three situatioiis could give grounds for concern: the size, nature and frequency of exogenous defects produced by the manufacturing processes, damage introduced by handling in shop environments, and inservice events that could also produce damage. Results on fracture behavior have been mixed: tests on TiAl alloy compressor blades in a military test engine demonstrated that quite severe rubs (against an abradable seal) could be absorbed successfully, but that at least one type ofimpact event could not. The test run in 1992 by the General Electric Company on CF6 low~pressureturbine blades appeared very successful at has been limited follow-up. In the , B2) alloys, even though turbineengine vanes were successfullyengine tested, low impact
tolerance at the projected operating temperature appears to have limited further pursuit of these materials for turbine blades (Pope and Experience indicates that the small, but finite ductility in many inte~iietallicalloys may be sufficientto absorb the local strains at notches in components that often occur during initial service, caused by misali assembly problems. A few designs that recognize low ductility have been successfully accomplished, the best example being the change from the standard 'fir-tree' attachment to a dovetail co~figurationin the lowpressure turbine blades iioted above. Eveii in this case, the design tools remain relatively primitive, and component tests were needed to validate the configuration. One ,way conclude that the activity in developing nnd i ~ ~ ~ l ethe~ new e ~~?~alytical t ~ n ~ et hods necessary for ~ ~ t e r ~ e t a has l ~ iyet c , ~to achieve ~ ~ i t i cass. ~l ehavior under cyclic (fati~ue) loading brings her set of complications to the life prediction of intermetallic components. (See also the chapter by Stoloff in this volume.) If parts are free of defects and undamaged, the use of a maximum-stress criterion rnay be applicable, but quite minor damage can rapidly reduce capabil~ty. ~ h i l ethere is some under~tanding of the mechanisms that lead to reduced capability, there is little ability to quantify the nature, ~ a ~ n i t u d e s , and probabilities of damage occurring in the engine environment. An example of the dilemma are the attractive fat~gue-crac~-growth-threshold values measured in lamellar TiAl structures that imply excellent performaiice capability, but may be a~~~-conservative when used to predict life in damaged material. Smallscale defects that can cause life reductions are of a size that cannot readily be detected by standard nondestructive inspection methods, and therefore, more frequent inspection of in-service parts by more sophisticated techniques could be a pertinent strategy, but not a cost-effective one. More work is clearly needed to understand the cyclic behavior of intermetallic alloys in general, but especially the issues associated with the behavior of small defects at the earliest stages of da under realistic service stresses. It has been sted that microstructural manipulation rnay be a path to improved performance; but, in the long term, surface treatments aiidlor coatings rnay prove to be better control approac~es. (See also the chapter by Datta et al. in this volume.)
Intermetallic alloys are in the early stages of application, and hence full-scale qualificat~on and precise
information on production costs lie in the future. These days, before the needed investment is made, rigorous cost/benefit, net present value (NPV), internal R), etc. analyses are conducted; and, although other approaches to assess value have been suggested, to our knowledge these have not been used. Another recent modi~cation to the equation is the inclusion of previously expended development costs that, even if amortized over a relatively long period, can tip the outcome. Input parax~etersto the economics calculation include: Market size and growth projections. ~ u a l i ~ c a t i ocosts n i~clLidingcapital expenditures. Production and process control costs (yields, scrap rates, repair, etc. are included here). An approach that has been employed is to compare the step-bystep cost structure of a current process with that envisaged for the new material. In-service costs. ~ a r r a n t ~exposure, r inspection, repair etc. The calculated values of order of merit (IRR, MPV) are compared with the hurdle rates needed to justify the investment. As pointed out by Williams (1997), the decline in government funds availab~efor trans~tioning a technology to production has made it increasingly difficult to move forward. Tnteriiietallic alloys have been a victim of this new environi~ent;and, although elements of quali~cationprograms have been performed, usually aimed at a restricted application, there has been no concerted effort. As one could estimate the needed investment would be in the $30 x 106 range perhaps this is not surprising. (See also the chapter by Busch and Goodrich in this volume.) In summary, it is reasonably clear why intermetallic alloys are not in production today: non-optimized alloys, unsolved scale-up issues, design and reliability uncertainties, and the lack of clear iiivestment strategies and resources. With this in mind we will next survey the status for several candidate systems, looking for progress in recent years and seeking to define the way forward. The intermetallic systems selected for discussion are those which were reviewed in Volume 2 of this series. The material-specific section begins with a look at alloys based on y-TiAl, as this is the i n t e ~ e t a l l i cwhich has shown the most systematic and sustained progress (albeit not without interruptions) toward widespread engineering use. This is followed by much shorter reviews of refractory-metal silicides, aluminides, Ti,Al-based alloys, transition-metal trialuminides, and Fe-aluminides. For these systems,
there are essentially no component or prototype production data from whicli to measure progress.
3.
YS
Gamma titaniui~-alu~inide alloys (gamma alloys), based on TiA1, are closer than any other intermetallics to making a pervasive impact as advanced structural materials. There i s a well-documented history of research and en~ineeringof these alloys; however, only after more than four decades of development are the first specialty uses taking place. This article does not review the history, nor is there intent to describe the physical ~etallurgyof this alloy system. For these the reader is referred to several recent reports: Uamaguclii and Umakoshi (1990); (1991); Uamaguchi and Inui (1 Chesnutt (1995); ~ i ~ i d u(199 k (1998); Appel and Wagiier (1998); Dimiduk (1998); Dxmiduk (1999); and Dimiduk et al. ( ~ 0 0 0 )and ~ the conference proceedings G a ~ ~~i t~a nai ~~ml u i ~ i n i ~ ~ s (Kirn et al., 1995), and Gamma Titmium ~ l ~ 1999 (Kim et al., 1999). Rather, this chapter assesses gamma titanium aluminides, relative to performance, affordability, and reliability. 3.1.1 P e r f ( ~ r ~ a n Factors ~e
The characteristics which make TiAl alloys attractive are: a high melting point (- 1480 "C), low density (- 3.8 Mg/ni3)),high modulus, good burn resistance, and competitive speci~c-streiigthlevels. These translate into potential performance adva~~tages in each of the candidate TiAl alloys selected for this review. Detailed critical assessments of inany features of TiAl were published previously ( Dimiduk, 1999), so only key attributes are given here. Alloy ~ o ~ ~ p o ~ i tand i o n~s ~ c r o s t ~ ~ ~ t ~ r ~ s al. (2000) recently reviewed the coin gamma alloys that have emerged world-wide, along with their relative merits and limitations. Alloys are available that have advantages over polycrystalline superalloys and conventional titanium alloys. Gamma alloys are polyphase, polyconstituent alloys consisting of a primary L1, y-TiAl phase, a secondary DO,, phase of a,-Ti,Al, and usually one more other phases, including a Ti-A1 beta phase in multicompone~it
alloys), borides, carbides, or silicides. Aluminum contents fall within the range from 45 to 48 atomic 740 Al, while other alloying elements serve to enhance their performance. The predominant role of A1 is to control strength and oxidation resistance, the former increasing and the latter decreasing as the A1 content is reduced. However, the oxidation resistance of gamma alloys does not depend solely on A1 content; rather there are complex synergistic eEects among the constituent elements Nb, W and Si that improve oxidation resistance, e.g. Nb becomes synergistically more eRective when combined with W additions. Cr and Mn additions at the 1-2% level increase ductility, irrespective of microstructure. Carbon improves creep strength even at sinall strains (stage I), especially under high applied stresses, and this improvement is associated with carbides present on the lamellar boundaries where a2 phase has dissolved during aging or creep testing (Gounia et al., 1998). The most efkctive way of refining grain sizes of both cast and wrought alloys is with boron additions; through constitutional supercooling (Cheng, 1999) in cast alloys, and by boride particles pinning grain boundaries in wrought alloys. There are few studies of the ~ a r t i t i o i ~ iof ~ ~elements g between phases for et al. (1997) oEer a qLia~itative, gamma alloys. but useful, sum f some partitioning trends. Most of such data have been obtained through ternary phase"diagram studies, with few efforts looking at the ts in iiiulticomponent alloys (Kini , Larson. and Miller, 1999; Menand, erature reports of thermodynamic modeling of the phase formation and partitioni~ig effects for gamma alloys are limited to ternary systems (Saunders, 1999). Much work is needed on phase equilibria to establish element partitioning, and the quantitative aspects of phase transformations for process control. ~ o m m o i ~icrostructures i for gamma alloys are the "duplex', nearly lamellar,' and 'fully lamellar' forins. The duplex form consists of nominally equiaxed yphase and occasional a,-phase grains, together with yand a2-containing laniellar grains in a dual microconstituent structure. Volume fractions of each microconstituent vary with process route and heat treatment, but there will be typically less than 50 of lamellar grains in a duplex microstrucplex microstructures are common in pro~essingof cast material and represent the lower e of strength and creep resistance, together with the highest ductility for the alloys. At another extreme i s the n o ~ i n a l l ysingle-microconsti~
tuent, fully lamellar form, having no equiaxed y or a2 grains. This niicrostructure provides outstanding strength and creep resistance (as described below), typically at lower ductility levels which scale inversely with grain size. Because producing useful fully lamellar microstructures requires a controlled thermal transient above the alpha-transus temperature, this f o m is typically limited to wrought-processed alloys, although there has beeii some exploration of these structures for castings (Huang and ~ h e s n u t t ,1995; ~ i m i d uet ~al., 1998). Interi~ediate to these extremes is the nearly lamellar microstructure. This form is much like the fully lamellar form, but has a discontinuous necklace structure of y grains surrounding the lamellar microconstituent. The necklace keeps the lamellar grain size small (< 100 pm) and permits good strength. However, creep resistance and fracture tou~hnessdrop rapidly as the necklace becoines continuous and welldefined. These microstructures are not well understood, but are common to grain-refined castings and are being evaluated in wrought-processed alloys. Further details regarding inicrostructural evolution, understa~di~ig, and use are described in the reviews previously referenced.
-
StiJkess and ~ t r e ~The ~ relastic ~ properties of TiAl alloys are generally favorable for mechanical design of many aerospace components. Figure 2a shows that the modulus relative to density for gamma alloys is intermediate between that of common structural alloys and the structural ceramics. Never before (except for the use of b e r y ~ l i u ~have ) designers had the opportunity to exploit an alloy system having a specific modulus, which is nearly double that of other elastic modulus ~tructural alloys ~ ~ i g2b). ~ r The e compares favorably with that of nickel alloys on an absolute basis, and far cxceeds thc density-normalized modulus for both nickel and titanium alloys. Unfortunately, it is only infrequently that elastic properties doininate design choices in the absence of other aspects of the full balance of properties. Further, to take advantage of such a diRereiit property requires redesign of components, for which there are few opportunities and which present separate obstacles to the ev~lutionarynature of system design. For aeroengines, casings and support structures require high stiffness for shape retention, and several smaller ringshaped structures that require high stiEness exist within an engine. However, many of these also require some ability to absorb or contain the energy release that occurs through possible failures of rotating
High- Temperature S t r ~ c t u ~ Applicatioizs al
449
Figure 3 shows the specific stress, as a function of the Larson-Miller parameter, for creep rupture of TiAl alloys and includes data for the superallay IN100 for comparison, The two alloys shown very nearly define the limits of creep behavior for gamma alloys. Generally speakmg, at teniperatures below 600 "C, the gamma alloys offer no load~carryingadvantage in creep over conventlonal titanium alloys; however, the gamma alloys oifer greatcr oxidation resistance, and resistance to burning in the turbine-en~~ne co~pressor environment. In fact, as shown in Figure 4, the niost recent laboratory investigations indicate tlmt alloying can significantly reduce weight gain during oxidation of gamma alloys. The best alloys oxidize about twice as fast as Ni-based alloys at temperatures as high as 870°C (Singheiser et al., 1999; Yoshihara and 1999, 2000). Naturally, nickel supera~loys operate effectively in the engine compressor environment, but at a weight penalty relative to gamma alloys. Thus, it is the density-nor~ali~ed properties of gamma t i t a n i u ~ aluminides that are the continued driving force behiiid their development. N
Al-
TOO
T
stic properties of selected materials: (a) roomtemperature Young's modulus versus density: and (b) Specific modulus versus temperature for gainma alloys, NiA1, a Nbsilicide-based alloy and two common aerospace alloys
hardware. During operation, engine airfoils sample vibrational modes that are a function of material stiffness and aerodynamic design, thus requiring a redesign for gamma alloys rather than simple substitution. The yield strength of gamnia alloys can cover very wide ranges depending on composition and processing; but even at the highest strength levels are only equivalent to other aerospace alloys, Strength properties become niuch more competitive relative to other metals when the density of the material is also important, as in rotating machinery. The specific strengths of some gamma alloys, Figure lb, meet or exceed those of polycrystalline nickel alloys at all temperatures of interest. The values are also superior to conventional titaniuii2 alloys at teiiiperatures greatcr than 250 "C. Turning to time-depeiident properties, N
Fatigue, crack growth, fracture and ductility Structural design for aerospace, automotive, or powergeneration systems places great emphasis on fatigue and fracture resistance; current methods seek control and a predictable response of components subjected to oscillatory loads. For gamma alloys there are data for stress versus cycles-to-failure (3-N curves) at various temperatures, crack-growth characteristics, and other aspects of durability and lif~-limitin fatigue conditions. The results show that the intrinsic fatigue resistance of gamma alloys (as measured by run-out stresses as a percentage of ultimate strength) under high-cycle loading conditions are outstanding. Another attractive characteristic found for some TiAl alloys that have the lamellar microstructure is a high threshold-stress intensity for cyclic crack growth. However, relative to current nickel aiid titanium alloys, crack-growth rates increase rapidly as a function of cyclic st~ess-intensityrange, Consequently, current work is focused on understanding the niicrostructural and metallurgical aspects of long-crack and short-crack growth thresholds. Results show that, provided the cyclic-stress amplitudes are low, and cracks which lead to failure are those intrinsically nucleated, then the behavior of short cracks having lengths below or comparable to the grain size becomes important (Larsen et al., 1999a; Chan, 1999). Thus, the grain-size distribution is the key material parameter for
.......................
...................... .....................
i~~~~ 3 Density~norm~~l~zed stress for 0.2% creep strain versus Larsen- ille er parameter for selected h i g h " ~ ~ r f o r m a alloys. nc~ ~ o ~ ~ a r i iss made o n of two gamma alloys, one wrought and the other cast, against two processing variations of the nickel alloy bN100, oiic optimized for high yield strength, the other for extended creep resistance. The data for the Ti-6242 alloy are for stress rupture, not ofhet creep strain, but still are only roughly equivalent to the Ti-47XD gamma alloy. The superiority of the wrought gamma alloy at high Larsoti-Miller parameter values is evident. Other properties for the gdmma t~t~iiiurn"a1urnini~~ alloys plotted here are given in Table 1
durability~and this factor, together with the high crack-growth threshold, could lead to fatigue~damageresistant materials with a pervasive impact on system design. ~ n f o r t ~ n ~ t e lsuch y , behavior is not yet ~einonstrated on production-scale materials, for which extrin~ic or process-related defects typically control cyclic crack nucleation. As such extrinsic defects are kiiown to occur, coupled with poor impact resistance, the promise of gamma alloys may not be fully realized. These issues are discussed below as aspects of 'Reliability'. For TiA1 alloys and other i n t e ~ e t a l ~ i cquestions s, regarding the app1ication"specific minimum levels of ductility and fracture toughness required for successftil component operatioii are still not resolved. Experience shows that ~echanical-design and manufacturing methods demand some capability o f materials to diminish local stress concentrat~onsthrough plastic flow, TiAl alloys exhibit only between 0.5 and 4% ductility at low te~peratures(and statistical minima inay be lower), but even these modest ductility levels are shown to be ample for reducing stress concentrations conimon in structural design (Wright, 1993; naul e f al., 1999).
Experience shows that Failure initiation for both inonotonic and cyclic loading in ~ ~ alloys m is~ a overwhelniingly domi~atedby the grain size, especially for lamellar microstructures (Kim and Dimidmk, 1997a; Larsen et al., 1999b). Current understanding indicates that plastic s t r a i n i ~leads ~ to internal stresses that nucleate cleavage cracks on glide planes, with a size equal to or greater than the grain size. Failure proceeds by cle~va~e-crack propagati~nonce a critical stress intensity is reached. A measured ductility of a few percent elongation iiicludes both a finite plastic regime and a negligibly small, crac~-propa~ation ime. For duplex structures, such cracks typically extend beyond a single grain and rapidly reach critical size, since the inultiplicity of slip planes suggests that a favorably oriented cleavage plane is likely to exist in one or more neighboring grains. Relative to duplex i~i~rostructures, lamellar microstructLires intrinsically constrain the cleavage-crack initiatio~size by reducing the likelihood of a favorably oriented cleavage plane (parallel to the laiiiellar planes) being available in ail adjacent grain, and possibly by raising the intrinsic toughness of the grain. Unlike the case of duplex microstructures, for sufTiciently small lamellar grain
H i g h ~ e ~ p ~ r aSt ~u r~ ~e c t Applications ur~~
sizes (-
-
-
uch of the ~omparativeinfor~ationpresented above and in the literature was generalized for clarity; however, from a practical point of view one inust focus on the best properties that can be achieved in a
48 1
production eiivlronment. The issue is addressed separately for cast and wrought product forms. Table 1 shows a set of typical mechanical properties for selected alloy compositions, each prepared as a single heat by a given process. For castings, alloy and process ~ e v e l o p m ~ nhas t occurred in ~~cilities close to the production environment. However, this is not true for wrought-processed materials, for which develop~ent has taken place using sub-scale ingot stock in laboratory Facilities. Only recently is wrought processing being evaluated in a production environment (Dimiduk et al., 1998). (See also the chapter by Appel et al. in this volume.) This distinction is important, since the developmental processes have iiot simulated the representative defect ~ ~ ~ u l aassociated t i o ~ with actual production melting and ~ ~ ~ c e s s i nInitial g. attempts at large ingot making (42-66 ciii dia.) indicate that severe segregation may be a problem for these large ingot sizes. The alloyjprocess with the best all-round perforniance in the wrought form is the ' by Kim and Dimiduk nominal composition: Ti0.2C, and processed to a had an average grain size of less than -200 ym. discussed previously, grain size in l a ~ e l l ~ r - ~ r o c e s s e d alloys xs controlled by adding sniall amounts of boron ( < 0.3 at,%). In laboratory-scale samples containing
App Iicnt ions Property comparlson for selected gamma TiAl alloys and processes Properties
K5-FL
IC5-AEL
K5-AFL
47XD
ABB-2
IN-718
~-~
RT YS (MPa) 495 UTS (MPa) 560 Plastic strain at failure (%) 0.8 Fracture, K,, ( ~ ~ a ~ m ) 17 Fracture, ISQ (MPaJm) 22 Fatigue strength (MPa), R=O.1, N=107 495 Threshold, AK,, (MPaJm) 8.5 Hrs, 0.2% Creep at 138 MPa Hrs, 0.5%) Creep at 138 MPa Hrs, 0.2% Creep at 276MPa Hrs, 0.5% Creep at 276MPa T ("C)
760 462 540 1.6 19 25 358 8.8 650 2100 -40 250
RT 750 855 2.1
760 575 745 5.0
22
RT
645 805 3.3 14 16
160 525 725 34
RT 402 482 1.2
760 344 458 18
7.5 __
--
-
> 400
-
-
8.8
60
-
~
RT 1049 1311 18
760 800 852 5
75 559
>zoo
8.5
-
__
55
760 405 495 2.1
22
17 340
380
RT 485 555 0.4
60 600 13
K5-FL (Fully Lamellar): GS-200 pm; K5-AEL (Alpha-Ex~rudedLamellar), GS w 90 pm; K5-AFL (Alpha-Forged Lamellar), GS 25 pin; 47XD (Cast, ~ca~ly-Laniellar), GS 10,urn; ABB-2 (Cast, Nearly-Lamellar), Large, Non-uniform Lainellar Grains.
-
N
boron, the avcrage lamellar grain size can be smaller than 100 ,am; however, it has been found difficult to achieve these fine-scale grains in scaled-up ingot material. Further, when considering uniformity of grain size, even these ~ne-grained materials often contain large grains with sizes greater than three times the mean - an issue requiring further development and control. Comparison of the creep properties 5 material with a reference wrought nickelbased superalloy was shown in Figure 3. Table 1 and Figure 3, show that, apart from room-temperature du~tility,the titanium a l u ~ i n i d eis superior in the projected operating temperature range of 650-800 "C. One downside is that these properties have only been demonstrated 011 interniediate-size heats of 30 kg. Casting alloys, although they have a longer history, have experienced less systematic developmeiit based on metallurgical variable^ and the control of micros~rucowmet's alloy 47XD in fine-grained, nearly fully-lamellar form, and ABB's alloy, in a largegrained lamellar form, are probably the front runners for aerospace use, but are not yet in service. The properties of these alloys are also summarized in Table 1. As for niaiiy casting alloys, relative to wrought ~ a t e r i a l sthe , proper tie^ are reduced.
the opportunities to develop and experientially mature structural materials, including ganima alloys. For automotive applications of materials, such cost and cost-risk constraints have always dominated materials selection and development. Only the unique balance of properties offered by ganima alloys, together with the realization tlolat processing is mostly within the capabilities of conventional metal manufacturing equipment, permitted emergence of these alloys. Progress made over the last decade suggests that ganima alloys are a viable class of titanium alloys, whose introduct~on is paced primarily by the wor1d"wide business climate, and possibly by the emerging design knowledge for low ductility alloys, rather than limited by specific technical barriers or niaterial ~erformance. As of late 1999, many thousands of automotive valves have been made by casting methods and tested throughout the US, Japan and Europe, using both p e r m a n e n ~ ~ ~ oand l d investi~ent-casting processes. At least 10 000 Lancer automobiles with TiAl turbochargers have been produced and sold. indicate that the casting yield for the it sub is hi turbocharger rotors is about 70% for an alloy known to be very difficult to cast. Low pressure turbine blades have been evaluated for both aircraft engines and power~generationturbines (Austin et al. 1997; Rugg, 1999; Perriii, 3999). Even late-stage 3.1.3 ~ ~ o t ~ or o td ~~c ~~and i o~n~ ~ o r ~ ~ ~ ~ l i t turbine-engine y compressor blades have been successfully tested and survived a severe case-rub event 1990s the As discussed previously, throughout the (Fecke and Davidsoii, 1998; Sargent and Huffman, world-wide aerospace business community has 1999). This accumulating experience further indicates changed its focus toward addressing the life-cycle that gamma alloy casting technology holds significosts associated with engineering technologies. This cant promise when the business coiidibions are riglit marked departure from a historical focus on perforfor implementation. mance gains associated with new technologies reduced N
-
Two comprehensive reviews of the state of production technolog~esfor these materials are iiiciuded in this volume in the chapters by Mc casting, and by Appel et al., for wrought processing; tlius, these two subjects are not extensively reviewed here. From their discussions, one may summarize that the cost of gamma-alloy parts will likely be equal to or greater than the costs o f typical titanium-alloy products, since many of the same processing difficulties exist, or are even worse for these alloys. Also, at least for tlie next decade, the relative number of gammaalloy parts in production via common processes will be few, such that the economies of scale cannot be realized. An assessment of cost relative to nickel alloys, the main competitors to gamma alloys, is more difficult to establish. This is simply because of the diKerences in perfor~anceand part design that must enter into the aiialysis. Assessments must be made on a part-by-part, desig~-by-designbasis, and aKordability can only be defined once the value of performance factors is quantified. Further, there are several other aspects of a ~ o ~ d a b i l iwhich ty are not yet addressed in a production environment. Technologies for joining and repair are emerging in the casting trade; however, surface treatments and properties, env~ronmentalprotection tech~olo~ies, and ~ a c ~ n i and n g finishing technologies exist only as quite specialized techniques.
not exclusively tied to stage-ICE c r ~ ~ c ~ ~ ~ ~ o w t h - r a t e properties. Fortunately, crack-growth thresholds, as measured from lo~g-crack-arrestmethods, are superior for the gamma aluminides ( comparison to those for nickel a ( 2 4 MPaJm) (Larsen et al., 1999 point to distinct paths for suggest that components requiring under modest stress amplitu succeed. This feature i s illustra map (‘Kitagawa’ diagram) coordinates of alternating stress versus crack length, for a typical gamma alloy subjected to cyclic loading. For flaw sizes below 100 pi, crack initiation under high-cycle conditions dominates life. The high-cycle, c~ack~nLicleationstress is typically enhanced by iiicreases in yield strength of the all flaw sizes become greater than ~ropagation(partic~larlyin the neardominates life. The diagram also shows that any increase in yield streiigth (load-carrying capabili ty) leads to a decrease in the size of the minimum propagating flaw, unless there i s a concomitant increase in threshold stress iiitensities (sliifl the staiited line to the right). It must be emphasized that these
-
1
3.1.4 ~ ~ l i a ~ i l i t y
The use of TiAl alloys is not only limited by a ~ o r d a ~ i l i tbut y , also by a lack of proven reliability. In tlie aerospace business, reliability information is used to set inspection and warranty intervals, as well as for achieving certi~cationsfrom various flight regulatory agencies (FAA, Do etc.) However, readiness and reliabilit~of gamma-alloy technologies are still limited by variations in compositio~,rnicrostructure and properties for a given process and product, Gamma alloys are known to be sensitive to sLibtleties in composition, especially the levels of aluminuin. Further, experience in production is so limited that ~aiiufacturers cannot yet guarantee composition spec~fications~ Even on an experimental basis, composition specification has been d i ~ c u lfor t gamma alloys. No chemical reference standards exist for the compound, or even any certified compositions~thus, alloy speci~cationlimits are emerging from consistency of chemical analyses rather than from intrinsic accuracy. ~ e s e a r c hon the relia~ilityof gamma-al~oycomponents centers on develop in^ a design system which is ~
4
Figure 5 Alternating stress (AS) versus crack length (a) life map, or Kitagawa diagram, for the K5 alloy containing machined flaws, from Worth et al. (1997). A safe zone is expected for stresses below the horizontal line and to the left of the slanted lines. The A&, and AK&, th lines are constructed from long-crack threshold measurements. and crack-closure corrections to threshold, respectively. Black squares for initially machined notches of depths shown. The points at anotch=8 x 10-2 mm, and qIotch=2x 10-I niin are not expected to fail, based 011 machined notch length. E x a ~ i n ~ t i oofn a failed specimen revealed an additional in~rementof rapid crack ‘popin’, effectively raising the i n i t i a ~flaw size (see text)
alloys are damage intolerant; thus one must strive to produce the highest quality products and avoid inservice damage by careful ~ o ~ p o n e selection, nt especially for parts that inay experience high vibratory loads. There i s little precedent for widely utilizing such a material under high tensile loads, although the ~ntermet~~llics could benefit from critical examination of the experience base emerging from application of strLLctura~ceramics. Figure 5 also shows what inay be termed ‘anticonservative behavior’ for specimens that were preflawed by machined semi-circu~arnotches (Worth et al., 1997). The material bad a fully lamellar microstructure and a mean grain-intercept length (grain size) of -280 pm. Failure occurred at stresses below those predicted from the cyclic stress intensity threshold, especially for small flaw sizes in the transition regime from h i g h - c ~ ~f ~a et i g ~to~ c r a ~ ~ - g r o dominated ~th failure. Inspection of the failed specimens revealed that for machined notches sinaller than the grain size, cracks rapidly grew to a length set by a small multiple (-2x) of the grain size, depending upon the local orientation o f the neighboring grains. Such behavior may be anti~ipatedfrom the available understanding of the cyclic behavior of single~grainlamellar TiAl, that shows low fatigue resistance for crack paths parallel to the lamellae (Yokoshima and Yamaguchi, 1996). The results can be interpreted as indicating that all flaws or defects behave as if they are several grain diameters in size irrespective of actual dimensions. For the experiments shown, this translates to apparent flaw sizes for the 80 pm notches of a -400 pm defect (after ‘pop-in’ to the notclied grain and a neighboring grain). A similar bebavior exists for inonotonic fracture and imiduk, 1997a). These findings are encouraging. They indicate a possibility for a lifeprediction a~proachusing crack-growth thresholds, provided that damage and defects are understood (probabilistically) for a particular material and coinponent. Further, they also reinforce the need for grainsize control and unifor~~ity and process control. A key factor in the fracture mechanics formalism for design is the inspection limit for flaw detection, for this value sets a ~ ~ o t h ebo~ndary r in using the ~ i t a ~ ~ diagram. Inspection methods are somewhat dependent on geometry and frequently have a lower limit of flaw detection near to or greater than the typical threshold crack sizes. Further, current design schemes often demand a life-prediction method that assumes the existence of flaws at sizes equal to or greater than the inspection limit, and assumes that life will be determined by such flaws. Such design methods could B
clearly severely limit the use of g a m ~ alloys. a To offset such limits will require a combinatio~of developments to improve damage resistance, utilize high-quality process routes, and introduce probabilistic design and life-management schenies. As mentioned previously, extraneous damage, for example from foreign objects and debris (FOD) being ingested by a turbine engine, can damage the material, resulting in stress intensities beyond the threshold value for crack growth. For such a case, the residual cyclic life remaining after damage may be short, since for gamma alloys crack-growth rates in the Paris Law regime are high. Consequently, a barrier issue for ~ m ~ l ~ ~gamma e ~ t alloys ~ n gin highperformance, high-payoff hardware (such as turbineengine compressor blades or disks) is that of raising the quantitative understaiiding of the nature of FOD and other potential d a m a ~ eevents. For hardware design, one must bound the magnitude of damage occurring for particular damage events, preferably in terms of a flaw length or a stress-intensity factor, and by a pro~abilityof occurrence. Using this information, a probable-residual-life model may be developed and reconciled with nondestructive inspection techniques and inspection intervals to guarantee safety. However, this simple view, as difficult to acconiplish as it may be, avoids other complicating issues such as the residual stress state left by the damage event, and the need for coupling the probabilities of damage with the probabilities of processing~relateddefects in the material. Such probabilistic design methods are only now emerging in the aerospace arena. This progress, in fhct, is good news for the use of internietallics, since without such methods and design practice the li~e~ihood of use for gamma alloys would be low indeed. Finally, the ductility and toughness of gamma alloys are less understood under mixed-mode loading, or at high strain rates. Clearly, there are conditions for which aero-propulsion, automotive, and powergeneration systems can experience unforeseen and uncontrolled impact loads. The toughness and impact properties of gainina alloys are poor and will limit wtheir a use where such impact events have a high likelihood. The ‘accumulated learning’ regarding structural alloys suggests that a fracture toughness of about 20 MPaJm i s near a lower limit for general engineering use. The l i ~ i t sfor impact resistance are application specific aiid not well understood. Design iiiethods must continue to improve to move gamma alloys into wider use. Such an evolution should be based on probabilistics rooted in a clear understanding
High- Temperature Structural Applications of both microstructural defect occurrences and inservice damage events - both of which are difficult to obtain in the current climate for material development. 3.1.5 Gamma Alloy Summary
The last decade has brought dramatic growth in fundamental understanding of the properties of gaiiima alloys, and growth in the technological aspects of producing them. The properties of the gamma alloys strongly depend on processing (cast, wrought, etc.) which set distinct constraints on their utility. Cast alloys are not yet fully commercialized, in part because of the perception that performance benefits do not outweigh cost and risk. ~roLightalloys show greater performance benefits on a laboratory scale, but there i s no complete and ~emonstrated comi~ercial-scale technology. All gamma alloys exhibit low ductility and a low cleavage stress, and the levels seem insensitive to alloy coinposition. It seems unlikely that further research will substantially change their behavior. However, the laboratory-scale demonstration of an attractive balance of alloy properties opens revolutionary opport~nitiesfor weight reduction, and perhaps increased operating temperatures in aerospace systems, but at a price. That price may be the need for refined design approaches, and an involved, timeconsuming, and expensive sequence to calibrate the design systenis and to build confidence in the material. From the technical perspective, technology scale-up, maturing design practice, and machining difficulties continue to pace development. The cost of finished products continues to limit interest in their use (especially for automobiles); however, the cost barrier is tied directly to production volume, finishing technologies, and design limitations. The next decade should bring a systematic reduction of costs for gamma-alloy hardware as fktiniliarity with the alloys builds and more are produced.
This section will cover alloys and com~oundsof the high melting-point, refractory metals i ~ o l y b d e ~ uand m niobium that are candidates for use at high temperatures. Both metals have intrinsic problems at high temperatures - deplorable oxidation resistance and body-centered cubic structures that creep too rapidly. Silicon-containing refractory metal systems show promise for improving these characteristics; and, in tlie case of molybdenum alloys especially, additions of boron provide yet better properties. Research and
485
development on high-silicon materials is conceiitrated in three areas: the compounds oSi2 (tI6, cll,) and Mo,Si, (tI32, L>8,,); Mo-base all and 13; and materials based o A15) and/or Nb& (tI32, materials will be discussed separately.
3.2. I ~ ~ l y b ~ e nDidicide, um MoSi, Previous research on Volume 2 of this seri aiid Miracle aiid Mendiratta, 1995). Petrovic and Vasudevan (1994) pointed out that several issues needed to be overcome before its use in hightemperature structural applications. These included eliminating the intermediate temperature oxidation (pest) behavior, increasing fracture tou~hne§s,and iniproving high-temperature creep resistance. The most eEective approaches to eliminate the pest problem are to adjust the pro porosity or by formiiig compo (cF8, 133) or Si3N, (hPl4 or 11P28, (Hebsur and Nathal, 1997). The room-tei~perature fracture toughness is less than 5 MPaJm, far short of the miiiirnum value of 20 MPaJm typical for current structural materials. ~ncorporatin~ MoSi, as one component of a composite structure IS the usual approach to this problem. An addition of 30-50 vol.% Si,N, to the compound has an outstanding effect on creep r Sadananda, 1997, see Figure 6) addition has limited inAueiice on RT, although it increases to 15 The addition does eliminate behavior; a 3 0 vol.% coinposite can survive 1000 cycles between RT and 500°C while a 50 vol.% material can survive 1000 cycles between 1250"C. Unfortunately, without iinprovennen RT toughness, there is not a favorable balance in base properties from which further d e v e l o p i ~ e ~mt i ~ h t occur. Nano-scale composites of MoSI,-SiC h been reported to have very high strength and 1250°C (Suzuki and Niihara, 1997; 1998); however, given the ultra-fine particle size, these inaterials would be expected to show poor creep and stress-rupture behavior. Commercial grades of MoSi, contain oxygen that typically is present in a glassy rain-boun~~ry phase, rich in silica. The presence of this phase promotes ductility at intermediate temperatures (about 1000 "C), and this iinprovement could potentially enhance service p e r f o r ~ a n c ~ The . the oxygen
er
486
Applica t ions
ure 6 Secondary creep rates for selected interrnetallics and h i g h - t e ~ p e r ~ ~alloys. t ~ ~ r ePVVA-1484 and Rene-N6 are modern siiigle-crystal superalloys currently used for turbine-engine airfoils and data shown are for tensile tests. MO-TZM is a comnion refractory metal alloy, tested in an inert environment. The DS-MASC Nb-silicide based alloy (Bewlay et al., 1996) and the Nb10% alloy were tested in tension in an inert environment. The Mo-12Si-12B alloy was prepared by casting and data shown are for 1200 "C co~pres§iontests. The r e ~ i ~ ~ iinterinetallics nin~ were evaluated by ~ o ~ p r e s s i otesting n in various e i ~ v i r o n ~ e i i ~ s
content of the material, the lower is the britt1e:ductile :D) transition temperature, though even at an oxygen content of 0.61 a/o the D transition is only 1050°C (Aiken, 1993; Srinivasan et al., 1993). wever, any ductility advantage is probably offset the de 1 effect of oxygen on the creep resistance i2 ( ~ i b a ~eta al., 1992). A third I-Rect of the glassy phase IS that it facilitates the man~factureof shapes by hot pressing, etc. Further work on the best balance of characteristics of particu1at~"based MoSi, would be necessary if a viable application i s identified. Attempts to improve the creep properties have met with mixed success. Mason and Van Aken (1993) showed tbat a directiona~ly-solidified(DS) eutectic of i,, an it^ composite, had improved creep strength. At 1200°C and a strain rate of the flow stress of the DS eutectic was ile that of a standard hot-pressed powder
product was 20 MPa. Sadananda and Feng (1 994) and Feng and Sadananda (1997) showed there was a signi~cantreduction in creep rate for MoSi, ~ a t e r i a l , dispersion strengthened with >25% SIC; and a coniposite containing 50% Si,N4 exhibited even better creep resistance. French et al. (1994) studied the stressrupture behavior of M0Si2 containing a dispersion of SIC powder; as in the Case of creep, the rupture times increased as the fraction of dispersoid was increased, However, even in material contai~ing40 vol.% Sic stresses of only 40 MPa resulted in a rupture life of less than 100 hours at 1150°C and less than 20 hours at 1200 "C. Qn a strength/density basis the creep properties of these composite materials are competitive with single-crystal superalloys. However, only limited ~rac~-propagationresults have been reported for a MoSi2-SiC composite ~Ramamurthyet al., 1994) and the data indicate that Pttigue crack growth at 1200 "C occurs at very low values of AK.
High- Temperature StrLictural Applications
In summary, at this time, even the best properties achieved for MoSi2-base materials, including composites, do not provide a balanced property advantage over current materials used to construct the engineering systems reviewed in this article.
3.2.2 Other ~
o
l
y S i~l i ~ i~d e ~~
~
u
~
Research on Mo,Si, is nieiitioned in a few places in Volume 2 of this series, but no alloying studies are included and only a few properties are given (Kumar, 1995; Fleischer, 1995; Miracle and Mendiratta, 1995). (For clarity, alloy compositions discussed in this and the next section are plotted in Figure 7, a schematic isothermal section of the Mo-Si-B ecpilibriuni diagram at 1600 "C takeii from o et al., 1997.) The compressive creep rate Si, at 1200°C and 10-* s-I, a fivefold 69 MPa was reported to improvement over MoSi, (2. 1 x IO-?' s-l), competitive with superalloys (Anton and Shah, 1991). these workers also reported that catastrophic oxidation of unalloyed M05Si3 occurred after only 20 cycles to 1200°C. Meyer et al. (1996), Meyer and Akinc (1996), and Akinc et al. (1999) have studied the creep and oxidation belzavior of Mo,Si, and were the first to r improvement of oxidation resistance additions. Compared with unalloyed Mo,Si,, a composition o f Mo-37Si-7.5B formed a protective scale, exhibited parabolic oxidation kinetics in the temperature range of 1050-1 300 "C, and exhibited a decrease in oxidation rate by five orders
487
of magnitude at 1200 "C - dramatic improvements. In addition, quite small B additions ( < 2 wt.%) elirninated pest oxidation at 800 "C ( eyer et al., 1996).The proposed mechanism for these improvernents was the ability of the borosilicate glass scale formed to ~ a p i ~ l y siater and thus suppress the volatilization of' molybdenum oxide. Meyer et al. (1996) and Akinc et al. (1999) reported the conipressive creep behavior Mo5Si,-base composition (MO-31 tained three phases: Mo,Si,(T~) A1 5)) and Mo5SiB,(T,) (tT32) combination of phases is not consistent with tbe equilibrium diagram in Figure 7. The creep rate of this three-phase material was only slightly faster than that of monolithic MO$%,in the range 1240-1320°C and 140-180 MPa. Thus, even though soluble in Mo-Si compounds, and new a result, it markedly improves tbe oxidation resistance of Mo,Si, with only a slight reduction In the compressive creep behavior. The Mo,Si, and Mo-Si materials discussed above are brittle at room tempe ture and the brittle~ductile transition temperatures are not clearly estab~ished. Antoii and Shah (1 99 1))reported the unalloyed M05Si3 at 1200°C as 12 6% ductility. Some ductility in the tion temperature range is encouraging. However, there are no cyclic oxidation, fracture toughness, creep, stress-rupture or fatigue data for these base materials. Thus it is not possible to realistically assess any potential for use in turbine engines. 3.2.3 ~ o l - v b d e ~ ~ ~ Allojm -Sili~~~~
a Perepezko et al. (1997)
~ e n ~ i reta al. ~ a(1899) et al. (1999)
0 Schneibel
i~~~~ 7 Ternary isothermal section of Mo-Si-B phase diagram at 16OO"C, after Perepezko et al., 1997. Selected compositions of interest and investigators arc shown. Detailed compositions studied are described in the text
~olybdenum-basealloys are also b turbine-engine use. Although not conipounds, as can be seen from Figure 7 these alloys have compositions contiguous with the compounds discussed above. These alloys have long been attractive as high-temperature materials because of excelleiit strength properties that are maintained to elevated temperatures. However, they have received little attention in recent years because of extremely poor oxidation resistance. The oxide formed, MOO, (0P16), i s a solid with very high vapor pressure even at ternperatures <760"C, and above this temperature exists in the gaseous state and is therefore non-pr~)te~tive. Recently, Berczik (1997a) was granted a patent for Family of Mo-base compositions and B to promote the presence 01 in a metallic MOmatrix. A ~ t ~ o the u ~ precipitates b are
488 not identified in the patent, Figure 7 indicates that they are probably M o ~ ~ i ( B ) T[the 2 (B) indicates B in solid solution] andlot- Mo,SiB,. Alloys developed by Berczik (1997a,b, 1998) contain quaternary and higher order elemental additions. (Note, to include these alloys in Figure 7, all the metallic elements have been combined and plotted as a/o 0.) Under oxidizing conditions, the presence of thes termetallic precipitates provides the necessary Si to forin a coatizig of a borosilicate glass. Berczik (1997a,b) presents static oxidation data at 1093 "C for a series of ternary MoSi-B alloys containing from 0.5-5.0 w/o Si (1.6-15.2 a/o i) aiid from 0-7.0 wjo (0-39.4 a/o El), These data show that the best oxidation resistance, given as the rate, occurs in an alloy Mo-4.5Si-4B, 24.9B, a/o) and was reported as 0.02 mils/mi~(or 120 mils/lOO h or -3048 ~m/lOOh). ince a figure of merit for oxidation resistance, based on current alloy performance, sought for turbineengine airfoils is 1 mi~/lOOh (- 25 ~m/lOOh)(l3ewlay et al., 1996), this oxidation behavior appears to be two orders of inagnitude Fmt. In addition, the recession rczik, 1997%)is for a nearcomposition, so the data may idation behavior of the Mo-base alloys. A later patent (Berczik, 1997b) claims maxima for the oxidation resistance of complex alloys at a recession rate ~ 0 . 0 1inches when heated to either 2000 "F (1093 "6)or about 2500 "F (1371 "C) for two hours. This rate translates to 0.500 inches or 1.25 x 1 0 4 ~ ~ ~h1 (although 00 this depends on the shape of the kinetic relationship), a rate that is even faster than for the ternary compositions above, The alloy compos~tion for which the best oxidation 1093 "C and at 1341 "C was (MO-1OTi-6.3Si-8,1l3, a/o). 98) reported oxidation data for newer alloy compositions that are shown in loy Mo-0,3Hf"2Si-ll3, w/o (Moa/o), showed a recession rate of r about 450 ,um/l00 h at 1093 "6. This rate is still about 18 times faster than the goal rate. ore encoLiragingly, alloys containing 3Si and 3Hf hibit recession rates of only 0.003 in/100 h or 75 ,um/ 100h, much closer to the goal. The alloying strategy described by Berczik (1997a,b) i s as follows. Alloys are designed to have matrices strengthened by elements such as Ti, Zr, Hf, and/or A1 that may also increase oxidation resistance. A small amount of C may also be added to promote carbide particle stre ng. An alloying level of 2 wt.% Si and 1 wt.% -6.1%-7.9B a/o, assuming a ternary
-
alloy), results in a 30-3S0/0 volume fraction of intermetallic Mo,SiB, and Mo,Si(B). This large fraction makes processing difficult, and to facilitate processing alloys were made from blended elemental powders or from rapidly solidified powder, then consolidated by extrusion or HIP. The difficulty in producing homoge~eouscast material was shown by Perepezko et al. (1997) on a Mo-7Si-14B a/o alloy. It was shown that a homogenization treatment of 150 h at 1200°C did not significantly change the solidification structure and that even an annealing treatment of 150 h at 1600 "C did not completely equilibrate the microstructure. These results clearly support the need to utilize powder methods or extensive thermomechanical processing to produce uniform material. Berczik (1997a,b) reports tensile properties for the composition ~0-0.3Hf"2Si-IB,w/o (MO-0.14Hf-6.1Si7.9B, a/o), shown in Table 2. Strength levels over the t e m ~ e r a t ~ rrange e reported are superior to singlecrystal nickel alloys over the same range. The low ductility at lower temperatures may present some concern, alt~ough the ductility in the potential operating range i s more than adequate. Creep, fatigue, fracture toughness, cyclic oxidation, and stress-rupture data for these materials have not yet appeared. Mendiratta et al. (1999) have studied similar materials, of which the most interesting are MO-1 . The alloys were triple arcmelted from elemental mixtures and homogenized in an inert atmosphere at 1600"C/24 h i- 1700 "(748 h producing an essentially homogeneous structure. Electron microprobe analysis of the ~0-12Si-12Balloy showed a three-phase structure with a 37% volume fraction of Mo(Si,B). The composition o f the MO( phase was Mo-6B-3,6Si7 that of the T, phase was Mo-26.5B-12.3Si and that of the Mo,Si phase was MO-1B-24Si, consistent with the phase compositions reported by Perepezko et al. (1997). The microstructure contained approximately equal amounts of the solid solution Mo(Si,B) and T2 phases. At equilibrium, this co~positionshould only contain 30 vol.% metallic phase, 21% Mo,Si and 49% T2; and it can be concluded that even extended high-temperature anneal in^ did not achieve full equilibrium. The fracture strength of this alloy was about 600 MPa at RT and varied between 650-700 MPa from 800°C to 1400 "C. The fracture toughness, Kq, was about 15 NPaJm at RT, decreased to a value of about 10 MPaJm at 8OO"C, rose gradually to a peak of about 13 MPaJin at 1300 "C, and decreased above that temperature. At 1200 "C/100 MPa the compressive I
~ i g h - T e ~ p e r a l ~~ tr re ~ c t uApplicatiorzs ~al creep rate was 1 x 10M9/sand at 1300 "C/lOOMPa, was about 1 x 10W8/s.These are outstanding results as can be seen in Figure 6, which illustrates the advantage of these materials over superalloys and niobium-based systems. Cyclic oxidation was studied for a range of coinpositions at 1200 "C over inore than 100 h. It was determined that the compositions most resistant to weight loss (and without weight gain) contained equal coiicentrations of Si and of about 1 1a/o. Cyclic oxidation testing, for up to 200 h, showed significant weight loss at 800 "C, very slight, essentially linear weight loss at both 1200°C and 1300"C, and
4139
catastrophic weight loss occurred at 1400 "6. The average recession rates were 1.8 jm/h at 800 ' C , 0.4 pm/h at 12OO0C, and 1.2 pm/h at 1300°C. These values are within the goal range for turbine-engine materials, as can be seen from Figure 8. It was also shown that if a Si coating were applied to the alloy prior to o ~ i d a t i o ~the , o~idation rate de~reased dramatically aiid the material behaved as if it were solid MoSi, for periods up to -200 h. Schneibel et al. (1999) studied the oxidation behavior and the mechanical properties of several compositions in the same three-phase field: Mo(Si,
Figure 8 Recession rate versus temperature for selected high-temperature interrnetallics and alloys. Results for Mo-Si-B -i- xHf are from Berczik (1998). The band representing Nb-Silicide + Law-phase alloys is from Jackson (1999). Data for MO-11Si-11 are from ~ e n ~ i ~ aett al. t a (1999)
490
Applications 2). Arc-cast buttons were prepared ts and then annealed in vacuum for ported oxidation behavior for alloys
recorded modest weight losses. ~ ~ c r o s t r u c t u r evalal uation of these latter alloys showed that the cast structure was not conipletely were not fully equilibrated. temperature threeoint flexure tests showed re strengths were number of a ~ b i g ~ ~ i texist i e s in the data in the
sitronal aims, others by the specific processing methods
oxidation. resistance at high temperatures, but gives no intermediate temperatures, whereas the data iratta et al. indicate that there may be an ate-temperature, pest-like oxidation problem ven at higher temperatures, Berczik's alloys only show oxidation resistance about equal to ~ ~ ~ s e whereas ~ e n d i r a t t a ' s that of a ~ i i c k ~ lsuperalloy, material is better by a factor of 100. Schneibel's oxidation results for the higher o alloys (which were h in air at 1200 "C or 1300 "C) are t with the other oxidation data s ~uaternary alloys show static strength at about the levels expected for molybdenum-base alloys with some ductility; Mendirattak similar strength, but are brittle. in ~eibel'salloys are both weak and brittle. he rather low values of toughness and ductility remain concerns and additional creep, fatigue, and fracture tou~hnessdata are necessary. It is important to resolve these inconsistencies, as the results of endiratta imply that a useful property nce may be achievable in this composition range. this point it appears that powder-processing nxthods are preferred, as cast material requires very high homogenization t~m~erat~Lres. Much more information on other aspects of manufacturability are necessary before this system can be considered for scale-up.
Miracle and Mendiratta (1995) reviewed early studies o f N b - N b ~ composites ~i~ in Volume 2 of this series, so we shall deal only with recent developments here. Bewlay et al. (this volume) have reviewed the most recent data including the efforts to develop materials useful in turbine engines in another chapter in the present volume, so this section will only deal with issues that fit within our proscribed framewor~. The historical problem with b-base materials has been the lack of sufficient oxidation resistance at high temperatures, so that not only is a coating necessary, but in the event of coating damage the underlying alloy would suffer very severe recession rates. Subramanian et al. (1994) and Jackson et al. (1995) showed that the partial su~stitution of Ti or Hf for Nb in these materials improved the oxidation resistance. mall amounts of Cr and A1 produced further ii~~rovements; but since these elements lower the melting temperatLire, can only be added in small amounts. results a six-elenie (Nb-2STi-8Hf-2Cr2Al- 1623, a/o) was designated as the site). It is typically produced as directionally-solid~~ed castings and in this form has demo~strateda good balance of properties coupled with much improved oxidation resistance. Jackson et al. (1995) reported the oxidation resistance was comparable with Ni-base superalloys, in the temperature range 1 0 0 ~ 1 2 0 'C, 0 as shown in Figure 8. ~ecently, Jac showed, as suggested by the research of (1996), Meyer and (1999) that a small cyclic oxidation resis partial substitution o reduced the recession rate to about 12.7 piii/l00h at t 200 "C. Oxidation rates at lower teniperatures are not so attractive, Figure 8, and the i~plicationsof this will be discussed later. The physical and mechanical properties of the MASG are thoroughly reviewed in volume); thus, only pertinent properties will be abstracted here for comparison with other materials. ature f ~ ~ c t ~ r e ~ t ~values u ~ ~ of n eKis lie s 18-23 MPa&n for monotonic loading, ve (R-curve) measure~~ents yield similar values of K,, and the composite displays increasing crack-growth resistance with increasin~crack length. The fracture strength of the MA rooni temperature and the material is brittle. The onset of ductility is at -1000'6 and the yield stress at
Nigh-~ e ~ p et ure r a Struc tural Applications a ductility of 19%. These rczik's (I 997a) results on molybdenum alloy above. Figure 6 includes the creep data for this material and shows equivalence with a single-crystal nickel-based superalloy, but a lower capability than several other candidate materials. On a density-corrected basis the MASC material i s much more competitive as the density is 7.35 Mg/m3 compared with that of superalloys at > 9 Mg/ni3 and shows a t~mperature . The modulus of the at room temperature and it decreases linearly with te~perature to 140 at 1200 'C (a relatively modest decrease). T igh modulus value implies poorer thermal fatigue resistance conipared to s alloy single crystals, as is probably also true for age c o e ~ c i e n of t th~rmalexpanis sufficiently close to the thermal expansion of the metallic phase of the composite that no problems with structural stability are anti~ipated. Tliere are limited crack-growth data on the MASC that are shown in Figure 9 which also coinpares performance with other selected materials (Zinsser and
49 1
Lewandowski, 1998; Zinsser et 3 999; Lewandowski, 1999). The average tlizesliold AK of 7.-5 9.7 ~ P a for ~ the m composite was lower than that of b-base solid~solutio~ialloy, but signi~cantly higher than for any o f the silicide matrices reported in the literature. In contrast to the results of other studies on ~uctile-phase-tou pounds (TiAl 3- Nb,Ti, MO slopes for the composite were similar to those obtained for metallic specimens. These results contra those of Venkateswara Rao et al. Nb composite for which no signi eiiing due to the presence of ductile measured and threshold levels were only 1 It was concluded that the spherical morph Nb particles allowed them to be ~~rcumvented by propaga ting cracks. uch be~avioris sn contrast to the blunting aiid bridging characteristics of the lamellar MASC structure. The progress, processing, and potent~alof this class of directi~)iially-solidified, composites has also been reported in a number et aE., 1996; ~ u b r a ~ ~ a n i a n
iyure 9 Goinparison of fatigue-crack propagation rates for selected interinetallics and solid-solution refractory-metal phases. The DS Nb-silicide alloy data are for the MASC alloy discussed in the text (Zinsser and L e ~ a n d o ~ s l1998) ~l,
Applications
492
ewlay et al., 1999). The oxidation behavior of aterials is satisfactory in the temperature range from 1 0 0 ~ 1 2 5"C; 0 however, the oxidation rates at lower temperatures f < 1000"6) are a concern. Turbine-engine co~ponentsspend a substantial portion of the system life, even for liigh-teniper~~ture com~onentssuch as airfoils, at temperatures below this level. ~nfortLInately,no data have been reported for this temperature range, although the trend is not encourag~ng.The fracture toughness of these materials appears to be adequate for turbine engines, although there is much to learn about damage resistance and durability. The lack of ductility at room temperature suggests possible problems with handling and a need r new design practices, as noted in the TiAl section. ost preliminary creep, fatigue, and stress-rupture properties are encouraging. At the present time the material has only been made in small quantities and very little processing and heat-treatment information is available, so it is not possible to begin an a~ordab~lity assessment. One encouraging recent finding from atta et al. (1999) is that near-net-shape casting ogies may be feasible for this system. Using s~andardtitanium casting and proprietary mold and owmet Corporation was able to fill plates having a n i i n i m u ~thickness of 4 mm with essentially no inold reactions. The Nb-Si-base in-situ composites discussed above, base alloys, are two promising paths to higher turbine operating tei~perature~ and greater efficiency. At present, these materials have reached the stage where they are competitive with current alloys. However, the real payoff is at higher material temperatures ( > 1300 "C) and at these temperatures tlie oxidation resistance is inadequate, although work continues (Jackson, 1999). Only if such progress is made will the engine businesses make the necessary investment to bmig these materials into use.
Volume 2 of this series included reviews by Anton (1995) on the role of the Ni,Al phase in superalloys, and by Liu and Pope (1995) on the development of ~onolithicNi,Al-based structural materials. At the end of their review, Liu and Pope listed research and development goals critical for the successful application of such materials. These included u~ders~anding and eliminating the environmental e ~ ~ b r i t t ~ e mthat ent occurs at 600-800 "C in oxidizing atmospheres, and the control of moistur~-induce~ grain~boundaryembrittlernent near room temperatLIre. Im~rovementsin hot cracking and weldability of B-doped Ni,A1 alloys and the need for the development of strong single-crystal and directionally-~olidified Wi,Al alloys for turbine engines were also cited. Progress has been made in some of these areas. owever, the bottom line is that Ni,Al-base alloys still do not offer a s u ~ c i e n t performaiice advantage in aerospace components to justify their use. Some cast alloys are affordable and have found niche applications, although not in turbine engines. Application in inill equipment, l~eat-treat~ient fixtures, forging dies, etc. has been successful. Large heats, up to 2267 kg have been melted of the casting alloy IC221M (Ni-15.9Al-8.OCr-0.8Mo Zr-O.04B a/o) (Mi$.OA1-7.7Cr-1.43M0-1.7Zr-0,00 wt%), which is believed to be the best cast compositioii. The properties are given in Table 2 (Sikka et al., 1997). Tri-nickel alumini~ealloys are margiiially workable, but the ~ r o u g h t processing is uneconomi~al for current applications and probably the use of superalloy powder technology is needed. The elevated temperature embrittle~entof Ni,Al can be minimized by the addition of 6-9 at.% Cr. Liu (1993) gives a general formuh for elevated temperature structural Ni,Al alloys to be used in hostile environments and notes that up to 20% of CO-I-Fe may be substituted for Ni to iniprove corrosion resistaiice. Such alloys may contain up to 15% of the disordered y phase. Alloying with MO, CO,and Fe causes a large lattice parameter mismatch between the y and y' phases, resulting in increased short-term strength, but long-term strength Alloys based on the compound Ni3Al have created loss, because this lattice mismatch is the driving force much interest over the last two decades because of the for particle coarsening (Stoloff, 1987). George et al. potential for creating a material with increasing yield (1993) showed that the B addition to Ni,AI teiided to streii~thwith temper~~ture. Some years ago, Aoki and suppress the environmenta~effects of moisture in the Izumi (1979) showed that a small B addition to air and tended to promote transgranular fracture; it subs~oichiometric(A1 deficient) compositions resulted was also shown that Zr considerably improved grain in considerable ductility at room temperature. Liu et boundary cohesion. Li and Chaki (1993) reported that al. (1985) showed that the ductilizing effect was tungsten-arc welded pl es of 1C-396M (Ni-15.8Alrestricted to these compositions and was ineRective in 8.3Cr-1.7Mo-0.5Zr-0.05 a/o) were susceptible to stoichionietric or superstoichiometric A1 compositions, severely restricting possible alloying opportun~~ies. cracking in the he~t-affectedzone (HAZ).
-
493
High- Tempera ture Structural Applications
Properties of nickel alunmide cast IC-22lM (reproduced from Sikka, 1997) with permission from the Materials Research Society Temperature ("C) Property Deiisity (g cm-') Hardness (R,) Microhardness (DPH) Modulus (GPa) Mean coeff. of expaiision OC-') Thermal conductivity (W m-' I C * ) 0.2% Tensile yield strength (MPa) Ultiinate tensile strength (MPa) Total tensile elongation (%) 102h Rupture s~rength(MPa) 103h Rupture strength (MPa) 104h Rupture strentgh (MPa) Charpy impact toughness (J) Fatigue strength 106cycle life (MPa) Fatigue strength 107cycle life (MPa)
Room 7.86 30 260 200 12.77 11.9 555 770 14
40
200
400
600
800
900
270 190 13.08 13.9 570 800 14
280 174 13.72 14.7 590 850 17
290 160 14.33 20.3 610 850 18
280 148 15.17 25.2 680 820 5 252 172 I24 1s
230 139 15,78 27.5 600 675 5 124 83 55 10
40
40
35 630" 550"
1000
120 126 16.57 30.2 400 500 7 5s 36 24
1100
114 200 200 10 28 28 11
"650 "C, investnient-cast test bars.
Sikka (1997) reviewed the commerciali~ationstatus of Ni,Al-based alloys and concluded that cast components are tlie primary applications. He mentions the use of casting to produce furnace fixtures, forging dies and pipes. The pipes being produced today are 36.2 cm dia and 685 cni long (National Materials Advisory Board, 1997), have served as transfer rolls since 1997 and recently an order was placed for 100 more (Furey, 1999). Although the initial cost of these transfer rolls is high, the lifecycle cost is reduced considerably. In a later paper, Siklca et al. (1997) reviewed the state of compositional optimization, melting process development, casting process, welding process, weld repairs and thermal aging response of the composition discussed above, lC-221M. More than 50000 kg was cast into a variety of coi~ponents in 1996, and several suppliers continue to cast components. Recently, Han et al. (1997) reported on a strong Ni,Al-base alloy produced by directional solidi~cation: Ni- 16.3A1-8.2Mo-0.2B (at. " 0 ) (Ni-7.8121-13.9Mo-0.04.B wt.%). The static yield strength of the alloy is quite high (990 MPa at 700 "C, 600 MPa at 1000 "C and 390 MPa at 110OoC).The stress-rupture life is about six times that o f PWA 1422 at 1100°C and 88 MPa. The excellent mechanical properties of this Ni,Al alloy are attributed mainly to solid-solution hardening by MO and second-phase strengthening by 15 to 20% y phase with a lattice mismatch of 1.185% between the y' and y phases. As noted above, such large mismatch implies a
large stored energy, which can drive microstructural coarsening. ~ h ~ t o ~ ~ c r o ginr athe ~ hgaper s by Man et al. (1997) clearly indicate that tlrre precipitate microstructure coarsens by about an order of magnitude after 190 h stress-rupture exposure at 1100 "6. No data are given for room-temperature properties following high-temperature tests. In any consideration of alloys for high-temperature structural service to replace today's Ni-based superalloys, the complete property set for the new material must be available for comparison. Such a complete range of properties for a single Ni3Al alloy is not available thus far, although Table 2 provides many useful data for comparison, So where do we stand'? First, some of the goals set some years ago by Liu and Pope (1995) have iiot been met and may not be achievable. The elevated-te~peratureductility loss and the room-tempera~uresensitivity to moisture have been lessened, but not eliminated. Some alloys liave been shown to be workable, but the co~merciali~ation of the alloys has proceeded almost solely with cast materials. A family of established superalloys exists that show static and dynamic properties superior to those thus far developed with Ni,Al-base materials. These standard materials have an extensive property base and a long history of reliability, agordability, and durability. Thus, it would appear that the standard superalloys should continue to be most useful in turbine engines unless a major breakthrough in Ni3Al-base systems occurs.
iAl has been considered as a potential liigh-teniperature structural material for d on a melting point of 1638"C, a and an elastic modulus niore 1000 "C. In addition, since NiAl is the basis for several oxidat~on-resi~tant coatings applied to conveiitional nickel-base super resistance is quite acceptable. been plagued by two persistent resistance, and brittleness at low and intermediate
Also in Volume 2, Miracle and
while significant improveinent s had been made in tensile strengtls and ductility of single crystals, a single alloy with both ducti~ityand sufficient high-temperalure streiigth had yet to be developed. In addition,
componeizt and covered a of impact parameters. Tests conducted on both single crystals and a NiAl eutectic alloy showed that for impact velocities that might be encountered by a vane, some conditions produced failure, while others caused no apparent damage. However, failure occurred under nearly all conditions typical for blades > 2.00 mm thick. The authors concluded that NiAl alloys were not good candidates for turbiiie blades using today's design ~ e t h o d ~ l o g i ehowever, s; ~ ~ p ~ l i c a tas i o an turbine vane may be feasible. Noebe and Walston (1997) reviewed recent develAl-base materials. Again it was concluded that polycrystalline NiA1"base alloys were not good candidates to replace Ni-base superalloys and that ~ber-reinforced NiAl composites had not lived up to expectations for lack o f a compatible fiber. Limited ductility, low fracture toughness and poor impact resistance were cited as fkctors hampering widespread application. The authors pointed to TiAlbase materials, that had been successfully engine tested, with ductility (1-2%) and fracture toughness ( > 15 ~ ~ a j as m us ) goals, but also concluded that the creep strength i-based superalloys must be attained .
Z Z ~ ~ ~ ~ oxide or other ceramic particles in NiAl can significantly improve creep behavior to temp of 1200°C or higher (Arzt and Grahle, 1995 et al., 1995; Carg et al., 1997). Arzt and G 5) have shown that such particle insertion also raises the e toughliess from about 4 M Y a j m to 8aJm. These techniques, however, do not result terial with both r o o ~ - t ~ ~ p e r a tductility L ~ r e and acceptable lii~h~temperature creep resistance. Lee et al. (1995) studied the strain-rate sensitivity of both single crystals and fine-grained polycrystalline rain sizes in the range 3.4-62 pm. pecimens were tested in tension at 877°C. Ductile fracture resulted at the slower strain rates of 0.0001 s-' and 0.1 s-l, but all grain sizes showed essentially zero ductility and c~eava~e"typetransgranular fracture The authors conclude surkces at a strain rate of 1 SKI. that ~ i s $ o c a tnii~o ~ t i ~ nis severely restricted at the in rate. Walston et al. (1997) and Walston a ( 1 conducted ~ ~ balli~tic ~ impact ~ tests using condi~ions to simL~latethose in the turbine section of aircraft engines. The tests were conducted at about 980 "C using cold, ~ l ~ ~ m i i i u ~ - aspheres l l o y to simulate the strength o f an impacting s u p e r ~ l l ~ y
Ti,Al-base alloys were developed niore quickly than other i n t e r ~ e ~ ~ lsystems, lic because ~icrostructLires and heat treatments were s i n ~ i ~to~ rs t a ~ d a ~a-,8 d titanium alloys. However, as more beta-stab~~izing elements are added this connect~vitybreaks down as new phases, such as orthorhonibic Ti,AlNb, are formed and complex reactions become prevalent. Ti,Al"ba~ealloy develop~entwas reviewed by Lipsitt al. (1993) review reflected the oi-thorlmmbic ~ ~ ~ A l N b - ~ ~ s e
coinprehensive account of the physical ~ e ~ ~ ~ofl ~ ~ r g y such alloys was subse~uently published ( ~ a n e ~ j e e , 1997)' Nandy and Banerjee (1997) reviewed the mechanical beliavior of Ti,AlNb and most recently Gogia et ul. (1998) reviewed the microstructure and mec~anicalpro~ertiesof orthorhom~icalloys in the jee (1995) noted that alloys based on Ti,Al d farthest of the {also called az) had ~ r o ~ r e s s ethe engineering inter~etallics,with several compositions achieving specific-streiigth d rupture capabilities superior to I ~ 718Cand~1 I 834 or Ti-1100 in the
temperature range 500-700 "C, Two such alloys were being produced in production mills, and aiicillary processes such as sLiperplastic forming and diffusion bonding were well established. e cautioned, however, that creep resistance, fracture toughness and impact tolerance were not adequate. These alloys were readily to surface enibrittlement and cracking under stress, a characteristic found in even the newer high-Nb alloys. Banerjee et al. (1993) concluded that the understaiiding of the metallurgy of Ti,Al-base alloys had progressed sL~~ciently that it was clear that the best balaiice of properties was achieved when there added so that some orthorho~bic se was present. This phase imparted increased toughness, liigh-temperat~restrength and stress-rupture resistance. Gogia et al. (1998) expanded upon the state of development o f these alloys. Significant da tabases exist on the tensile and creep behavior of this class of alloys (in the range 222-2641 and 15-30Nb) and demonstrate that reasonable coinbinations of ductility and strength can be realized provid the final aging temperaNb levels improve the ture is above 650°C. combination of stre ctility, toughness, and creep resistance, but also increase density and su ibility to the destabili~ation of a fine 0 - k structure by a cellular decomposition reaction optimum Wb contents in ttiesc orthorlionibic-phasebased alloys in terms of alloy stability and the balance of properties has not been established. In addition, sigtna phase appears at Nb levels beyond 30%. Solidification-inducedmicro-segregation also increases so that it i s difficult to homogenize alloys, and processing becomes more problematic at high aluminum contents. Sucli high Al levels also result in losses ness and ductility. The fracture toughness of these alloys lies in the range of 2540 MPaJm aiid the density-conipensated yield strength is better than that of IN 718. However, the impact properties of these alloys are very low, so dynamic applications inay iiot be advisable. Specific stress-rLipture properties and steady-state creep rates are comparable to those of IN 718, but primary creep strains are significantly higher in the orthorhoinbic alloys at comparable specific stresses and temperatures. In addition, coatings might be necessary to prevent the dynamic e ~ b r i t t l e ~ ~phenomena ent that occur at high temperatures Several studies (Smith et al., 1992, 1993; Chatterjee et al., 1997) used orthorhombic alloys as the matrix material for a fiber-reinforced composite. The presence of the fibers improves all the tested properties
and removes the mechanical property deficits that would keep these iiiaterials froin use in aircraft turbine engines. However, such composites still suffer dynami~~mbrittlementand may need a coating to prevent it. Although these alloys have progressed significantly, and many properties are now adequate for turbineengine service, problems remain. oxidation behavior, for oxygen ingress into the surface can cause severe cracking. Thus, coating protection would be mandatory for service in oxidizing atmospheres. Moreover, the coatiiig would also need to be a diffusion barrier for oxygen to avoid embrittlement beneath the coating. The final question then becomes one of deciding whether the temperature advantage gained (perhaps as much as 100 "C) over conventional titanium alloys is worth the cost. point in time, is negative. its
S
The microstructure and properties of AI,Ti (t18, and its alloys have been reviewed twice in th decade, by Kumar (1993) and ~ a ~ a g u c and h i lnui (1995) in Volume 2 of this series. c o ~ m e n t ~on d the problems to be solved before these materials could serve as strL~ctura1elements. Yaniaguclii and Inui state that microalloying with some ternary elements does cause the base compound to become somewhat more deformable, but the ductility has not been improved sL~bstantial~y. They also discuss the properties of A1,Ti inacroalloyed to convert the form to a stable Llz s t r L ~ c t uWhile ~~. some slight ductility in bending and considerable ductility in compression have been observed, these cubic com~ositionsare still brittle in tension. Unfortunately, RT yield strengths of this class of materials are low, and so they are iiot suitable for consideration for high-temperature service. V reported that the fracture to cubic trialuniinide increased value of -4.5 ~ P a J m to between 200 and 500 "C. The to a value of -4.5 MPaJm at 1000 "C. These values are much lower than the standard (a n i i n i m u ~of 20 ~ ~ a used ~ in m turbine ) engines, (1991) and Whittenberger et al. ( compression studies of coinposites of A1,Ti-based L1 compounds containing Ti (W3> C32) particles. Below 600 "C, their best specific strength was approximately four times that of a superalloy. This static strength advantage is not maintaiiied to higher
496
Applications
temperatures, although the creep behavior of the material rivals that of superalloys up to about 830°C. Alloys from this system are presently not considered useful structural ~aterials.Further development could yield useful oxidation-resistant coatings, but an unanticipated breakthrough would be needed to make them suitable as load-bearing materials for turbine engines.
hese two compounds and their alloys have some interesting properties, but offer no strength or stiffness advantage over current materials. Vedula (1995), in Volume 2 of this series, reviewed characteristics of the base compounds and some derivative alloys. When correctly processed these alloys possess adequate low temperature ductility, but the low density and excellent oxidation and corrosion resistance of these materials are their strong points. Creep and stress-rupture resistance are poor. For this reason, Vedula (1995) concluded that the potential application temperature range must be more modest than for nickel-base su~eralloys.Attempts to improve high temperature properties have been made. Sikka (1991) reported a minimum creep rate for tric Fe3Al of about 1 V 2% h-' at a stress of nd 500°C and for Fe-35A1 a ~ i n i n ~ creep rate of about 0.5% h-l at the same stress and 600 "C. Morris and Cunther (1997) reported that a 1O/O echanically alloyed material, Fe-40Ai ), showed a strain rate of about 5 x 10-5 s-l at 700 "@ at a stress of about 250 MPa. Although this is a considerable improvement, it is still not suflicient to be attractive for turbine engines. In s u m ~ a r y ,the mechanical-property balance of these materials is useful, but the specific properties are not better than many standard eiigine materials. No large heats of this material have been made, so scale-up behavior and a ~ o r d a ~ i l i tremain y to be established. These materials may be useful in situations where oxidation and corrosion resistance are needed, but high s t r e ~ ~ ist hnot required. It seems unlikely that these materials will be further developed for turbine engines.
In this chapter we have tried to describe the steps necessary to prepare a new material for commercial usage in demanding applications. Our objective was to provide insight into the myriad factors and require-
ments that must be satisfied before the ~~rtification and transition to service of any new material can occur, ~e have also reviewed the present status o f development of a nuinber of i n t e ~ e t a l compounds l~~ and alloys, and we have shown where the present developinental status of these niaterials is deficient. It is o ~ i hope r that this information can help provide a more robust frameworl~for the continuing development of new high-temperature materials.
Aikcn, R. M. (1993). In Slructural Internzetallics, (eds R. Darolia, J. J. Lewandowskr, C. T. Lm, P. L. Martin, Miracle, and M. V. Nathal) TMS, ~arreiidale,PA, p. 791. Akmc, M., Meyer, M. IS., Kramer, M. J., Thom, A. J., Huebsch, J. J., and Cook, B. (1999). Muter. Scz. & Engr+ A261, p. 16. Anton, D. L. (1995). In I n ~ e r ~ e f a l l~i c o ~ ~ o -u Practice n d ~ ~ (eds J. H. Westbrook and R. L. Fleischer). Wiley, Chichester, UK, p. 3. Anton, D. L., and Shah, D. M. (1991). In Nigh ~ e ~ 2 ~ ~ r a t u r e Ordered Inte~nze~a[lic Alloys IV (eds L. A. Johtisoii, D. P and J. 0. Steigler). Mafer. lies. Soc. Symp. Proc.,
u ~
umi, 0. (1979).
on ~
~ i ~ ~z ~o~ ~ ku ~
ishi,
Appel, F. and Wagner, R. (1998). Mat. Sci. & Eng., 268. Arzt, E. and Grahle, P. (1995). In High T e ~ ~ e r a t u O r er ~ e r ~ ~ r ~ t ~ r ~ Alkoys e ~ ~V l (cds ~ ~ J.c . Hortoii, X. Baker, S. Schwartz). Mat. Res. t
Austin, C. M., Kelly, T. J., McAllister, K. G., and Chesnutt, J. C. (1997). In Struct~ralInternze~all~cs 1997, (eds M. V. Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. Yarnaguchi). TMS, Wnrrendale, PA, USA, p. 413--425. Banerjee, D. (1995). In I ~ ~ t e r ~ ~ ~~ a lol i c$ ~ - ~Practice ~ o u ~ ~ ~ (eds J. €3. Westbrook and R. L. Fleischer). Wiley, Chichester, UK, p. 91. Banerjee, D. (1 997). Progress m ~ a t e r i Science, ~~l~ Banerjee, D., Gogia, A.K., Nandy, T. K., M~~raleed~aran, IS., and Mishra, R. S. (1993). In Struct~ra[I n t e r ~ e t a [ l ~(eds cs R. Darolia, J. J. Lewandowski,C. T. Liu, P L. Martin, D. B. Miracle, and M. V. Nathal). TMS, ~ a r r e n d a l ePA, , p. 19. Berczik, D. M, (1997a). U.S. Patent Number 5,595,616, United Technologies Corporation, Issued: January 21, 1997. Berczik, D. M. (1997b). U.S. Patent Number 5,693,156, United Technologies Corporation, Issued: Dccember 2, 1997.
High- Temperature Structural Applicatiovts
497
Vasudevan, and H. A. Lipsitt). Mater. Res. Soc. Symp. Proc., 322, p.203. Furey, S. C. (1999). Private Communication, Septcmber 17, 1999. Garg, A., Whitteiiberger, J. D., and Luton, M. J. (1997). In High Temperature Ordered Intermetallic Alloys VII (eds G. C. Koch, C. T. Lm, N. S. StoloR, and A. Wanner) Mat. Res. Soc. Symp. Proc., M. R., and Subramanian, P R. George, E. P., Liu, C. T., D. P. (1993). In High Temperuture Ordered ~nte~metallic Alloys V (eds I. Baker, T i t ~ ~ n i uAm~ u ~ ~ n iI999 d e s (eds R. Darolia, J. D. Whittenb~rger,and M. H.Yoo). Mater. Y.-W. Kim, D. M. Dimiduk, and M. Loretto). TMS, Res. Soc. symp. Proc., 2 Warrendale, PA, USA, p. 517-525. Gibala, R., Ghosh, A. K., Van Aken, D. C., Srolovitz, 23. J., Chatterjee, A., Roessler, J. R., Brown, L. E., Heitman, P. W., Basu, A., Chang, H., D. P., and Yang, W. (1992). and Richardson, G. E. (1997). In ~tructurulIntermetallies Mat. Sci. & Eng., AI 1997 (eds M. V Nathal, R. Darolia, C. T. Llu, P. L. Gogia, A. K., Bandy, T. IC., Banerjee, D., Cansey, T., Martin, D. B. Miracle, R. Wagner, and M. Yainaguchi). Strudel, J. L., and Franchet, J. M. (1998). Infer~etallics, TMS, Warrendale, PA, p. 905. p. 741. Chen, G. L., Zliaiig, W. J., Liu, 2;. C., Li, S. J., aiid Kini, Y.-W. Gouma, P. I., Mills, M. J., and IGm, Y.-W. (1998). Plzil. Mag. (1999). In Gamma Titanium Ahinimides 1999 (eds Y -W Ltrs., 78, 59-66. K h , D.M. Dirniduk, and M. Lorretto). TMS, ~ a r r e n ~ l e , Nan, Y. F., Xing, Z. P., and Chaturvedi, M. C. (1997). In PA, USA, p. 371-380. S~ructuralInter~etallic.~ 1997(eds M. \I.Nathal, R. Darolia, Cheng, T, T. (1999). 11%G a ~ Titunitkm ~ a A/uminides 1999, C. T. Liu, P L. Martin, D. B. Miracle, R. Wagner, andM. (eds Y.-W. Kim, D. M. Dimiduk, and N. Lorretto). TMS, Yaniaguchi). TMS, Warrendale, PA, p. 713. Warrendale, PA, USA, p. 389-396. Hebsur, M. G., Whittenberger, J. D., Lowell, C. E., and Clemens, Ha, Kestler, H., Eberhardt, N., and Knabl, W Garg, A. (1995). In High Temperature Ordered (1999). In Gamma Titanium Alummides 1999, (eds Y.-W. I n t e r ~ e t a l l ~Alloys c V I (eds J. A. Horton, I. Baker, S . Kim, D. M. Diniiduk, and M. Lorretto). TMS, Hanada, R. D. Noebe, and D. S. Warrendale, PA, USA. p. 209-223. Soc. Svntp. Proc., 79. Darolia, R., Lahrman, D. F., and Field, R, D. (1 992). Scripta. Hebsur, M. C., and M. V. (1997). In Structurul Inter~etallics1997 . V Nathal, R, Darolla, C. T. Dlmiduk, D. M. (1995). In ~ a m ? n Ta i t u n z u ~Aluminides, (eds Liu. P. L. Martin, Miracle, R. Wagner, and M. Y.-W. Kim, R. Wagner, and M. Yamaguchi). TMS, Yamaguchi). TMS, Warreiidale, PA, p. 949. Warrendale, FA, USA, p. 3-20. Huang, S. C. and Chesnutt, J.C. (1995). In I~termetallic Dimiduk, D. M. (1998). I ~ t e r ~ e ~6 ,( ~ ~ ~ ~ c ~ ~ , Compounds: Principles and Practice, Volume 2 (eds J. W. Dimiduk, D. M. (1999). Mat. SCE.& Eng. Westbrook and R . L. Fleischer). Wiley, Chichcster, UK, Dimiduk, D. M., Martin, P. L., and Kirn, p. 73-90. Jackson, M. R. (1999). Private Communi~a~ion, 27 October Dimiduk, D. M., McQuay, P. A., and Kim, Y.-W. (2000). 1999 ~ i t u n i u’99 ~ Science and Technology,p ~ o c e e d i ~ofthe g s 9th Jaclrson, M. R., Rowe, R. C., and Slielly, D. W. (1995). In Titanium World Confi.rence, St. Petersburg, Russia, Juiie High T ~ ~ p e r a t u Ordered re I n t e r ~ e t a l l ~Alloys c VI ( 1999. CRISM “Prometey7’Publications, Saint Petersburg, A. Horton, I. Baker, S Han Russia, pp. 259-268 (ISBN 5-900791-06-8). Schwartz). Mater. RYS.Soc. DiPietro, M. S., Kurnar, K. S., and Whittenberger, J. D. Jackson, M. R., Bewlay, B. P., Rowe, R. G., Skelly, D. W., (1991). J. Muter. Res., and Lipsitt, H. A. (1996). J. Fecke, T. and Davidson, D. E. (1998). JO Kirn, Y.-W., aiid Dimiduk, D. Feng, C. R. and Sadanaiida, K. (199’7). In IGm, Y.-W., a i d Dimiduk, D. M. (1996). U.S. Patent Ordered Intermetullic Alloys VII (cds C. C. ICoch, C. T 5,558,729, 24 September 1996. and A. Wa~ner).Mar. Res. Soc. S-vmp. Kiin, Y.-W., and Dimiduk, D. M. (1997a). In Cleavage Fracture, George R. Irwin Symposium, (ed M.S. Chan). Fleisclier, R. L. (1995). In Infermetallic Compounds - Practice TMS, Warrendale, PA, p. 305-317. (eds J. H. Westbrook and R. L. Fleischer). Wiley, Kim, Y.-W., and Dimiduk, D, M. (1997b). In Structura~ Chichester, UK, p. 237 French, J. D., Wiederhorn, S . M., and Petrovic, 5. J. (1994). I~termetallics1997 (eds M. V. Nathal, R. Darolia, C. T. In High Temperature Silicides and Refractory Allovs (eds Lm, P. L. Martin, D. B. Miracle, R. Wagner, and M. C. L. Briant, J. J. Petrovic, B. P. Bewlay, A. K. Yamaguchi). TMS. Warrelidale, PA, USA, p, 531-543.
Berczik, D. M. (1998). In Proceedings of the U.S. - Japun ~ ~ o r ~ons Very ~ o IHigh ~ Temperature Structural Materials, Turtle Bay, Hawaii, 9-11 December 1999. Private Cominunicatiou. Bewlay, B. P., Jackson, M. R., and Lipsitt, W. A. (1996). Metall. & &later. Tram. A., ski, J. J., and Jackson, M. R. (1997).
ini, U*-W., Wagner, R., aiid Yai~aguchi,M., eds (1995). Gumma ~ i f a n z uA1umz~~~e.s. ~ TMS, Warrendale, FA, USA. im, Y.-W.. D i ~ i d u k D. , M., and Loretto, M., eds (1999). ~ u n z r nTituni~m ~ ~ ~ ~ ~ ~ 1999, n i dTMS, e , s Warrendale, PA, USA. . A., Beuth, J. L., and Milke, J. G. (1999). ~ e t u l l . aid Mater. Tram A, umar, K. S. (1993). In Str~cturulZ'nter7~e~ffllics (eds R. Darolia, J. J. Lewaiidowski, C. T. Liu, P. L. Martin, D. B. d M. V. Nathal). TMS, Warrendale, PA, p. 87 (1995). In Interwteta~~jc C o m p ~ ~ ~-~Practice d,s Westbrook and R. L. Fleischer). Wiley,
Larsen, J. M., Roscnberger, A. H., Worth, B. D., Li. K., Porter, W J. (1999a). In G a t m a 1999 (eds Y-W. Kim, D. M. 5 i ~ i d ~ kand , M. Loretto). TMS, Warreiidale, PA, USA. p. 463-472. ILarsen, J. M., Roscnberger, A. Id., Li, K,, Porter, W. J., and John, R. (1999'17). In ~ ~ ~ i g'99u-ePro~eedingsof the Se ~e n t lInternu~Io~al ~ ~ a t i g u eCovlJlerence (eds X. R. Wu, and 2. 6. Wang). Higher Education Press, Beijing, P R. China, and Eiiginecring Materials Advisory Services Ltd.,
Mills, and M. Uamaguchi). Mat. Res. Soc. Symp. Proc., , p. ~ K ~ . 9 . 1 - K ~ 2 . 9 . 6 . Lee, J. I., Hecht, N. L., and Mah, T. (1998). J Am. Cer Soc., 81, p.421. Lee, T.-S., Hon, W.-P., Yang, S.-C., Hsu, S.-E., Chin, S., and Anton, D. L. (1995). In High T e ~ p e r ~ ~ t uOrdered re ~ n t e r ~ e t a l l iAlloys c VI (eds J. A. Horton. I. Baker, S. Hanada, R. D. Noebe, and D. S. SchwartA). Mat. Res. L e ~ a n d o ~ s kJ.i ,J. (1999). In Fatigue '99 - Proceedings ofthe S e w t t h Internutioizal Futigue Confermce (eds X. R. Wu, and Z. 6. Waiig). Higher Education Press, Beijing, P. R. Clitna, and ~ngincerii~g ~ a t e r i aAdvisory ~s Services Ltd., West Midlands, UK, p. 1471-1477. Li, H., and Chaki, T. K. (1993). In High Temperuture Ordered ~ ~ ~ e r ~Ailoys e ~ aV l (eds ~ j I. ~ Baker, R. Darolia, J. D. Whittelibe~ger,and M. H. Yoo). ~ u t e r Res. . Soc. Symp. Lipsitt, E. A. (1993). In High Temperuture Ordered rnierwtetu~licAlloys V (eds I. Baker. R. Darolia, J. D. . H. Uoo). Mut. Res. Soc. S,ywtp.
Temperu~i~re Ordered Interrne~allic Alloys V (eds I. Baker, R. Darolia, J. D. Whittenberger, and M. H. Yoo). Mater. Xes. Soc. Sjmp. Proc., L., and Horton, J. A. (1985). Acta
Mason, D, P., and Van Aken, D. C. (1993). In ~iglz Temperature Ordered I n t e r ~ ~ e t a ~AIZOJAS lic V (eds I. Baker, R. Darolia, J. D. Whitte~berg~r, and M. H, Uoo). at er. Res. Soc. S y ~ Proc., ~ . Menand, A. (1999). In ~ a ~ T~tun~um n ~ a Y.-W. IGm, D. M. Dimrduk, and M. Loretto). TMS, Warrenda~e,PA, USA. p. 3 11-124. i ~ nParthasarathy, , T. Mendiratta, M. G., ~ u b r ~ ~ ~ aP.n R., A., and Dirniduk, D. M. (1999). Unpublished, Air Force Research Laboratory, 29 October 1999. Meyer, M. K., Kramer, M. J., and Akinc, M. (1996). Meycr, M . I<., ;md Akiric, M. (1996). J. Arnev. Cer. Soc., 79(4), 938. Miracle, D. B. (1993). Actn ~ e ~ a lMater., l. Miracle, D. B., and Darolia, R. (1995). C o ~ p o u -~ Practice i ~ ~ (eds J. H. Westbrook and R. L. Fleischer). Wiley, Chichester, UK, p. 53. Miracle, D. B., and Mendiratta, M. G. (1995). In I n t e r ~ e t ~ ~ l l i c Compounds - Practice (eds J. FE. Westbrook aiid €3. L. Fleischer). Wiley, Chichester, UK, p. 287 Morris, D. G., and Gunther, S. (1997). In ~i~~ T e ~ ~ e r u t ~ r e Ordered Z'~ter~~eta11ie Alloys VZ'I (eds C. C. Koch, C. T. and A. Wanner). Mut. Res. Soc. Syrnp.
Naka, S,, Thornas, M., Sanchez. C., and IChan, T. (1997). In S t r ~ ~ c t u r aZl ' n t ~ r ~ i e1997 ~a~ (cds ~ ~M. ~ ~V. Natlial, R. Darolia, C. T. Liu, I". L. Martin, D. B. Miracle, R. Wagiier, and M. Y a n i a g ~ ~ ~TMS, ~ i ) . ~ a r r e n d ~ l PA, e, p. 313. Nandy, T. K., and Banerjoe, D. (1997). In ~ ~ r ~ ~ ~ ~ r a I ~ ~ e r ~ e ~ u 1997 l l i c ~ (eds s M. V. ~ a t h a l R. , Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. Ytunaguchi). TMS, Warrendale, PA, p. 777 National Materials Advisory Board (1997). In Z~ztermetallic Alloy ~ ~ v e l o A~ Program ~ ~ e ~ ~~v n, l u a ~ National i~~. Academy Press, Publication N ~ ~ ~ - 4 8 7Washington, -1, D.C., p. 23. Noebe, R. D., and Walston, W S. (1997). In S ~ r ~ c ~ u r a l lnter~etallics1997 (eds M.V, Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. Ya m a g ~ ~ c hTMS, ~ ) . Warr Olson, G.B. (1997). Scierzce, Percpezko, J. H,, Nunes, C, (1997). In High T e ~ ~ ~ e r uOrdered ~ u r e Z ' ~ i t e r ~ e ~ ~Allop 1lic VIZ' (cds C . C. Koch, C . T. Liu, N. S. StoloflF, and A. Wanner). Mater. Res. Soc. Symp. Pro
Petrovic, J. J., and Vasudevan, A . K. (1994). In High Tem~eratureSilicides and ~ e f r a c t o r yAlloys (eds C. L. Brtant, J. J. Petrovic, B. P. Bewlay, A. K. Vas LIU,C. T. and Pope, D. P. (1995). In I n t e r ~ ~ ~ tC~ o~ I~~ pi co ~ ~ n H. ~ ~A.~Lipsitt). Mater. Res. Soc. Symp. Proc., Pope, 5. P., and Darolia, R. (1996). M R S Bulletin, 21, p. 30- Practice (cds J. €3. Westbrook and R. L. Fleischer). 36. Wiley, Chichester, UM, p. 17.
Vedula, K. (1995). In ~ ~ ~ ~ e r ~ Ce t ao l l~~ c~- Practice ~ ~ i(eds ~ ~ s R a m ~ ~ ~ u r t U., h y ,Kim, A. S., Suresh, S., and Petrovic, 5. J. J. H. Wcstbrook and R. L. Fleischer). Wilcy, Chichester, (1994). J. Am. Ceram. Soc., 97, 1). 2681. UK, p. 199. Rugg, D. (1999). In G u m ~ l aAlloys 1999 (eds Y.-W. Kirn, D. M. Dirniduk, and M. Loretto). TMS, Warrendale, PA, USA, p. 11-14. Sadaiianda, I<., and Feng, C. R. (1994). In High Temperature Silicide~~ and Refkaciory Alloys (eds C. L. Briant, J. J. Petrovic, B. P. Bewlay, A. I(. Vasudevan, and H. A. Lipsitt). Muter. Res. Soc. S-yinp. Proc., 322, p, 157. Sargent, IC., and Huffman, M. (1999). Air Force Research Laboratory, Private Com~unication. USA, p. 387-404. i~~ e~s Saunders, N. (1999). In G u ~ ~~ f if ~ Af l f~ ~~~ iin 1999 (eds Y.-W. Kim, D. M. Diiniduk, and M. Loretto). TMS, Warrendale, PA, USA. p. 183-188. Schneibel, J. H., Liu, C. T., Easton, D. S., and Carmichael, C.A. (1999). Moter. Scr. & Engr., A26 Sikka, V. JS. (1991). In Heat Resi.~tunt M~terials(eds I(. Yamaguchi). TMS, p. 613. Natesan and D. J. Tillack). ASM I~ternational,Materials Park, OH, p. 141. Sikka, V. R. (1997). In High Temperature Ordered I n ~ e ~ i ~ e t a ~ l i c Alloys I'II(eds C . C. Koch, C. T. Liu, N. S. StolofE, and A. TMS, Warrendale, PA., p. 234. Wanner). Mater. lies. Soc. Syvnp. Proc., W i l ~ i a ~J.s ,C. (1994). 111~ t r u c t u r ~ B l t e r i ~ ~ ~ t a199 ~lics Sikka, V E=., Santella, M. L., and Orth, J. E. (1997). In High M. V. Nathal, R. Darolia, C. T e ~ ~ e r a t u Inter~etallics re (eds D. P. Pope, C. T. Liu, S. Mirac~e, R. Wagner, and g, and M. Yarnaguch~).Muter. Scr. dt E ~ g r . . Warrendale, PA, USA, p. 3-8. , p. 564. Worth, B. D., Larsen, J. M., and Roseii~)erge~, A. H. (1997) Sin , Quadakkers, W. J.. and Sheniet. V. (1999). In G u ~ i ~i~uniuvn ~ a ~ l u ~ ~ i n1999 ~ d e(eds ~ s Y -W. Kirn, D. M. Darolia, C. T. Liu, P. L. arti in, Dimiduk, and M. Loretto). TMS, Warrendale, PA, USA. p. 743-752. USA, p. 563-569. Smith, P. R., Graves, J. A., and Rhodes, C. G. (1992). In I n t e r ~ e t a l Matrix l ~ ~ ~ompositesII (eds D. B. Miracle, D. L. Anton, and J. A. Craves). at. Res. Soc. Sypnp. Proc., Miracle, aiid M. V. Nathal). TMS, Warrendale, PA, USA, p. 43. p. 885. Smith, P. R., Graves, J. A., and Rhodes, C. @. (1993). In Yaniaguchi, M.. and Inui, €3. (1993). In ~ t r u c t u r ~ Z S ~ r u c ~ ~ r uI nl ~ ~ r ~ e t a l l i c(eds s R. Daro Intervnetall~cs(eds R. Daroliia, J. J. L e w a ~ d o ~ s lC. ~ lT. , Lewandowski, C. T. Liu, P. L. Martin, D. Liu, P. L. Martin, D. B. Miracle, and M. V. Natlial). and M. V. Nathal). TMS, Warrendale, PA, p.765. TMS, WarrendaS~,PA, USA, p. 127-142. Srinivasan, S. R., Schwarz, R. B., and E ~ b u r yJ., D. (1993) In Yamaguchi, M., and Inui, H. (1995). In ~ n t e r ~ e t a l l i c High ~ e ~ p e r u t u rOrdered e Interinetallic Alloys I' (eds I. ~ ~ ~ p - oPractice ~ ~ (eds ~ dJ. s€3. Westbrook and Baker, R. Daralia, J. D. Whi Fleischer). Wiley, Chichester, Muter. Res. Soc. Svmp. Proc chi, M., and Umakoshi, Stoloff, N. S. (1987). In ~ ~ p e r o l l -148. Stoloff, and W. C . Hagel). Wiley, New York, p. 80-85. ma, S., and Yantaguchl, Subrainaman, P. R., Mendiratta, M. G., and Diimduk, D. M. 873-88 3. (1994). In High ~ e ~ p e r a ~ uSilic~des re and Se~ractory Yoshihara, M., and Kim, Y.-W. (1999). In G a ~ p ~n ~~ ~ f f ? z ~ u i n Alloys (eds C. L. Briant, J. J. Petrovic, B. P. Bewlay, A. A l u ~ ~ i ~1999 i ~ (eds ~ ~ ,Ys -W. Kiln, D. M. ~ i m i d u kand , M. IC. Vasudevaii, and H. A. Lipsitt). Mater. Res. Soc. Synp., Loretto). TMS, Warrendale, PA, USA, 322, p. 491. Yoshihara, M., and Kim, Y.-W. (2000). In Subrarnanian endiratta, M. G., and Dimiduk, D. M. High T e ~ p e r a t ~ ~r eo r r o ~ s i o r l - 2in~ p~ ~ , (1996). J. S(l), 33. Zinsser, W. A., Solv'yev, S., and L e w ~ ~ ~ ~ ~ oJ.wJ.s (1999). k i , In Suzuki, Y., a a, I<. (1997). Set. Eng. Compos. MGt., 6, 85. High T e ~ p e r u t ~ r~er d e r re n~t e r ~ ~ t a lAlloys l ~ c VII (eds Varm, R. A., and Zbromec, L. (1997). In High ~ e ~ p e r u t u r e E. George, M. Yamaguchi, and M. Mills). MRS Symp. Ordered Intermetall~eAlloys I'II (eds C. C. Koch, C. T. Proc. 552, p. ~K6.10.1-K~6.10.8. MRS, Pittsburgh, PA. and A. Wanner). Mut. Res. Soc. SJWZ~. ski, J. J. (1998). Metall. &
This Page Intentionally Left Blank
inany of the production heat-treating processes and in the ope~dtionof certain special ~ e v i c e § ~ Inter~etallicsof interest for structural ap~licationsfor general use include Ni,Al, iAl, Fe,Al, FeAl, Ni, oSi,. The general properties such as melting point, density, crystal structure and failure mode for each of the compoui~dsare summarized in Table 1 (Schulson, 1996). All of the intermetallic coinpounds listed in Table 1 have been of interest for ions for specific attributes that they the use of each of the interrnetallics their brittleness at room temperature and high brittl~-to-duct~le transitio~te~peratures. the last 20 years, significant progress has been made towards identifying the ~ausesof brittleness of the various c o ~ p o u n d sand in some instances, useful solutions have been developed, The causes for brittleness and proposed solutions for various coinpounds are discussed in later sections. ~ompositionsdeveloped from solutions to the brittlenes~of the ~ntermetallicbased alloy are described along with their processing and applications for general use in this chapter. The s t ~ c t uapplications ~~l for general use deal with all other than in aerospace and turbines. The general use refers to noii-critical applications such as furnace fixtures for heat treating or heating elements. Although non-critical in nature, the applications described in this chapter have made aiid are inaking a significant improvement in *See List o f C o n t r i b ~ t €or ~ r ~current address.
The intermetal~icc o ~ p o u n ~( s defined as ordered alloy phases metallic elenients (Anton et al., 1989), where aii alloy phase is ordered if two or more s ~ ~ ~ a t are t i cr ~e ~s ~ i r e d to describe its atomic structure (Fleisc~eret al., 1989, Fleischer and Taub, 1989). The ordered structure exhibits attractive elevated-tenipera~ure ~ r o ~ e r t i ~ s
than the coiiimon feature of long-range order, the crystal structure and the constit~entsof each intermetallic yield them unique attributes, which are described (Table 2) in the
tion at temperatures temperature ductility good high-temperatu
502
A ~ ~ l ~ c ~ t i o ~ ~ ~ asic propert~esof iuter~eta~libc compounds under d e v e ~ o p ~ efor n t structura~applications'
Inlerimtallic compound
Crystal structure
Density, p (g/cc)
Melting temperature ("'C)
Ni,A1 Ni,Si Fe& TiAl Ti,AI FeAI
Ll,, feeC,(CP4) Ll,, fcc', (CP4) DO,, lscc', (CF16) Ll,, tetra., (tP4) hexag., (hP8) 2, bcce, (cP2) B2, beee, (cP2) C1 I,, tetra., (tI6)
7.50 7.30 6.72 3.91 4.20 5.56 5.86 6.24
1400 1140 1540 1460 1600 1300 1640 2020
NiAl OSi,
Fracture mode* IGd IGd TGJ TG~ TCf I G+ ~TG f lGd+ TG IGd+ TGf
"From E.M. Schulson. "At r oom tein pera t ure. 'fcc=face-centered cubic. ~I~~~ntergranuI~r. %bcc=body-centcred cubic. JTG=transgraiiular cleavage.
~de~itificationof interinetallic-based alloys aiid weld-~llerwire for structural ~~pplications hiternietallic
Alloy identi~~catioiiCornmelit
Feature
et^^
Ni3A1 NiAl Ni,Si Fe,Al FeAl TiAl
Ni,Al
IC-50 3C-221M 16-438 -
--
FAS P GE-48-2-2 45XDTM 47XD'" IC-221LA IC-221w
available commercially from seveii commercial sources in the United States (see Section 6.1.1).
and wrought Gdst Cast No alloy for com~ercialization No alloy for commercialization Powder metallurgy Powder n~eta~~urgy Cast alloy Cast alloy alloy wire Weld wire Weld wire
a h c
d e
'~~uctility and fracture toughness. !I'ensile arid fatigue strengths and castability. 'Elevated temperature strength and castability. "Weld wire for cast repair of Ni,Al-based alloys for application temperatures < 800 "C. "Weld wire for cast Ni,AI-based alloy for application ~ e i ~ p erat~ires ~ 8 0 "G. 0 Wire is also useful for weld overlay deposits on steels, stainless steels. and other nickel-based alloys.
resistance to carburization, (c) an increase in yield strength with temperat~rethat peaks at about 850 "C, and (f) the possibility of creating nonconductive and wear-resistant surfaces by ~reoxida~ion at S 100 "C in air that results in the formation of A1,0, (Deevi and ikka, 1999). All of these attributes have resulted in many possible applications for Ni,Al-based ~tlloys. F u r t h e ~ i o r e various ~ products of these alloys are
lloy~ "lie NiAl-based alloys offer many interesting attributes: (a) a iiielting point of 1640 "C, which is nearly 200 "C higher than that of nickel, its higher meltingpoint constituent, (b) its density is 25% lower than that of advanced superalloys, (c) its thermal conductivity is two to five times that of advanced superalloys arolia, 1993), (d) excellent oxidation resistance (Nesbitt and Barrett, 1985; Nesbitt et al., 1992, Barrett, 1988), (e) carburization and coking resistance, and (f) lower cost tkan Ni,A1 and advanced superalloys. There are two major concerns with NiAl: (1) very low ductility at room tem~eratureand a brittle-toductile transition temperatu~eof 2 500 "C (Hahn and Vedula, 1989; Grala, 1960) and (2) low h i ~ i ~ t e m p e r a ~ ~ i r e strength (a800 "C). Even with very large research efTorts, the NiAl-based alloys have not reached a stage of any useful applications, and there are no commercial suppliers. However, they have potential to be used as injection molds, dies for glass manufacture, high-temperature heat excliangers, and as sensors based on their shapemeniory property.
2.3
Alloys
The most attractive attribute of these alloys is their low cost (McKaniey et al., 1991) because of the low-cost constituents. These alloys also possess (a) excellent
Structural Applications for General Use high-temperature oxidation resistance, (b) resistance to high-temperature gaseous sulfidation, (c) lower density than cast irons and steels, and (d) rust resistance at ambient temperature. Being of DO, crystal structure, the Fe,Al-based alloys are limited in their higher temperature strength to 600 "C (McKamey et al., 1991; Sikkta et ul., 1991; Sikka and Baldwin, 1992; Sikka, 1997). The Fe,Al-based alloys, in the wrought condition, havc a room-temperature ductility value of approximately 10% and a brittle-toductile-transition temperature of approximately 800 "C. One of the Fe,Al-based alloys is being used as a hotgas clean-up filter for coal-conversion systems. There is a commercial supplier of the powder for this alloy, and a commercial manufacturer produces tubular, porous metal filters by centrifugal casting of powder slurry followed by a sinterjng process (see Section 6.1.2). 2.4 FeAl-Based Alloys The FeAl-based alloys possess several interesting and useful attributes of: (a) high-temperature oxidation resistance (McKamey et al., 1991; Natesan and Tortorelli, 1997; Tortorelli and Natesan, 1998), (b) high-temperature sulfidation resistance (see the chapter by Natesan and Datta), (c) carburization and coking resistance, (d) low density, (e) good combination of electrical resistivity (Lilly et al., 1998; Reddy and Deevi, 2000), oxidation-resistance and strength for heating-element applications, and (0 low cost, based on the constituents. The FeAl-based alloys have the B2 (cP2) crystal structure as opposed to DO, (cF16) for Fe,Al-based alloys. The brittle-to-ductile transition temperature for FeAl-based alloys is close to 600 "C (Maziasz et al., 1997). They possess higher creep strength than Fe,Albased alloys. The FeAl-based alloy is in commercial use and there is both a commercial producer of powder and fabricator of sheet in the United States. There are also commercial producers of centrifugal cast tubes, although the intended applications are not fully developed (McQuay and Sikka, 2002) (see Section 6.1.3 for suppliers).
2.5 Ni,Si-Based Alloys These alloys have several very interesting attributes: (a) excellent resistance to sulfuric acid (Oliver, 1989), (b) high-temperature oxidation resistance (Ulvensoen et ME., 1993), (c) very high strength up to the iiiterniediate temperature of 600 "C, and (d) the selected
503
Ni,Si-based alloys compositions have good ductility at room and high temperatures (Oliver, 1989). There are potential applications of Ni,Si-based alloys in chemical industry; however, in-plant trials and accumulation of property data are still under way.
2.6 TiAl-Based Alloys These alloys are very attractive (Froes and Suryanarayana, 1996) because (1) their lowest density as compared to all other intermetallics, (b) high-temperature strength and the high specific strength. These alloys have L1, structure, possess very low ductility at room temperature, and have brittle-to-ductile transition temperatures exceeding 700 "C (Huang, 1993). Because titanium is a major constituent, the TiAlbased alloys are expensive and become even more expensive because of difficulties in thcir processing (Williams, 1997; Dimiduk. 1999; and chapter by Appel, Kestler, and Clemens in this volume). The structural applications for several uses of these alloys include turbochargers and engine valve seats. The recent TiAl alloys appear to possess a most promising combination of ductility, high creep resistance, and fracture toughness, and are candidates for turbine applications (Loria, 2000). Several turbine components such as low-pressure blades, radial diffusers, and compressor cases have been successfdly tested. Although extensively evaluated, neither of the applications stated above is being pursued in the United States. A report in the business and technology section of a major daily newspaper in Japan (1999) indicated that the TiAl-based alloy turbochargers are in use in Japan by Mitsubishi Motors (Deevi and Zhang, 2000). ABB and BMW successfully tested the turbochargers and exhaust valves (Stoloff et al., 2000).
3. Processing Technologies
The Exo-MeltTMprocess (Sikka, Deevi, and Vought, 1995; Sikka and Deevi, 1995; Deevi and Sikka, 1997) is the unique method for melting of Ni,Al-, Fe,AI-. and FeAl-based alloys. All three of these aluminides can be melted in air using the Exo-MeltTMprocess. The TiAlbased alloys require vacuum melting, details of whlch arc given in the chapter by McQuay and Sikka in this book. Details of large-scale melting of Ni,Si-based alloys have not yet been worked out. The processing of intermetallic-based alloys into shapes and product forms such as sheet, bar, and tube is described below.
Applications
504
3.1 Processing into Shapes by Casting Casting is the most economical method of processing intermetallic-based alloys into shapes. Sand, centrifugal, and investment casting are the most acceptable methods for casting of aluminide-based alloys. Details of casting methods and limitations, as well as examples of various castable shapes, are available in the chapter on casting of aluininides by McQuay and Sikka in this volume. 3.2 Processing ofshapes by Powder Metallurgy
The Ni,Al-. Fe,Al-, and FeAl-based alloys can be produced as powder by both the water- and gasatomization processes (Shaw and Reinshagen, 1997; Sikka, 1988a; Sikka et al., 1990). The water atomization process results in irregular shaped particles with high oxygen contents. The high oxygen content results from the formation of A1,0, on the surface of powder particles, which cannot be reduced by the conventional hydrogen-reduction process used for water-atomized iron powders. The water-atomized powder particles exhibit excellent packing and compaction behavior. The subsequent processing of the compacts breaks the surface oxidc and incorporates it into the matrix and, in most cases, with no detrimental effect on properties. Examples of the water-atomized powder particles and of the transfer of oxide particles into the matrix in thc fully dense product are shown in Figure l(a) and (b). Limited work has also been reported on the polymeratomization of iron aluniinide (Straws et al., 1998).
The polymer atomization uses a polymer instead of water for breaking the liquid metal stream into powder. The use of polymer is expected to reduce the powder’s oxygen content. The gas atomization of aluminide powder is carried out by either nitrogen gas or by argon (Shaw and Reinshagen, 1997). The compaction of gas-atomized powders is poor but is still useful in production of shapes such as sheet by tape casting and subsequent processing (Deevi et al., 1999a,b; Hajdigol et al., 1998; Mistler et al., 1998) or extrusion consolidation (Sikka et al., 1986; Wright and Sikka, 1988; Sikka, 1988a; Knibloe et al., 1990). The gas-atomized powders are also extremely useful for plasma spraying of powders to form shapes on a substrate (see the chapter by Natesan and Datta). The powder processing of intermetallics is also reviewed in the chapter by Seetharaman and Semiatin in this volume. The water-atomized powder of Fe,Al-based alloys is used for fabrication of porous metal filters. The process developed by PALL Corporation involves the following steps: (1) production of a powder slurry with an organic binder, (2) centrifugal spinning of the slurry into tubes for shaping, (3) heating of the centrifugally spun tubes to release them from the mold, (4) the binder burnout step, and (5) the sintenng step. Tubes, up to 60-cm (-24-iii.) long and 51-mm (2-in,) OD, are produced by this method. The sintered tubes typically contain approximately 50% porosity for use as gas filters. The pore size is controlled by the selection of the proper range of powder size. The porous tube
Figure I Photomicrographs showing ( U ) water-atomized powder particles and (b)roll-compacted and sintered final sheet of 0.2-
mm thickness
Structural Applications for General Use
505
Figure 2 Photograph and micrograph of a co-extruded tube of Fe,AI on type 304 stainless steel: ( a ) Fe,AI cladding on tube outside diameter and (b) interface
sections are subsequently fabricated into filters by welding solid flanges. The powders can also be used for production o f multi-layered tubular products. Examples of such products are Fe,Al on the outside of stainless steel for improved oxidation and sulfidation resistance or FeAl on the inside of a lxgh temperature alloy such as Tncoloy 803 for resistance to carburization and coking. In either case, a cast billet o f substrate material is packed with the aluminide powder (either on the outside or inside) and extruded. The extrusion process
accomplishestwo steps: (1) densification of the powder and (2) bonding of the densified aluminide layer to the substrate. Photographs of multi-layered tubes of Fe,Al on the outside and FeAl on the inside of substrate alloys are shown In Figures 2 and 3. The thermal-expansion mismatch and strength differences between the aluminides and the substrate material can cause interfacial stresses that can result in debonding at the interface. Special intermediary interfaces may be required between the substrate and the aluminide layer in order to prevent debonding.
Figure 3 The photograph and mcrograph of a co-extruded tube of alloy 803/FeA1: (a) FeAl claddiiig on the tube’s inside diameter and (b) interface
506
Applications
Figure 4 Photographs of superplastically formed disk of Ni,Al-based alloy IC-218: (U) top and (b) bottom
The bar stock produced by the extrusion-consolidation of powder can be formed into complex shapes by superplastic forming (Sikka. 1988b; Choudhury et al., 1990). The Ni,Al-based alloy IC-50 and the Fe3Al- and FeAl-based alloys can all be superplastically formed (Sikka, 1988b; Choudhury et al., 1990; Lin, 1997; Chu et al., 1998). A superplastically formed disk of IC-218 IS shown in Figure 4. Fine grain size is required for superplastic forming and coarse grain size is required for high-temperature strength. Thus, the superplastically formed components require post-treatment for grain coarsening.
3.3 Processing Sheet The processing of sheet of Ni,Al-, Fe,Al-, and FeA1based alloys through conventional ingot casting and processing is not possible. The Ni,Al-based alloys are not easy to hot work, because of their very high strength even at high temperature and their drop in ductility in the intermediate temperature range 600 to 900 "C. However, Ni,Al-based alloys have significant ductility at room temperature and thus can be cold worked. Even during cold working, the Ni,Al-based alloy requires frequent intermediate anneals because of its rapid work hardening. The work-hardening rate of Ni,A1 is three times more than tlyat of stainless steels and four times more than for carbon steels (Schulson, 1996). Furthermore, because of the rapid work-hardening rate, the cold rolling produces non-uniform deforma-
tion between the surface and the center of the sheet and thus limits the thicknesses that can be rolled without surface cracking. Typically, a cast thickness of up to 2.5mm (0.1 in) can be cold rolled. Sheets of such thicknesses can be cast by direct casting processes such as single roll or twin roll. The Ni,Al-based alloy IC50 has been cast into large coils by the single-roll process (Sikka, 1988c; Sikka, 1989). Sections of such cast sheet have been successfully cold rolled up to 50% prior to an intermediate anneal. An annealing temperature of 1100 "C results in recrystallizing the cold-rolled structure. The direct roll-casting process has the limitation that only sheets of thickness G2.5mm (0.1 in) can be produced. Furthermore, currently there is no commercial caster in the United States of sheet by direct casting. Although the Ni,Al-based alloy sheet is cold rollable from the cast structure, similar possibilities do not exist for Fe,AI- and FeAl-based alloys because of their limited room-temperature ductility. Thus, sheet fabrication of Fe,Al- and FeAl-based alloys requires powder metallurgy-based processes (Deevi et al., 1999a-b; Deevi, 2000). These include ( I ) roll compaction, (2) tape casting, and (3) plasma spraying. All three processes and the associated technologies for the processing of these intermetallics have been patented by Philip Morris USA, and are currently owned by Chrysalis Technologies, Incorporated. Each process is briefly described here, See chapter by Appel et al. for review of IMC forming.
507
Structural Applicattons for General Use 3.3.1 Roll Compaction for Sheet Production
In this method, water-atomized powder of a specific size is mixed with an organic binder, and the blend is formed into a green sheet by compacting it by cold rolling. The green sheet is subjected to a binderburnout step at approximately 500 "C (it varies with the nature of the binder), and a pressntering step gives the sheet sufficient strength for handling during subsequent cold-rolling processing steps. The presintered sheet is cold rolled and given two or three intermediate annealing and sintering treatments. The sheet so produced is fully dense with a grain size of approximately 20 to 30pm. A micrograph of a final sheet of 0.20-mm thickness (0.008-in) is shown in Figure l(0). This process is used commercially for the production of FeAl-based alloy sheet and over six tons of FeAl sheet of 0.2-mm thickness were produced during the last three years for chromium-free heating elements. The processing technologies allow production of a ductile FeAl sheet with ductility in the range of 4 to 6% with an equiaxed microstructure. The sheet has virtually no texture, and the properties are identical in the rolling and transverse directions. In addition, the sheets can be stamped at room temperature as shown in Figure 5(a), and cold formed into various configurations as shown in Figure 5(b). Although not in use, the roll compaction process can also be successfully employed for the production of sheet of Fe3AI- and Ni,Al-based alloys.
3.3.2 Tape Casting for Sheet Production In this method, gas-atomized powder of a specific size is mixed with an organic binder to form slurry. The slurry, cast into continuous sheet by tape casting, is of a thickness comparable to that of roll-compacted sheet. The tape-casting process involves uniformly spreading the slurry onto a moving conveyor and its on-line drying and coiling. The tape-cast sheet also goes through the same steps of binder burnout, presintering, and cold rolling with intermediate annealing and sintering steps as the roll-compacted sheet. The final thickness of the tape-cast and processed sheet is also fully dense. However, since it is produced from the gas-atomized powder, the grain size of the final sheet is coarser than that of the roll-compacted sheet. The grain coarsening of tape-cast sheet occurs because its grain boundaries are free from oxide particles. which are present in the roll-compacted sheet due to the use of water-atomized powder.
The tape-casting process has a big advantage over the roll-compacted process in that it allows the production of a continuous green sheet that can be coiled. Tape casting may be of great commercial interest for the production of TiAl sheet. 3.3.3 Plasma Spraying In this process, the gas-atomized powder is plasma sprayed onto a substrate. The powder splats form a sheet of nearly full density. The sheet so formed can be cold rolled and annealed to produce a final sheet of uniform thickness and microstructure. The process was demonstrated (Deevi et al., 1999b)and is an alternative method, similar to spray forming, for the manufacture of sheet.
3.4 Coating Processes .for Intermetallie-Alloy Surfaces
In many applications, advantage can be taken of the special attributes of intermetallic alloys by just using them as a surface coating. For example, chromium-free FeAl compositions of Chrysalis Technologies, Incorporated may be used as a coating for protection against oxidation, carburization, or sulfidation. Chromium-free FeAl alloy powders have the potential to replace many plasma-spray powders of superalloys containing large percentages of chromium that may lead to the formation of toxic chromium oxides. Similarly, Ni,Al-based alloy can be used for oxidation and special wear properties at high temperatures. In such a case, three processes can be used successfully: (1) weld overlay, (2) thermal spray coatings, and (3) co-extrusion of a layered composite. See chapter by Datta et al. for an over-view of coating technology. 3.4.1 Weld Overlay
The weld overlay process involves depositing a layer of the intermetallic alloy on a substrate by a commercially known process such as the gas tungsten arc (GTA) or an inert gas-metal arc known as MIG. The first deposited weld layer is diluted by mixing with the substrate and thus requires a second layer to approach the targeted intermetallic composition. The thickness of the second layer is typically 3 to 6mm (0.125 to 0.25 in). The Ni,Al-based alloy IC-221M has been successfully weld-overlaid onto carbon steel, 300 Series stainless steels, and chromium-molybdenum steels containing 2-1/4 to 9% chromium and 1% molybdenum. FeAl weld overlays are in the early stages of
Applications
Figure 5 Roll compacted sheets of FeAl and stamped sheets of FeAl and coiled configuration illustrating the room temperature formability of roll compacted and sintered FeAl sheets
509
Structural Applicationsfor General Use
development. Similar possibility exists for weld overlays of Ni,Si-based alloys on various substrates, for improving the corrosion resistance to sulfuric acid. Weld overlays of TiAl-based alloys do not provide any unique benefit and thus are not being pursued. 3.4.2 Thermal Spray Coating
In this process, gds-atomized powders of the selected intermetallic powder size are deposited on substrate surfaces by a high-velocity oxy-fuel (HVQF) process. Trials with Ni,Al-based and FeAl-based powders have resulted in thin, smooth coatings. The quality of the bond between the coating and the substrate still requires optimization. 3.4.3 Co-extrusion
Co-extrusion is a good technique for developing coatings of intermetallic alloys on either the inside or outside of the tubular products. Details of this process were described in Section 3.2.1, and examples of coextruded tubes are shown in Figures 2 and 3. 4. Fabrication Methods
This section deals with the operations needed to fabricate individual parts into a complex assembly. The operations needed for fabrication include: cutting, machining, welding, brazing, and mechanical joining. Each of the operations for intermetallic-based alloys described in this chapter is discussed here. 4.1 Cutting
Intermetallic-based alloys are significantly more difficult to cut than conventional alloys. One of the reasons for their resistance to cutting is their high workhardening rate (Schulson, 1996). The work-hardening rate of FeAl is six times that of 301 stainless steel and seven times that of carbon steel. Similarly, the workhardening rates of Ni3A1 are three times that of 301 stainless steel and four times that of carbon steel. The second difficulty in cutting is the increase in yield strength with temperature (Liu et al., 1985), which is especially strong for N&Al-basedalloys. The most common methods used for cutting intermetallic-based alloys are (1) abrasive wheel, (2) band saw, (3) laser cutting, and (4)EDM. In order to minimize environmental effects related to hydrogen generation from the reaction of moisture with alumi-
num, it is recommended that an oil-based coolant be used during abrasive-wheel cutting. The abrasivewheel cutting is commonly used for Ni,A1- and FeAlbased alloys. The Fe,Al-based alloys are easier to cut by a band saw, but can also be cut by an abrasive wheel. EDM is a commonly used process for all of the intermetallic alloys, and laser cutting is used for creating special shapes of FeA1. Laser cutting and chemical milling were extensively used by Chrysalis Technologies, Incorporated to obtain intricate shapes of Fe&, and are indeed excellent methods for specialized applications. 4.2 Machining
Machining of intermetallic-based aUoys is significantly more dificult than for conventional alloys. As for cutting, the high work-hardening rate and increase in yield strength with temperature are the two causes for difficulty in machining of intannetallics. Grinding is the easiest machining method for removal of material for most intermetallic alloys. Single-point machining is also possible, but it requires slow speeds and small depths of cut. High-strength steel tools, tungsten carbide tools and ceramic tools are all used for machining intermetallics. EDM and lasers are used for cutting special shapes and making holes. The machining of intennetallics is an area of research needing additional work to realize cost reduction in machining by optimizing cutting tools, speeds, depth of cut, and coolants. 4.3 Welding
Besides cutting and machining, welding is the next most important fabrication process for assembly of components. The GTA and MIG processes are the most common methods applicable to the welding of Ni,Al-, Fe,Al-, FeA1-, and Ni,Si-based alloys. Friction welding has also been employed to weld FeAl. Thin sheets of FeAl were welded to FeAl and to dissimilar materials by Chrysalis Technologies, Incorporated. Joining of TiAl turbocharger wheels to their steel shafts can be accomplished through friction welding. More details of joining of individual intermetallic alloys are given below.
4.3.1 Joining of Ni,Al-Based Alloys The thin sheets of IC-50 alloy can be joined to form welded tubes. Such joints can be made without a filler metal, but require a special cleaning treatment (Sikka,
510
Applications
1991) of the joint prior to welding. The cleaning is done with FeCl, solution, which removes the A1,03 film from the surface of the sheet in the weld region. Cast components of the nickel aluminide alloy IC22 1M can be weld repaired and welded using the CTA and MIG processes. Compositions of IC-221LA and IC-221W filler wires are available (Santella, 1997). Nickel aluminide components have been successfully welded to components of the same composition (similar metal joints) and components of different composition (dissimilar metal joints). A welded radiant-burner-tube assembly of nickel aluminide is shown in Figure 6. In this assembly, there are joints of nickel aluminide to nickel alurninide and to carbon steel plate. A roll assembly of nickel duminide to an HK-40 (cast stainless steel) trunnion is shown in Figure 7. This assembly shows the nickel aluminide to cast stainless steel joint. Deciding what filler wire to use between IC-221LA and IC-221W depends on the application. Cosmetic weld repair of castings and welded components operating below < 1000 "C is effective, but a use limit of 1150 "C is set by the formation of Ni5Zr, eutectic that melts at 1172 "C. Both IC-221LA and IC-221W filler wires are produced by using powder as a core in a nickel sheath. Such filler wire is commercially available from Stoody Company, 5557 Nashville Road, Bowling Green, KY 42101. Significant effort is still needed to develop the weldability of IC-438.
Figure 6 Photograph showing welded radiant burner tube assembly of Ni,At-based alloy IC-221M
to identify the best filler metals and weld parameters for both similar and dissimilar joints.
4.3.2 Joining of Fe,Al- and FeAl-Based Alloys Welding of Fe,AI- and FeAl-alloys is significantly harder than for Ni,AI alloys. These alloys require preheating to approximately 300 "C to keep the moisture away during welding and a postweld heat treatment of approximately 700 "C to reduce welding stresses (Santella, 1997). The welds can be made by using a filler metal of the base-metal composition. The filler-metal wires are produced either by filling powders of correct composition between steel strips or by filling the aluminum wire core between steel strips along with alloymg elements between the aluminum wire and the steel strip. Thin strip components of Fe,Al- and FeAIbased alloys can be laser welded without filler metal.
4.3.4 Joining of TiAl-Based Alloys Most of welding research on TiA1 has focused on joining cast turbocharger wheels to a 4340 steel shaft. Friction welding has been chosen as the first process to investigate. However, details for successful friction joints have not yet been worked out. 4.4 Brazing
Brazing is a joining process considered for special applications.For example, tungsten carbide cutting tools were successfully silver-brazed to nickel aluminide blocks in air. Based on this observation, no major problems are anticipated in brazing of other intermetallics.
4.3.3 Joining of Ni3Si-Based Alloys Based on limited data, the wrought sheet of Ni,Si can be welded using GTA. Most of the initial welds are made without filler metal. Additional work is required
4.5 Mechanical Joining
For certain applications,components can be fabricated by mechanical joining. One successful method of
51 I
Structural Applicationsfor General Use mechanical joining for Ni,Al-based components has been pinning. The pins used for joining the Ni3A1based pipe of Figure 7 to the cast stainless trunnion were of Inconel 617. The pins are held in position by two mechanisms: (I) the difference in thermal expansion of pin material with respect to the stainless steel trunnion and (2) the peening of edges of the hole in the nickel aluminide. These pins have successfully operated at ternpcratures of 950 "C for five years.
6.1.I Ni3Al-Based Alloys The Ni,Al-based alloys, currently in use, are patented and thus, only licensed suppliers can produce it for commercial use. The licensing process has limited the number of producers for Ni,Al-based alloys. Furthermore, at the present time all of the producers are located in the United States. It is anticipated that more producers will come on board as the demand for these alloys increases. The current suppliers include: Product form
5. Applications
Current and potential applications of intermetallicbased alloys are presented in Tables 3 through 6. These tables also include the materials that will be replaced by the intermetallic-based alloys.
Alcon Industries (Cleveland, OH) Static castmgs Static and centrifugally Alloy Engineenng & Casting Co. cast (Champaign, IL) Sandusky International (Sandusky, Centrifugally Cast OH) Static and centrifugally United Defense (Anniston, AL) cast Ametek (Eighty Four, PA) Powder Stoody Company (Bowling Green, Weld wre
6. How to Purchase Intermetallics
Ku) Weld wire
Purchasing of intermetallic-based alloy components is not expected to be any different than conventional alloys. However, for that to happen, the following must happen: Suppliers for the intermetallic-based alloys need to be identified. Alloy compositions need to be finalized and ASTM specifications need to be approved and available for designers and users. Design data sets are needed for each of the intermetallic-based alloys. A recycle program may be required to compensate for the higher cost of intermetallic-based alloys. Each of the items is further described below. 6.1 Suppliers for the Inteumetnllic-BasedAlloys The following are suppliers for various intermetallicbased alloys.
Supplier
Polymet (Cincinnati, OH)
6.1.2 Fe,Al-Based Alloys The Fe,Al-based alloys that are currently in use are patented and thus are only produced by licensed producers which include the following: Product form
Supplier
Powder Porous metal filters
Ametek (Eighty Four, PA) PALL Corporation (Cortland, NY)
6.1.3 FeAl-Based Alloys
The currently used FeAl-based alloys are patented by Chrysalis Technologies, Incorporated, and thus are produced only by Chrysalis Technologies, Incorporated and its vendors. The powder and sheet of chromium-free FeAl is manufactured by Chrysalis Technologies, Incorporated's principal vendor, Ametek Specialty Metals Products Division (Wallingford,
Figure 7 Photograph of a welded roll assembly of Ni,Al-based alloy IC-221M to HK-40 trunnions
512,
~p~li~utio~s Current and potential appl~cationsof NiSAl-based alloys
Component
Application/Objective
Replacing
Thin-wall tubes and statically cast return bends
Radiant burner tubes in heat-treating furnaces To increase life via higher creep, oxidation, and carburization resistance
Cast stainless steels such as HU, HT, HK, etc.
Thick-wall tubes and pipe and statically cast trunions
Furnace rolls in austenitizing and hydrogen annealing furnaces To iiicrease life through eliniinatioii of blisters, nonsticking of iron oxide to the roll and higher creep and high~temperaturewear resistance
Cast stainless steels such as HU and HP modified
Statically cast trays and fixtures Trays and fixtures for holding coiiipoiients during Cast stainless steels such as HU, HT, HK, etc. carburization and annealing processes To increase life from higher creep strength and better resistance to carburization and oxidation Statically cast die blocks
Dies for the hot-forging process To increase life due to its peak in yield strength at 850 "C and good oxidation resistance
21-13 and other die steels
Statically cast tube hangers
Hangers to support tubes operating at high temperatures in chemical plants To increase life due to high creep strength and oxidation resistaiice
~ U P E R T H and ~ ~other ~ T ~ h~~h"temperature alloys
Stat~~ally cast ~omponents
Variety of applications for heat-treating furnaces To increase life by higher creep strength, oxidation, and carburization resistance
Alloy steel and cast spainless steels
Preoxidized sheet
MonQ~ithicAI@, boards Electronic circuit board applications The electrically insulating surface with a metallic core will absorb more heat from the circuit bourds than monolithic AI,Q, boards. This will extend the electronic boards c~pabilityfor higher power capacity
Cast-shaped rods
Fe-Cr-A1 alloy heating elements Industrial €urnwe heating elements Increase the life of the heating elements by preventing their oxidation and sag~ing
Ni,Al powder
Binder for tungsten and chromium carbide as tool and die materials Improves hardness and wear resistance and aqueous corrosion resistance in certain acid solutions
Expensive cobalt that is currently used as the binder materia1
Cast hot-pressing dies
Dies for hot pressing of permanent magnet materials Improved die material because of excellent chemical compatibility and peak yield strength at the hot-pressing temperature of 850°C.
IN-7 18 dies
-
Coniiecticut). The following provides a list of product forins and the associated supplier: ~ r o f or^ ~ i ~S ~~ ~ ~~ l i ~ r
Powder ChrysalisTechnologies, Inc. (Richmond, VA) Sheet ChrysalisTechno~og~es, Inc. (Ric~mond,VA) Chromium"free ChrysalisTechnologies,Inc. ( ~ c h ~ o VA) nd~ FeAl alloys and other FeAl alloys
ChrysalisTechnologies, Inc. ( ~ c ~ o nVA) d , Sititered and reactioiis y n t ~ e s parts ~z~~ ChrysalisTechnologies, Inc. ( R i c ~ o n dVA) , Injection molded metal parts ChrysalisTechnologies, Inc. (Richmond, VA) Heaters C h ~ s a l iT~chiiologies, s Inc. (Richmond, VA) Structural co~pone~ts for corrosion aiid sulfidation atmospheres
513
S ~ r u c t ~ r Applications al f o r General Use Current and potential applications of Fe3Al-based alloys Conipoiient
ApplicationlObjective at gas clean-up filter To enhance the filter's corrosion performance in sulfur-containing gases
Porous metal filter for hot gases
Replacing Stainless steel porous metal filter
~ o n o l i t h stainless ~c tube with duplex tubes
Extruded outside cladding on coal-fired boiler tubes
Boiler tubes in coal-fired plants To improve the corrosion performance in both oxidizing and sulfidizing eiivironnients
~taticallycast or fabricated parts
Variety of applications in coal-fired power plants Alloy steel and stainless steel To improve the component life from enhanced resistance to oxidation, sulfidation, and rust
Thermal spray ChrysalisTechnologies, Inc. (Ichmond, VA) coatings ChrysalisTechnologies,Inc. (Richmond, VA) FeAl porous filters and foams FeAl ChrysalisTechnologies, Inc. (Ichmond, VA) hon~yco~bs
6.1.4 Ni,Si and ~ i ~ l - B a Alloys s~d
Currently there are no commercial producers of Ni,Sibased alloys. NiAl-based alloy powders can be obtained from the following supplier: Prodi~ctform
S ~ p ~ ~ i ~ ~
Powder
Xform Incorporated (Cohoes, NY)
specific compositions were iiicluded in cations. The cast Ni,Al-based alloy TC-221 inte~etallic-based alloy that is approved in A ~peci~cations. Its A ~ T Mdesignat~oni s AlOO2-99 and the specifications are for castings. The chemical analysis of TC-221M alloy as specified under A 100299 i s given in Table 7. Having approved AST specifications will encourage designers and users to start specifying Ni,Al-based alloys for either new or ~eplacem~nt applications. Efforts are also under way by Chrysalis ~echno~ogies, Incorporated to have FeAl-based alloy compositions included in ASTM sp~cifications.
Its
6.1.5 T i ~ l - Alloys ~ a ~ ~ ~
Most of the production of TiAl-based alloys has been in support of aerospace ~ ~ p l i c a t ~ o Although ns. no significant market is currently developed for general use, the aerospace suppliers also are producing these products. These include: ~ ~ o ~ ~ ~ ~ f o r r lie n r Investment castings Investment castings Powder Powder
Wowmet Corporation ( ~ i t e h a l lNET) , PCC Airfoils, Inc. (Portland, OR) CM Crucible Research (Pittsburgh, PA) Pratt &Whitney (West Palm. Beach, FL)
Physical and mechanical property data are required for the successful design and implernentatioii of any new alloy. Although such data are becoming available for i~termetallicalloys 1996), no c o m p ~ e ~ e ~ s i v e database exists at . Among the interrnetallicbased alloys described in this chapter, data for the ~ i 3 ~ l - b a s alloy, e d IC-221M are most c o ~ p l e t e( et al., 1997). A s u m ~ a r yof current IC-221 data is given in Table 8.
cost
Significant developments have occurred in the composition o f iiitermetallic-based alloys. However, the co~positionsstill continue to change as new developments occur. For com~~ercial production, it is essential that specific compositions bc finalizcd with permissible ranges for each element. It would help further if the
When choosing a replacement or a new ~ ~ t e r ifor a l an application, cost is a significant issue. This is especially true of the interinetallic-based alloys, because of following factors: 1. Higher cost of constituents. This is true for Ni,AI, Ni3Si, and TiAl-based alloys.
~ ~ ~ ions l i ~ a t
514
Current and potential applications of FeAl-based alloys Component Thin sheet, wire rods of chromium-free FeAl alloy
Applicatioii/Objective Heating elements with 50% higher electrical resistivity To increase heating element life by enhanced oxidation resistance and creep strength and avoid generation of toxic fumes of chromium oxide
Replacing FeCrAl alloys and nic~e~-based alloys Consumer and industrial heating elements, free of chromiuni
Thin sheet chroniiu~-freeFeAl alloy
As a honeycomb filter for cars and diesel trucks
FeCrAl alloys and steels
Thin sheet of chromium-free FeAl alloy
As a substrate for catalytic converters
Steels and FeCrAl alloys
Inside-coated tubes of chromiuni-free FeAl co-extruded with iron-based alloys
Ethylene cracker furnace tubes To improve life Limited by carburization and coking by an order of magnitude
Monolithic wrought Alloy 803 and cast HP-modified
Centrifugal rolls
Continuous caster transfer rolls To improve life through enhanced oxidation and creep resistance
Steel alloy or weld-overlaid rolls
Centrifugal~ycast closed-end containers
High-temperature salt baths for carbonates and nitrates To improve the salt bath life by an order of magnitilde
Stainless steel and nickel-based alloys
FeAl powder
Binder for tungsten and chromium carbides as tool and die materials To improve hardness and wear resistance and aqueous corrosion resistance
Expensive cobalt that is currently used as binder material
Chro~ium-freeFeAl alloy powder
Powder components for use in aggressive atmospheres and at high temperatures To improve hardness and wear resistance, and to provide corrosion resistance To eliminate the generation of toxic fumes of chromium oxide
Expensive 3 16 series and 304 steels
Centrifugally cast closed-end container
Baths for sulfuric acid solutions To improve container life through enhanced oxidation and creep resistance To improve container life by improvements in its res~sta~ce to sulfuric acid solutions
Iron- and n ~ ~ ~ e ~ - balloys ased
Components of different sizes and shapes
Coinponents such as mixers for corrosive solutions To improve component life through enhanced resistance to corrosion and wear
Iron- and nickel-based alloys
FeAl components
Nonmagnetic hip and knee joints
Non-magnetic alloys based on iron with s ~ g n i ~ c amounts ~t of chromium
Metal injection-molded components of chromiumfree FeAl alloy powder
Wrist-watch cases with no chromium or nickel To eliminate the allergies associated with nickel- and chromium-containing alloys
Coiiventional steel alloys
Investment-cast engine valves
Lighter automotive valves for engine application
Steel valve seats
Table 6 Current and potential structural applications for general use of TiAl-based alloys
A pplicalion/Objective
Go~ponen~ __-_
____
~
____
-
Investment-cast t urbocbargers Investment-cast engine valves
_
_
_
_
_
_
_
_
_
I _ _ _ _ _
_ _ I _ _ _ - -
placsng __-____---__---
IN-713G rotors
Turbocharger rotors for autoniotive engiiies To improve e n g m weight savings and turbocharger life Valves for autoiiiutive engine application To improve the wear resistance of the valve seats and reduce weight
Steel valve seats
-
le 7 Chemtcal requi~enients specified under A 1002-99 for castings of Ni,Al-based alloy ~_ _ _ ~ _ _ _ Element Composition (wt.%) ~-
0.08 0.02 7.3-8.3 7.5-8.5 1.20-1.70 I .6&2.10 0.003--0.012 0.20 1.OO balance
C" Sb
A1 Cr MO
Zr I3 Si' Fe" Ni
~ ~ ~ a x ~ m u i ~ . h M a x i ~ ~For u ~ welding . applications, the sulfur shall be 0.003% by weight or less and silicon shall be 0.05% by weight or less.
2. Higher cost because of very l i ~ ~ i t edd e ~ at~ ~ d present. Because o f limited demand, internictallic alloy components are still produced under experimental or pilot-scale o p e r a t ~ o ~ s . 3. Higher costs because of d i ~ c u l ~iny cutting and machining, weld-wire cost, and limited experience with recycling o f i n t e r ~ e ~ a l l i c - b aalloys. ~~d 4. r cost because of limited processing ence and crsss-coiitamiiiatjoi~ issues during production. 5. Limited experience in weldin based alloys.
One method o f reducing cost is to credit to the users. For example, in the case of applications such as radia~t-burner tubes, rolls, or
Physical and i~eclialiicalproperties of Ni,Al-based cast alloy IC-221M (Sikka. 1997) Temperature ("C) Property Density (g/cnn") Hardness (Rc) Microhardness (DPH) ~ o d u l u (GPa) s Mean Coefl'. o f thermal expansion (10 6/oC) Thermal conductivity (w/m 0.2Oh tensile yield strengt~(MPa) Ultimate tensile strength (MPa) Total tensile elongation (%) 102h Rupture strength (MPa) 103h Rupture strength (MPa) 104h Rupture strength (MPa) Charpy impact toughness (J) Fatigue strength 10' cycle life (h4Pa) Fatigue strength 107 cycle life (MPa)
"Room temperature to 100°C. 'Room temperature to specified temperature. 'Data at 65O"C for invest~ent-casttest bars.
Room 7.86 30 260 200 12.77" 11.9 55.5 770 14
200
270 190 13.08" 13.9 570 800 14
400 -
280 174 13.72h 16.7 590 850 17 __
__
40
40
600
800
900
290 160 14.33h
280 148 1.5.17'
230 I39 15.7ah
120 126 16.57'
20.3 610 850 18 -
2.5.2 680 820 5 252 172 124 15
27.5 600 675 5 124 83 55 10
30.2 400 500 7 55 36 24
__
40
35 630" 5 SO'
-
-
__
1000
1100
-
_I
200 200 10 28 18 11 -
heat-treating fixtures, a credit offered for recycling of the components at the end of their life could reduce the cost by l0 to 20%. Of course, the recycle program will require excellent segregation practices, which is not the case today at large maiiufacturing facilities. The total life cycle shows other cost reductions. This Factor iniplies that the initial higher cost can be made up by sonie of the follow in^ factors: (a) reduced production down-time, (b) improved quality of the manufacture^ ~ r o d u c t ,(c) extended co~lponentlife, and (d) red~~cedsize of the needed replacement inventory, and (e) in some instances, nothing else works as well.
available, the use of ~ e ~ l - ~ a alloys s e d is expected to nearly double in the next three to five years,
s There are significant needs to enhance the structural applications of inter~etallic-basedalloys for general use. These include: n designers and users about the 1. ~ d ~ c a t i oof attributes of i n t e r ~ e t ~ ~ l l i c - b ~ salloys ed as opposed to conventional alloys. 2. Alloy compositions for the commercial application iieed to be fixed and included in ASTM specifications. 3. The supplier base for i~termetallic-basedalloys needs to be expanded, which in time will be taken care of by the economic factors of supply and demand. 4. ~ o n l p ~ e h e n s i v edatabases are needed for intermetallic-based alloys. 5. Costs of intermetallic-based alloys need to be reduced. This can hap en through increased understanding of methods for cutting, machining, welding, lower-cost processing methods, and good r e c y ~ l i ~programs. g
s based interinetallic alloys are the ir development and applications. ection 8 on future needs lists several items that are needed for further enhancement iii application of these mat~rials.This section lists some very specific iterns that are needed and potentia~ fund~ng sources to accomplish tliem. 7.
The most crucial needs for these alloys are in three areas: (a) development of low-cost cutting and maclii~ing pr~~cesses,(b) widely accepted welding procedures and weldinent properties for both similar and dissimila~~ e t a welds, l and (c) p~oduction of wrought product forms. for all three areas is continuing from the nt o f Energy. As a ~ ~ i t ~ o ndata a l develop in the above-mentioned areas, applications of ~ i ~ A l - b a s ealloys d will further enhance. The quantity of Ni,Al-based alloys in coniinercial applications is expected to double in the next three to five years. 7.
~
~
~
y
Thc critical developments for these alloys iizclude (a) e n l i a ~ ~ ~ine rnoto ~ - t e ~ p e r a t uductility, re (b) further iniprovement in h~gh”tem~erature strength, (c) welding procedure development, and (d) methods for coniponent fabrication. i n many cases, FeAl can significantly e ~ i h a applic~tions n~ when used as rnultilayers. Funding most of the areas listed above is provided by the Eiiergy aiid Chrysalis Technologies, s results of these develop~entsbecome
0~
ts
The authors thank Professor D. H. Sastry of Department of ~ e t a l l u r g y ,Indian Institute of Science, presently ~ i s i t i ~Scientist g at ~ h r y s ~Tec~nologies, l~s Incorporated, for his valuable suggestions, and G. Carter and M. L. Atchley for typing the chapter.
Anton, D. L., Shah, D. M., Duhl, D. N.,and Ciamei, A. F.
s
Barrett, C. A. (1988). Oxid. Met., 30, 361. Cahn, R. W. (1989). ~ e t ~ l~ s~, t ~ r i aand l s Processes, , 1, 1. Mukherjee, A.K. (1990). J.
Kai, W., Waiig, J. Y., and Inoue, K. (1998). ~ a ~ e Scierzce r i ~ and ~ ~~ ~~ ~ i ~ e e r i ~ ~ ~ 2 ~236-242. g , Deevi, S. C. and Sikka, V. K. Deevi, S. C., Hajsligol, M. R and Scorey, C. R. (1999a). eds. Easo P. George, Michael J. Mills, and Masahslru Yamaguchi, KK4.6.1-ECK4.6.9.
Structural Applications for General Use Deevi, S. C., Hajaligol, M. R., Lilly, A. C., and Fleischauer, G.S. (1999b). Tram. Nonferrous Met. Soc. China, June. 309-317. Deew, S. C. and Sikka V. K. (1999). Electronic Circuits Having NiAl and Ni,AI Substrates, U.S. Patent No. 5,965,274, granted on October 12, 1999. Deem. S. C. (2000). Intermetallics, 8, 679-685. Deevi, S. C. and Zhang, W (2000). The Encylopedia of Materials Science and Technology. in press. Dimiduk, D. M. (1999). Materials Science and Engineering A , A263, 281. Fleischer. R. L.. Dirmduk, D. M., and Lipsitt, H. A. (1989). Annu. Rev. Mater Sci., 19, 231. Fleischer, R. L. and Tanb, A. I. (1989). JOM, 41(9), 8. Froes, F. H. and Suryanarayana, C. (1996). Titanium aluminides. In Physical Metallurgy and Processing of Intermetallic Compounds (eds N. S. Stoloff and V. K. Sikka). Chapmdn & Hall, New York, 297-350. Grala, E. M. (1960). In Mechanical Properties of Intermetallic Compounds, (ed. J. H. Westbrook). Wiley, New York. 368. Hahn, K. H. and Vedula, K. (1989). Scr. Metall., 23, 7. Hajaligol, M. R.. Deem, S. C., Sikka, V. K., and Scorey, C. R. (1998). Materials Science and Engineering, A258,249-257. Huang, S. C. (1993). In Structural Intermetallics (eds. R. Darolia et al.). TMS, Warrendale. PA, 299-307. Kear, B. H., Sims. C. T., Stoloff. N. S., and Westbrook, J. H., (eds) (1970). Ordered Alloys - Structural Applications and Physical Metallurgy, Claitors Pub. Div., Baton Rouge, LA. Knibloe, J. R., Wright, R. N., and Sikka, V K. (1990). In Advances in Powder Mefallurg,y, Vol. 2 (comp. E. R. Andreotti and P J. McGeehan). Metal Powder Industnes Federation. Princeton, NJ, 219-231. Lilly, A. C., Deevi, S. C., and Gibbs, Z . P. (1998). Malerials Science and Engineering, A258. 4249. Lin, D. (1997). In International Symposium on Nickel and Iron Aluminides. Processing, Properties. and Applications (eds S. C. Deevi et al.) ASM International, Materials Park, OH, 187-197. Liu, C. T., White, C. L., and Horton, J. A. (1985). Acta Metall., 33, 213. Loria, E. (2000). Intermetallics, in press. Maziasz, P. J., Goodwin. G. M., Alexander, and D. J., Viswanathan, S. (1997). In International Symposium on Nickel and Iron Aluminides: Procawmg, Properties, and Applications (eds. S. C. Deevi et al.) ASM International, Materials Park, OH, 157-176. McKamey, C. G., DeVan J. H, Tortorelli, P. F., and Sikka, V.K. (1991). J. Mater. Res., 6(8), 1779-180s. McQuay, P A. and Sikka, V. K. (2002). Casting of aluminides. In Intermetallic Compoundr - Progress (eds. J. H. Westbrook and R. L. Flnscher). Wiley, Chichester. UK. Mistier, R. E.. Sikka, V K.. Scorey, C. R., McKernan, J. E., and Hajaligol, M. R. (1998). Materials Science and Engineering, A258, 258-265. Natesan, K. and Tortorelli, P. F. (1997). In International Symposium on Nickel and Iron Aluminides: Processing,
517
Properties, and Applications (eds S. C. Deem et al.). ASM International, Matcnals Park, OH, 265-280. Natesan, K. (1998). Mater. Sci. Eng., A258. 126-14. Nesbitt, J. A. and Barrett, C. A. (1985). In High Temperature Ordered Intermetallic Alloys (eds. C. C. Koch. C. T. Lm, and N. S. StoloftJ. Mater. Res. Soc. Symp. Proc., 39,601610. Nesbitt, J. A., Vinarcik, E. J., Barrett, C. A.. and Doychak, J. (1992). Mater. Sci. Eng., A153, 561-566. Oliver, W. C. (1989). In High Temperature Ordered Intermetallic Alloys 111 (eds C. T. Liu, A. 1. Taub. N. S. Stoloff, and C. C. Koch). Mater. Res. Soc. Svmp. Proc., 133, 397402. Reddy, B. V. and Deevi, S. C. (2000). In Intermetallics, in press. Santella, M.L. (1997). In International Symposium on Nickel and Iron Aluminides: Processmng, Properties. and Applications (eds S. C. Deevi et al.). ASM International, Materials Park, OH, 321-328. Schulson, E. M. (1996). Brittle fracture and toughening of intermetallic compounds, In Physical Metallurgy and Processing of Intermetallic Compounds (eds N. S. Stoloff and V. K. Sikka). Van Nostrand Reinbold, New York, 56-94. Shaw. K. G. and Reinshagen, J. H. (1997). In International Symposium on Nickel and Iron Aluminides. Processing, Properties, and Applications (eds S. C. Deen et al.). ASM International, Materials Park, OH, 301-31 1. Sikka, V. K.. Liu, C. T., and Loria, E. A. (1986). In Structural Metals bv Rapid Solidification (eds F. H. Froes and S. J. Savage) ASM International, Metals Park, OH, 417-427. Sikka, V. K. (1988a). In Modern Developments in Powder Metallurgy - Vols. 18-21, comp. D. A. Gustafson and P. U. Gummerson, Metal Powder Industries Federation, Princeton, NJ, 543-557. Sikka, V. K. (1988b). In Interdisciplinary Issues in Materzals Processing and Manufacturing, Vol. 2 (eds S. K. Samanta et al.). ASME, Boston, MA, 451-467. Sikka. V. K. (1988~).In Casting of Near-Net-Shape Products (eds Y. Sahai et al.). TMS, Warrendale, PA, 315-333. Sikka, V. K. (1989). Mat. Res. Soc. Symp. Proc., 133, 487492. Sikka, V. K., Baldwin, R. H.. Howell, C. R., and Reinshagen, J. H. (1990). In Advances in Powder Metallurgy, Vol. 2 (camp. E. R. Anderson and P J. McGeehan). Metal Powder Industries Federation, Princeton, NJ, 207-218. Sikka, V. K. (1991). U.S. Patent 5,016,810. May 21, 1991. Sikka, V. K., McKamey, C . G., Howell, C. R., and Baldwm, R. H. (1991). Properties of large heats of Fe,Al-based alloys, ORNL/TM-11796, Oak Ridge National Laboratory, Oak Ridge, TN. Sikka, V. K. and Baldwin, R. H. (1992). SAMPE Quarterly, 24( 1). 2-9. Sikka, V. K and Deew, S. C. (1995). Mater Technol., 10(5/6), 97-100. Sikka. V. K., Deevi, S. C., and Vought, J. D. (1995). Advanced Materials and Processes, 147(6), 29-3 1.
518
Applications
Sikka. V. K. (1996). In Oxidation and Corrosion of Intermetallic Alloys (eds G. Welsch and P D. Desai). Purdue Uiiiversity, West Lafayette, IN, 1-1 18. Sikka, V. K. (1997). In International Symposium on Nickel and Iron Alumznides: Processing, Properties, and Applications (eds S. C . Deevi et al.). ASM International, Matenals Park, OH, 361-375. Sikka. V. K., Santella. M. L., and Orth, J. E. (1997). Mater. Scz. and Eng., A239-240, 564.569. Stolo& N. S. and Dames, R. G. (1966). Prog. Mater. Sci., 13, 1. Stoloff, N. S., Liu, C . T., and Deevi, S. C. (2000). Intermetallics, in press. Straws, J. T., Scorey, C. R., McKernan, E., and Hajaliyol, M. R. (1998). Materials Science and Engineerzng, A258, 29 1-297.
Tortorelli, P F and Natesan, K. (1998). Mater. Scr. and Eng. A258, 115-125. Ulvensoen, J. H., Roruik, G., Kyvik, T., Pettersen, K., and Estrade, L. (1993). In Structural Intermetallics (eds R. Darolia et al.). TMS, Warrendale, PA, 707-713. Walston, W. S. and Darolia, R. (1993). Mater. Res. Soc. Symp. Proc., 288,237-242. Westbrook. J. H. (ed.) (1960). Mechanical Properties of Intermetallic Compounds, Wiley, New York. Westbrook, J. H., (ed.) (1967). Intermetallic CompounB, Wiley. New York. Williams, J. C. (1997). In Sfructural Intermetallics 1997 (eds. M. V. Nathal et al.). Seven Spnngs, Champion. PA, 3. Wright, R. N. and Sikka, V. K. (1988). J Marer. Science, 23, 43 15-43 18.
Chapter 25 efrigeration ner, Jr. and V. K. Becharsky Ames Laboratory and Department of Materials Science and Engineering, Iowa State University, Ames, IA, USA
1. Introduction Magnetic refrigeration is based on an intrinsic property of magnetic solids - the magnetocaloric effect (MCE). When a magnetic material near its magnetic ordering temperature (called the Curic temperature, T,, for a ferromagnetic substance) is placed in a magnetic field, the magnetic moments (unpaired 3d or 4f electrons) are aligned, and the magnetic entropy, ,‘A is lowered by ASM. Under adiabatic conditions the total entropy of the solid remains constant, and thus the sample must heat up increasing the lattice entropy by AS,= -AS, giving rise to the adiabatic temperature increase, AT,,. When the field is removed or reduced. S, increases, the lattice entropy decreases, and ATdd becomes negative. Typical AS, and ATad values for various magnetic field changes, A H , are shown in Figure la and lb, respectively for Gd metal. Gadolinium metal has been the most thoroughly studied material for its magnetocaloric effect and is considered the standard material against which other materials with near room-temperature MCE are compared. Gschneidner and Pecharsky (2000a) noted that 11 independent measurements have been reported for Gd through 1999. As is seen in Figure 1 the MCE (either as -ASM or AT,,) increases with increasing A H , and for a normal ferromagnet the MCE vs. temperature has a ‘caret’-like shape with its peak at Tc. This is the most comnion type of behavior, but other shapes - ‘table’-like and ‘skyscraper’-like have been observed, see Figure 2. The broad ‘table’-like behavior i s due to multiple magnetic transitions in the (Gdo,,Er0,,JNiAl phase (Takeya et al., 1994, Korte et al., 1998a,b), while the ‘skyscraper’-likebehavior is due
to a first-order magnetic transition in Gd,(Si,,,3Ge3 67) (Pecharsky and Gschneidner, 1997a, 1998). The MCE can also havc the opposite sign - the magnetic entropy increases when a sample is inserted into a magnetic field causing the sample to cool down. This behavior is quite rare and has been observed for some materials undergoing antiferromagnetic ordering, such as GdMnSi at -50K (Figure 3a), FeRh at 325 K (see Section 4.2), and in some superconducting materials (Gschneidner and Pecharsky, 2000b). It has also been predicted theoretically and observed experimentally in PrNi, by von Ranke et al. (1998) and is due to the crossing of the two lowest magnetic energy levels of Pr (Figure 3b).
-
2. Magnetic Cooling
About 75 years ago both Debye (1926) and Giauquc (1927) independently suggested that the magnetocaloric effect could be used to reach temperatures below 1 I<, by cooling a paramagnetic substance in a large magnetic field to as low a temperature as possible by pumping on liquid He. After reaching 1.2 to 1.5K, the sample (e.g. a paramagnetic salt) is thermally isolated and the magnetic field is removed and the sample cools well below 1 K . Giauque and McDougal (1933) were the first to demonstrate this: they reached a temperature of 0.25K starting at 1.5K using 61g of Gd2(S0,),8H20 and reducing the magnetic field from 0.8T to OT. This technique is known as adiabatic demagnetization. It is still used today in ultra-lowtemperature studies - e.g. nuclear adiabatic demagnetization has been used to reach micro Kelvin
-
Intermetallic Compoundv: Vol. 3, Principles and Practire. Edited by J.
0 2 0 0 2 John Wiley & Sons, Ltd.
H.Westbrook and R.L.Fleischer
Applications
520 140
120
100
0
2 80 2 3
U
60
-? 40
20
0
200
230
260
290
Temperature (K)
320
350
200
230
260
290
320
SJ)
Temperature (K)
(a)
350
Figure 1 The magnetocaloric effect in polycrystalline, high-purity Gd metal for several different magnetic field changes AA? (a) the magnetic entropy change and (b) the adiabatic temperature change. Also shown in (a) and (b) are peak values, AS, (max), or AT,, (max), and the full width at half maximum (FWHM) along the corresponding TL and T, values used to numencally characterize the relative cooliiig power (RCP(S) and RCP(Q, respectively) of Gd
temperatures (see Section 2.1). However, adiabatic demagnetization is a one-step process which cannot be repeated without first warming the volume to be cooled (i.e. due to the reapplication of the magnetic field). In order to make magnetic cooling practical, one must use a continuous, cyclic process in which one must reject (dissipate) the heat generated when the material experiences a magnetic field rise (just as the heat is rejected in the compression step of a gas-cycle refrigerator). The cooling is achieved by removing the magnetic field experienced by the magnetic refrigerant, which is equivalent to the gas-expansion step in the gas-cycle refrigerator. These steps are then continuously repeated. One of the first continuously operating refrigerators was designed and operated by Heer et al. (1954) who were able to cool from 1 to 0.2K using a magnetic field of 0.7T and a two-minute cycle to magnetize and demagnetize the iron-ammonium-alum refrigerant. In the next 40 years work on magnetic refrigeration slowly moved forward with scientists and engineers building devices to operate in various
temperature ranges from 4 K to near room temperature. More details can be found in the reviews by Gschneidner and Pecharsky (1997) and Pecharsky and Gschneidner (1999a) and will not be repeated here. In 1997 Zimm et al. (1998) designed, constructed and
of
0
' '
" "
a ' ' '
20
' ' ' ' ' a
40
80
60
Temperature (K)
Figure 2 The three types of magnetocaloric effects
0
Magne tie Refiigera tion
521
10
5
5
A
0
0
4 -5
-5
-10
-10
0
20
40
60
80
100
0
5
10
15
20
Figure 3 The iiiveise (negative) magnetocaloric effect in GdMnSi(a) and PrNi,(b)
successfully tested a proof-of-principle, near-roomtem~eraturemagnetic refrigerator that showed for the first time that magnetic refrigeration is competitive with vapor compression technology (see Section 2.2).
where W,= TV+ Wl is the reversible work. Therefore, when an isotropic solid is m a g n e t i ~ ~reversibly d at constant pressure and volume, t change its niagnetizatioii by dM field strength, H , is given (e.g. se dWr = HdlM
The basic t h e r ~ o d y n a ~ relationships ic involved in magnetic refrigeration/adiabatic demagnetizatioii are easily derived from the first (equation la) and second (equation lb) laws of the~mody~amics for a closed system
(3)
In the absence of the P-V work, the enthalpy ( E ) IS defined as ~ = U - ~ l M
(4)
Rewriting equation 4 as the full differential of the enthalpy and combining with equations 2 and 3 we have d E = TdS - Md
where U is the internal energy, Q is the heat, T i s the absolute temperature, S is the entropy, and Wand rY; are the work and the lost work, respectively. By combining both equations l a and lb, and eliminating term we obtain the well-known (e.g. see Ragone, 1995) thermodynamic relation
(5)
The Gibbs free energy of such a system is G=E-TS
(6)
which after evaluatin the full ~ i f f e r e ~ t iand a l combining with equation 5 gives dG I=L~ --SdT- M d H
(7)
the s t a n ~ a r d thermodynamic (7) yields the differential equation relating the change of the entropy of a magnetic solid with magnetic field at constant temperature and the change of its magnetization with temperature at constant magnetic field
other paraniagnetic salts is detrimenta~for adiabatic demagnetization applications, aiid therefore, paranzagnetic inter~netal~ic compounds have attracted attention with respect to their magnetocaloric properties. FLirthermore, as suggested by ti et al. (1956) and discussed by Andres (1978), the erfine enh~ncement of the field at the rare-earth nuclei, combined with the dQ, or dST = ( ~ ~ / d (8) q ~ dmetallic ~ conductivity, enables cooling of both the nuclear spin system and the conduction electrons; and is one of the Maxwell equations. therefore both the refrigerant and a load could be rrish (I 965) we consider the entropy to cooled to very low temperatures. Intermetallic Pro f li' and H at constant pressure and containing compounds including PrBi, PrT13, PrCu6, PrCuz, 12 (see review by Andres, 1978) as u s e f ~ lrefrigeraiits. Intermetallic PrNi, (Craig et al., 1972; Andres and Darack, 1977) is still used successfu~l~ in nuclear ut at ~onstantpressure adiabatic deinagneti~ationdevices. In eonjuiiction with Cu as the low tenigeratiire stage and PrNi, as the upper stage, a record was set by Tshimoto et al. (1984) for the lowest working t e m ~ e ~ a t (27 ~ r ep experiments could be performed on materials, otber than on the refrigerant itself. (C,,dq, = - l i ' ( ~ ~ or / ~ ~ ~ d ~ (1 1) dT,y = - ~ T / ~ ~ ) ( ~ ~ / ~ ~ ~ Acfive d ~
uatioiis (8) and (I 1) explain the fundamentals o f by adiabatic demagnet~~ation. First, a paramagnetic solid i s magnetized (i.e. magnetic field is increased b where HF and HI are final and initial s, respectively, and HI is typically ze t constant temperature, TL, wlibch can be reached by conventional nieans (ty~ically
isolated froin the siirroLindings and niagnetic field is resulting in the adiabatic reduction of by ~~~~, which can be calculated by integratin~ e ~ ~ ~ ~(1 t1)i if o i(TIC,,) ~ in addition to refore, effective cooling by adiabatic demagnetirids with the largest values of and TIC, at te~peraturesclose to the abso~utezero. magnetic for the first demonstraof the netization refrigerator. wever, the law thermal conductivity of this and ~~~~~~~~~~
The success of today's magn~ticrefrigeration with the exception of the lowest temperatures (see previous section) rests concept intro have been use to transfer heat from the cold part of a device to the hot part. Steyert s~ggestedthat the standard highheat-capacity solid used as the regenerator be replaced by a magnetic material which has its inagnetic ordering temperat~~re near that of the hot temperature (i.e. the heat rejection) end of the device. equation (10)) all magnctic solids have iz high heat capacity near their magnetic ordering ten~peraturea t~erefore,could serve as an effective regenerator. magnetizing and demagnetizing this magnetic regenerator material at the appropriate time, the amount of heat transferred per cycle can be increased, i.e. the regenerator becomes magnetically active, thereby giving rise to the term act (
~
~
~
)
.
This refrigeration process help of Figure 4. The 'AMR composed of fine particles (usually spheres with a diameter o f 0.3 nim) of the magnetic refrigerant material having an appropriate Curie temperature. One could use thin plates or a wire mesh of the same
-
523
Magn e t ic Reffrige ra timz
I
I
I
I
Schematic re~rese~~ation of an active rnagiietic regenerator (AMR) magnetic refrigerator material in place of the tightly packed bed of spheres. A heat transfer fluid (a gas or a liquid) is used to remove tlie heat from the part of the device to be cooled and to carry it to the hot heat exchanger that rejects the heat to the ambient. This is performed using two displacer pistons to move the heat-transfer fluid from one heat exchanger to the other via the A bed. The final major component is the magnet, w can be either a permanent magnet or a superconducting magnet. The magnet must move relative to the AMR bed (or vice versa), so that at one part of the cycle the magnet imposes a niagiietic field 011 the AMR material, while at the other part of the cycle the magnetic refrigerant experiences no magnetic field. The AMR cycle is com field i s applied to warm up (equation en the cold-heatthe bed material to cool down while the fluid picks up heat from the bed transferring it to the hot heat exchanger; (3) then the magnetic field is removed, regenerator material to cool down further (i.e. below the point where it was cooled by the fluid) and, finally, (4) the war~-heat-transfer fluid passes through the AMR bed causing the magiietic refrigerant to warm up, while at the same time the fluid i s cooled by the bed material. During the highmagnetic-field part of the cycle the fluid exits the bed at a temperat~rehigher than that of the hot heat
exchanger ( TH),and during the 1ow"~agnetic-fieldpart of the cycle it exits thc regenerator at a temperature lower than that of the cold heat exchanger (TL). Therefore, the heat is removed from the material or volume that is to be cooled at TLand is rejected at TH. A proof-of-principle device based on this cycle has been operated for more than 1500 hours over an 18-month period (8 hours a day, 5 days a weck), and during that time the Ast~onautics Corporation of A~erica/AmesLaboratory team has report~dsome inipressive results. They achieved a cooling power of 600 watts, a maximum COP (coefficient of performance - the heat removed at the cold end divided by the work required to operate the refrigerator) of 16, a Carnot efficiciicy of 60% and a t e i ~ ~ ~ e r a tspan u r e (the difference in the t and cold beat exchanger temperatures) of 38 for a i~agiietic-fieldchange of 0 to 5 T near room temperature using Gd metal spheres. More details are found in the papers by et aE. (19%) and Gschneidiier et al. (1999). It should be noted that the temperature in the AM bed is not a constant value across the length of the bed from tlie cold end to the hot end when in op~ration. The steady-state temperature profile varies from T12at the cold ciid to T, at the hot end. When tlie magnetic field is applied the whole temperature profile is shifted by AT' correspondi~g to the AT':,, at the current temperature, which is a function of the bed length, and conversely when the field is removed. This
accounts for the fact that the temperature span noted above (38K) is mLIcli larger than the maximum AT,, for Gd (or any other ma etic refrigerant material) for a 5 T field change (- 10 There are some limi~~tions and difficulties. As is evident from Figure I and equation (11) the larger the magnetic field c h a n ~ ethe larger the AT’,,, and so the efficiency of a magnetic refrigerator increases proportionally to the increased magnetic field. This was evident from the results presented by Zimm et al. (1998). Furthermore, for consumer applications (refrigerators and air conditioners) the magnetic fields will be generated by permanent magnets, thus about 1.5 to 2 S T is the maximuni currently available magnetic field. Hence the COPS and the percent of Carnot efficiencies for a reciprocating device will be significantly smaller than the above noted values, One possible way to improve the performance of the magnetic refrigeratorlair conditioner using permanent magnets is to move away from a ~eciprocating apparatus to a wheel-based device. For large-scale applications, such as cooling large buildings, superm a r ~ e chillers, t food-processing plants, one would use superconducti~gmagnets, and lar e COPSand Carnot efXciencies would be realized even with the reciprocating machine. Another limitation is the nature of the typical caretlike MCE (Figure 1 and 2) - the fall-off of AT,, (or AS,) as one moves further away from T,, Thus the useful practical temperature range is k2OK for lowmagnetic”fie1d devices to k 4 0 K for those that use superconducting solenoids. To improve the efficiency and increase the temp~raturespan, the AMR bed can be inade up of two or more layers of different magnetocaloric materials covering the entire required temperature span. The substance with the lowest T ‘ would be located at the low teniperatL~reend and the material with the highest Tc at tlie hot end of the regenerator. Those materials with intermediate T,s would be located in the center of the AMR bed (Figure 4).
E be measured directly or it can be The ~ C can calculated from the measured magnetization or heat ~ ~ p a c i t yboth , as a funct~on of temperatLire and magnetic field (indirect techniques), The direct and indirect techniques each have advantages and disadvan~ages.
The direct techniques provide only one measure of the magnetocaloric effect, AT,,, and are usually time consuming and difficult to perform, especially for small temperature intervals. If the direct-nieasurement apparatus is not properly calibrated, or if the material is not properly isolated, large experimental errors ( > 20%) become ine~itable,es~eciallyif the ATad values are large (i.e. > 1OK) (Gschiieidner and Pecharsky, 2000a). The two indirect ~easurements(hcat capacity and ~ a g n e t i z a t i ~ both n ) yield AS, while the heat capacity measurements also yield AT,,. The advantage of the ~agnetization measurenients is thdt they can be carried out over a short period of time (few hours to one day) compared to the direct and heat capacity measurements. Thus this technique is used as a screening tool, to see if the ma netocaloric effect is sufficiently large, or unus~al,or of interest to justify further studies. Also ~agnetizationstudies can be used as quick check of the quality of samples. Heat capacity ~easurementsas a function of both magnetic field and temperature take considerably more time, since one needs to determi~e the absolute entropy, S, from 0 to a te~peratureof Tc + 50 I(. in zero and various applicd magnetic fields. The advantage is that one can obtain both AT,, aiid ASM at the same time. Furthe~more,the values of S as functions of temperature and magnetic field are required by the scientist or engineer to design magnetic refrigerators. An extensive analysis of the magnetization and heat capacity methods for determining the has been carried out by Pecharsky and Gsclineidner (199%). In general, quite reliable results can be obtained by both methods for most magnetic materials. However, in the case of a niagnetic first-order transition, one cdn obtain erroneous values for the MCE unless one is quite careful in making the measLire~entsand properly carrying out the calculations. Such procedures hme been discussed in detail for both G d ~ ( ~ i 2 G eand 2 ) Dy, both of which exhibit a first-order magnetic/st~u~tural transition, by Pecharsky and Gschneidner (1999d). Several studies have been carried out to check the reliability of the three methods relative to each other to see if measured data are in agreement. That is, are the A Tad values obtained from direct measureiiient in accord with those calculated from the heat capacity as functions of teniper~ture and magnetic field; and likewise for ASM values obtained from ~ a g I i e t i z ~ t ~ o n vs. those from heat capacity ~ e a s ~ r e m e n tIn s . general the values are in excellent a ~ r e e ~ e ifn tthe measurements arc made on tlie same sample (Gd, ErAl, and DyA12), aiid often even if the samples are different and ~~~
Magnetic Refr igera tim
525
increases. Thus when the A T , ~ * / and A ~ A ~ , / Avalues ~ the I~easurei~ents were made at different laboratories, presented in Tables 1 and 3 are compared to one for example, Gd and GdPd (Pecharsky and Gschneidanother for different substances, one needs to be ner, 1996, l999b; Dan'kov et al., 1997, 1998), and careful, since the numbers may be skewed because of ErCo, (Ciguere et al., 1999; ada et al., 1999a,b). this curvature, even if the AH values are the same. Sometimes the agreement is only fair when different these values are useful for an initial screening or samples, supposedly of the same co~position,were examination of two materials that have about the same measured at different laboratories, e.g. different ErA1, TC, If they differ significantly, one can feel quite and DyAl, samples which were prepared and measured confident that the material with larger A independently in Japan and the USA (Pecharsky and has the better roper~ies(i.e. larger ~ s ~ ~ ~ e i 1996). ~ n e rAs , noted earlier, ~ r s t - o r ~ ~ value r they are close, then o needs to examine t magnetic t r a ~ s i t i o present ~s some difiticult challenges; papers (data) to determine whicli one is the better and, as m e might expect, differences can occur magnetocaloric material. values in the literature for the same As we noted in Figure 2, there are several different (Pecharsky and ~schneidner,1996). temperature dependerzcies of material with the skyscraper-like have a larger AT,, or AS, value than a material whicli has the normal caret-like or tab1 does not mean that the skyscraper-like material will be a better magnetic refrigerant, because in general the temperature range over which it can be so utilized i s The iiiagnetocaloric properties of intermetallic coinsmaller than that of either the caret-like or table-like pounds (either AS, or ATdd,or both) are summarized material. Likewise for the table-like vs. the caret-like in Table 1 for compounds that have only one ordering compounds: the temperature range for the former is temperature, and in Tables 2 and 3 for intermetallics significantly wider than for the caret-like substance that exhibit two or more magnetic transitions. Table 2 (see Figure 2). For the effective operation of a lists the various m a g n ~ t ~transjtion c temperatures, and magnetic refrigerator (MR), both the MCE value Table 3 lists the MC values associated with the and the temperature over which it works are iniporhighest magnetic-transition temperature. Also tant. Thus, the design engineer needs to know the included in Tables 1 to 3 are the corresponding values for some of the magnetic metallic elements, p r i ~ a r i l y refrigerant capacity of a given rege~erator~ ~ t e r i a l , The refrigerant capacity is a measure of how much for comparison purposes. Knformation on the MCE in heat can be transferred between the cold (TL)and hot lanthanide~lanthanide solid solution alloys, and ( T H )sinks in one ideal refrigeration cycle, and is ~anthanum-~anganese perovskites is not considered defined as: in this review, but it can be found in the review by Gschiieidner and Pecharsky (2000a). Since ASM is reported in various units in the literature, we have converted all of the data to mJ/cm3 IS units for uniformity, so the ASM values can be readily where ASM(r ) is the refrigerant's rnagiietic entropy compared to one another. Also, since many of these change as a function of temperature. materials are potential candidates for magnetic refrigeration, the cooling power per unit volume (see Tables Many times the ASM(T') is not well established, since one needs the absolute entropy curves as a function 1 and 3) is the most meaningful parameter for the temperature at several magnetic fields, including zero engineer designing magnetic refrigerators. field. Thus we have introduced i;l new parameter called Another difliculty in comparing the MCE values for one substance with another is that the results are the relative cooling power (RCP). This value is reported for a variety of applied magnetic Gelds, ranging d e t e r ~ i n e dby specifying the I~agnitude of peak (AS, or AT,,) and the full width at half from -0.5 to 107'. So we report AS, and AT,, values in units per tcsla, i.e. ASM/AZf and AT,,/AH. This, of the MCE peak, as shown in maximum (6TFwHM) Figure 1. I n the case of the magnetic entropy change, however, presents another problem in that the ASM and ATad values are not linear functions of H , see the product of the maximum value o f ASM and Figure 5a for AT,, vs. H and Figure 5b for ASM vs. N. 6TF,HM=TH-T'L (see Figure 2a) yields close to 4/3 The curvature in general is larger as the inagnetic field tiines the cooling capacity (equation (12)) in the
-
526
A ~ p ~ ~ c u ~ ~ o ~ s
The magnetocaloric properties and the Curie temperatures (or temperatures at which the niagnetocalo~icefkct is a maximuin) for materials which exhibit onSy a single inagnetic transition atid for which some information exists on the i~~~netocaloric Material
5 5.6 6 (14)" 6.6 (8)" 7 13 (13.5)" 13.6 I6 17.7 20.5 & I .5 21 24 24.4
33.6 22.1 50.4 46.8 48.1 36.4 39.2 42.4 13.9 28.1
30k3 30 31.6 32 (35)" 38.2 39i1 40.8 44 46 47.5
39.0 24.0 25.6 25.2
__
-
472 807
5 4.3 5
__
-
8137 722 873 785 959 599
5 5 10 5 5 8
-
-
-
843
5
1010 839 I070
4 9 5
__
-
909
5
__
-
25. I 45.0 26.6 23.1
902 770 839 900
5 7 5
55.7 63.9 7s 72 82 (85)" 94
22.9 22.0
963 95 1
5 5
-
__
-
19.6 32.0 13.9
86 I 83 1 1180
7
130 136 142 158 I67 196
48.0 107.5 21.4 5.0 8.5 7.6
960 968 729 285 895 260
5 5 7 1,4 5 1.4
214 214 217 225 23 1 232 234 (237)" 247 26 I 265
6.3 8.8 44.6
25 1 246 1030
1.4 1.4 5
-
I
27.2
624
-
-
59.5 9.3 8.7 10.7
833 260 26 1 300
7 9
-
1.2 5 1.4 I .4 1.4
1.50 1.26 0.69 1.17 1.69 1.55 1.64 1.09 1.27 1.37
29.3 29.0 17.5 49.0 52.7 38.5 40.2 53.4 58.9
10
1.50 1.17 1.25 1.40 1.26 1.36 1.25 1.69 I .20
50.4 68.8 16.8 68.0 32.6 66.2 21.1 69.6
8 9 10 6 10 5 10 7 10
10.1lb 12 5 13 5 14 5 15 10 5
1.15 1.10 I .06 0.83 0.85 0.74
74.8 73.7 46.8 40.1 14.4 71.4
10 E0 7 7 6 9
5 5 9 16,9" 16,17' 12
2.10 2.36
46.2 23.6
5 5
-
__ __
__
0.76
68.4
5
-
__
18,1gb 20,19b 16 21 22 21
__
__
-
-
9.8 5 4.3 __
9.8 I _
__
10 5 8 7 10
__
-
-
3.10 0.18 1.49 0.27 3.76
99.0 10.0 89.1 13.5 56.4
5 7 7 7 5
__ __
__ __ __
-
-
__
1 2 3 4 1 4 5 6 5 7 8 9 5
21 21 18 23 24,25b 23 20,19" 21 21 21
527
Mcignet ic ReJi.iger.ution Table 1 (continued)
268 270 272 276 (281)' 278 29 1 294 298
8.4 8.6 6.9 28.0 8.0 7.5 15.4 6.4
823 1120 38 1 672 864 788 1110 418
10 10 1.4 5 10 10 6 1.4
1.06 7.06 3.00 1.09 1.08 2.07 -
118.7 111.3
306 306 313 310 313 (307)" 315.6 (292)" 323 323 33 1 333 336 342 (325)"
14.1 8.9 14.1 7.2
959 312 933 626
5 1.4
96.0
10
__ __
__ __
-
13.1 13.6 11.0 6.7 12.3
888 949 704 703 888
5 5 5 10 5
-
-
-
1.60 1.70 1.08 -4.62 -3.32 1.72 1.88 1.46 1.18 1.76 -1.95
93.6 132.8 - 60.0 - 66.4 103.2 116.6 81.8 145.1 109.2 - 87.7
530 533 631 1-2 665 697 1038s-4 1385.7
-
-
-
__ __
--
0.79 0.90 0.74 0.32 0.46 1.73 1.40
19.8 34.2 38.3 9.7 9.1 95.3 78.4
__
-
-
__
5
10 10 -
63.0 134.1 128.5 161.2
5 10 10 6
__
-
I
5
26 27 21 28 26 26 29 21
1.95
18 21 18 26 30 31 18 18 18 26 18 30
1.58 1.58 1.78 1.58 1.58 3 2.32
32 32 6 33 33 34 35
-
s
10 1.95 2.5 S
5 5 10 5
"Value listed in pareiitheses is the magiietocaloric peak temperature at high magnetic field. hIf two reference numbers are listed for an entry, the first number is for ASM and the second is for ATad. 'Quenched. 'Anneaied. e ~ e ~ toe ~~~~e ~ ~ 1~ s 13. Tishin (1994). 1. Pecbarsky et al. (3996). 25. Green et al. (1988). 2. Korte et al. (1998a, b). 14. Pecharsky and Gschneidner (1996). 26. Niu et al. (2001). 15. Giguere et al. (1999). 3. Tokai et al. (1992). 27. Pecharsky and Gsch~eidner(1999~). 4. Tomokiyo et al. (1986). 16. Foldeaki et al. (1998). 28. Pecharsky and Gschneidiier (1 997b). 5. Gschneidner et al. (1996). 17. Nikitin and Tishin (1991). 29 Spichkiii (1999). 6. Hashimoto et al. (1981). 18. Pecharsky et al. (2000). 30. Wikitin et al. (1990). 7 von Ranke et al. (1998). 19 Pecharsky and Gschneidner (1998). 31. A n n a o ~ ~ ~etoal. v (1996). 8. S~mpathkumaranet al. (2000). 20. Pecharsky and Gschneidner (1997a). 32. ~ i k i t i net al. (1973). 9. Zimin et d.(1 992). 21. Maeda et al. (1983). 33. Nihtin et d.(1975). 10. Hashirnoto et al. (1986). 22. Dan'kov et al. (2000). 34. Hirschler and Rocker (1966). 13. Daudin and Bonjour (1982). 23. Kuhrt et al. (1985). 35. Rocker and ~ o h ~(1967) ~ a ~ s 12. Canepa et al. (1999). 24. Daii'kov et al. (1996).
t e ~ ~ e r a t u range re from TL to TH (~schneidnerand Pecharsky, 2000a). Therefore, we will call the product
sured as the adiabatic t e ~ ~ ~ r a t change, ure AFdd,can be numerically characterized by the product
the relative cooling power (RCP) based on the magnetic entropy change. imilarly, the MCE mea-
(see Figure lb), and it will be callcd the relative cooling power based on the adiabatic t e ~ ~ e r ~change. t ~ r e It
Magnetic ordering temperatures for materials which exhibit multi magnetic orderings and for which some information exists 011 the magnetocaloric effecta
21 22 32 35.5 37.5 40.5 43.5 45 56 58 84 12s 127 128 131 135 179 230 28 1 323 323 340
11 8 22 24.5 26.5 30 34.5 40 20 28 52 20 40 68 90 121 w 90 220 39 (51) 265 265 219
13 15 15 15.5 16 16.5
9 9 7
23 19
-
389 400 490 250 450
%ec references given in Table 3 for the corresponding compounds.
has the dimension of and no physical meaning, but may be useful for n erica1 comparison of different magnetocaloric nuterials, especially when no AS, values are available. A large RCP(T') for the same AH generally indicates a better magnetocaloric material. Both RCP(S) and RCP(1]3generally increase with increasing AH, but they also exhibit a non-linear response witli increasing magnetic field as noted for AS,, and ATad(Figure S), and thus similar cautions to that noted above also apply here. Although RCP(T') does not have any physical significance, Gschneidner and Pecharsky (2OOOa) have shown that there is a oneto-one corres e between RCP(S) and RCP(T), and thus the is a useful and reliable indicator of the cooling power of a magnetic refrigerant in the absence of AS, values. The relative cooling power values have been normalized by dividing by AH, and
these are tabulated in Tables 1 and 3 as R C ~ ( ~ / A H and RCP( T)/AH. elow we have divided into three groups the various i~termetalliccompouiids for which the ties have been reported: compounds of the magnetic 36 transition metals, compounds of the 4f lanthanide metals, and those compou~dswhich contain both magnetic 3d and qf metals.
There are just a few 38 transition metal i ~ ~ e r m e t a l l ~ c compounds which have ordering temperat~iresless than 300 M, These include several ~ i ~ ( ~ n ~ - ~ ~ ~ phases (Tc varies from I58 K for .x=0.4 to 306 K for x=O. l), the (Mn3-,-""Cr~C")AlC series of alloys (Tc varies from 196K for x=O.i6 and y=O, 1 to Tc=298 for x=ji=O), the ,E",) phases (Tc=225, x=0. and 231, X= (HfO 83TaO.171F% +.k (TC=219 for x=O, and both x= i- 0.09 and -0.09) (see Table 1 for the first three series of alloys and Table 2 for the Hf-Td-Fe phases). In all cases the RCP(S)/A,N and R C P ( T ' ) / A are ~ about f to of those of the lanthanide-containing intermetallic compouncl with corresponding Tcs.
4
pending upon the exact composition and heat treatment (three different composit~on~ are listed in Table 1). This unusual compound exhibits a giant negative AT2,,, but because it is a sharp, narrow peak (i.e. it has a negative ) 1 that of the skyscraper-like MCE), its ~ ~ P ( Tis' about Gd,(Si,Ce,-,), where the 2.5 <x< 4.0, normal caret-like MCE. The MCE in an irreversible first-order f e r ~ o ~ a g n e t i ture form)-antiferromagneti~(high temperature form) transition. Tlie peak AT,, values are extremely rocessiiig history of the FeRh phase. in addition to the pure 3d metals Ni,Fe and (510,the only iron-based intermetallic phases that have been studied with respect to their magnetocaloric properties are UFe2 and YFe,. The latter, UFe,, has a MCE comparable to that of pure Ni, while the former has a value about half that of Ni. The small size of the values for the 3d transition metal compounds is not surprising, since the theoretical magnetic entropies (S,) for the 3d metals are one-half to one-fourth those of the lanthanides, because SM=Rln(2J+ 1) for the lanthanides and SM=Rln(2S+ I) for the 3d metals, where the R is the gas constant, and J and S are the total angular
529
Magnetic Refrigeration
T ~ ~ 3l eThe ~ ~ ~ g n e t o c a lproperties or~c and the niagnetic ordering temperatures for materials which e~hibitrnultl ~ a g n e t i c
orderings and for which some information exists on the magnetocaloric effect. The values listed below are for the upper ordering temperature, T,,,( I), unless otherwise noted Material -300 (Gd, 2Ero,,)NiAl Nd
21 22 32 35.5 37.5 40.5 43.5 45 56 58 84 125 127 128 131 135 179 219b 230 265' 265b 2s 1
22.2 -
19.0 18.0 17.2 20.4 15.6 14.2
-
734 855 857 862 937 86 1 866
10
__
_I
10 10 10 5 10 10
__
__
15.2
869
-
-
-
25.6 35.4 57.4 14.4 33.2 20.1 18.3 27.2 6.1 11.7 4.7
69 1 885 1210
5 5 5 6.02 5 6.02 0.9 1.2 0.9 0.9 I0
5
__
664 804 274 624 190 282 542
570 575 575 595 61 8
-
__
0.71
71.0
5 1 7 2 5 1 5 1 10 1 10 1 10 1 10 1 7 3 10 1 6.02 4 5 5 5 5 55 7 4,6" 5 5 6.02 4 7 7 8'6" 7 7 €0 9
0.35 0.47 0.20 0.29 0.14
3.5 16.8 2.0 11.5 4.2
1.58 1.5 1.58 1.5 1.5
2.30 0.36 2.02 1.91 0.91 0.89 0.8 1 0.74 0.42 0.66 0.55 1.44 1.76 2.24 0.86 1.84 1.41
75.9 7.6 92.9 87.4 47.3 49.8 48.6 49.2 12.6 52.6. __
50.4 38.8 40.4 29.1 27.6 91.9
-
__
1.49
85.1
10 I1 10 11 11
"If two reference numbers are listed for an entry, the first number is for ASM and the second IS for ATd, 'T,,,(2): ferromagnetic-antiferrornagnetic transition.
e ~ ~ to ~ ~~~1~ ~ 3~ c 1. Korte et al. (1998a,b). 2 . Zimm et al. (1990). 3. Z i m et al. (1989). 4. Nikitin et al. (1985).
~
s 5. Pecharsky et al. (2000). 6. Green et aZ. (1988). 7 Herhst et al. (1996). 8. Dan'kov et al. (1996).
momentuin and spin quantum numbers, respectively. The J value varies from to 8, while S for the 3d metals is of the order of $ to 1 in the 3d intermetallic compounds. Furthermore, Mn atoms tend to align ai~tiferroma~~etically , and this behavi~rcould also account for some of the low values.
2
S
Tlie largest amount o f information on the MICE lies in this group o f intermetallic compounds (see Tables 1 elow 150 K all of the known magnetocaloric materials contain a lanthanide metal, and some of the ternary phases also contain a 3d metal (see Section
9. Pecharsky et al. (1999). 10. Nikitm et al. (1973). 11. Nikitin at al. (1975)
4.4). Most of the compounds have A ~ ~between. ~ / A ~ 1.25 and 1.50K/T, except for pure Gd (ATad/ AH=2.07K/T), and the Cd,(Si,Ge,-,) phases which exhibit the giant magnetocaloric effect (i.e. have skyscraper-~ikeMCE peaks). Also as seen in Tables 1 and 3 only three compounds have R C ~ values ( larger ~ than ~ that ~ of~ pure C d metal (1 110 m ~ / c i ~ ~ Tc=294 T) (1180), 7;.=94K; G d , ( ~ i ~ . ~ ~(1210), G e ~ , TC=128 ~ ~ ) K; and CdZn (1 120), Tc=270 K. W i l e for RC'P(T)/AH values, none exceeds tkat of Gd (161.2 closest being that for C d , ( S i ~ , * ~ e * * (116,6), ~) Tc=323 K. Thus either GdNiIn or C d , ( ~ i * . ~ ~ G e ~ . ~ ~ ) would make an excellent AMR material for cooling
The magnetic field dependence of the MCE of Gd, Gd,Al,, GdAl,, GdZn, Gd5(SizGez)and Gd,(Si,Ce): (d) ATadand (b) AS,
dZn is about 25K lower than that of Gd, and the former could serve as the lower-temperature d material in a two-layered rege~erator with being the higher-teinperature layer phase to cool below the freezing point of H,O (273 K). Pecharsky and Cschneidner (1999~)have shown that the magnetocaloric properties of the GdGdZn eutectic can be adjusted by varying the C d to Zn
over-all composition of the sample. Large R C F ( ~ / values A ~ have also been observed in Gd-based solid solutions doped with other lanthanide metals (Tb, Dy and Er), e.g. a value of 1570 mJ/c1n3 T was calculated by Gsclineidner and Pecharsky (2000a)
4.3.1 ( ~ ~ ~ - x~ ~~ ~r ~x ,~ ~ ~e ,l s~ The pseudo-~inary (Dyl -xErx)A1, alloys exhibit a linear decrease in the Curie temperature from pure ) to ErAl, (TC=13.6K). The ATd,
and ASM values have an inverse dependence on x, they are highest for ErAl, and lowest for DyAl, ( ~ s c h n e i d ~ e r et al., 1996). Since the .f quantum numbers for Dy and Er are the same (15/2), the theoretical SMis constant as Er is substituted for y in the ( D Y ~ - ~ E ~alloys, JA~~ Thus 111 the absence strong crystalline electric-field eEects one would expect the RCP(5') and RCP(7-') values to be essentially constant from Examination of Table 1 indicates that this is not observed, i,e. there is a nearly linear decrease on going from DyA1, [ R C ~ ( ~ / A ~ 1=mJ/cm3 9 5 T] to ErA1, [722 mJ/cm3 T]. The RCP( ZJ/Afrl values are also consistent with this trend (73.7 Since AS, and AT,, have the respect to R C ~ { ~ and / AR~C P ( ~ / A the ~ , width of increases as the irC the caret-like MCE peak STFwHM increases and as x decreases. The Dy-rich (Dyl-xErx)Al, (0 G x G 0.45) alloys exhibit a spill-reorientation transition from (100) at low te~peraturesto (1 11) above 40 K ( ~ s c h n e ~ d net er al., 1996). Theoretical calculations (von Ranke et al., 2000) have suggested that a negative MCE ina at 20K would be observed in a single crystal of
Magnetic Rejiigera lion with the magnetic field parallel to (1 11). The negative MCE is expected to be largest for small field changes and might be expected to be about zero for a A H > 7 T. No such effect has been reported in the measurements of polycrystalline DyAl, or the Dyrich (Dyl -xEr,)Al, alloys, but the field changes may have been too large for the effect to be observed in poly cry stalline alloys.
4.3.2 Gd,( Si,Ge,-,)
53 1
in Figure 6. Both the and Ge atoms inside the layer (dark gray) are close partially covalent Si(Ge (black) are located on the slab surface connect in^ different layers (slabs) to one another, and forming,
At room temperature in the silicide layers are connected to one another by covalent-like
-based ~ ~ t e ~ ~ e t a l ~ i c s
The Gd,(Si,Ge,-,) compounds were originally discovered by Smith et al. (1967a) and Holtzberg et aE. (1967), who reported extended solubility of Si in Gd,Ge, (0.0 < x < 0.8) and Ce in Gd,Si, (2 d x< 4) and acknowledged the existence of a ternary intermediate phase with unidentified crystal structure extending between 3.2 < x 2. According to Smith et al. (1967b) both Gd,Si, aiid GdsGe4crystallize iii the Srn,Ge4-type orthorhombic structure. Their atomic parameters were later refined from the single-crystal diffraction data for Gd,Si, by lglesias and Steinfink (1972) and from the powder diffraction data for both Gd,Si4 and Gd5Ge4 echarsky and Gschneidner (1997c), which revealed a small difference between the crystal structure of the silicide and that of the germanide (see below). Pecharsky and Cschneidner (1997c) also established that: (1) the intermediate ternary i ~ i ~ e ~ m e t aphase l~ic Gd,(Si~Ge,-~),where x varies from 0.96 to 2, has a monocli~iccrystal structure that is different from but closely related to, the crystal structures of Gd,Si, and GdsGe4; (2) the Gd5Ge4-based solid solution has a narrow homogeneity range that extends from OGx60.8; and (3) a broad homogeneity range from 2 6 x 6 4 for the Gd,Si,-based solid solution. The netocaloric properties of the Gd,(Si,Ge,-,) intermetallics are closely related to their composition and crystal structures, aiid therefore, a brief suininary of crystallographic results is necessary to better illustrate the subject. As shown by Choe et al. (2000) both GdsSi, aiid GdsCe, are built from essentially equiva~entlayers (slabs) formed by five almost Aat atomic nets that are infinite in two dimensions (a and c), as shown in Figure 6. Most of the Gd atoms are located on the vertices of the cubes and rhornbic prisms sharing rectangular faces (the cubes and the prisms are traced but the Gd atoms are not shown in Figure 6 for clarity). The remaining Gd atoms (the smaller and the lightest shaded spheres in Figure 6) are located inside the cubes and are coordinated by octahedra formed by Si or Ge atoms, which are dark rzrav and black shaded srheres
scheniatically in Figures 7a and 7c, respectively. The crystal structure of the Si-rich Gd,( solid solution remains the same as that of the parent silicide with all layers interconnected via the Si(Ge)? bonds (Figure 7a). In the intermediate phase one half of the inter-layer Si( e)z bonds are broken and the room-temperature crystal structure of the alloys with 0.96 6 x < 2 becomes monoclinically distorted and i s composed of alternating stron ly and weakly interacting layers (Figure 7b). The room-temi~eraturecrystal structure of the Gd,Ge4-based solid solution restores
Figure 6 The basic btiildiiig block (a slab) o f the Gd,(Si,Ge,-,) intermetallic compounds. Most of the Gd atoms are located on the verticcs of the cubes and rhonibic prisms (not shown), and the re~aiiiingC d atoms (the smaller and lightest shaded spheres) are inside the cubes. The Si and Ge atoms are shown in dark gray and black (the larger spheres)
532
ME 7 The relationships between the room-temperature crystal structures of the Gd,(Si,Ge,-,) alloys: (a) 2 < xg4, orthorhombic Gd,Si,-type structure; (b), 0.96 G x g 2, monoclinic Gd,(Si,Ge,)-type structure; and (c) 0 <x <0.8, orthorhombic ~ d ~ ~ e , ~ structure. t y p e The layers (tilted by 90” with respect to Figure 6) are shown in light gray. The black circles represent exterior Si and/or Ge atoms ( d e p ~ n d on i ~ ~the coniposition) connecting thc layers. The thick lines indicate the distances consistent with ~ovalent~y bonded pairs and strongly interacting layers
the orthorhombic s y ~ m e t r y(Figure 7c), where the layers remain essentially the same as in Gd,Si, and in the intermediate phase (except for the Si to Ge ratio) but they (the layers) are no longer interconnected. This phase sequence shows that each intermetallic phase in the pseudo-binary Gds(Si.xGe4-x)system is formed by a periodic arrangement of well-defined ~ u i ~ d i nblocks g - the layers of strongly interacting Gd, nsidering that the thickness of the layer res 6 and 7) is of the order of 0.5nm (5A), these compoui~dspresent a unique class of naturally layered systems, where both structure and properties depend on how the sub-nanometer layers are assemble^ to form the bulk material. The drastic changes in bonding, crystal structures, and rnagnetocaloric properties (see below) occur because one of the t ~ e ~ o d y n a ~ ai rc a ~ e t e r(chemical s compos~tion) aEects the i n ~ e r ~ c t i o between n~~ the Iajws (inter~ ~ ~ the layers (intralayers), while i n t e ~ a c t i oinside l ~ ~ remain e ~ ~ essentially s ~ the same. Overall, the properties of bulk Gds~Si~Ge4 -. J intermetallics are determined by both types o f interactions (inter-layer and intra-layer).
The crystal and magnetic structures of the intermediate Gd,(Si,Ge,-,) intermetallic phase, where 0.96 d x d 2, depend on temperature, magnetic field, and pressure. According to Choe et at. (~OOO), who studied the temperature dependence of the crystal structure of Gd5(Si,Ge2)which is a representative alloy from the intermediate phase region (x=2), its crystal structure is monoclinic and the material is paraina netjc at room temperature (Figure 8b, also see Figure 7b). On cooling below 270 IS GdS(Si,Ge2) undergoes a simultaneous magneticlcrystallographic transition, forming an orthorhombic (Figure 8a), ferromagnetic phase that has nearly the same crystal structure as that of GdsSi4 (see also Figure 7a). Therefore, heating and cooling causes the reversible breaking and reforming of half of the covalent Si(Ge), inter-layer bonds. The same ~ r y s t a l l ~ ~ r a pclmnge hic occurs as Gd,(Si,Ge2) is magnetized-demagnetized or pressurized~de~ressuriz~d above 270 K. The change in crystal structure is associated with large shear movements of the slabs relative to one another, as shown by the thick horizoiital arrows in Figure 8. The 0.8 A, while the inter-layer distances change by N
N
N
Magnetic Rq fiigera t ion
533
second-order magnetic phase transitions from a ferromagnet to a paramagnet on warming (Figure 9) are found in the orthorhombic Gd,Si,-based solid solution alloys that are built from strongly interacting layers (Figure 7a). Curie temperatures of these alloys are higher than that of pure elenieiital Gd (294 as the inter-layer interactions of the roomintermediate monoclinic phase (0.96 <x < 2) undergo a change due to the increased con (Figure 7b), so does the magnetism tures the alloys are orthorhombic a (Figure 9), with strong interactions between all layers (Figure 7a). As temperature rises, both the crystal (Figure 8) and magnetic structures and half of the covalent iiiterlayer been broken in the monoclinic paramag~eticmaterials. The crystal structure of (a) the orthorhornbic and This phase transforniation i s a first-order transition. (b) the monoclinic modifications of Gd,(Si,Ge,) phases. The Based on the cominunicatioii by doublcheaded arrows indicate the movements of the layers the low-temperature crystal stru during the rnagnetic-~artensiticphase transition relative to solid solution (Odx60.8) is the same as that of one another Gd,Si4, i.e. all layers are interconnected, but the roomtemperature paramagnetic phases have only weakly intra-layer distances hardly change at all. The monointeracting layers (Figure 7c). On heating, the lowclinic to orthorho~biccrystallographic t r a n s f o r ~ ~ t i o n temperature ferroma netic phase first transforms into in response to changing temperature and magnetic a ferri- or antiferroniagnetic phase via a simultaneous field was also noted for Cd,(Si,~,Ge,~,)by Morellon et magnetic/c~ystallographicfirst-order phase transition, al. (1998a,b), but no crystallographic details (i.e. and then into a paramagnetic phase through a secondatomic parameters) were reported. order phase transformation (Figure 9). Therefore, The large shear displace~entof the atomic layers in magnetic interactions in the Gd,(Si, Gd5(Si,Ge2) and in ~ d , ( S i ~ G e ~ where - ~ ) , 0.96 <x < 2, are especially dependent on the bonding between the coupled with the change of crystallographic symmetry layers (Figures 7-9) creating unique structure-property and magnetic order, characterizes this transformation as magnetic-martensiti~,According to Levin et al. (1999), the start and the end of the magnetic350 martensitic transition depend strongly on the direction of change (i.e. increasing or decreasing) of both 300 temperature and magnetic field. What makes this transformation uiiusual and unique are the facts that (1) it is closely associated with the reversible breaking and forming of covalent inter-layer Si(Ge), bonds; (2) both the crystallographic aiid magnetic structures change si~ultaneously~ a behavior that previously was mainly associated with strong inagnetoelasticity and non-spherical symmetry of the lanthanide 4f orbitals (for Gd the 4forbitals are spherical), and (3) similir t~ansfori~atioi~s are unknown for other Gd16 24 32 40 00 08 containing materials. x (Si) Based on the results reported by Pecharsky and ~ i ~ 9~ The r e magnetic and c~ystallogra~hic phases in the Gschneidner (1997 a-d), and Morellon et al. (1998a,b), Gd,(Si,Ge,-,) system in zero magnetic field. The the magnetism and c~stallographyin the Gd,(Si,Ge, -J orthorhombic Gd,Si,-type structure is labeled 0-1. the system are closely related with the chemical composiorthorhombic Gd,Ge,- type structure i s labeled 0-2, and the monoclinic Gd,(Si,Ge,)-type structure is labeled M tion, as shown by the phase diagram (Figure 9). The
Appliccr t ions
534
re~ationships~ h i c hresult in the extraordinary magnetocaloric (see below), as well as magnetostrictive oselloii et al., 19984 and magnetoresist~~nce o r e ~ ~ oetnal., 199%; Levin et al., 1999) ~ehaviors, Since the magnetic phase transition in the rooinrature orthorhombic phase region of ixGe4-,v), where 2<x,<4, i s a second-order phase transfor~ation,the MCE in these alloys has e and is slightly lower than that o f which is expected since Gd,Si4 and solution alloys represent a dilution magnetic Gd sublattice by nonmagnetic p elements. The i~agnetocaloriceffect in one of the alloys from this phase region, Gd~(Si~,5Ge~,5), as calculated from both ~ ~ a ~ ~ and ~ magnetic ~ ~ ~ field ~ t and i o temperature-dependen~ heat capacity is shown in ure 10. Although the MCE is reduced by about red with that of Gd, it is still quite large, together with other Cd5(Si,Ge4-J compositions with 2 < x ,<4, are the best magnetocaloric inaterials known today for temperatures between -290 and -360K. The characteristics of the MCE for several other alloys from this phase region can be found in Table 1. The first-order magnetic-martensitic phase transition that is ob~erved in the room~temperature monsclinic phase region of Gd5(Si,Ge4-J, for 0.96 ,<x,<2 shows much greater change of magnetization with temperature comp~redto alloys in the 2 < x < 4 range and results in a considerably larger I(aM/8T)191and in a considerably larger MCE (see equations (8) and (11)). The MCE in all alloys from
this phase region exceeds that of pure Cd despite the dilution by non-magnetic elements. Its magnetic entropy change is shown as a function of temperature for ~ d ~ ( ~ i (Pecharsky ~ G e ~ ) and ~ s c ~ n e i d n e1997b) r, in Figure 11 together with that of Cd. The MCEs have been determined for several other alloy compositions froin this phase region, and the corresponding numerical data are given in Table 1. The MCE for these alloys is considerably greater in magnitude, but exists over a narrower temperature ran e. ~ i t the h exception of GdZn, Gd and some Gd-R alloys, the Gd,(Si,Ge,-,) intermetallic alloys from this interniediate-phase region are among the best magneto~~loric materials known today for ~ a g n e t i ccooling between 130 and n 290 K (see Table 1). The first-order nature of the magnetic/crystallographic order-order phase transition in the roomtemperature orthorhoni~ic Ce-rich solid solution Gd5(Si,Ge4-,) alloys, where 0 < x ,<0.8, also results in a large ~ ( a ~ over / a a~narrow ~ l te~peratnrerange and, consequently, in a large MCE (see Figure 12 based on Pecliarsky and Gscliiieidner (1997a) as a r~~resentative example, and Table 1 for other available numerical data). Just as in the r o o ~ - t e ~ p ~ r a t u r e monoclinic interniediate intermetallic phase, the MCE in this phase region is quite large over the narrow temperature range in the vicinity of the firstorder phase transition. Despite a second magnetic ordering observed at higher t e ~ p e r a t ~ r e(Figure s 9), which is a magnetic order- order phase t r a n s f o ~ a tion, the MCE in the vicinity of this high-temperature phase transitio~is ~onsiderablysmaller. The inter-
-
-
$0
80
140
70
120
3
9 30 20
40
n ~ z a t i o ndata
10
20
0
240
re
calculated from both heat capacity and magnetization data for a magnetic-field change from 0 to ST. The magnetocaloric eEcct of Gd 1s shown for ~ o ~ ~ a r i s o n
260
280
300
80
11 The magnetocalori~effect (AS,) for ~ d ~ ( S i ~ G e ~ ) calculated from magnetization data for a magnetic-field change from 0 to ST. The magnetocaloric eRect of Gd is shown for c o i n ~ a r i s o ~
500
400
300
200
100
0
0
20
80
100
The magnetocaloric effect (AS,) for Gd,(Si, ,,Ge, 67) calculated from ~ ~ ~ g n e t ~ z adata t l o nfor a i ~ a ~ n e t i c - ~ change eld from 0 to 5 T. The magnetocaloric effect of DyA1, is shown for cornpanson
metallic Gd,(Si,Ge, J alloys, with 0 < x < 0.8 are among the best magnetocaloric materials known today for magnetic cooling between -20 and 130 K (see Tables 1 and 3). Therefore, a series of ds(SixGe,-x) alloys, where 0 d x 6 4, represents an extremely important class of magiietocaloric iiitermetallic materials covering a large temperature range and they potentially provide eficieiit magnetic cooling between -20 and -360 Adjustment of their chemical composition, more precisely the adjustment of the atomic fraction of Si in their chemical formula from 0 to 1, enables precise tuiiing of their respective magnetic ordering temperatures and, tlierefoi-e, the maximum MCE and operating temperature for a particular inagizetic refrigerant material. Much work still lies ahead before they are successfully commercialized in magnetic refri~erationapplications (iiicl~dingfinding the most suitable preparation and processing conditions, and fiiiding potential s u b s ~ i t ~ ~for t e s the most expensive component, Ge), however, this series of intermetallics is currently a b e ~ c h ~ for ~ athe r ~development of new magnetocaloric materials.
-
The most popular 3d element in the 3 d 4 f containing intermetallic compounds which have been studied for their MCE properties is Ni (17). Ni is followed by Fe n (1). The numbers following the chemical symbols indicate the number of compounds
magnetic, the 3d spins align antiparallel to those of the heavy lanthanide metal, giviiig rise to ferrini~gnetic structures and thus reduced moments ( 9). This eflkct in turn should lead to reduced es, as is evident in the values listed in Table 1 or YFe, (in which U i s a no~i-magnetic earth metal) compared to the values reported for Fe3 (Tables 1 and 3), i.e. the from 20 to 35, for the former, while those for
-
~ * i
the magnetic laiitha magnetic Y. The non-magnetic behavior of Ni, Laves phases, which have The RCo, Laves phases also have similar ~ ~ values, suggesting that the magnetic moment of the atoms is close to zero. The quaternary (Gd,- ~ E r s ) ~compound ~~l series has an interesting set of ~ ~ a g n e t properties. ic compositions exhibit multiple magnet~c transi with a maxiimm of four transitions for 0,5<xd0.6 (see Table 2), only ErAlNi has a single transition (see Table 1). The multiple ~ a g n e t i ctransitioiis give rise to broad-peaked MCEs (both ATaCjand A~~~~ for most alloys. The three exceptions are x=0.55, 0.60 and 0.80, and these alloys exhibit broad caret-like peaks which are about 50% broader than that of the Er compound.
~
4.4.1 Coniparison of (Gd,-,yEr,) ~ i A aizd 1 (Dyl-AEr,y)A12Phases
It is informative to compare compositions in each series that have nearly Id tures. The compositions chosen are and (Dyo.ssEro 45)A1, which have Tcs 4. Froin the MCE properties, AS, a magnetic regenerator than when one examine especially the RCP( , it would appear that the fornier i s not much better than ( To evaluate these two materials bet tioii capacity (equation (12)) for the two ~ o i ~ i ~ o u ~ ~ d s was d e t e r ~ i ~ efor d va~iousfield ~ ~ in d ~a ~ e r ~e ~ t ~
(
temperature ranges, see Table 5. In Table 5 the values in the boxes are higher ones for the equivalent field change and temperature range. For low field changes y~,55Er*.45)A12 phase is better and also for the smallest temperature range (20 to 60K). But for other temperature ranges and higher field changes the picture is not as clear. However, the differences in q values are generally only a few percent; so the filial choice of the material to use niay well depend upon other properties. These include: (1) the cost of the raw materials; (2) cost of manufacturing the material in the desired f'orm; (3) physical properties such as thermal conductivity; (4) mechanical properties; and (5) eiivironmental concerns. (Dy, - xEr,)A12 is clearly the choice inaterial because A1 is much less expeiisive than Ni and the cost difference between Gd and Dy is fairly small, with Gd being cheaper (item 1); the RAl, phases melt congruently and the RNiAl phases do not, and the latter may need to be heat treated before they are used (item 2); and Ni i s a carcinog~n,the other metals are benign (item 5). With respect to item (3) the thermal conductivity of (Gd*.~~Er*,~~)NiAl would be expected to be lower than that of RAl, because of the fourth element (Ni), which in turn would be expected to increase the plionon scattering, leading to a lower thermal cond~ctivity,which would mean lower longitudinal thermal losses in the regenerator. The iiieclianical properties (item 4) for both phases are not known, but they are both brittle and the degree of brittleness could be a factor.
esearch on magnetic refrigeration and especially on magnetic materials for active magnetic regenerators has rapidly expanded in the past five years. Progress has been accelerated by two breakthrouglw that were announced in 1997. First was the announcement on February 20, 1997 that scientists at Astronautics Corporation of America (Madison, Wisconsin) and
The ~agnetocaloricproperties of (Gd, $3-, ,,)NiAI and (Dyo55Ero45)A12 for a magnetic field change of 0 to 5T (for ASM and RCP(S))or 0 to 10T (for ATad and RCP(T))
40.5 102 4680 8.9 498
T, (E0 -ASM (IIIJ~CIII~K) - RCP(S) (mJ/cm") AT,, (K) R W T ) (K2)
40.8 I26 4510 12.5 662
Ames Laboratory, Iowa State University (Ames, Iowa) had successfully demonstrated magnetic refrigeration to be a viable and competitive technology with gascycle refrigeration (see Section 2.2). Second was the June 10, 1997 report of the discovery of a reversible giant magnetocaloric effect by the Ames Laboratory, Iowa State University group (see Section 4.3.2). Several new inte~metallicmaterials with substaiitial magnetocaloric effect properties have been discovered over the last five years. These include: (1) the (Dy, -,xEr,)Al, series of inter~etallics (see Sections ~ i A l which 4.3.1 and 4.4.1); (2) the ( G ~ * - ~ E r ~ )phases, have inultiple magnetic transitions and thus uiiusual MCE temperature dependences from 5 to 65 Sections 4.4 and 4.4.1); (3) the giant MCE Gd5(Si,, G e 4 4 intermetallic phases which have tunable magnetic ordering temperatures from 40 to 290 K for x < 2, and the largest known normal caret-like MCE for the 300 to 360 I(range for x > 2 (see ection 4.3.2); and (4) the large MCE materials GdNiIn. (Tc=94K) and GdZn (Tc=270 K), which have ~ C ~values( slightly larger than that of Gd metal (see Section 4.3). Magnetic refrigeration has come a loiig way since the pioneering work of Giauque (1927) and Debye (1926). Tremendous advancements have been made in the development of the adiabatic demagnetization process, enabling one to approach more closely the unreachable absolute zero and to discover many novel low-temperature physical phenomena. What was once
The ~ e f ~ i g e r a t capacity io~ of (Gd, 54Er,,,)NiAl and (Dy,,5,Er, ,$Al, (q is in units of J/cm3)* (Gd, 54EroI",, Field changes
"K T e i ~ ~ e r a t u range re 10-60 5-70 20-60 *~efri~eration capacity
0-5 T
0-7.5 T
4.07 4.61 3.24
[".751 6.52 4.60
(DY" 55Ero 4 w 2 Field changes 0-10T
m1
g r5 3 J .
5.89
0-5 T
0-7.5 T
0-10 T
~
/
thought to be curiosity-driven research has been transformed into a real techiiological possibility that energy-efficient and environmentally-friendly magnetic refri~erationand air conditioners will soon be offered for sale around the world. The road ahead is not paved, however, it is only traced. We know the d i r e ~ ~ i obut n , we also have to realize that much more basic and applied research is needed and will be carried out along the way, no doubt resulting in the discovery of better materials, better magnetic field sources, and ingenious magnetic refrigerator designs. ~ntermetallics are by Par the most effective magnetic refrigerants and the permanent magnet material for magnetic field sources. Magnetic lan~hanidesplay a vital role for both, and it is possible to predict that they will remain at the forefront as their complexity is further uncovered and put to use, benefiting basic and applied science in general, and magnetic refrigeration in particular.
The authors wish to thank their colleagues Jessica Anderson, Jake Auliff and Paul Toinlinson for assistance in compiling the numerical data used in this review; and Alexandra Pechnrsky and Youri Spichkin for allowing us to use some of their unp~iblished results. Different aspects of this work were supported by both the Materials Sciences Division, Office of Basic Energy Sciences and by the ~ a b ~ r a t o ~r y~ c h n o l o gResearch y Program, Office of ~ o m p u t a t i o ~and a l Technology Research, both of the US Department of Energy.
rences
Andres, E;. (1978). Cryogenics, 38, 473. Andres, K. and Darack, S. (1977). Physica B+C, 8 1071. Annaorazov, M. P., Nikitm, S. A., Tyurin, A. L., Asatryan, IS. A., and Dovletov, A. Kh. (1996). J. Appl. Phys., 79, 1689. Canepa F., N a ~ a ~ e ~ M., ~ n Palenzona o, Cirafici, S. (1999). J . Phys, D: Appl. Choe, W., Pecharsky, V. K., Pecharsky, , Jr., and Miller, G. J. (2000).
Craig, R. S., Sankar, S. G., Marzouk, N., Wallace, W. E., and Segal, E. (1972). ,I. Ph-ys. Chem. Soli~s,33, 2267. Tisliin, A. M. (1996). Dan’kov, S. Yu., Spichltin, J . M a p Magn. Mater., Dan’kov, S. Yu., Tishin, charsky, V. I<., and Gschneidner, K. A., Jr. (1997). Rev. 5ci. r n . s t r ~ ~ ~ . , 2432. Dan’kov, S. Yu., Tishm, A. M., Pecharsky, Gschneidner, I(.A., Jr. (1998). Phys. Rev. B, Dan,kov, S. Yu., Ivtchenko, V. V., Tishi Gschneidner, K. A., Jr., aiid Pccbarsky, V. M. (2000). Adv. Cryog. E~ig.,46, 397. Daudin, B. and Bonjour, E. (1982). Compt. Rend. Acud. Sci. Paris, 295, 535. ~ 81, 1154. Debye, P. (1926). Anti. P h y . ~ iflerpzig),
768. Giguere, A., Foldeaki, M., Schnelle, W., and Gmelin, E. (1999). J . Phj’s.: Condens. A4utiter, Green, G., Patton, W., and Stevens, J. (1988). Adv. Cryog. Eng., 33, 777. Gschneidner, K. A., Jr. and Pecharsky, V. K. (1997). In Rare Earths: Science, Teclznolqy and A p ~ ~ l i c u i i111 o ~(eds , ~ R. G. Bautista, C. 0. Bounds, T. W. Ellis and TMS, Warrendale, PA, p. 209. Gschneidner, E;. A., Jr., and Pecharsky, V. IL (2OOOa). A m . Rev. Mater. Sci., 30, 387. Gschneidner, K. A., Jr, and Pecharsky, V. K. (2000b). m at er. SCE.Eng., ~ 2 8 7 301. , Gsclincidner, IS. A., Jr., Pec (1996). Adv. Cryog. Erg., Gschneidner, E;. A.. Jr., Pecharsky, V.K.> and Zinirn, C. B. (1999). In Proc. 50th Ann. Intern, Ap~liffnce Tech. Con$, West Lufayette, Indiana. International Appliance Technical Conference, Inc., Largo, FL, p. 144. Hashirnoto, T., Nurnasawa, T., Shino. M., and Okada, T. (1981). Cryogenics, 21, 647. Hashirnoto, T., Matsurnoto, K., Kurihara, T., Numazawa, T., Tomokiyo, A., Yayarna. H., Goto, T., Todo, Sahashi, M. (1986). Adv. Cryog. Eng., 3 Heer, C. V., Bariies, C. B., aiid Daunt, J. G. (1954). Rev. Sci. I ~ ~ s ~ r25, u ~1088. ., Merbst, J. F., Fuerst, C. D., and ~ c ~ i c h a eR.l ,D. (1996). J . Appl. Phys., 79, 5998. Hirschler, W. and Rocker, W (1966). Z. ungew. Phy.~., 368. Holtzberg, F., Garnbino, R. J., and McGuire, T. R. (1967). J Phys. Chcm. Solids,28, 2283. Iglesias, J. E. and Steinfink, H. (1972). J . ~ e s s - C o m m oMet., ~ 26, 45. Ishinioto, H., Nishida, N.. Furubayash, T., Shinohara, M., iura, Y., and Ono, E;. (1984). J . Low Temp.
. and Poldy, C. A. (1979). In
~ a i ? ~ on ~ o o Pecharsky, ~ V
K. and ~schneidner,K. A., Jr. (1999~).In Cryocoolers 10 (ed. R. @. ROW,k.). Kluwer Academic,‘ ~Io~land, Plenum P~blishers*New York, p. 629. Pecharsky, V. K. and Gschnei~ner,E;, A., Jr. (1999d). J .
the ~ h y s arid ~ t~~h i~~ ~~~ i iofs tRare r ~ Earth, Vol. 2 (eds. K. A.
~ ~ ~ n e 3r. i ~ and ~ eL.r Eynng). ~ North
Pecharsky, V. K., Gschneidn (1996). Adv. Cryog. Eng., -,Cschneidner, K. A,, Jr., Dan’kov, S. Yu., and Tishin, A. M. (1999). In Cryocooler~~ 10 (ed. Ross, Jr.) Kluwer Acadern~c/~le~urn Publishers, York, NY, p. 639. Pecharsky, V. K., Pecharsky, A. O., Tomlinson, P., and A., Jr. (2000). u~publishedresults. 5). ~ h e r ~ ~ d ~qf ~~ ~a at ~ ~i ciWiley as ~ ,, ~ . Rocker, W and Kohlhaas, R. (1967). Z. nngew. Phys,, 146. Sampathli.clmaran,E. V., Das, I., Rawat, R., and Majumdar, S. (2000). Appl. Phys. Lett., 7’9, 418. Smith, G. S., Tharp, A. G., and Johnson, Q. (1967a). Actn
~orelloii,L., Blasco, J,, Al~arabe~, P. A., and Ibarra, M. R. (2000). Yh.J’S. Rev. B, 62, 1022. . The P~ysica~ Pri~cipleso ~ ~ a g ~ e t i s ~ . Nikitin, S. A. and Tiskiii, A. M. (1991). Cryogiw.ic,s, 31, 166. ikitin, S. A., Talalaeva E.
Q.,and Tharp, A.
6;. (1967b). Acta
Arnes Laboratory, Iowa State University, Arnes, IA, private co~rnunic~tion.
., Chernikova, L, A.,
S., Tishin, A. M., Arklarov,
Tishin, A. M. (1994). J. Adv. ~ n t @ r1, . , 403. Tokai, Y., Takahashi, A., Sahashi, M., H a s h i ~ ~ T., ~o, nd Tomokiyo, A. (1992). Jpz. J. Appl.
Tyurin, A. L. (3990). Phys.
Tornokiyo, A., Yayarna, H., ~ a k a b a y a s hH., ~ , Huzufiara, T., Xnomata, E;. (1986). Adv.
947; Engl. transl. Ph.~w.Me
, Is., Pecharsky, A. O., and Pecharsky, V. K. (2001). J. Magn. Magn. Mater., sc~neidner,K, A., Jr. (1996). Adv.
Phys. Lett., 70, 3299.
G s c h ~ e i ~ nI( ~.rA., , Jr., and Kortc, B. J. (1998). Phys. Rev. B, 5 von Ranke, P, J., de Oliveira, I, G., G Silva, X, A. (2000~.Phys. Rev. B, Wada, H., Tornehawa, S., and Shiga ~ ~Marer., g 196-197, ~ . 689. Wada, H., ~ o ~ e h aS.,~ and a , Shiga, M. (1999b). ~ r ~ o g e ~ i c , s , 39, 915. Zirnrn, C. B., Bttrclay, 3 . A., Harkness, H. H., Green, G. I?.,
A., Pecharsky, V.,
., and Anderson, I.
al., 2001). The authors obtained a cooling power of 100 watts using a 40 kOe magnetic field with a 25 Cschneidner, K. A., Jr. aiid Pecharsky, V. K. (2000). temperature span. The results obtained by the Jap Magnetocaloric materials. In Annual Review cfl ~ ~ ~ e r I a Z ‘ ~ nese team compare favorably with those obtained on Science, Vol. 30 (ed. E. N. Kaufinanii). Annual Reviews, the A/AL apparatus. Palo Alto, California, p. 387. The other near-room-temperature magnetic refrigPecharsky, V. K., Gschneidner, K. A., Jr., Pecharsky, A. O., and Tishin, A, M. (2001). Thcrmodynainics of the erator has been designed and built iiiagiietocaloric effect. P ~ y s ~ c Review ul B, 64, 144406. University of Victoria (Victoria, Pecharsky, V. K. and ~schneidner, K. A., Jr. (1999). operate at a frequency of 1 Magnetocaloric effect and magnetic refrigeration. In field of 20kOe (Rowe and Magnetism Be~vund2000 (eds. A. J. Freeman and S. D. approach is to increase th Bader). ~ o r t h - ~ o l l a n d , Elsevier Science B.V., minimize the amount of magnetic refrigerant utilized Amsterda~,p. 44. in the refrigerator, but as of July 200 L no e x p ~ r i ~ e n t a l Pecharsky, V. K. and Gzchnerdncr, K. A., Jr. (2001). Some results were available. common misconceptions concerning magnetic refrigerant In October 2001 Astronautics ~orporation of materials. J . A ~ ~ l~ i~ ey ~~ ~ 90, i c 4614. s, America announced that they had ~uccessfullyoperTishin, A. N. (1999). Magnetocaloric effect in the vicinity of phase transitions. In Handbook of Mugnetic Mutenals, ated the world’s first room temperatu Vol. 15 (ed. E;. H. J. Buschow). Elsevier Science B.V., magnet, rotary magnetic refri~erator011 A m s t e r d a ~p.~395. 2001. This represents two major advarx powered by a permanent magnet; and a rotary rather than a reciprocating one. et
In mid-2001 several major new developments occurred regarding near-room-temperature magnetic refrigeration. At the Cryogenic ~ n g i ~ e e r i nConference g in Madison, isc cons in, July 2001, Hirano et al. (2001) (2001) reported on their h devices are reciprocating design to the Astronautics/ Ames Laboratory (A/AL) proof-of-principle magnetic ection 2.2). The most advanced of these two is the Chubu Electric Power Company (~agoya)/Toshiba (Uokoharna) apparatus (Hirano
Herman, R. (2001). News release from Astronautics Corporation of America, Milwaukee, WI, October. Hirano, N., Nagaya, S., Takahashi, M., Ito, K., Kuriyama, T., and Nomura, S. (2001). Paper No. C-11E-OS, Cryogenic Engineering Conference, Madison, WI, July 16-20. Rowe, A. M. and Barclay, J. A. (2001). Paper No. C-1lE-01, Cryogenic Engineering Conference, Madison, WI, July 16-20.
This Page Intentionally Left Blank
it
The requirements for greater aircraft-engine performance, greater thrust-to-weight ratios, aiid greater file1 efficiency have resulted in signi~cantincreases in turbine gas-path temperatures. Present-day aircraft engines have combustion-gas temperatures well in excess of the melting temperatures of the airfoil alloys, and therefore rely on sophisticated cooling methods to keep the alloys solid. Both materials and airfoil blade designs have evolved to sustain these increasing demands (Bewlay et al., 1999a; Subramanian et al. 1996). Advances in high-temperature materials have had a niajor impact on the efficiency of gas-turbine engines, so that currently superalloys provide a maxiinum surface temperature capability of 1150 "C. Tlie evolution iii HPT (high pressure turbine blade) cooling technology is illustrated schematically in Figure 1. In the 1960s, equiaxed-gr~inNi-based superalloys were cooled with radial cooling passages and with film cooling holes at the leading and trailing edges to reduce the interaction with the combustion gases. This advance over uncooled hardware improved blade durability, and allowed an increase in turbine inlet temperatures to greater than 1100"C ( ~ 0 0 "E;). 0 Once cooling of hardware became routine, and tlie ability was developed to cast airfoils with more complex cooling schemes, these gains in cooling effectiveness were coupled with improved investment casting technology, and with the introduction of directional solidi~cat~onfor the production of HPTBs with either columnar or single-crystal microstructures. In the past decade, the potential of new alloys strengthened with intermetallic compounds with low
-
densities, high elastic moduli, and high melting ranges (Diiniduk et al., 1993; Subramanian et al., 1997) has been explored. Intermetallic-based compound materials, such as Nb or MO silicides, have been combined with metallic secoiid phases in order to generate composites with a combination of attractive hightemperature properties and acceptable low-temperature properties. Nb-silicide based i ~ - s composites i~~ with Nb$i and/or NbsSi, silicides have been shown to have great potential because of their attractive balance of high- and low-temperature mechanical properties (Mendiratta et al., 1993; ewlay et al., 1996, 1997). These materials have th potential to surpass the performance of Ni-based superalloys. This chapter will describe directional solid~fication and single-crystal technologies, with ~articLilarconsideration to their r in present and future aircraftengine applications. eas that will be covered include Ni-based superalloys strengthened by Ni,Al, DS eutectics of Ni-based superalloys, and DS Nb-silicide-based i ~ - s i t composites. ~ This chapter will compare the role that Ni-based superalloys and Nbsilicide composites play in improving tlie performance of gas-turbine engines. ~ i c r ~ s t r u c t u r ephase s , compositions, and mecha~icalbehavior will be reviewed.
ir~ctional
catio Alloys
The performance of many h~gh-temperaturestructural materials that are used in rotating applications can be improved by directional solidification. The microstructures which result have a mini mu^ number of boundaries perpendicular to the principal stress axis;
I n t e ~ m e t a ~C l ~ c~ ~ p a Vol. ~ n3,~Principles s ~ and Practice. Edited by J . H. Westbrook and R. L. Fleiscber. @ZOO2 John Wiley & Sons, Ltd.
542
(a) Diagram showing the tempcrature limit of Ni-based superalloys as a function of year o f introduction (schematic), (b) HPT inlet gas t e ~ p e r ~ ~as u raefunction of blade cooling designs
so creep failure can be substanti~~llydelayed by minimizing these early-failure initiation sites, In tlie extreme, solidification can be controlled so that a single crystal of the structural material is produced, with no transverse bou~daries. Directional s ~ ~ ~ i d i ~ ~ c aprocessing tion has been applied to many i~termetalliccompounds, particul~~rly Ni,Al, TiAl, and eutectic fractory phase reinforcement, cLeaii, 1983; Coldman, 1992). The high melting points of the alloy classes and the high stabi~ityof oxide phases based on the intermetallic alloy constituents limits the choices for niolds. The most c o m i ~ o n l y ~ u ~mold e d systems are based on the alumina-ni~illiteportion of the alumina-silica phase zircon-silica portion of the zirconiaam. For the al~niina-ni~llite system, the melting point is around 1890“C (Klug et al., 1987); and for tlie zircon-silica system, the melting point i s about 1687C. The relatively low upper-use temperatures for candi~aternold ceramics limit the amount of superheat that can be used during directional solidification, and hence there has been a continual effort to improve casting quality at low superheats by increasing methods of heat input and removal. Directional so~idification was first used at the begi~ningof the twentieth century to obtain samples for physical ~ r o ~ e r tmeasurements. y Block (191 I) measured the volume change on solidification by directionally solidifying metals and organics in small
glass tubes that were pointed at the bottom. The entire tube would be slightly undercooled, and this could be considered an early application of the present-day ‘power down’ technique. idgman (1925) also developed a novel technique obtain single crystals by lowering a melt out of a furnace. Bridgrcnaii used a two-cha~bered crucible with a constriction in the middle to select a single grain, an early example of the so-called ‘selector’ common1 used for casting single crystals of superalloys. The ridgrnan approach, was modified by Stockbarger (1936) to contain a chilling device at the bottom of the casting mold to initiate solidification. The improved h~~h-temperature m ~ ~ h a n i properca~ ties that could be obtained from directional grain structures provided a major driving force to improve directional solidi~c~tion t e ~ h n i ~ u e sVer~nyder . and e d property improveGuard (1960) d e ~ ~ o ~ i s t r a tsuch ments in a model superalloy. They produced c~lumnar~grained Ni-2 1Cr-3.SA1 by casting cylindrical rods using the ‘ e x o ~ o ~ technique. d’ They conducted creep tests on specimens oriented in ~ o i ~ ~ i ~ ~ d i n ~ ~ transverse and oblique directions relative to grain orientatio~;loiigit~dinalspecimens displayed longer life and higher ductility. ~ircraft-e~gine turbine blades were produced by a variant of the exomold t e c ~ n i ~ uby e Piearcey and VerSnyder (1966). They djrectio~allysolidified blades of alloy ~ a r - ~ 2 0by0 casting the alloy into a preheated mold positioned on a copper chill. Piearcey
~ i o ~Silicide i ~ iIn~Situ Composites
and Terkelsen (1967) tested ~ a r - M ~ O 0 €3-1900, , 100, and TRW 1900 in both conventionally cast DS conditions; samples taken parallel to the longitudinal DS direction had longer rupture lives, higher rupture ductility, and better thermal-shock resistance than conventionally cast alloys, while Mar-M200 samples taken perpendicular to the longitudinal direction possessed rupture lives equivalent to condemonstrated mechanical advantages due to the absence of grain boundaries, but they were found to be highly anisotropic. Piearcey (1967) observed high tensile and creep strengths near [OOl] and [111] directions, and low strengths in orientations that can provide for single direction slip. Erickson et al. (1971) adapted the €3ridgman-Stockbarger technique to directionally solidify castings of superalloys, and they called the technique ‘high rate solidification7.Their approach was to lower out of a furnace a mold containing a cluster of blades. Giamei and Tschinkel(1976) devised a method to improve the heat removal during DS by ower ring the casting into liquid tin. This liquid metal cooling ( L M C ~techniqu~, which lias been described for right circular cylinders, produced higher thermal gradients and finer dendrite arm spacings. For the e~uipmentthat was used, the LMC growth rate was limited to about 4 cm/min, because at higher rates there was too much solid-liquid interface curvature. The LMC approacli demonstratcd with shell molds in airfoil shapes by lov et al. (1995). They used liqui~-tincooling with both shell rnolds 0.8-1 cm thick, as well its alundum tubes 0.1 cm thick, and they achieved tliennal gradients as high as 200°C/an. Blades in shell inolds could be directionally solidified at rates up to 3 cm/ i i n , and cylinders were cast in alundum tubes at rates up to 8 cin/inin. The mechanical properties of LMC cast material were compared with those of conventionally cast alloys; the LMC alloys provided better tensile and rupture strength. Tlie Russian-developed LM C fbrnaces were evaluated in detail by Hugo et al. (1994) and Betz et al. (1995). They conducted extensive thermal modeling, and concluded that the LMC approach had merit, particularly for large castings. s been adapted for the The LMC approach casting of very large parts, as described by GroDmann et al. (1997). Microstructural analyses demonstrated fine dendrite arm spacing - indicative of high cooling rate, finer-scale carbides, and finerscale eutectic pools than in the equivalent conventioiially cast alloys. Fitzgerald and Singer (1997) developed
543
a closed form model of L G by putting the solidliquid interface within the radiation baffle. They evaluated the effects on gradient and directional solidification rate, cooling-bath temperatLire~ mold thickness, and rnold conductivity. Coolant teniperature was found to be the dominating factor, with lower coolant temperature produc~nghigher thermal. gradients and permitting higher casting rates. Processing rates for producing single crystals are very different for intermetallic materials, sucli as GaAs, used for electronic applications, and for AlNi, which has been evaluated for structural applications in aircraft engines. For the electronic materials, very low rates, e.g. < 1.5 x 10-2cm/min, are employed to achieve planar front solidification, which results iii minimLLm defect density (dislocations and point defects) and well-controlled high purity. For str~~ctural materials, typical growth rates of 0.15-0.5 cm/min are used and these result in a non-planar liquid-solid interface. This allows economic prod~ctionby rapid throughput of near-net-shape components. Generally, a multitude of grains will nucleate at the initial solidification site, and through competitive growth, the fastest-growing grains will expand laterally first to produce a coarse-grained elongated microstructure. If the initial solid~ficationis controlled so that a grain grows from a single-crystal seed, then a singl.e-crystal coinponent can result . Alterna tively , when the liquid is contained in a ceramic mold, growth of the initial multitude of grains can be choked off by using a geometrical grain selector between the nucleation site and the actual component region of the mold. The grain selector configu~ation,such as a ‘pig-tail’ spiral or a multiple right-angle turning selector, perinits only a single grain to continue to grow through the tortuous selector path to the corn region of the mold, so that the final component produced is a single crystal. Ni-based superalloys are used in a wide ran applications as directional or single-c~ystalstructures, particularly in the high-pressure turbine blades of advanced turbine engines. Al~houghmany of these superalloys are more than 50% intermetallic phase by volume (L12 AlNi3, y‘) at service teiiiperatures, they solidify as single-phase solid solutions based on y-Ni (AI), with minor fractions of carbides and borides. Some y/y’ eutectic can form interdendritically, depending on the alloy cheniistry and solidification conditions. However, the bulk of the i n t e ~ e t a l l i c phase that performs the strengthening is formed by solid-state precipitation during cooling from solidification or by subsequent heat treatment. The y’ phase is
formed peritectically from the liquid plus most compositions of structural inter nal solidification of y’ is not possible. h of the experience gained in directiona~ solidification of superalloys is directly transferable to intermetallics, particularly to cubic materials, such as AlNi. Control of the unidirectional temperature gradient, and optimizing the temperature profile in the furnace, are critical to the production of directional or sing~e-crystai intermeta~lic structures, Operating within a ‘process window’ of the ratio of thermal gradient to growth rate is potentially even more restrictive for inte~etallics,compared to superalloys. The undesirable creation of sub-grain boundaries can be generated by growth of diverging dendrites from a common nucleation site. In liquids of intermetallic alloys, this effect may be enhanced due to local perturbations in the liquid of major-element stoichionietry, minor-element concentrations, and trace impurities. Even for cubic interi~etallicphases, faceted dendritic growth can occur, and the nucleation of secondary dendrite facets can lead to greater volmnes of micro-shrink porosity where secondary arms make feeding of liquid to the starved interdendritic regions very difficult. This can lead to generation of stressraising notches at micro-shrinkage, a serious concern for inte~etallicsthat have a low fracture toughness. Compositional control in directional solidification of interme~allic alloys can also be difficult due to elemental constitutio~~l partit~~ning between liquid and solid. Substantial transverse chemical gradients may result. The interaction of liquid-solid partitioning with local liquid density modification from the parent alloy can lead to defects in superalloys called ‘freckles’. If liquid of high density (rich in W, MO, Ta, Re) r e p ~ ~ cthe e s i~terdendriticl i q u i ~of low density (richer in AI, Cr), strings of fine grains can be nucleated in the interdeiidritic area. An analogous inversion can be envisioned for inter~etallics,even for a high~purity binary composition, if the composition is noncongruently melting . Ni-base superalloys typically have liquidus temperatures of about 1350°C. They have bene~tedfrom low interaction rates between their liquids and the ceramic materials that are generally used for molds and cores. This has allowed net-shape processing of direction~lly solidified and single-crystal components with the generation of minimal interactions between the liquid 2 AlNi alloys with liquidus temperatures =. 1600 “C have also been processed to net-shape in cerainic molds. However, for m.a;ny of the potentially useful intermetallic-based composites, the high
liquidus temperatures and high reactivity with oxide ceramic molds require the use o f containerless directional solidification, so that net-shape processing is no t possible for most compone~tsof interest with the present techniques. The area of containerless solidification processing requires considerable further attention.
The development of DS eutectic irz-situ composite turbine blades was derived from a rich knowledge of Ni- and Co-base superalloys for the strong, ductile composite matrices that were employed. Strengthening of these matrices was accomplished with both lamellar intermetallic cornpounfls and fibrous monocarbides TaC, TiC and NbC development of mo site materials conc~~itrated on CO-base systems with TaC fibers, but as designs required greater strength at intermediate temperatures, CO was replaced by Nibased y’-strengthened matrices. Aligned eutectics were produced in alloys that were essentially multi-element superalloys. Studies deiiionstrated that carbide svability in thermal cycling was excellent for a broad range of composite compositions (Jackson et al., 1982). These eutectics possessed very high-temperature strength ( > 120 MPa for 1000 hours at 1100 “G), with creep-rupture capability far beyond any superalloy at that time. Excellent high-tei~perature strength in the aligned monocarbide eutectics was not their only advantage over the superalloys of the time. Through iterative alloy approaches, oxidation and fatigue behaviors were improved, transverse ductility was improved with boron additions, and matrix stability during cyclic thermal exposure was improved, providing excellent property retention (Jackson et nl., 1982). Engine tests were successful (Cigliotti et al., 1982; Menzies et al., 1988), both for solid and hollow owever, there were mass-production di~cultiesdue to the solidification-~rocessingtechniques that were required to generate aligned structures. The low solidification rates required for aligned growth (~ypically < 3 x 10-2 c ~ ~ m i nand ) , the small numbers of parts that could be ~anufacturedin a single furnace cycle, resulted in a large number of furnaces for the necessary production volumes. The financial investment was therefore very high and the return on investment too low to malte this technology an attractive business proposition at tkat time.
~ i o b i uSilicide ~ In Situ Corq?osites
Although eutectic-based superalloys are not actually used today, the spin-offs from the eutectics research in single-crystal superalloy chemistry and casting technologies are at the heart of today's jet engines. Even today, SX superalloys do not possess the strengths of the DS eutectic alloys from the early 1980s.
The evolutionary improvement in superalloy capability has been remarkable (Figure 1). The single crystals of today operate at high stresses within a few hundred degrees of their melting temperatures. However, the gains in superalloy capability have begun to slow. With a highly alloyed gamma prime at -65 v/o in the strongest superalloys, there is little room for further precipitation strengthening. The refractory metals that have been previously judged as being most suited to the engine environment kinan, 1988). Niobium is ductile has wide solubility for a number of strengthening additions, and is relatively low in density (lower than Ni) at 8.56 g/cm3. ~ o l y b d e n u m has good solubility for strengtheners, but it has a high density (10.2 g/cm3), and relatively low ductility at room temperature. It also has substantial oxidation limitations. The density of MO is such that wiless very uced, the minimum requireness will yield an increase in superalloys, and will therefore pose a severe penalty on rotor weight. For this reason, the Nb-based refractory metal-intermetallic composites represent the most promising materials for surpassing superalloys. This general family of composites is referred to as refractory metal intermetallic composites ( ~ ~ I Cins this ) chapter. With bulk temperatures in current HPTBs of 1000 "C, the superalloys are highly stressed at nearly 80% of their melting temperatures, and the maximum surface temperatures in these blades are -90% of the melting temperatures. A Nb-based composite system, with a melting temperature of 1800°C or more, may allow a substantial increase in both bulk temperature and surface temperature. ulk temperatures of 1200 "C (-2200°F) for such composites are ~ 7 5 %of the melting temperatures, and a surface temperature of 1370 "C (-2500 "F) i s < 85% of the melting temperature. Developments based on Nb offer significant gains in temperature ca of aerospace components (Dimiduk, 1993; 11, 1988). There are two principal deficiencies of niobium-based systems at
-
545
high temperature: (1) inadequate high-temperature iiiechnnical behavior, and (2) limited oxidation resistance. Nb-silicide based in-situ composites, with Nb3Si and/or Nb,Si, silicides have been shown to have great potential because of their attractive balance of highand low-temperature mechanical properties al., 1996). From model binary Nb-Si allo quaternary and higher order alloys have been developed to generate in- U composites with improved oxidation properties wlay et al., 1999a; S~bramanian et aE., 1996).
A range of processing approaches lias been used to generate Nb-based RMICs, includin et al., 1994, 1995a), directional solidification ( physical vapor deposition et al., 1994), forging, and extrusion (Mendiratt , 1993; Weiss et al., 1994). Solidification processing of these high-temperad the ture, Nb-based RM Cs is severely l i ~ i ~ eby capability and availability of existing mold materials. As a result, cold-crucible methods have been developed in conjunction with arc melting, and directional solidification. Directional solidification techniques that have been employed include Boat zone, and Czochralski methods. Tliese approaches will be described in the following sections.
formed using an optical iniaging float zone process (OIFZ) by Pope ef al. (1994)' as shown in Figure 2(a). This Boat zone method is similar to the zone-melting technique of Pfann (1966); a small volume of the alloy, in a relatively large rod-type charge, is melted and then the molten zone is translated in a controlled manner along the rod. The molten zone is retained in position by surface tension between the two co-linear rods of the same alloy, and as a result a crucible is not required to retain the melt. The principal advantages of this approach are that it is very clean and it is capable of working with a range of different materials on a small scale. Induction heating and electr~n-beamheating sources have also been used for directional solidification of high-temperature composite materials (Jolinson et al., 1993).
546
Applications I
L
Schematic diagrams of directional solidification approaches for RMICs. (a) OIFZ processing (Pope et al., 1994), (b) cold crucible Czochralski directional solidification, and (c) cold crucible Bndgman-type directional solidification
Directionally solidi~ed(DS) i
ts have been progrowth from an induction levitated melt using higher growth rates than are used for Ni-based alloys. This process is chematically in Figure 2(b). A range of NbMICs consisting of Nb,Si, and other silicides in a Nb solid solution matrix with melting points up to 0 "C has been directionally solidified. ectional so~idificationprovides excellent control o f microstructure and composition in samples with low defect concentration and size. The maximum ingot dia~neterproduced to date is 30 mm. Directional solidification can also be used as a small solid-airfoil manufacturin~techniq~e.The ingots can be machined
into airfoil con~gurations,as is the practice for ODS Ni alloys. At present there is limited experience in providing components with cooling channels using this technique.
Bridgman methods have been used for dire~tional solidification o f RMICs. Segmented water-cooled copper crucibles have been used to produce DS ingots with diameters up to 50min. In the Bridgman approach the water-cooled copper crucible is withdrawn in a controlled manner through the electromagnetic field that is used to melt the alloy b y
induction levitation. This process is shown schematically in Figure 2(c).
The basis for phase stability in niobium-(niobiumsilicide) composites is the Nb-rich side o f the Nb-Si eutectic between Nb, ndiratta and Dimidu (Nb)-Nb?Si and (Nb)1991; Sclilesinger et al., 19 es have been prepare iiniduk et al., 1993; 1995b) wjth compositions from 30 to 2523; all compositions are given in atom per cent throughout the present paper. The microstructure of the composites from binary hypoeutectic alloys consists of (Nb) deiidrites with an interdendritic Wb,Si-(Nb) eutectic. ~ extrinsic In these high-strength i ~ - s i t composites tou~liening is provided by the (Nb); there is no
iiitrinsic ductility of the silicide. The i, and Nb,Si have tI32 and t 32 crystal structures, r ~ s ~ ~ ~ " tively. As pointed out Schlesinger et al., 1993, there are high- and 1ow"teinperat~ire all~tropes of the Nb,Si,. A typical microstructure of a Ti-Si alloy composite i s also sbowii in (Bewlay et al., 1995b): it consists of inte dendrites of the (Nb) and Complex Nb-silicide composites eleinent Nb alloys have also been 3(b) slzows a typical micrograph (ba (BSE) inzagiiig) of a 1 ~ ~ i ~ t Lsection ~ d i ~ofa a~ specific inetal and silicide composite microstr~cturefor which a broad range of properties have been repo (Bewlay et al., 1996). This s
24.7Ti-8.2I-Sif-2.0Cr-l.9A1-16. MASC throughout this chapter. It con~ained both llic faceted dendrites, metallic and M3Si int intermetallic (where sniall amount of an , and an interdendritic represents Nb, Ti a eutectic of silicide and metal.
(Bewlay and SutliE, 1998). At these lower concentrations the eutectic is between the
(b)
igure 3 Scaniiiiig electron micrograph of the longitudinal section of (a) a DS Nb-silicrde based composite Nb-27Ti16%Si, and (b) the DS MASC
Typical microstructure (BSF, Image) of the section of a L)S co1nposltc generated from a quateriiary ~ b - l 6 T i - ~ ~ f - ~alloy. 6 S i The (Nb) i s the dark phase and the M,Si i s the light phase
548
i pole figure, arid (b) [OOl] (Nb) pole figure of the ~b-16Ti-8Hf-16~i alloy
rowth direc~ion, but the (N ) was not strongly textured. Composition can have a substantial impact and the texture o f the ies on the definition of ~ i ~ u i d - s o ~and i d solid-solid phase stability in Nb. There has been additions to Nb-Si resistance of these composites can be improved by partial substitution of ~ b ~ a i i i a n ~eta nal., 1996; Jaclsson et aE., added because it is also a strong solid gthener of the (Nb). base stability ~nformationis a critical requirement for definitio~iof both the ~anufactureof these hightemperature ia-situ composite systems and their service. In the fo~~owingsections phase stability in the Nb-Ti-Si and Nb-Hf-Si ternary systems will be briefly described.
defined, and an i s o t ~ e r m a ~ section at 1500°C is shown in Figure 7. In the Nb-Ti-Si system, phase equilibria involve five phases: (Nb), Nb,Si (T,= 1975 "C), Nb5Si3 (TM=2515"C),Ti, i, (ir,= 2130 "C) and Ti,Si with particular focus on the first four of these phases. The binary Nb-Si phase diagram possesses a eutectic E-+Nb,Si -i-(Nb) at 1880 T, and also a peritectic L + N b ~ S i ~ - + N b at ~ S1980 i "C. In the ternary phase diagram a eutectic groove ext ds between the binary (Ti) and L-+Nb, L-+Ti,Si, ithere is a change in the equilibria describing the liyuidus surface, and the eutectic groove, with decreasing Nb and increasing Ti con~entration.In the liquidus projection one peritectic ridge intersect^ a second one to generate one composition, and the resulting peritectic ridge intersects the eutectic groove at another composition. There are two transition reactions, as shown in Figure 6.
erna phase equilib~~a have been reported for rn ions. A liquidus surface projection of the metal-rich region (up to 37,5%Si) of the Nb-
The liquidus surface projection of the Nb-Hf-Si system is shown in Figure $ ewlay et al., 1999b, 1999~). tectic trough between the There is a shallow
. Figure 6 also shows the eutectic between Nb,Si and ( ). Isothermal sections at te~peratures of 1650 and 115OC have been
549
10
210
30
so
40
60
70
80
w
Diagram of the liquidus surface projectroil of the metal-rich end of the Nb-Ti- i system showiug the two transition reactions (fine solid lines show temperature contours)
e3, generate three invariant reactions in the Nbsystem. The peritectic ridge, p2, intersects the eutectic groove, el, to ~enerateU,. The peritectic ridges, p1 and p,, intersect to generate U2, and the subsequent ridge intersects the eutectic groove, e2, to generate U,, as shown in Figure 8.
mposites of quatern~ry and higher orcler systems has also been investigated. A1
I0
30
0
50
0
~ b - T i ~ H f system. -~i The inary phases are omitted, since Hf influerices stability, but the other Hfobserved in ~ b - s i l i c i ~based e composites, because alloy coinposi tions of interest generally have less present than needed for stabili~~tion of these phases. The (Nb) of most composite c o ~ p o s ~ t i o nofs c ~ r r e n t
concentration can move the alloy into a three-phase region. A1 and Cr, which are important alloyin additions for oxidation resistance, ~ a ~ t i t i op~iniaril n to the (Nb) when present at c o n c e ~ t r ~ ~ i of o n sc 5at%. MO, Ta and W have also been explored as alloying additions, but there have been l i ~ i t evaluat~ons e~ of the effects of these elements on phase st~bility. Higher Cr additions ( 35%) have also been studied in detail, and they can lead to a s ~ b s t a n timprovei~~ ment in the oxidation resistance. Fi
70
rind section of the Nb-Ti-Si phase diagram at a temperature of 1500C. The compositions that were heat treated . EMPA ~ e ~ s u r e ~ i eofn tphase s coinpositions are shown as %s
550
~t ions ~
~
~
i
~
a
Coiiiposition ranges for the silicide and (Nb) phases observed in the DS MASC (~b-24.7Ti-8.2Hf-2~0Cr-1.9A1-16.0Si) (Rcwlay et al., 1996) ~
Phase (Nb) M,Si 5%
b 58-6 1 48.3 49.0 25.5-27.9
____
Ti
HE
Si
A1
Cr
27.2 -29.4 18.2 22.2-23.2
5.0-5.3 7.8 12.5-12.9
0.9--1.3 24.7-25.4 35.4-37.5
2.5-3.0 0.1 1.0-1.5
2.8-3.9 0.1-0.2 0.4-0.5
e regions for the ~ b - T i ~ C r - ~ i Average phase compositions for silicide and Laws phases that have been observed in complex alloys that were itions of > 5 % can lead to the heat treated at 1200 "C before oxidation treatments (Jackson ves phase in the composite. The and Bewlay. 1998) Laves phase has been found to . Typical enhance oxidation resistance significa~tly Nb At HE Ti Cr Si (at.%) (at.%) (at.%) (at.%) (at.%) (at.%) coiii~~osition~ of the phases present in a high-Cr, silicide-based ia-situ coniposite are shown iii Table 2. It 13.6 25.4 1.0 36.5 21.2 2.3 M5Si.3 phases coexist with a SiHigh Ti ) (Jackson and Bewlay, 1.9 11.4 19.3 1.4 36.5 M,Si, 29.5 1998). The specific crystal structure that is stabilized is Interniediatc depen~enton the Ti concentration, although this is not Ti 6.5 14.0 0.7 35.9 40.5 2.4 M,Si, expected from the quaternary phase diagrams. The low Ti stability of these Merent silicides will depend on both 3.0 30.2 10.6 0.7 51.4 4.1 (Nb) te~peratureand alloy composition of the composite. Laves 19.8 1.4 7.3 12.3 52.6 6.6 ~
ies ere emphasis will be placed on four niechanical properties that are essential for hi~h-temperature mat~rials:lo~~temperature damage tolerance (fracture toughness and room-temperature fatigue strength), l~-te~perature strength, atid creep behavior. The of the art of these mechan~calproperties based superalloys and Nb-silicide based co~positeswill be coiiipared in the present section.
A m i n i ~ a llevel of fracture toughness is required to provide damage tolerance in order to make compoiients that cau survive the final assembly into the turbine engines and tolerate iinpact loading in service froin events such as foreign object damage. A fracture toughness of 2OMPafi is considered a minimum value for critical components. There have been extensive studies of the fracture mechanisms in composites from binary Nb-Si alloys ( 1995; Mendiratta and Dimiduk, 1991; Rigney and
Schematic diagram showing the projection of the metal-rich end of the Nb-Hf-Si liquidus surface, The projectloll shows the peritectic ridges, p,, pz, p,, the eutectic valleys, e,, c2, e3, and the invariant reactions, U,, U,, U,. The alloy Compositions that were investigated are shown as the solid points
551
Single-phase fields are shown for (a) Nb-Ti-1%-Si and, (b) Nb-Ti-Cr-Si phase equilibria at 1200 "C for cornpositlolls up to 38% (single-phase fields ori~inatingfrom Hf-Si and Cr-Si binaries are omitted for s i ~ p l i ~ c ~ t i o n )
552
1000
00
0
0
400
1200
800
1 Fracture t o ~ g ~ n e sof s ~ b - t o u ~ h e n e silicided based composites from binary Wb-Si, ternary Nb-Ti-Si, quateiiiary N b ~ T i - ~ ialloys, - ~ l and the MASC
Lewandowski, 1994; Neltkanti and Dimiduk, 1988). Rcurve measurements have also been performed by several researcliers. Fracture tou hness for a series of MICs is shown in Figure 10. The effect of Ti on fracture toughiiess o f a range o f DS Nb-Ti-Si com~ositesi s also shown in Figure 10. Generally in those systems with the highest toughness, ductile rupture o f the (Nb) is observed. Fractographs show that (Nb) dendrites fail in a ductile manner, and are pulled out o f the matrix on failure of the composite. Crack bridging and crack blunting can also be observed. This su~geststhat the majority of the toughxiess is provided by the ( b), although microcracking and interface de-bonding may also make ficant toughening ~ontributions. odels of ductile phase toughening have provided significant insight into toughening niechanisms that are operative in this family of composites. However, they do not completely describe the tre behavior that has been reported enshall et al., 1997). In particular, the eEect of voluine fraction of the Nb solid-solution phase on composite t o ~ ~ ~ h n is e s sonly p ~ ~ r t i a ldescribed ~y by models for ductile phase tougliening. 7. b si~icide-based in-sit24 composites have higli-te~~perature strengths, their intermediate-temperat~re strength requires improve~ent ewlay et al., 1996; ~ ~ ~ b r a i ~ a ieti ial., a n 1997). The strength of a range o f liig~i-teiiiperaturecomposites is shown as a function of temperature in Figure 11,
ure 11 Tensile strength of a range of Nb-silicidc based insitu composites, and Nb alloys, as a function OF temperature, showing the i ~ p r o v ~ i n e noft h i ~ ~ - t e ~ p e ~sat tr ue r~ ~g tof~ ~b-silicidc based composites over Ni-based superalloys, Intermediate temperature strengths are also compared
including Nb-Si-based in-situ composites alloyed with Ti, Hf and other elements. It can be seen that at temperatures above 1000 "C the strength of the M i s higher than that o f the Ni-based super PWA1480, but at temperatures from room temperature to 1000 "C the strength of the Niis substantially higher than that of th The tensile fracture stress of the -800MPa at room temperature, and the yield stress at 1200"C. M~nolithicalloys of similar to the metallic phase of the MASC have yield strengths of less tkan 55 MPa at 1200"C. Thus, C possesses substantially improved tensile properties. There are essentially two strengthen in^ mechaiiisnis available for improving the strength o f the (Nb): Solid-solution stren~thening and d i s p ~ r s i o ~ stren~th~ning, for example b efractory MC carbides and/or nitrides (e.g. WfT\I, Z ) (Anton et al., 1988). The effects of adding elements such as W, Ta, or MO on solid~solution strength~ning~r o o m - t e ~ p ~ r a t u r e fracture toughness, and high-temperature creep have been explored. Ni- base^ superalloys have been employed in many high-temperature applications because of their high intermediate temperature strength and their strength retention at temperatures up to 1100°C. Thus, materials that are suitable for use at temperatures of 1200°C and above must also be able to compete on a de~isity- normalize^ strength basis with ~ i - b ~ s e d superalloys at intermediate temperatures; otherwise, they will exclude themselves from some structural applications. For example, in the lower-temperat~re,
~ i o ~ Silicide i u ~ In Situ C o ~ p o ~ i t e ~
553
1
rupture performance that results from the s~bstantial reduction iii density of the second-~enerationsin~le-crystal proposed that in the case of tlie behavior is compromised (Ti,Hf),Si, type silicide at li tions. There are, therefore, two areas for improvement of the creep perforniaiice of these materials. The first IS to improve the creep rupture ~ e r f o ~ ~ a to n cae level greater than that of M S ~ ~ 1 and 0 , the second is to reduce the stress sensitivity of the creep rupture performance. The rupture strength that will be req~iredfor future applications will be well beyond any behavior observed thus far. Current estimates are that, ultimately, aii 80fold increase over current rupture lives may be required for application of the silicide-based compo-
s
stress rupture behavior of of advanced single crysta
C is similar to that d s~peralloys,and
ure 12 Stress rupture behavior of the DS MASC i s compared to that o f DS and SX Ni-based superalloys in Larson-Miller plots (C = 20) where the t e m ~ e r a t u ~ e - t i ~ e parameter is plotted against (a) rupture stress, and (17) alloying additions, such as Hf, Ti, Cr, AI and MO, on rupture stress/material density the compression creep behavior have also been
high-stress regions of a turbine airfoil, if the yield strength of an in-situ composite is 50% lower than that of a Ni-based s~iperalloy,as suggested by Figure 11, then the i;n-situ composite airfoil may have to be lowest creep rates. At any selected stress level and redesigned in order to reduce the stresses. T ~ L I S , concentration, the creep rate increased RMXCs require further inter~~ediatetemperature Ti concentration. At stress levels of s. strength improve~~ent N
7.3 Tension and compression creep behavior of Nb silicide-based i ~ - composites ~ ~ ~ z ~ at ten~peraturesin the range 1000 to 1200 "C have been reported (Bewlay et al., 1999d). The creep rupture behavior of the MASC i s compared with the creep rupture behavior of two single-cry$talNi-based superalloys in the LarsonMiller plot in Figure 12(a). These data indicate that the creep rupture behavior of the MASC is similar to that of CMSX-4 at low stresses, but inferior to ~ M ~ ~
Nb-7.5Hf-2lTi-16Si composjtion~. At higher stress levels the Ti additions have a detri~entalefkct on creep performance. At any selected stress level, increasing the Hf concentration Leads to an increase in the creep rate. These creep data suggest that the Ti:Hf ratio should be maintained at a level less than 3 aiid the Ti concent lion should be kept below 21%. At high Ti and/or f concentrations th type silicide is stabilized in pre Nb,Si, type or tP32 Nb,Si type silicides - l1997a, O 1999d). The hP16 phase hm poor creep
554
Secondary creep rates at 1200 "C and stresscs of 140-280 MPlt for a range of Nb-silicide zn-sifu composites based on
erformanc~and its presence is detrimental to creep p e r f ~ r ~ a n of c ethe co~posite, L~l~ranianianet al. (1997) have also examined coinpression creep behavior at for stress levels of 70 to 210 complex co~posites similar t type composites modified with up 14Gr. En general, the secondary slightly higher than those of the C (which was similar to that of the composite ewlay et al., 1999d). Creep of composites is controlled by a combined in the silicide and metallic phases. ling has indicated that the creep properties of these composit~s are insensit~ve to volume fraction o f metallic phase (Henshall et al., 1999, although the ~ x p e r i ~ e n ~ data a l do not comletely support this proposal. The role of the volume fraction of silicide 111 c o i ~ t r ~ ~ l icomposite ng creep behavior is shown in Figure 14, which shows the effect of stress on secoiidary creep rate for a range of
quaternary alloy coinposites with Si concentrations from 12 to 22%. This Si con cent ratio^ range provides volume fractions of the metallic phase from -0.70 to 0.30 respectively. Increasing the S i concentration from 12% leads to a i ~ i n i n i uin~the creep rate at 18% Si and a volume fraction of silicide of -0.60. At coiicentratioiis lower than 12%, the composite creep performance is dominated by the creep behavior of the metallic phase and is therefore relatively poor. At Si concentrations greater than 20%0,the composite creep erformalice is dominated by damage of the silicide ay et al., 1999d). Previous work on monolithic suggests that creep deformation in Nb,Si, is controlled by diffusion of Nb in the Nb,Si, phase aiIian et al., 1995). The creep expoiients for the ~ b - 7 . 5 H f - 1 6and ~ i the ~ b ~ 8 ~ f - 2 5 ~ i - were 1 6 S i 1 and -2 respectively. At higher Elf and/or Ti concentrations, the exponents were increased substaiitia~ly. The change in slope suggests a change in creep echanism. For example, in the ~ b ~ 3 M o ~ 8 H f - ~ S T i,-the 1 6 creep exponent was
-
-
555
Niobium Silicidg In Situ Conzp7ositt.s 10""
+Creep Rate 140 MPa
Metallk Phase De
4 0-5 c1
\
I"-
10
14
12
16
18
20
on secoiidary creep rate of Wb-Si based composites for stresses of 140-28OMPa at a temperature of 1200 "C. At low Si conceiitr~tions,deformation is controlled by creep of the Nb aiid at high Si concentrations, composite deformation is controlled by cracking of the silicjde
-
1; but at SMo, the exponent was 5. These data suggest that at higher alloying levels, the creep rate is controlled less by the behavior of the monolithic silicide for which the creep exponent is -1 (Subranianian et al., 1995). The goal for creep performance is that there must be no more than 1% creep in 100 hours at high temperatures and stresses, such as 1200°C and > 170 MPa. If there is minimal primary creep, this corresponds to a secondary creep rate of 2.8 x 10-* s-]. This has been demonstrated in binary Nb-Si, ternary Nb-Ti-Si, and quaternary Nb-Ti-Wf-Si composites, but not in higher order systems. Anisotropy of creep properties also needs to be examined in greater detail. N
ertie
ysie
Elastic moduli of this family of composite materials have also been measured. Dynamic elastic moduli at room temperature and at temperatures up to 1200°C have been reported by ewlay et al. (1996) for the
MASC. The modulus was 165 GPa at room temperaGPa at 1200°C. i, has a modulus of 1 room temperature. Typical modulus values for a thirdgeneration Ni-based superalloy at 1100 "C are 7583 GPa in (001) and 210GPa in (11 1). The anisotropy of the inodulii of the monolithic silicides and Nbsilicide composites needs to be examined further, for these influence the thermal stresses that can be generated during operation. The thermal expansion o f the MA Bewlay et al. (1996), IS approximately linear from room temperature to 1200°C with a coeficient of 10.45 x 10-6C-". This value i s typical of compositions of the Nb-silicide based composites tbat have low ratios of Nb:(Ti + Hfj, with values of 1.3-1.6. Compositions have been studied recen oxidation resistance with ratios of have lower linear expansion rates Ti,Hf concentrations and absence silicide. Such composites with high oxidation resistance have increased Cr levels, aiid Laves-type phases can also be stabilized. For some composites, boron is added to form a T2 ~ n i o b i uborosilicide) ~ phase since this provides improved oxidation resistance. Table 3 shows O h expansion from room teniperature to 1200 "C for typical phase co~positions for the individual monolithic phases of these composites (MsSi, silicide data are for the tetragonal structure tI32). Generally, the (Nb) volume fraction of the composite is 0.40-0.50, the silicide volume fraction is 0.45455, and the balance of the microstructure consists of Laves and/ or T2 phases. Measured densities were 6.6 and 7.2g/ cm3 for the metals, 6.7-7.4g/cm3 for the Laves phases, and 5.7-6.5 g/crn3for the silicides. Densities of arc cast RMICs of nominally 30% (Nbj were intermediate between metal and intermetallic values. Composites of current interest exhibit 1-1.06% total expansion from room temperature to 1200 "C (8.5 x 10-6-9.0 x 10-6C-'j, which are comparable to reported values for monolithic (Nb,Ti),Si, (Jackson and Bewlay,
able 3 Thermal expansion for a range of intermetallic phases up to 1200 "C
Phase
Laves M,Si, Silicide T2 Nb
M,Si silicide
Nb (a/o)
Ti
Elf
Si
Cr
A1
€3
RT-12Q0"C Expansioii (%)
21 .0 38.5 41.5 55.3 49.0
11.0 16.0 13.0 28.2 18.2
5.5 6.0 3 .O 2.0 7.8
8.5 37.0 12.5 1.o 25.0
53.0 1 .o 4.0 10.0
1 .o 1 .o 0.5 3.5
-
0.5 25.5
__
__
-
1.07 0.78 0.94 1.06 1.05
-
monolithic binary Nb, i3 and ternary (Nb,Ti)~Si~ silicides were reported to have expansion coefficients of 9.0 x 10-6 C-' over this t~mperatLirerange. The similarity in expansion behaviors of the di~erent phases which can be present in these Nb-silicide based composites suggests that the expansion mismatch between the phases i s relatively small. Thus, it is expected that the thermal ratcheting between the phases will be negligible, and the interfaces between composite phases will not experience excessive stress during 1 cycling. will be subjected to a variety o f thermal The fatigue cycles during operation. Thermal fatigue can occur for cyclic exposure of a structure when a themal gradient is created within the material in transient and/ or steady-state temperature fields during component operation. Thermal stresses are generated by differences in modulus, linear thermal expansion coefiicient (a) and temperature ( A T ) , between the hotter and cooler regions of the structure and due to crystal an~sotropy.Single-crystal Ni-based superalloys benefit in thermal Fatigue from the low values of modulus that can be obtained for the { 100) orieiitation of the crystal, and from the absence of grain boundaries, where thermal fatigue can lead to cracking. The Nb-based
-
ICs must also be able to withstand thermal stresses that will result from similar temperature gradients within airfoils. The above physical properties have been measured because their values have a major influence on the stresses generated in rotating turbine hardware for the silicide composites. For the silicide composites, the values of (the modulus) x (percent expansion) are approximately equivalent to those values for current Ni-base single crystals, so thermal stress that is generated as a function of the temperature excursion will be similar to that for superalloys. However, the reduced density o f the MASC compared to Ni alloys can lead to --25O/0 reduction in centriftigally generated stresses.
vior The composites from binary Nb-Si alloys have very poor oxidation resistance, as shown in Figure 15. The oxidation resistance at 1200 and 1300 "C of silicidebased composites is substantially improved by additions such as Ti, AI, and Cr ( ewlay et al., 1997; Cockeram 1994, Cockeram and Rapp, 1997), as shown in Figure 15. External and internal oxidation are the
The oxidation behavior of silicide-based composites is shown as a function of temperature. A comparison with both Ni-based s~peralloysand monolitli~cNb alloys is also provided
two principal concerns. ith regard to internal oxidation, additions of Hf caii reduce oxygen solubility and diffusivity and thereby slow e ~ ~ b r i t t l e i ~ eatn t elevated ~emperatures (Subramaniaii et al., 1996;
C shows oxidation rates intermediate id losses of an older ~ i - b a s e dsuperalloy, like IN 738, and the improved oxidation behavior of t h i r d ~ ~ ~ n e r a tsing~e i o n crystal superalloys. The dashed lines in Figure 15 indicate the goals, wliere component surface temperatures may be 1315 "C. This goal is derived from current superalloy capability. If the oxidation behavior of superalloys i s acceptable for surface teniperatures of 1150 "C, then the rate of metal loss for the best superalloys at that tei~perature, 25pm/lO0 hours, is a suitable goal for the refractory metal in-situ composites at their anticipated maximum surfhce temperature. The oxidation data for the C at 120OC show a s ~ b s t a n t ~ aimprov~ment l
-
i, composites, but the over that of binary ( oxidation resistance of this composite at 1200-1315 "C requires further ~ m p r o v e ~ e n t . The addition of Cr-rich Laves phases can further improve the oxidation resistance. The oxidatioii resistance at 1204 "C and 1315 "C of a Nb-18Ti20Cr-2A1-18Si alloy is compared to that of MAS Figtire 16. The improvement in oxidation resistance is sLibsta~tia1for the Cr-rich alloy, with -33% phase (by volume), -25% (Nb), and -42% silicides. However, the low (~b)-phaseconte result in relatively low fracture toughness. A recent study examined the response of oxidatio~ resistaiice to bulk alloy compositian for ~b-silicide based composites (Jackson and Bewlay, 1998). These results for the effect o f compo~ition on oxidation behavior have been characterized by r yses for major element effects (Nb, Ti, Al), and by direct comparison for other addition
ure 16 Comparison of 2200 "F and 2400 "F oxidation resistance of ~ b - l ~ T i - 7 ~ f - 2 O C r - ~ A l with - l 8 Sthat i of the MASC. Data are for four samples o f MASC and nine of the modified alloy of (Nb-l~Ti-7Hf-2OCr-2A1-18Sijat 1204 "C (2200 "F),(a), and two samples of baseline coinposition and four o f the Laves phased modified alloy at 1316 "C (2400 "F), (bj, are shown. The higher Cr on cent ration leads to s~abilizationof a Laves phase
and V>. Oxidation at 1204 "C (2200'T) and 1315 "C (2400 as measured by weight change per unit area, can be related to major elements by: ~
OF),
+
at 1204 "C, dwt/area = Cl,,, AI,,, for (1.OSi + 0,7Cr + 0.5Ti + 0.3A1+ O.OlI-If), at 1315 "C, dwt/area = C1315c + A,,,,, for (1.OSi + 0.761-+ 0.4Ti + 0.8A1+ 0.5Hf) here C and A are temperature-dependent constants. At 1204°C c' was 473 a d A was 11.5. At 1315 "C C was 1741 and A was 39.1. These relations showed Si to be most beneficial in reducing losses by oxidation, followed by Cr and Ti. A1 plays an increasingly important role as the oxidation temperature i s increased. minor addition elements offers a benefit to the he composite; Ta and Zr are the strengthening elements with the least damage to oxidation (at 6 a/o addition levels), wliile V, MO and W idation behavior, In some alloys replacement, has been shown by Subramanian et al. (1996, 1997) to improve the oxidation resistance further. The results of the study of Jackson et al. (1998) also show the effects of different phase volume fractions on oxidation. ~ecreasingthe volume metal fraction of (Nb) improves oxidation behavior, and there is an optimum balance between Laves fraction and silicide volume fraction, for a given volume fraction of (Nb). owever, when Laves phases are added to these composites, it is difficult to maintain the balance of high and low temperature mechanical properties with r e ~ ~ t i high ~ e ~volume y ~ r a c t ~ o of n sLaves phase ( 0.2). xidation studies suggest that a minimum Laves volume fraction is required to provide the composite with adequate resistance. Although oxidation behavior of these alloys was substantially better than binary Nbi composites, the best present composites only meet the oxidation goal of < 2 5 p n lost in 100 hours at teni~er~turesup to 1204 "C. Further temperature capability is still desired.
This overview has compared the state of the art in ~iobiumsilicide based i ~ - ~ composites i t ~ with Nibased superalloys. DS aiid SX Ni-based superalloy parts are routinely ~ a n u ~ ~ c t using u r ~ ceramic-based d
melting systems and molds at a scale of -50kg. RMICs can be produced at solidification rates similar to Ni-based superalloys, but they require containerless processing techni~i~es. This currently has limited the MIC c o ~ p o n e n t sthat have been produced to Niobium silicide based ~ o m p o s i t ~offer s exciting opportunities for structural applications up to 1200 "C. Microstructure, phase eyuilibria, ineclianical behavror and oxidation performance of niobium silicide based coni~osites and ~ i ~ b a s e superalloys d have been compared. The Nb-based RMIC's have much higher
higher than that of current single-crystal superalloys. These ~ b - s i ~ i c i dcomposites e have densities as low as 6.5 &/em3, 25% lower than for advanced singlecrystal superalloys ( 9.1 g/cm3). Although other intermetallic systems have been considered for applications above 1000 "C, the Nb-silicide based composites appear to offer the best balance of properties. Fracture toughness has restricte use of intermetallic-based systems. toughness values in excess of 2 0 M P a 6 have been reported in silicide-based composites toughened by (Nb), but these are lower than the toughness levels enjoyed by Ni-based superalloys at their time of installation. Creep perfor~~ance and environinental further i~provement. Studies Ti-Si alloys and even illore ate that additions of Ti, Hf and MO can increase the composite secondary creep rates. Thus, careful selection and control of the alloy clieiiiistry is required. ~ x i d a t i o n and creep r ~ ~ a ithe n most serious challenges for the future use of the Nb-based RMTCs. Alloying additions to Nb-based RMICs have increased oxidation resistance substantially, but considerable further improve~entis needed. ~ a t e r i a l loss rates due to oxidation are still only comparable current superalloys. The addition of has a beneficial impact on wever, super~~lloys operate with m ~ ~ x i m surlace u~ tem~eratL~resof 'only' will be expected to operate equally successfully at temperatures up to 1315 "C. For Nb-silicide based composites, further evaluation of the effects of both alloying and processing modifications on the properties is necessary in order to achieve the required oxidation and creep goals. Alloying additions that are introduced to improve one specific N
-
property must maintain a full balance of properties. The required process developments will probably include cold-wall casting schemes and vapor depositian approaches. A s u ~ c i ~ n conibination t of chemistry, microstructure and propetty control has to be developed in large-scale components to provide the required balance of high and low temperature mechanical properties, and environmental resistance.
The authors would like to thank D.J. Dalpe, R.R. ishop, W.J. Reeder, L Peluso, P. Whiting, A.W. Davis, S. Sitzman, W. sser and E.H. Hearn for their contributions to experimental work. This research was partially sponsored by AFOSR under
.J. Grylls, Prof. J.J. for very helpful discussions.
rences
Bewlay, B.P. Jackson, M.R., and Subrama~iiaii,P.R. (1999a). Bcwlay, B.P., Bishop. R.R., and Jackson, M.R. (1999b). Z. ~ e t u L l ~ u n d9e0, ~ 6 041 ~ ,3-422. Bewlay, B.P., Sutliff, J.A., and Bishop, R.R. (1999~).J. Phase Equil, 20(2), 109-1 12. Bewlay, B.P., Whiting, P., and Briant, C.L. (1999d). MRS Proceed~n~s on Nigh Temperat~rsOrdered IiztervnetaZlic Allr>j’sVIII. I~K6.11.1-KK6.11.5. Bibring, H. (1973). C o ~ fon . In-Situ C~~~zpo~si~es-I, Eds. F.D. Leinkey and E.R. Thompson, National Academy of Science NMAB-30811, Washington, pp. 1-69. Block, W. (191 I). Z. Phys. Chem. 7 Bridgnian, P.W. (1925). Proc. Am. Bucltman, R.W., Jr. (1988). AlZo-vzng, pp.419-445. Coclteram, B.V (1 994). PhD Thcsis, Ohio State University, Columbus, OH. Coclteram, B.V. aiid Rapp. R.A. (1997). In Pr(~cessi?zgand Design Issues in High Teinperature ~ a t s r i a 1 s ,Ed N.S. Stoloff and R.H. Jones, TMS Publications, Warrendale, PA, pp. 391-402. Dimiduk, D.M., Mendiratta, M.G., and Subramanian, P.R. (1 993). In Structural ~nterrnetul~ic.~, Eds. R. Darolia, J.J. Lewandowski, C.T. Liu, P.L. Martin, D.B. Miracle, and M.V. Nathal, TMS Publications, Warrcnd~~e,PA, pp. 619-630. Ericksoii, J.S., Owczarski, W.A., and Curran, P.M. (19’71). Met. Prog. 99,(3) 38-60. Fitzgerald, T.J., and Singer, R.F. (1997). ~ e t a ~and l . er T T ~ 2~ 8s4. 110. 6: pp. 1377-83. Giarnci, A.F., and Tschiiikel, J.G. (1976). et all. Trans. A 7A: 1427-34. Gigliotti, M.F.X., Jackson, M.R., Yang, S.W., Henry, M.F., and Woodford, D.A. (1982). Cord: orz In Situ Composites-
Anton. D.L., Snow, D.B., and Cianiei, A.F. (1988). AFOSR Annual Report, May 1988. Bctz, U.,Hugo, F., and Mayer, H. (1995). 3rd Iizternational Char1e.s P(~rson~s Turbine Conference: M a t e r ~ ( ~ 1 ~ ~ Engineering in Twhines and Comprss,sors, pp. 557-65 Institute of Materials: London, Uaitcd Kingdom. (Eds D. Miracle, J. Craves, and D. Atiton). Mat. Res. Soc. Bewlay, B.P., Lipsitt, N.A., Reeder, W.J., Jackson, M.R., and Syvnp. Proc., 273, pp.461-472. Sutliff, J.A. (1994). In Processing and ~ a b r i c a t i ~onf GroBmann, J., Preuhs, J., E h - , W., and Singer, R.F. (1997). Advamed Materials fiw High Temperature Applications Proceediiigs ofthe I997 International Symposium on ~ i ~ ~ i d 111, Ed. V.A. Raw, T.S. Srivatsan and J.J. Moore (TMS Metal Processing and C a ~ ~ tEditors ~ ~ g , A. Mitchell, and P. Publications, Warrendalc, PA) pp. 547-565. Auburtin, pp. 3 1-40, Vacuum Metallurgy Division, Bewlay, B.P., Jackson, M.R., Reeder, W.J., and Lipsitt, H A . American Vacuum Society. (1995a). Mat. Rcs. Soc. Symp. Proc. 364,943- 948. Henshall, G.A., Strum, M.J., Subramaniaii, P.R., and Bcwlay, B.P., Lipsitt, H.A., Jackson, M.R., Reeder, W.J., and Mendiratta, M.G. (1995). Mat. Res. Soc. Sywip. Proc. Sutliff, J.A. (1995b). Mater. Sci. Eng., A ~ 9 2 1 ~ 9 534-543. 3. 937-942. Bewlay, B.P., Jackson, M.R., and Lipsitt, N.A. (1996). 11, G.A., Strum, M.J., Bcwlay, Metall. atid Muter. Twns.. 27A, 3801-3808. (1997). Metall. and Mater. Trans., Bewlay, B.P., Jackson, M.R., and Lipsitt, H.A. (1997a). J Hugo, F., Mayer, H., and Singer, R.F (1994). Invest~ent Bewlay, B.P., Lewandowski, J.J. and Jackson, M.R. (1997b). Custirig Institute 42nd Annual Meeting, pp. 9: 1-9:5 Dallas, TX 75206-1602: Investment Casting Institute. Jackson, M.R., Gigliotti, M.F,X., Yang, S.W., and Walter, J.A. (1998). Microscopy and ment 2, 278-279. J.L. (1982). Conf on In Situ Composites-IV,Eds F.D. ., and Jackson, M.R. (1998). J. Lenikey, H.E. Clinc and M. McLean, Mat. Res. Soc. Symp. Proc (12), Elscvier, pp. 155-365.
Jackson, M.R., Rovde, R.G., and Skelly, D.W. (1995). Mat. Res. Soc, Symp. Proc. Jackson, M.R., and Bew order report, AFML, Sept 1998. Johnson, D.R., Joslin, S.M., Revierc, R.D., Oliver, B.F., aiid Noebe, R.D. (19931, In P r o c e ~ s s ~and i ~ ~F a b r ~ c a t i oof~ ~ d v a n c ~e a~t e r i a lfor s ~ ~ Ternperat~re g h Applicatio~~s II, Ed. V. A. Ravz and T. S, Srivatsan. TMS Publications: Warrendale, PA, pp. 77-90. Kablsv, E.N., Ge~asimov,V.V., Shalimov, A.S., Kupre V.P., and Dubrovsky, V.A. (1995). Lzt. Proizvodsvo, 3@-32. 967). Trans. et all. Soc.
Nekkaiiti, R.M., and Diniiduk, D.M. (1988). Mut. ReS. Soc.
Khan, T. (1979). COP$ on In-Situ Composites-ZII, Eds J.L. Walter, M.F.X. Gigliotti, B.F. Otiver and H. Bibring, Cinn, Lexington, MA, pp. 378-388. Klug, F.J., Prochazka, S. and Doremus, R.H. (1987). ,I. Am. Uercrn. Soc. 70: 750 759. Massalski, T.B. (1991). Binary A11o.y Phase Dzagrai~,ASM Metals Park, Ohio. . (1983). ~ i r e c ~ i o n a ~l ~~y o / i ~Materials, i~ed The Metals Society, London. Mendiratta M.G. and D i ~ i ~ d uD. k , . (1991). Scripta M e t ~ l l . 237-242. Me .G., and Diiniduk, D.M. (1993). Metall, Tram, 501-504. Menzies, R.G., Bruch, C.A., Gigliotti, M.F.X., Smith, J.A., and Haubert, R.C. (1988). ~ ~ ~ ~ e r a / l19K8, o y , s Champion, Pennsylvania, USA, 18-22 Sept. 1988, AIME, pp. 355-364. ~ a r r e n d a l e~~nnsylvania, ,
Schlesinger, M.E., Gokhale, A.B., and Abbaschian, R. (1993), J . Phasc Eyuil., 14(4), 502-509. Stockbar~er,D.C. (1936). Rev. Sci. Inst. 7, 133--36. Subra~anian,P.R., P a r t h a s ~ ~ ~ t T.A., h y , Mendiratta, M.C., and Diniiduk, D.M. (1995). Scripts Met. 32(8), 12271232. ~ u b r a ~ a n i aP.R., ~ i , ~ e n d i r a t t a M.G., , and ~ i n i i d u D.M. ~, (1996). Jozrrnal OJ Metals, Subramanian, P.R., Mendiratta, M.C., Diiniduk, D.M., and Stucke, M A . (1997). Mater. Sci. Eng., A VerSnyder, F.L., and Guard, R.W. (1960). Tram. A m . Soc. ikfct. 52: 485-93.
nmg, Wiley, New York. Piearcey, B.J., and VerSnyder. F.L. (1966). SAE Journal 7 84-87. Piearcey, B.J., and Terkelsen, .E. (1967). Trans. M t ~ ~ u Soc. ll. A I 239,~ 1143-50. ~ Pope, D.P., Shah, D.M., Romaiiow, (1994). Mat. Res. Soc. Symp. Proc.
., Heathcote, J., Lucas, G., and Odette, G.K. (1994). Mat. Res. Soc. Symg. Proc.
of AIME, pp. 175-204: pp. 585-594.
Strong, predominantly nielallic, bondiiig between unlike atoms leads to the formation of intermetallic compo~ind(IMC) phases. From bondin structure ordering, high strength at low and high temperature, and low ductility particularly at low temperature. The major obstacle to the widespread use of intermetallics has been their poor ductility and low owever in recent years understanding of the defor~ationand fracture behaviour of intermetallics has increased significantly. The main critical Eactors - the complex crystal structures, the large urgers vectors, the high lattice stress, the inade~uate slip systems, the inability to promote cross slip and the lack of grain boundary cohesion, responsible for low fracture strain and low KIc- have been identified. Such sub-structure defects are remedied by adjusting the macroscopic parameters, e.g. grain size, stoicliiometry, grain boundary de ro-alloying, second-phase incorporation^ incr e number of slip systems and altering the n slip. Although improved uiiderstanding of the deformation and fracture behaviour has allowed the development of some useful interinetallics, the low ductility and low K,, problems still remain. With IMCs in coating form, the problems of low ductility and low kr,, scarcely arise. The use of intermetallic coatiiigs in many applications is already established? e.g. in electronic and decorative areas, in power generation and in aerospace.
ments with IMCs as structural m a t e r i ~ ~ass well as their current problems and limitations are reviewed by Lipsitt et al. and Bewlay et al. elsewhere in this volume. Still further detail on this topic is provided by many chapters in Vol. 2 of this t~eatise.Use of ~~~s as semiconductors is elaborated elsewhere in this volume by Ramanath et al. in their review of thin films of IMCs and in Chapter 15, Vol. 2 of this treatise by ~ a s u m o t oet al. Use of IMCs as decorative coatings is covered in this volume by WoIE on precious metal IMCs and by Steinemann et al. on coloured I In the power generation and aerospace fi approach of applying IMC coatings allows the use of both superalloys with inadequate ii~eltiiigpoints and bulk intermetallics with high melting points but poor environinental resistance and toughness properties, thereby increasing the temperature capa substrates. Here we will con~entrateon I applied in these two fields, where three situations can be distinguished: (ij their use to protect conventional hi~h-temperature alloys; (ii) their use to protect bulk (iiij their use as an inter composites.
This approach not only allows the limitations of the bulk inter~etallicsto be overcome, but offers opportunities to produce intermetallic coatings/surfaces with non-equilibrium structures including extended solid solubility, amorphous structures, stru~turesco~taining
l ~ ~ t ~C ~ ~~ ~ i ~ Vol. ~o u~ 3,n ~Principles d l~ ~~ ~~andc Practice. Edited by J. H. Westbrook and R. L. Fhscher. 0 2 0 0 2 John Wiley & Sons, Ltd.
multiiayers, ~laiiientarycomposites, and structures in niany other fornis such as in-situ composites. However, the development of coatings in the field of intermetallics lags behind the development of bulk materials. At present there is incomplete understanding of microstr~ictures and defect structures governing the properties of the coatin s and the way these fundamental properties are influenced by process parameters. This gap in knowledge has a strong bearing on the contents and treatment of the subject ~ a t t e r in s this chapter. This chapter is not meant to be a catalogue of tlie infinite varieties of intermetallic coatings. Instead the chapter has been written with emphasis on generic principles, i ~ o d e ~ ~and i n gscientific theories, where possible, which will fkilitate basic understaii~ing of the structure and properties of coatings systems based on intermetalli~s and for inteiiiietallics. In writing this chapter, heavy reliance has been made on the inforination provided in the excellent two preceding volumes of this treatise. In particular it builds on Ch. 22, Vol. 2 by Nicholls and ~ t e ~ h e n s oon n high temperature coatings for gas turbines, on Ch. 43, Vol. 1 by Doychak on IMC oxidation and is c o m ~ l e ~ e n t ebyd the chapter in this et d.on processing and properThis chapter is structured in six sections. Section 2 deals with the main issues involved in the design of intermetallic coatings and coatings for intermetallics. Section 3 discusses recent observations on the production and behaviour of aluminide and silicide coatings for conventional alloys. Coatings for bulk inter-
metallics and for intermetallic composites are considered in Sections 4 and 5, wliile Sectioii 6 presents a forecast of future areas of activity.
This section briefly describes the approach to design of coating systems for interm~ta~~ics and coating systems using intermetallics. In writing this section, examples from other coating systems have been used to deinonstrate certain generic principles. coating system involves consideration of the working environment, coating surface, coating/substrate interface and the body of the coating itself. In inany cases an additional interface is created due to the coating/ e n v i r o n m ~inter~ctions ~t (scale f o ~ a t i o ~The ) . integrity of the coating system requires its chemical and mechanical stability. Of particular importance in discussing coating/surface e n g i ~ e e r i ~are g issues involving the design of coating/substrate interfaces. The best point to start this section is to examine the various failure modes of the coating systems. For many high-temperature, d~gradation-resistantcoatings, coating failures occur by two maiii modes - (a) and (b) (Figure 1) - both of which deiiude the coating of elements intended to contribute resistance to the high-temperature corrosion processes. The oxidation process continuously consumes the elements responsible for forming a protective scale; unless there is a sufficientreservoir of these elements in the coating, the concentration at Iiiterfdce 1 will be reduced to such an
Schematic diagram lndicating various high-temperature coating degradation mechanisms. p is partial pressure; a is activity.
563
Couting Technology
extent that scale formation cannot take place and thus allows base metal oxidation. The second mode of degradation - Mode (b) - occurs by outward diffusion of damaging substrate elements to be incorporated at the scale/coating interface or within the scale, thus underiiiinii~gthe integrity of the scale. At Interface 2 the inward diffusion of the scale-formi element will also affect the regeneration of the p tective scale following spallation. The problem of interdiffusion is illustrated for an alum~nidecoating in Fig 2. The coating consists of A1 layer and a thinner layer two layers -- a thick oute containing a mixture of NiAl and Ni,Al. The elements from the substrate accumulate in the inner layer which inhibits further secondary diffusion. The influence of the substrate type, composition and microstructure on the 0
r
development of an aluminide coating is illustr~tedin Figure 3. Increasing Ni increases the diffusional growth rate of the outer NiAl phase, and also suppresses, with benefit, the development of the solid solution or duplex layer. ~inimizatioiiof interdiffusion is critical, and there is a need to consider the nature of the base metal and alloying elements. It may be necessary to i n ~ ~ o d u c e a separate anti-diffusion layer at the coating/substrate interface. In the choice of a suitable anti-diff~sio~ coating, empiricism predomi~ates over un~erstanding and modelling of the solution for interd8usion and partitioning of eleni ts, even in simple systems, are poorly understood. ere is a ~ a r t i ~ u l adearth r of interdiffusion data. In recent years (Datta et al., 2000) interdiffusion modelling work has been undertaken. at Duplex layer
1 0
0
0
Q
1
10
0
0
0
60
80
The distribution of eleinents 1n an alurninide coating on nickel-based superalloy (reproduced Institute of Metals)
permission of the
564
1 ~ 0 0 1050
100~ 9
-7.0
7.5
8.0
7.5
8.0
l r r x 10 -4 (IIK) igiire 3 Temperature depencfence of the growth of the NiAl phase (left) and the solid solution zone (right) in aluininized substrate alloys with various nickel 4- chromium contents between 20 and 25 wt.% (reproduced by permission of Applied Science ~ u ~ l i s hs)e r
orthumbria. Such work, although at the initial stages of development, will significantly facilitate the design of complex coating systems. n modelled using the lly and ~anielewski, 994). The model allows calculation of the concentration profiles in singlephase, mu~ti~compone~~t and multi-layer systems even when the con~ponents’diffusivities vary with composiaiiielewski and Filipek, 1996), e.g. in thermally treated and/or oxidized substrate/coating systems, interdiffusion in a Pt-modified pAR MO02 has been studied (Datta et al., 2000). The computed and experi~entally measured deiisities of Al, Pt and Ni at 1073K after 200 h of difiusional annealing in an argon atmosphere are shown in Figure 4. Other applications of the ode1 including calculations of the intrinsi~~i~usivities and modelling of the intersystem-selective oxidation of the coating on MAR MO02 are presented further in this section. Apart from chemical effects of coinpositional radients, ~ e c ~ a n i c astability l is affected by the
presence of stresses at the coating/substrate or coatin scale interfaces and within the coating itself. The most important sources of stresses are: (i) the external stresses producing deformation; (ii) the thermally induced stresses arising from differences in thermal expansion coefficients across the coating/substrate, coating/scale or substrate/scale interfaces; and (iii) the stresses in the coating associated with the growth processes, phase separation and precipitation. In recent years considerable research work has been undertaken to obtain the quantitative i n f o ~ ~ a t i o ~ needed to model the behaviour of the protective scale in the presence of mechanical stresses. Nevertheless uncertainty remains in this area. lion
Recent work at the University of ~ o r t h u ~ b r(Griffin ia et al., 2000) has niodelled the deformation and fracture behaviour of AI@, scales formed on cylinders of Fe,AI
inter~etallics with chamfered edges giving further
565
Coatirzg Technology
30
30
0
90
60
AR MOO2/coat~n~ interface (pm)
3 4 -90
.
. , -60
.
. , -30
.
. 0
,
.
. 30
,
.
. , 60
.
. 90
1
Distance from MAR ~ O O Z / c o a tinterface i~~ (pm)
-90
-60
-30
0
30
60
90
~OOZ/co~t~ interface ng (pm) The calculated and measured density of Al, Ni and Pt in the Pt modified P-NiAl coating on MAR MO02 after difYhonal annealing at 1073K for 200 b
2. When the same defects as in 1 above were introduced for a perfectly plastic substrate: (0 the plastic deformation of the substrate caused the oxide near the edge to behave like an elastic beam in bending and at 1109 crack growth began in the oxide layer; (ii) the oxide started to detach from the substrate at a temperature of 728K; a much lower propagation temperature than in the linear substrate case, and was due to stress relaxation caused both by plastic deformation of the substrate and by oxide fracture; (iii) the oxide itself split at 668 K, and the oxide/ substrate interface spalled at 528 K. 3. With only a crack at the interface and no crack at the oxide edge, again for a plastic substrate: (i) the temperat~reat which d e l a ~ i n a t i o ncommenced was 874 K; and (ii) there was less stress relaxation than in the previous scenario, because the oxide remained intact and under tension at the outer edge of the chamfer, as the matrix deformed plastically. Clearly, corners or edges are the locations of failure, which demonstrates the need to design alloys and coatings which will produce defect-free scales or which are capable of producing crack-healin One approach is to introduce an interlayer. The function of such an interlayer is to promote coating/ substrate adhesion by providing a composition/stress gradient interface.
insight into the scale failure phenomena. The bonding between oxide and substrate was included in the model as was the ability of a through-surface crack to develop at a cylinder edge. The model assumed uniform cooling at a rate of 100 K/s from a temperature of 1273K to 293 K without creep; CTEs were assumed to vary with temperature. On running the model the following results were obtained:
for enhancing the high-tempe~dturecapabilities of conventional alloys. ere we discuss some particular coating systems for this purpose.
I. When embedded defects were introduced at both the scalelsubstrate interface and the oxide edge, the following sequence of events was shown for an elastic analysis: (i) the oxide did not fracture; (ii) scale spallation commenced at 1091 K, beginning at the oxide~substrateembedded defect; (iii) the scale completely spalled, in one piece, at
Nickel aluminide coatings are the most well~known and often used of the intermetall~ccoatings. Nickel aluminide is an ordered intermetallic and exists over a composition range of 45-40 at.% A1 ( Xts high oxidation resistance combined with low density and excellent electrical and thermal conductivities have extended the use of nickel aluminide for many applications - structural coatings for enhancing high-temperature corrosion resistancc, electronic metalization, epitaxial overlayer and thin contact
As previously mentioned, IMC coatings offer potential
3.1 ~ i c kAlu ~ l
s
electrodes in semiconductor devices (Sands et al., 1990; Charmers arid Loebs, 1990; Lee ct al., 1995). However, atteiition here will be selective~yfocused on nickel a l u ~ i n i d eas a structural coating to prevent the hightei~perature environniental degradation generally eiicountered in aero gas turbines. Such consideration will highlight the development and performance of these coatings in relation to their processing methods, the evolution of iii~crostructures,and the processes of protective scale formation and scale breakdown (Datta et al., 1998). The compositioias of gas turbine superalloys, optimized to provide a high content of the precipitation (y ') in order to achieve adequate creep stre~gthei~er resistance, adversely affect the high-temperature corrosion resistance o f these materials. Nickel alumiiiide coatings have been designed to impart the required corrosion resistance to the superalloy substrates. These uced by two types of pack processes her et aE., 1998). In one variant the com~onentis placed in contact with a pack consisting minium, a halide activator (NE-I,CI, NaCl or and subjected to a diffusion within a temper~turerange of 750 to 1000°C. The A1 halide fornicd, aided by thc decomposition of the halide activator, undergoes dissociation on the surface allowing A1 diffusion into the substrate and leaving halide ions free to react with 1 and cause the cycle to continue. A modification of this technique, termed 'out-of-pack' is similar to the chemical vapour deposition (CVD) process in that the coiii~onentsto be coated are not in contact with the powder of the pack, the aluminium halide gas generated being transported over the component/ ate surface by a carrier gas (Strieff, 1993). minide coatings grow by two mechanisms depending on the use of a low or high activity pack. The high aluminium activity process involves the inwa~ddiffus~on i n i u ~to form 6-Ni2A1, with a ininor amount . The coating is heat treated to iffusion from the substrate to form /I-NiAI, c2 e ductile and oxidation resistant phase than phase. A three-zone structure develops comprising an outer zone containing a-Cr and other a /3-NiAl matrix, a middle zone of a A1 with Cr, Ti, MO and CO in atrix inner zone containing carbides and o phases (Strieff, 1993; Goward and Boone, 1971). he low aluminium activity process leads to the forn~ationof a two-zoned strLicture by the outward usion of Ni from the substrate reacting with Al. outer zone consists of a single phase P-NiAl with
,
alloying elements diffusing out from the substrate along with Ni. The inner zone contains carbides and/or phases formed by the Ni witlidrawal from the 8matrix (Strieff, 1993). The protectivity of alurninide (NiA1) coatings sterns from their ability to form an a-alurnina scale. aalumina has an hcp structure of oxygen anions with two-thirds of the octahedral sites filled by trivalent cations. The high ~emperatureprotection afforded by the aalumina results from the oxide having low concentrations and rnobilities of both ionic and electronic defects. The slow growth rate of the oxide is related to its highly stoich~ometricstructure and large band gap width, which makes electronic conduction difficult. In the main, a-alumina acts as an ionic conductor in which both oxygen and a l u ~ i n i u mare mobile. There appears to be sonie contention ovcr the growth mechanisms of the aluniina scale. It has been reported that the columnar st~uctureof a-alurnina is iiidicative of oxide formation at the scalelsubstrate interface. Tracer studies of the oxide formed by overlay coatings and bulk fl-NiAl have indicated alumina growth by inward d i ~ u s ~ o of n oxygen. However, the growth rate cannot be accounted for by the rate of oxygen diffusion along alumina grain boundaries. It has been proposed that the growth o f the oxide formed by ovcrlay coatings is either exclusively by the outward diffusion of cations or is a c o ~ b i n a t ~ o of n anion and cation transport. The morphology and microstructure of the scales formed by the P-NiAL have been used to suggest growth ~echanisms.The f o r ~ a t i oof~ a characteristic ridge morphology of the scale has been ascribed to the transformation (at 1100 "C) of transitional aluminas, formed by the outward diffusion of alu~inium,to aalumina where growth occurs by both inward and outward diffusion. The alumina cells nucleate and then grow laterally across the surface consuming the transient phases until the cells impinge. The different growth mechanisms shown by diffusion and overlay coatings have been ascribed to the differences in ~icrostructures-grain size. Smaller grain size such as in ~CrAlX-typecoatings promotes the formation of inwardly grown oxides by s~ort-circuitdiffusion of oxygen. The large grain size of P-NiAl promotes outward diffusion. The views concerning the growth of tlie oxides are not consistent, The effectiveness of aluminide coatings in preventin high-temperature corrosion is underi~ined by the incorporation within the scale of the outwardly diffusing, damaging substrate elements. The high
-
growing a-A1203 scale, pr~venting spa11 cracking, is a major factor. This overall iiii may be associated with a number of key factors:
activity aluminide coatings being inwardly grown facilitate this iiicorporation. Such incorporation of the damaging elements i s more difficult in the outwardly grown, low activity coatings (Strieff, 1993). The effectiveness of alurninides is further coinpromised by the increased attack by impurities in combustion gases caused by engines operating on lower grade fuel and in harsher environments. The l i ~ ~ t a t i o iiii i s the use of conventioizal nidsel aluminide coatings, depositcd on Ni-base superalloys, e.g. 002, to provide oxidation and hot-corrosion resistance, are well knowii (Clian et al., 1997). ~
~
C
~
~
~
1. Promotion of the selective oxidation of 2. An ability [or the oxide to reform followi~~ spallation, probably associated with e n h a n ~ e diffusion in the coating. 3. The creation of an A1 rcscrvoir tlirough the 4. The inhibition of the cQating/substrat intcrdiffusion. 5. The exclusion or limitation of cQncentration of substrate refractory elements in the outer zone of the coatings - such elements undermine the integrity of the coatings. 6. A reduction in the oxide growth stresses.
S
A major advance was made by the addition of nickel alumin~decoatings; such coatings outperform unmodified coiiventional alurninide coatings tta et aE., 1999). Two main types of Pt-A1 coatings have been identified:
Even the improved Pt-aluminide coatings eveiitually suffer failure. The hilure mec~anisms i d ~ n t i ~ e d include:
1. A single-phase structure with a continuous PtAl, surface layer, and 2. A two-phase PtA&+ ( ~ i , P t ) A l structure with varying amounts and rnorphologies of Pt-rich phases and with varying aniounts and extent of substrate inter~entionin the ( ~ i , P t ) layer. ~l
1. Theriiial stress generated within the coatin during thermal cycling. 2. Depletion of A1 in the coating and the failure to regenerate A120,. 3. Associated growth of oxide coatiiig/substra~e interface and their eventual inclusion in the scale, loweriiig scale adhesion. 4. Outward diffusioii of substrate eleme~tssuch as from the diffusion zone to the coating/oxide interface (Figure 5 ) increasing the scale ~ r o w t h
The Pt-atuminide coatings have displayed improved oxidation resistance and a greater resistance to type-I trieff, 1993) hot corrosion and a marginal improvement in type-II hot corrosion resistance. There is iiow a consensus that improved adherence of a slow-
Schematic diagram of t
i
~
~
~
~
u
~
coating~sca~e interface
App~icut ions
568
rate leading to localized scale thickening and evcntual spallation. 5. For Hf-contain in^ superalloys (e.g. MAR M002) large Hf peg formation (10-2Opn deep) at the oxidelcoating interface. 6. Sulphur segregation at tlie oxidelscale interface. owever, there has beenland is, a continued demand to improve further the effectiveness of Pt-modified NiAl coatings. Several avenues have been explored: improved understanding of the processes of scale formatioii in oxidizing and hot corrosion environments; better insight of the mechanisms responsible for the loss of inte~rityof the protective A1,0, scale due to the incorporation of the substrate elements, e.g. Ti; through the modification of the coatings by incorporating other noble metals - Rh, Ir - in combination or singly and inclusion of semi-reactive elements such as Hf or Ir; and by improving the manufacturing methods for producing these coatings so as to minimize the sulphur content. There has been some renewed interest in the Iast two methods. One approach has been to incorporate iridium, with or without platinum, in the coating. Analysis of the Irafuminidc coating system in the as-processed condition shows a layer m o r p h o ~ o ~-ya Ni-rich outer layer and Tr-rich inner layer. A1 concentration decreases with depth through the coating and concentrations of W and Ta appear within the Tr-rich layer. Ir-Pt-aluminide
also produces, in the as-processed condition, n dual layered structure (P-NiAI outer layer), similar to that of the Ir-Pt system with Ir concentrated in the inner layer, and the outer layer rich in Ni and Pt; Ti is excluded from the outer layer. The oxidation (200 h at 1000 "C) of an Ir-Pt-aluminide on the formation of Alz03, which suffers less spallation than that shown by the ~r-aluminide system; the presence of voids at the coatinglsubstrate interface; the absence of Hf peg f o r ~ a t i o n ;internal oxidation and outward diffusion of W> Ta and Ti. The conclusion is that the reason for the beneficial effects of in corpora tin^ Ir over coatings based on Pt alone is unclear, Similar efforts considered the use of rhodium. The improvement in the hot corrosion behaviour of Ptaluminide and Pt-Rh-alu~~inidecoatings on MAR MO02 is illustrated in Figure 6. The beneficial effect o f this system is further demonstrated by the absence of the substrate elements (VV and Ta) in the outer layer.
The Generalized Darken Model (Holly and Danielewski, 1994; Panielewski et aE., 1994; Danielewski and Filipek, 1996) which fkcilitates a description of the interdifkion process in both open and closed systems and when component intrinsic difksivities vary with
400
0 0
50
I00
150
Time (h) Hot corrosion kinetics of a Pt-Rh aluminide system on MAR MO02 at 900°C
200
569
Coating Techrzology Table 1 Computed intrinsic d i ~ ~ s i v i t i in e s the substrat~/coatingsystem (PI-modified P-NiAl coating on MA Temperature K DAI
1073 1173 1273 1373
Intrinsic diffusivities, cm2s-'
.-2.73 x 9.28 x 2.49 x 5.08 x
10-12
10-12 10-l' 10-l'
Dc* 8.59 x 2.78 x 1.43 x 2.77 x
10.10-I2 10-" 10-"
composition, has been used for the interdiffusion t-modified P-NiAl coating on MAR M002. The average intrinsic diffusivities of Al, Cr, CO, Ni and Pt for the closed system (Pt-modified B-NiAI coating on MAR M002) have been calculated (Datta et al., 2000) - Table 1 - using the so called 'Inverse Method' (Danielewski and Filipek, 1996). The results of modelling the ~nterdiffusionin the Pt-modified p~ i A l l ~~ A0 diffusional ~ 0 ~ couple (closed system) at 107JK using the intrinsic diffusivities (Table 1) are presented in Figure 4. For the cal~ulationsof the concentration profiles in the oxidized coating/alloy system the following data were used:
Dco 9.42 x 4.17 x 1.05 x 2.27 x
10-" 10-'* 10-l' 10-"
DN*
DPt
2.50 x 10- l 3 1.39 x 10-I' 1.68 x 10-'' 1.06 x 10-l0
8.3 x 10-13 6.28 x 10-I3 2.17 x 10-" 4.22 x 10-"
values of the intrinsic diffusivities (Table l). Recent measurements of the concent and Pt in the PtlP-NiAl 60 minutes of annealing at 1 the asymmetric nature of the ~nt suggesting that the ~ s s u ~ p t of i o~~ o n s ~ average diffusivity is not valid. Progress in obtaining more precise intrinsic diffusivities in such complex systems
1. Atomic masses of Al, Cr, Go, Ni and Pt: 26.98, 51.99, 58.83, 58.71 and 195.09gmol-'. l (P-NiAl(Pt) 2. Thickness of the d ~ ~ u s i o n acouple M002): 2.A = 80 pm. on: c = 0. I 141x101cmW3. 3. 4. Annealing time t* = 100 h. 5. Constant (average) intrinsic diffusivities of Al, Cr, CO,Ni and Pt (Table 1). 6. Estimated values of oxygen uptake (kp= 1O-I2 g2 s-l) at 1173 calculate the flux of oxygen as a function of time Oo(t)) and the equivalent flux of A1 through the ) ,l ~ coating/scale interfaces: j ~ ~ ~j ( t~(t). Using intrinsic diffusivities from Table 1, the computer modelling of the interdiffusion in the selectively oxidized, Pt-~odifiedp-NiAl coating on MAR MO02 (i.e. AI is the reacting metal forming the Al,O, scale) was undertaken. The computed densities of Al, Ni and Pt in the Pt-~odifiedP-NiAl coating on MAR MO02 after oxidation at 1173 K for 100 h are shown in Figure 7. Satisfactory agreement can be seen in the case of A1 and Ni distributions, but the measured Pt densities exceed the The lack Of agreement between the measured and calculated values for Pt probably arises from underesti~ation of the
Figure 7 The calculated and measured density profiles of Al, ~i and pt the oxidized pt-modified P-NiAl coatillg on MAR MO02 superalloy after oxidation at 1173 KI for 100 h
80 '\
70 .
g
60
/
~
-U
.-$
50
5
40
0
3
-
-
30
AI
P
20
-
Pt +T-'l
I
0
2
4
6
8
'
I
10
'
I
I
12
14
'
I
16
'
I
----r--
18
20
Distance (pm)
C o ~ ~ e n t r a t i oprofiles n of AI, Ni and Pt in Pt pouple after 60mm of the diftusional
may be possible using the Morral-Thompson method of average composition.
In this section, consideration i s given to the issues surrounding the production, use, and performance of TiAl and TiAl, as coatings, primarily for conventional Ti alloys. The three ordered compounds in the Ti-A1 system display improved oxidation resistance in the l3 TiAl>.Ti3A1 in accordance with their ability to develop aii A1,0, scale dictated by the A1 contents of these compounds. Ti-aluminide coatings differ from the nickel aluiiiinide coatings in a number of ways: hile nicl~el-alui~inide coatings, unmodified and modified, represent well-establi~hedcoating systems, developed to protect Ni-base superalloys, the Ti-AI-based coating systems are still under development. (ii) Various metliods are used to produce Ti-A1based coating systems, whereas NiAl coatings are predominantly produced by the pack process. (iii) Ti-Al-based coating systems are being considered mainly to protect Ti-base materials. (iv) The effect of the Ti-Al-based coatings on the mechanical properties of the substrates still needs to be established. This is in contrast to the nickel aluminide coatings where a body of knowledge already exists concerning the mechanical behaviour of a l u ~ i n i ~coated e superalloys.
Ti-Al-based coating systems are, in the main, being considered for protection of disordered high teniperature Ti-aluminium alloys such as TIMET 1100 wt.%). Here the (Ti-6A1-2.75Sn - 4Z~~O.4Mo-O.45Si Ti-AI coating systems are designed not only to provide protection against oxidation but also to limit the interstitial dissolution, preventing the formation of an embrittled surface/subsur~dcelayer, with concomitant improvenient in ductility. Two main methods are used to produce Ti-Al-based coatings: magnetron sputtering (Leyens et al., 1996) and pack processing (Munro and Gleeson, 1996) for both TiAl and TiA1, coatings. The aluminizing pack process has been use mialek, 1993) to deposit TiA13, coatings of various kness (800/1038"C, 4 to 24h, sources =pure Al. Al-Si or A1-25Cr, activators = NaCl or NaF). The deposition of a duplex aluminide coating consisting of a tbiniier TiAl, inner layer and a thicker TiAl, outer layer (thickness depending on time) on TiAl has been achieved - 6 b at 1000 "C - (Munro and Gleeson, 1996). In this work the observed lack of depeiidence on the time of deposition of the aluminium content in the TiAl, at both TiAlITiAl, and TiAl,/ substrate interfaces indicates a steady-state growth of the TiA1, layer. Both layers grew by of A1 (Munro and Gleeson, 1996; Go 1971; Van Loo and Rieck, 1973). The growth kinetics followed a parabolic law. In variance with this finding, pack processing at 700°C for 10h produced a single TiAl, layer which transformed to a dual layer on subsequent annealing at 1050°C for 4 h (Mabuchi et al., 1989). The coatings produced by a pack process (Munro and Gleeson, 1996) suffered cracking during cooling from the deposition temperature ( 1000 'C)due to the mismatched coefficient of thermal expansion (CTE) between TiAl and TiAl, and showed very poor thermal stability at 1100°C. Aluininizing of pure Ti at 850°C (StreiK and Poize, 1996) and a diffusion couple study at 600" produced a single layer of TiAl, (Van Loo and 1973). Aluminizing pure Ti at >85O"C led to the formation of multilayered coatings coiisisting o f TiAl, and TiAl, (StreiR and Poize, 1996). In contrast to the well-established nickel aluminide systems (Sections 3.1 to 3.3), the Ti-A1 systems are new on the scene but are em~rgingas potential hightemperature structural coatings. For such emerging coating systems to be considered for high-temperature structural applications, the key requirements are oxidation resistance (discussed above) and adequate mechanical properties - mainly fatigue and creep strength. Consequently all recent work on Ti-A1
-
Coatirzg Technology systems has included not only studies of the environmental response of the coatings but also of their meclianical propcrties. (See Chapter 4, Huang and Chesnutt: Chapter 5 , Banerjee; Chapter 7, Yainaguchi and Inui in Vol. 2 of this treatise as well as several chapters in the present voltrme; Lipsitt et al., and Sikka, and Appel et al.) ~ a g n e t r o nsputtering has allowed the production of Tioatings on higli-tem~~er~ture 1100 - a gradient coatiiig with nt towards the surface and a multilayer coating consisting of Ti,Al, TiAl and TiAI, (Leyens et al., 1996). 0th types of coatings are observed to improve the oxidation resistance in air and in pure oxygen (Figure 9) of TIMET 1100, at 750°C. The lower oxidation resistance in air is associated with the presence of N2 (the nitrogen effect) which prevents the development of A120,. A mixed oxide laycr containing A120, and TiO, with (after prolonged exposure) an Al-depleted zone underneath characteri~edthe air oxidation morphology of the multilayer coatings. Improved parabolic behaviour was only displayed initially (70h); a transient paralinear stage was even~ual~y succeeded by breakaway kinetics. The gradieiit coatings undergoing transforination to multilayer coatings at higher temperature, with inferior initial resistance to oxidation, did not suffer breakaway corrosion. Pure oxygen exposure proiiioted mainly a-Al,O, foriiiation. The gradient coatings showed higher mass gain due to TiO, in the cracks formed during the high temperature gradient+multilayer transformation. Titanium alumini~ecoatings also affect mechanical properties. The fatigue behaviour from st~ain-controlled in,
.
Time (h)
Ti-A1 coatings: gradient (G:Ti-Al) and multilayer (ML:Ti-A1) systems. Tests were performed in air and pure oxygen at 750 "C
57 1
tests (R=0, ri =0.33 s-l) of the coated and uncoated materials has been fourid to be indi~tinguishableat rooiii temperature and at 600°C both in tcrms of the fatigue life and of fracture morpholo fatigue life was due to lower stresses caused by a decrease in Young's modulus at higher temperatt~re~ offset by the tendency to eiihanced crack formation promoted by environmental embritt~e~ent; thus the fatigue life at 600 "C was unaltered. life was invariant of the nature and thickness of the coating. It is important to bear in mind that the observed influence of the surface coating on fatigue life may differ from that on the damage-tolerant capability. A number of issues need attention in considering the propagation at high tei~peratureof pre-cxisting cracks at the coatiiig surface: (i) The crack closure effect reducing the stress intensity factor ( B K ) for crack propagation by oxide deposition at the crack tip. (ii) The coating/substrate interface actin temporary barrier to crack ~ r o ~ a ~ a t i oand n, the excessive crack-tip stress faditating tlic crack p r o ~ a ~ a t i othrough n the interface to tlie substrate leading to eventual failure of the system. Clearly soiiie of these factors will have conflicting effects. The neutral influence of such coatin behaviour of coated TiAl steins from two counteracting effects. The lack of oxygen ingress prevents the P+a transformation which occurs in the presence of oxygen and hence tlie generation of mobile dislocations promoting creep. This increase in creep resistance 1s offset by the overall increase in expected surface ductility.
Iron aluminide intermetallics are being developcd for use as structural materials and/or as claddin conventional e~gineeringalloys. n addition to their strength advantages, these inaterials exhibit excellent resistance to corrosion in single- and ~ulti-oxidant environments at elevated t e ~ p e r a t u ~ ethrough s the forniation of slow-growing, adherciit almnina scales. The oxide scales also act as barriers to the t corrosion-accelerating reactants such as alkalis, and they retard the scaling kine underlying substrate alloys when they are exposed to multi-oxidant environments. In the iron-aluminium system, the alloys of interest are ofco and FeA1. The crystal structure of Fe
572 2 structure. The melting temperatures of Fe,Al and FeAl are 1520 and 1250"C, respectively. The density values f these alloys are 6.72 and 5.56 g c~yl-~, respectlve~y. owever, the Young's modulus values for Fe,Al and FeAl are 140.6 and 260.4 GPa, respectively, iiidicating that the latter has a tendency to be much more brittle. Iron alui~inidesare of interest primarily because of their much lower cost when compared with nickel alu~inides,a lower density than stainless steel (with potentially a better strength-to-weight ratio), and hightemperatrrre corrosion resistance. However, limited ductility at ambient temperatures and a sharp drop in strength above 600'C have been major deterrents in use o f thesc aluminides as structural materials deau er al., 1987; Culbertson and Kortovich, 1986; amey et al., I99 1: and Vedula, Chapter 9 in Vol. 2 of this treatise). In general, other elements are added to these alloys to improve their mec~anicalandlor corrosion properties in differing environments. Corrosion resistance is generally imparted to structural alloys by in-siru development of c h r o ~ i a ,alumina or silica scales on the alloy surface. The slower the growth rate of the oxide, the better the oxidation resistance of the alloy. In this vein, alumiiia scales (which have ~nherentlyslower growth rates) can offer subs~antialadvantages over chromia scales, espccially in single-oxidaiit environments. The oxidation resistance of iron aluminides depends on the formation of a chemically stable A1,0, surface layer upon exposure to an oxidizing environment. Studies of the phase stability in the Fe-Al-0 system demonstrate that Al,Q, will form on the Fe aluminide class of alloys even at relatively low oxygen partial The A1 levels present in Fe aluminides pressure (PO,). (15.9 and 20-30wt.% in Fe,Al and FeAl, respectively) are well in excess of the critical concentration needed for the foimation of a continuous alrimina scale on the surface (Natesan and Tortorelli, 1997). Even though the corrosion r e s i s t a ~ cof~ Fe aluminides is significant in complex gas environments, the present use of these alloys as structural materials at elevated temperatures is very limited because of their inadequate strength properties and fabrication difficulties. Several approaches have been used to apply Fe,Al or FeAl onto the surface of conventional structural alloys, thereby maintaining the strength properties of the substrate and the corrosion resistance of the iron aluminide. A ~romisingap~roachfor application of iron alumini~eis the weld overlay process, in which claddings of Fe aluniinides are produced by electrospar~~deposition ( E S ~ ) gas , tungsten arc (GTA) and
gas metal arc (C Johnson, 1995; Coo The ESD process is a uses short~duration,high-c~~r~ent electrical pulses to deposit an electrode material on a metallic substrate (Natesan and Johnson, 1995). A principal advantage of ESD is that the coatings are fused to a metal surface with low heat input while the bulk substrate material reinains at ambient te~perature. This eliminates thermal distortions or chaii in the metallurgical structure of the substrate. cause the coating is alloyed with the surfiLce, i.e. allurgically bonded, it is inherently more resistant to damage and spalling than the mechanically bonded coatings produced by most other low-heat input processes, such as detonation-gun, plasma-spray and electrochemical plating. Overlay samples of F uminide were produced by GTA (Goodwin, 1997 c et al., 1997); however, the high A1 content of lay was found to lead to hydrogen~crackin~ susceptibility. Pre- and post-weld heat treatments have been applied to alleviate the problem. A pack cementation method was used to develop Fe-A1 and AI-Cr intermetallic coatings on ng and Rapp, 1997; Dionne and ferrous substrates ( Lo, 1996). In the pack cementation process, specimens of alloys to be coated are placed in a pack consisting of a mixture of A1 and Cr powders, a chloride activator salt and an inert aluinina filler powder and heated to elevated temperatures for a length of time dictated by the desired coating thickness. FeAl intermetallic coatings were also produced by a magnetron sputter depositioii technique in which a substrate bias voltage of 15OV and a target current of l 5 - 2 0 m A ~ m - ~ were used to obtain a coating of 28 pm in thickness after 6 h of deposition (Liu et al., 1998). FeA1 coatings have also been attempted using a thermal spray technique in which Fe-Al-Nb powder was deposited with a high velocity oxy-fuel spray using H, its er al., 1995; Blackford er al., 1998). temperatures transient iron oxides will be present and the thermodynamically stable alumina can develop a continuous scale only over long exposure times. The progressive oxidation of Fe,AI has been studied by Raman spectroscopy and 'ruby' fluorescence from 300 to 1200°C (Natesan et al., 1995). The results indicated that, by 500 "C, Fe2@ appears prominently in the scale and is replaced by Al,O, at T> 1000°C. With progressive oxidation an increasing conipressive strain develops in the scale. Strain relief (at room temperature) clearly occurs when reaction temperatures exceed 950 "C, probably because of crack forimttion in the scales, thus sign all in^ the onset of spallation.
Coating Technology Extensive studies have been conducted on the corrosion pcrforniance of alumina-forming alloys and Fe aluminides in several oxidizing environ~eiits (Natesan, 1993a; Natesan and Cho, 1994; Tortorelli and Dc Van, 1992). The results sliowed that a critical A1 content in excess of 12wt.%, which is present in Fe aluminides, is needed for the formation o f alumina on the alloy surface in single- and multi-oxidant environments. Furthermore, a comparison was made of the oxidation performance of bulk Fe aluminide with that of a coating of the same applied on a steel substrate. Comparison of Auger A1/0 peak-to-peak ratios for thc ternary Fe-Cr-A1 intermetallic alloy and for coatings of Fe aluminide on a Type 316 stainless steel substrate after oxidation in air at 650°C iiidicated that the coated specimens develop scales 200-600 thick whereas the bulk alloy developed a scale -1pm EM analyses of the surfaces of these specimens showed similar oxide morphologies in all of the specimens. Since the bulk aluminide and the aluminide coating exhibit similar oxidation characteristics, information obtained from oxide scales developed on bulk alloys would be a ~ ~ ~ l i c a btol e the scales developed on aluniinide coatings. Strains in the thermally grown oxides were determined by ruby ~~iorescence measurements for scales that developed on various substrates (Lipkin and Clarke, 1996; Ma and Clarke, 1994; and Renusch et al., 1996). The data provide a sensitive measure of strain accumulation in the scales and clearly indicate when strain relief occurs. The technique can be used to compare the strain values for specimens that have received diRering treatments or to compare the values obtained for various alloy speciniens that have been oxidized under the same conditions, although it is difficult to assess the importance of the measured strain values in scale fracture because little or no quantitative data are available in the literature for films of alumina. The trends identified in such an approach can shed light on the important variables that are res~onsiblefor the mi~rostructure,alloying-element effect, and adhesion and time-dependent variatio~sin the scale and at the scale/metal interface observed. The above-cited investigators studied five alloys oxidized for 100 h at the three tei~peratures. Alloy FA 186 is a ternary alloy that contains Fe, Cr and A1 and is considered a base alloy. Alloy FA 129 is designed to exhibit high ductility at room temperature while retaining its strength at high temperatures, whereas Alloy FAIL, is designed for improved oxidation resistance through addition of Zr. Alloy FAS is
A
573
designed to resist sulpliidati~n,and for improved resistance in aqueous environments by deliberate addition of MO. The results show that the strain values ranged between 0 . 0 0 ~and 0.012 for specimens oxidized at 800 "C. enerally, the scatter in the data among the three sets of meas on each of the specimens is fairly small specimens oxidized at 3000°C show 1 for FA186 and FA alloys, indicating significa~~t spallation of the scale. This observation confi~msthe §EM analysis of the fracture surfaces and thermogravirnetric test data for these alloys. The in-plane strain values for FA 129, FAL and FAX are consistently high and indicate that the scales developed in tliese alloys can accoinrnodate some deformation, as evidenced by the lack of spalling in the oxidized alloys. The results for specimens oxidized at 1200°C indicate low strain values for all of the specimens with the exception of FAL, which exhibited strain values in the range of 0.01-0.016. The higher strain values in FAL seem to correlate with the lack of a convoluted scale and absence of substrate deforniation in this alloy when compared with other alloys. To examine the adhesion of t~ermallygrowii scales to tlie substrate, Natesan et al. (3996) applied a tensile pull to separate the scale from the substrate. The technique they used involves attachiiig an epoxycoated pin to the scale surface at a temperature sufficient to cure the epoxy. The pin is subsequently separated from the sample at room tem~eratureby applying a tensile load. From the applied load and pin area of contact, stresses needed to pull the scale from the substrate were calculated. The results show a in maximum stress value for specimens of FA 186 and FA 129 exposed at 1000 "C. On the hand, maximum stress values for FAL and specimens are almost independent of exposure ternperatwe. The adhesion test results indicate that, even with a wide variation in the oxide layer thickness (resulting from oxidation for 100h at 800-1200 "C), the tensile stress needed to pull the saniple from the substrate is fairly independent of oxide thickn difference in the stress values for FAL, and F be due to differences in the chemistry in the in the scale/substrate interface, which are directly ~ n ~ u e n c eby d the initial composition of the substrate alloy. The results also show that Zr (in FAL) and (in FAX) additions e a similar effect at 1000 1200 "C, where a-A1 will be the stable oxide in the scale. A comparison of the results obtained from these two alloys after 800 "C oxidatioii indicates that rather than Nb, addition may stabilize the alumina
scale (i.e. minimize the transient oxides) on the alloy surface. S
or high~te~nperaturestructural materials with significantly higher meltin points than presently used alloys, attention has been drawn to silicides. Of the potential candidate systems, molybdenuni silicides are ctive owing to their high nielting mar, Chapter 10 of Vol. 2 of this treatise.) ~ o l y b d e ~ L idisificide m has particularly good
ng oxidation resistance at elevated irkus and Wilder, 1966; Regan et al., et al., 1992; Bartlett et wl., 1965). terial has been reported to show high creep rates at temperatures > 1200 "C ( 1992) and exhibits degradation by a plienornenon known as 'pestiiig' at 600 'C (Fitzer, 1956; The l o w - t e m ~ e ~ ~ ~pesting t u r e phenomenoii arises due to the extremely sl rowth rate of SiO? scale o oxides, While oxidation in c o ~ p a r i s with o ~ that on the surpace of the alloy is beneficial, if oxidation occurs in the interior of the sample, the alloy may be subje~tedto pesting attack because of a large volume increase of 340 and l8OYOin forming MOO, from Si, respectively. Porosity and preexisting cracks in the starting material have a large i ~ ~ ~ e ion i c oxidation e performance in that oxidation in the interior of the alloy can lead to widening of the cracks, e n l a r ~ e ~ eof n t the pores and further acceleration of the Oxidation reaction. At temperatures above 900 "C, a continuous silica scale is generally th a resultant enhanced resistance to tternpts to minimize the pesting degradation aim to develop the silicide as a crack-free coating
-
a laser beam to produce a dense material in the form of e layers with fairly good b o n ~ i n gto the substrate idouci and Peiletier, 1998). These coatings were developed by applying either a powder mixture of pure i or by a powder of The coatings developed with pure powders contained e and some unreacted MO and and Fe%, phases, formed by
reaction of substrate steel with Si powder, were observed. The coatings developed by applying MO powder also containe coating contained a preheating the substr were beneficial in minimizing cracks in these laserdeposited coatings. Cockeram et al. (1995) studied the growth kinetics and pesting resistance of MoSi, coatings and Ce-doped MoSi, diffusion coatings grown by the pack ce~entationmethod. They concluded that a protective layer of sodium silicate rather than silica can be formed rapidly by the addition o f a sodium oxide layer to the surface of Ivl postulate^ that sodium ions break up the ti network of Si-0 tetrahedral bonds in SiOz and also increase the diffusion rate in silica thereby iinprovin~ the growth kinetics of a sodium silicate or sodium aluminium silicate (the source of A1 being the alumma crucible reacting with N aF activator in the cementation pack). Addition of Ge to i,has been reported to accelerate the growth kinetics of the oxide scale (Fitzer LJ/ al., 1973), but Cockeram et al. (1995) concluded that most s i g n i ~ ~improvement ~nt in pesting resistance was achieved by the presence of sodium silicate. and Ge-doped titanium T i t a ~ i silicide i ~ ~ ~ and silicide coatings have n e~amined to provide oxidation resistancc for pure Ti, Ti-Al-Nb and pTiA1 alloys (Cockeram and Rapp, 1995a, b; Gleeson, 1996). In all these studies, a hali~e-activ~ted pack cementation method was used to codeposit Si and B, or Si and Ce, or silicide diffusion coatings. The Bor Ge-doped silicide coatings were grown by diffusive conversion of the Ti te to form either a i,/Ti,Si coating with a or else a multilayer of Ge-doped silicide solid solutions (Cockeram and Rapp, 199%). The purpose of the B or is to dissolve into and i m ~ r o v the e protect grown during hig~-tei~perature oxidation by forming y healing cracks at a low Ge-doped silicide coatings ion against cyclic and isothermal oxidation of pure Ti and Ti-Al-Nb alloys. Gedoped silicide coatings exhibited slower oxidation k~neticsthan those for -doped coatings. Both these coatings were reported as effective barriers to the inward penetration of 0 for pure Ti and Ti-Al-Nb silicide coatings were developed to TiA1 (Cockeram
Coating Technology
Gleeson, 1996). The mechanism for the silicide coating growth was attributed to the inward diffusion of Si, deposited by gaseous Si-halide species at the sample surface - the coating was compact and planar. On the other hand, the coating structure produced by pack siliconization of y-TiAl consisted of a compact multilayer structure which eventually developed an extremely porous middle layer, that was attributed to breakaway oxidation. The rate-controlling mechanism for the silicide coating growth was reported as the diffusion of A1 away from the coating~substrate iiiterfax due to the low solubility of A1 in tlie Ti silicides. The silicide coating integrity was improved when developed by diffusion annealing in a this improvement was attributed to a Xowe at the coating surface enabling sufficient time for rejection of A1 ahead of the inward-moving silicide/ TiAl interface.
4.
ics
The aim of this and the following section is to provide a discussion of the available surface engineering methods to overcome the limitations of structural intermetallics. The discussion focuses on activities relating to tlie developnient of: (i) Coatinglsurface modification processes to improve oxidation resistance (Sections 4.2 to 4.6). (ii) Barrier coatings to prevent fibrelmatrix interactions in intermetallic composites designed to improve fracture toughness of monolithic intermetalfics (Section 5).
One of the most promisi~g applications of bulk IMCs is with y-TiAl which is being pursued as a potential constructioiial material for component within both aircraft and industrial gas turbines. This section will therefore concentrate on coating systems to protect it. y-TiAl offers particular promise as a material for the fourth stage of the gas turbine, where good specific stiffness and stren~thcan give rise to large weight savings and therefore lower iiiechanical loads on these turbine components. However, the use of y-titanium aluminides at high temperatures is limited by their relatively poor oxidation resistance and their susccptibility to highte~peraturecorrosion. A further problem relates to
545
the formation of an oxygen-embrittled zone beneath the oxide scale. Early studies reported this embrittled layer as a,-Ti,Al, but more recent studies support the formation of an oxygen-containing cubic intermetallic phase (TiSA1,02).This oxygen-contain in^ i n t ~ ~ e t a l l i c degrades tlie mechanical properties of y-T high te~peratureservice. Also, to explo aluminides’ full potential as high-teinperat~restructural materials in energy conversion systems, it is necessary to characteri%etheir high-temperature degradation modes in aggressive oxygen-, sulpliur- and chlorine-containing mixed environments. (See also chapter by Natesan and Datta in this v o l u ~ e . ) I~provementin oxidation corrosion resistance may follow the addition ternary or even quaternary elements such as V, , Nb, Ta, W and Cr. It is well documented in the open literature that b and Cr significantly improve the oxidation resistance TiAl, but that their influence under mixed corrosion or in the presence of deposits is less documented. Data for Ti-44Al-2~b-2 gest tliat, even with the additions of Nb, this alloy has a significant hot corrosion problem. To alleviate these corrosion problems, the possibility of forniing other intermetallics as stable diffusion barrier coatings on y-TiAl is being investigat~d,For example, recent work has showii that the addition of noble metal elements (Pt, Ag, Au in addition to Nb) can stabilize cubic quaternary internietallics based on y-TiAl and therefore such phases may also show promise as intermetallic barrier layers.
The developi~entof coati~g$to protect Ti a l ~ ~ i n i d e based intermetallics a inst high te~peratLireenvironmental degradation, otably oxidation, requires a knowledge of the scaling processes in these alloys. A l t h o u ~the ~ oxidation characteristics of Ti aluminidebased materials have been studied extensively (Perkins and Meier, 1991; Smi umphrey, 1992; Welsh and Kaveci, 1989; aE., 1991; Fish and Duquette, 1993; Cha Coddett, 1987; DLIet al., 1994 and Doychak, chapter in Vol. 1) there are areas where ~~ncertainty still exists. Clearly the protectivity of Ti aluminide intermetallics can only be obtained through the formation of a thermody~amicallystable, slow-growing, continuous and adherent scale and Humphrey, 1 Welsch et aE., 1991). However, the develop~entof
such a scale on Ti aluminide ~ntermetallicsis more difficult than for disordered Ti-A1 alloys and other higher concentration of A1 is required to forni a pr~tectivea ~ u ~ i scale n a on Ti alui~inides ns and Meier, 1991; Du et itionally, the values of the gy of formation of A1,03 and the oxides of Ti are comparable. In Ti-A1 both A1,03 and med. Tlie generation of the other oxides retards the development of layer. The temperature, the alloy coinposition and the constituents of the oxidizing environment will affect the f o r ~ a t i o nof the initial oxides and subsequent scale development. Here a significant point that has emerged from previous oxidation studies concerns the role of N, as iiidicated by the d i ~ e ~ e noxidation t characteristics of Ti-A1 etallics in air and in pure oxygen (Perkins and , 1991; Meier and Pettit, 1992a,b; Chaze and tt, 1987). The higher oxidation t e ~ p ~ r a t u r e ( > 1000 "C) and the presence of N2 (air oxidation) Led to TiO, formation. Oxidation at lower temperature 0°C) in pure oxygen favours the fomation of the surface condition of Ti-A1 dictating the type of oxide to be formed. The A1,03 scale formed by reo oxidation remained protective even in the event of subsequent exposure to air. It appears that there are several important aspects of the oxidation of Ti-A1 that need attention, viz: the occurrence of layered scale ~ ~ o ~wo m t ~~ ~ i the ~ i alterna~~ve ng ~ iA 1 ,0~ 3scales, ~ ~ the nitrogen effect, the transitioii between the TiO,/ Al,O, kinetics and the way these aspects can be in~uencedby the transient state oxidation. work involving an in-situ (monoch ed X-ray photoelectron spec Y) study of the initial stages of oxidation of a Ti-54Al intermetallic provided further insight into the oxidaehctviour of Ti aluminide intermetallics. The spectra of a s~utter-cleanedsample at ambient ernperature reveal almost no presence of the oxidic part of the aluminium 2s core lcvel - that is, no separated peak or s ~ o ~ l d at e r higher binding energy. [1L ( L a n ~ u i r of ) oxygen corresponds to an exposure of one second at 10-6mbar (lL=IO-6rnbars-1)]. After a few Langmuir (12 L) of oxygen, the growth the oxygen 1s and the oxidic parts of the Ti core levels are seen. This is best visible for the aluminium 2s core level where at 119.3eV binding energy an oxidic peak occurs. Similarly, oxidation can also be obse~vedin the Ti 2p spectrum after a few Langinuir of oxygcn. It then consists mainly of four parts. First, there is the Ti 2p doublet (Ti 2p1', at
459.8 eV and Ti 2p3', at 453.7 eV) at the same binding energy as in the pure, sputter-cleaned alloy. Additionally, two broader features are seen corresponding to Ti in higher h~ndingenergy states at about 464 and 458 eV binding energy. With higher oxygen exposure there is 110 associated significant further change in the form of the spectra. The oxygen Is core level line of the sputtered, cleaned surface indicates the presence of a native oxide. The calculation of peak areas, taking iiito account the photoionization cross-sections (Scofield, 1978; Hiifner, 1996), allows d e t ~ ~ n ~ i n a t iof o n the concentrations of the elements, For the A1 2 core level, following a deconvolution in oxidic and metallic portions for the 12 L cover~ge,the AlO,/Al ratio is calculated to be 0.80. Althougli a deconvolution of the Ti 2p spectrum is very difficult, nevertheless, one can roughly estimate that the ratio of oxidic Ti to oxidic A1 is about 0.6. A remarkable fact is that, upon heating to 700"C, the oxidic part of the A1 2s level grows and shifts to a higher binding energy while the oxidic parts of the Ti2p levels decrease. The binding energy of the oxidic A1 2s core level ends LIPat a position close to the value found for the stoichiometric AI@,. It is important to note that the A12s binding energy of 117.1CV in tlic alloy is lower tliicn that of the pure element (1 17.9 eV), In contrast, the Ti 2p doublet remains almost unchanged in the alloy. The spectra from the iwsitz4 oxidation studies at 850 "C, reveal the absence of a saturation effect unlike that at ambient te~perature.The oxidic part of the A1 2s grows contiiiuously while the metallic part decreases. The increase is not as pronounced as it was with the oiiset of the arnbient temperature oxidation, and the position of the oxidic peak is at higher binding energy. After 1000 L of oxygen, almost all A1 at the surface is seen to be oxidized, presurnably to the A13+ state as in AI, 3. The Ti2p core level indicates almost no oxidation of the Ti in the beginning. This is evident from the fact that the typical broad features at 464 and 458eV binding energy are not visible, even at 430L of oxygen. At lOOOL, however, large amounts of oxygen are bound to Ti Zeadiiig to the characteristic four features, Unlike the ambient temperature case where the oxidation slows after a few Langmuir and only a certain amount of Ti remains in the oxidized state, at 850°C a pronounced Ti oxidation sets in and is only delayed with respect to the AI oxidation. At 850 "C the oxygen content is seen to constantly increase, while at ambient temperature the 01s level only shows an ini~ialrapid growth. In addition, the Al/Ti concentration ratio quickly increases at the surface as a few Langmuir of oxygen
Coating Technology
are supplied to the alloy surface at 850°C. This goes hand in haiid with a quick oxygen uptake. After about 30 L this development slows. At 720°C the relevant core level lines become enhanced with values up to 5800L of oxygen. The A1 2s core level shows that with the oxygen supplement to the surface the oxidic part increases quickly at the expense of the metallic part. But, in contrast to the 850°C case, the metallic part does not vanish totally suggesting that metallic A1 is still present at the surface , For the first 120L only AI rved, leaviiig the Ti2p level very similar to the situation at 85O"C, with higher oxygen load the Ti also reveals the typical features for oxidation. Coincidently, with the more pronounced oxidation of the Ti, a shift occurs of the 0 1 core level from near the positio~in Al@, to a higher binding energy as in Ti02. One way to demonstrate the evolution of the surface oxidation is to calculate the concentration ratios of oxidic to metallic AI and that of A1 to Ti. In the beginning, as the A1 is strongly oxidized, its surface concentration increases. As the A1 o ation slows and the Ti oxidation sets in, the All ~ o ~ c e n t r a t i oratio n at the srirface decreases again, even below the value at the teginning of the experiment. For the sample oxidized at elevated temperature, i.e. 850"C, the oxygen uptake of tlie surface was not limited to a supplement of a few Langmuir of oxygen. The oxidic 4 2 s core level part grew slower than at ambient temperature and the metallic part finally vanished after 1000 L. This leads to the sug~estion that the reaction depth was markedly higher than at ambient temperature. In addition, the oxidation of the Ti was delayed with respect to the Al. With the experiment at 720°C and higher oxygen exposure, what is found is that, compared to the 850°C case, the oxidatioii of the Ti is delayed with respect to the AI oxidation. As the AI oxidation slowed after 120L of oxygen, the Ti concentratioii at the surface increases again. This is concomitant with a stronger oxidation of the Ti. It is remarkable that after 5800L metallic A1 is still visible. From this experiment alone no conclusions caii be drawn as to whether this metallic A1 signal originated from the underlying alloy or from metallic particles embedded in an oxidic matrix. From air oxidation experiments on Ti-6A1-4 V it has been shown that nodular A1,0, nuclei develop on a sample at elevated temperature. In nodules a thin rutile (TiO,) film was identified by EDX ~ ~ e ~ s L i r e i ~Therefore, eiit. the most likely explanation for the still visible metallic A1
579
i s the forination of A1@, islands on the alloy leaving
parts of tlie metallic surface uncovered. mind, the retarded Ti oxidation is pre initial development of the rutile film and an indication of alternating layers of A120,/Ti
An i n - ~ ~MXPS i t ~ study was underta~eiito gain in into the very initial stages of Ti-54Al interme oxidation. This was done for different substrate temperatures using different i n - s i ~ cleaning ~ processes. While at ambieiit t e ~ p e r a t u ar ~very thin mixed oxidic scale of a few monolayers was formed after a few Langmuir of oxygen exposure and no further oxidation occurred, progressive oxidation was observed at elevated temperature. it was possible to identify the character of thc very early oxidic surface species. In addition, compatibility with earlier iiivestigations concern in^ scale growth for much higher oxygen supplenients was found. A distinctly different oxidation beliaviour occurred depending on whether the sample was pre sputter- or laser-cleaned. In the light of tlie measurement perforined here, it seems that the lasercleaned surface exhibits a higher stability against oxidation at elevate emperature. This needs to bc further investigated. orphological ii~vesti~ations on the laser-cleaned surface and gravimetric techni~uesto determine the oxidation kinetics are required. Three approaches have been adopted to design oxidation-resistant coatings for titanium aluminides (Taniguchi, 1994; Streiff, 1993; rady et al., 1996a,b): 1. Aluminizing (Streiff and al., 1989; Smialek et al., 1990; Uoshihara et al., 1991; Sinialek, 1993; L 2. Coating system based CO) ( ~ h i m et i ~al, ~ 199 Kee and Luthra, 1993). ide coatings (Cock see Section 3.6.
Alurninizing TiAl by a pack process has been used to develop oxidation-resistant coatings (Takei and Ishida, 1992). A duplex coating consisting of a thick TiAl, outerlayer and a thin TiA1, inner layer (Figure 10) was produced at 1000°C using a pack niixture of 10-50% Al, 5% Alp, the r e ~ a i n d e rA12 ,. This process has been found to be very ve in improving oxidation ever, oxidation at 1373 I(, resistance (Figure 11). ( 5 h) led to scale layering - Al,Ti, 6 and A1,Ti - by the
578 AI,Ti
AO ,I ,
Al,Ti
Ti alloy
AlTi AITI,
10 pm H
ure 10 Concentration profile on cross-section of alu~ini%ed Ti alloy after oxidation at 1173K
50
0
973
1073
1173
1273
1373
Temp (io
1 Mass gain of TiAl with and without alurniiiizing after 5 h oxidation at tem~eraturesbetween 973 and 1373K
inward diffusion of AI (Figure 12). Clearly this inward diffusion of Al will knit the protectivity of the coating. Improved oxidation resistance of TiAl intermetallics has been achieved by preoxidation in pack mixtures of various compositioiis (Taniguchi et al., 1993; Taniguchi and Shibata, 1992). The use of packs to carry out heat treatment ensures preoxidation at low PoZ.This idea is based on the fact that oxidation of TiAl at low Po2 leads to selective oxidation of Al, avoiding the difficulty caused by the similar equilibrium dissociation pressures of rutile and aluminium oxide, and promoting their siniultaneous formation. Both SiO, aiid TiO, powders have been used to achieve low PO, oxidation. Heat treatment of these
Figure 112 Coiicentration profiles on cross-section of a ~ u ~ i n i z eTiAl d after oxi~ationat 1373
powders under vacuum at 927°C for ZOOks allowed the format~onof Al,O, or ~ l ~ O , - r i coxides h c~~racterized by whisker-type structures. These oxides have been found to provide protection against cyclic oxidation at 1027°C for 400h. A variant of this method, involving preoxidation of specimens of Ti-AI and Ti-A1 f 0.24 wt. % Hfpacked in a mixture of chromia and metallic powders, also produced coatings with much improved oxidation resistance at 1124 "C (20 h) (Taniguchi and Shibata, 1992). Exploratory work to find the influence of pack composition (1 OO%Cr,O~, 7 0 ~ 0 ~ r ~ O ~Cr, -30~0 3 0 ° ~ C r ~ 0 , - 7 0 ' ~ ~ Caiid r ) teiiiperature (927-1 127 "C) has allowed establish~entof the optimum composition and temperat~re(70°/~Cr,0,-30~/~~r at 1127 "C) for the most effective resistance to oxidation at 1127°C. Two factors - the activity of oxygen and the kinetics of oxidation - d e t e ~ i n e dby the pack cornposition aiid processing temperature, were found to be responsible for the growth of a defect-free, protective Al,O,. All other packs produced either porous coatings or coatings with insufficient thickness. It has been reported (Taniguchi and Shibata, 1992) that the surface structure of the alloy after the pretreatments consisted of a thin top layer of oxide (mostly alumina with a small quantity of chromia and chromium) and a sublayer of Cr-rich alloy, suggesting the occurrence of alloy ~ x i d a ~ i oand n deposition of Cr during the pretreatrnent process. Since the surface structure of the ailoy after pretreatnient can be greatly aEected by the vapours generated inside packs at high temperatures, it would be extremely useful to the understanding of the observed pretreatinent effects if the vapour composition within the packs could be
Coa ling Technology 0
-4
+Cr (Cr203pack) -8 -4--0, W203pack)
,.-.
6
-a-- Cr (70~r20~-30Cr pack)
-12
v
a
+0, (70Cr203-30Crpack)
a,
2 -16 'C 3
13 5 -20 U1 W
-24
-28
-32 900
1000
1100
1200
1300
1400
1500
Temperature (K)
Figure 13 Equilibrium vapour pressures of O2 and Cr i n Cr,O,-Cr packs
estimated at the pretreatment temperatures. Figure 13 (Xiang et al., 2000a~compares the calculated equilibrium vapour pressures of 0, and Cr in the Cr,03 pack and in the pack coiitaining 30 wt.% Cr at temperatures from 1000 K to 1400K . The calculated equilibriuni vapour pressures of O2 and Cr iii the pack containing 70wt.% Cr are the same as in the pack containing 30 wt.% Cr. It can be seen that the vapour pressure of 0, i s Comparable to that of Cr in the Cr,O, pack, although the former is slightly lower than the latter. Adding 30 wt. % Cr to the Cr,O, pack greatly increased the vapour pressure of Cr, but, substantially reduced the vapour pressure of O,, creating a favourable condition for depositing Cr and for mildly oxidizing addiiig 2 wt.% of CrCI, into the .% Cr it is possible to create an even more ~dvourablecondit~onfor depositing Cr because a considcrably high vapour pressure of CrCl is ut, the oxidizing behaviour o f the pack way not be affected, since the vapour pressures of both 0, and Cr remain the same as in the pack containing no CrCl,. It is known that CrCl is the carrying vapour species that i s responsible for depositing Cr in metal surfaces. In this contcxt it is instructive to see the possibility of obtaining co-deposition of AI and Si which can be subsequently pre-oxidized. Recent theoreteical studies (Xiang et al., 2000b) on the pack cenientation process have denioiistratcd that it is possible to co-deposit AI and Si on metal surfaces
in a single process. Such c o - ~ e p o s i t ~ may o ~ occur wl~en the halide vapour pressures o f the two elements generated in the powder packs are of the same magnitude. Figure 14 shows that by carefully choosing the halide activators such conditions for ca-deposition can be achieved. Figure 14 illustrates the possibility of co-deposition of AI Si using powder activated by CrCl3.6 a commercially av and cost-effective salt; it reveals that at a coating temperature of 1000"C it is possible to co-deposit and Si oiily when the AI content in the packs i s between 2 to 2.05 wt.%. Such a narrow composition range for processing certainly requires delicate control of the process. However, it may be possible to widen the composition range for processing by carcfulully using two or more types of activators for the same pack.
protectivity against idation under cyclic conditions up to ~0OOhat 900 in air. This development stems from the observatio that Ti-Al-Cr alloys can form continuous A1,03 scales with a minimum 8 to 10% of eier, 1989). A s~utter-deposite~ coating optimized at the compositioii level Ti-44A128Cr on Ti-47A1-2Cr-2Ta has been observed to provide resistance against cyclic oxidatio~at 900 "C
580 0 1
-2 c-..
-g
-3
Q - 4 =t ._ .-
CTI = UJ
-5
-6 -7 -8
-9 -40
Depcndence on A1 content of vapour pressures of AI and Si chloride species at 1000'@ for the composition series 3 Si -k xAl -+4 CrC13.6H20+ (93-x) A1,03, (wt.%)
in air up to ~ ~ 0 (0 h arron et al., 1992). Two chemical iiicompatibility problems - brittleness (leading to the f o r ~ a t i o nof reaction zones of Cr-rich precipitates) - limit the usefulness of these coatings. ~undame~ital work based on the determination of phase r e ~ a t i o ~ s ~in i pTi-Al-Cr s systems (Brady et al., has allowed identi~cationof the main phases consistiiig inaiiily of z (Ll, phase centred on Gr) or yTiAl phases and the Ti(Cr,Al), Laves rady et al., 1996a, b). In the Ti-AI-Cr composition range favouring the formation of A1,03, the Laves phase, while i ~ ~ a r t i i i gm ~ u n i t yagainst oxidation, causes brittleness (Brady et al., 1995a, b). cracking resistance has been reported et al., 1994) to stem from the presence of a aves pkase with z phase or y phase. The susceptibility of these phases to transformation to ittle TiAlz phase and Gr,Al phase or P-Cr at adversely affects the improved cracking resistance derived from the preseiice o f the duplex structure up to 1000°C but imparts some degree of roomtem~erature~uctility . The phase in Ti-Al-Cr-alloys in the c o ~ ~ o s i t range ~ o n promoting Al,O, remains stable LIP to 1000°C and imparts some degrce o f rooiii~ O ~ - f o ~y + i Laves n ~ , Tishows the existence of a contin~ousy-phase (Figure 15) ( rady et al., 1996a, b). WO benefits result from this:
1. y-phase surrourrdiiig the brittle Lavcs phase eliminates brittleness by crack~bluntin
2. the presence of a y-phase ensures chemical compa tibility . A low-pressure-plasma-spray (LPP
y + Liives coating (Ti-51Al-12 Cr> on a Ti-48A1-2 Cr-
rady et aE., 1996a, b) has been reported to improved oxidation ~esis~ance (up to 500 h) at 800 to 1000 "C (from discontinuous weight-gain kinetics) as indicated by the low mass tion of the absence of cracks in the coatings and of interdi~usion,and by good coating/substrate thermal compatibility (Figure 16).
Alumini%~ngusing the pack cementation process to develop a TiAl, layer has been the obvious choice to protect Ti-A1 alloys from oxidation. scale developed on TiAl, affords l i ~ i t The reported disadvantages (Taniguchi, 1994; Takei and Tshida, 1990; Mabuchi et al., 1989) include rapid diEusion of Ti and Al, crack formation in TiAl, and the f o ~ a t i o i iof TiAl, phase during processiiig or subsequent treatment. A1 depletion by fast diffusion of A1 affects the ~egenerationability of the protective irkendal~voids a c ~ o m p a n i eby~ the fast diffusion of the alloying elements reduce the scale/ substrate adhesion. The introduction of a Ni barrier layer, to some extent, eliminates this problem. The brittleness of TiA1, and TiAl, when subjected to iiiteriial stresscs developed during aluminizing results in cracking and failure. These observations have led to
58 1
Coatiag Technology 1.0
I
I
0 0 /
200
400
Time (h)
\
Cr
AI
2.0
.
*O0
-.
,, ,
400
Figure 16 Interrupted weight-gain oxidation data for LPPS Ti-51Al-12Cr-coated and uiicoated Ti-48A1-2Cr-2Nb at (a) 800 "C and (b) 1000 "G iii air eter et al. alurnina f o f ~ a ~ j oboundary n (bf
F i ~ u r15~ (a) Schematic Ti-Al-Cr oxide map of Perkins atid Meier (3989), (b) Schematic partial 800-1000 "C Ti-AI-Cr phase diagram, showing the composition range of the y and NASA Lewis oxidat~o~-resistaiit coating alloys
the develop~~ent of MCrAlY types of coating to provide protection for TiAl. ~lasma-sprayedMCrAlY (M = Ni,Co) has been found to be not very effective in improving the oxidation resistance of TiAl (Furukawa, 1991; ~himizu et al., 1992); the €ori~ation and maintenance of a protective A1,0, was difficult. In contrast, a magnetron sputter-deposited, finegrained Co-30Cr-4Al layer (- 30 pm thickness) on TiAl was effective in improving the oxidation resistance by promoting a protective A1,O3 scale at 827°C (Tanig~ch~, 1994; Taniguchi et al., 1993). The activation energy value of 214kJ/mol far the parabolic stage is consistent with the valuc for tlie growth of an Al,0, her te~perature(- 1027 "C) the loss of
protectivity has been reported to occur by brea~away kinetics preceded by a short, transient parabolic stage. At high temperature, coating r~crystallization(leading to the formation of inicropores at scale/coating and coatin~/substrate interfaces) and the generation of Kirkendall voids accompanying CO diffusion into the substrate were responsible for the brea~-downof the protectivity. A su~p~ising observation is that the U addition did not improve scale adhesion. Sputtered coatings have also been used to improve the environ~ e n t a 1resistance of T ~ 3 A l - ~ ~~saetde r i a l sco~taining P-stabilizing elements, such as 11Nbat,%), added to enliance d ~ c t ~ l i tbut y lower oxidation resistance. The poor e~vironmenta~/oxi~ation resistance of TiAl materials inability to proiiiote protective from oxygen dissolution in th dissolution in Ti-oxides. The use of a single sputtered NiCrAlV layer has been observed to be not very effective in i ~ p r o v i n gthe
582 oxidation resistaiice at 850 and 950°C. The columnar crystal boundaries and other defects provided rapid e if fusion paths for the outward diffusion of Ti and inward diffusion of oxygen and Ni. The rapid at the scalc/substrate preventing the formation of a continuous A and allowing ~issolut~on of TiO, together eneration of voids accompanying the inward diffusion of Ni, u~dermincd the protectivity of tlie scale at teinperatures excceding 950 "C. imilar use of a duplex coating consisting of a eposited inner diffusion barrier layer (see de ter in Vol. 2) of W and Cr and a plasmaCrAlY (Fe-~4Cr-8Al-O.~Y) has been made to prevent environmental embrittlement of a2 orthorhombic, and ( a + fl) t i t a n i u ~alLimin~desubjected to thermal cycling up to 100Oli at 850*C. The prevention of interdiffusion by the iniier W layer, together with the ability of the outer coating to develop adherent Cr and A1 oxide scales, were responsible for limiting oxidation and oxygen-induccd eriibrittlement , .6
Ak
ee the ~ i s c ~ ~ sabove s i o ~i
petus for developing intermetallic composites stems orn the desire to overcome the inherent property deficiei~ciesof the single-phase bulk materials - inadee ~ow-tem~erature ductility temperature strength (creep). ites, the reinforcement/matrix interface plays a critical role in d e v e l o ~ ~ n high-performance g internietallic c o ~ ~ o s i t eand s needs to be designed to affect the matrix-to-reinforcement load transfer, to impart toughness, to provide high-teniperaturc strength, to provide protection against prolonged oxidation and to i ~ i n ~ m i zmechanical e and chemical incompatibility likely to occur at high processing and service temperat~re.In all of these roles, proper coating of the reinforcem~nt,can be important in the behaviour of the composite (see further below). Different classes o f composites use different typcs of reinforcements discontinuo~~ fibres, ~articulates,natural microstructures ( ~ ~ - , ~eutectic i I ~ composites), continuous fibres, and reitiforcements containing both fibres and part~culatedispersio~s.~liilosophy,design, develop-
ment and processing of compos~tes have been discussed more fully in Volume 2 and by Bewlay et al., in this volume. It i s important to emphasize that intermetallic composites are at their initial stages of development. Of the various factors which need to be considered, the dominant issues surround the bond strength, chemical stability and CTE inisniatch at the reinforcementlmatrix interface. A strong interface is required for off-axis loading, a weak iiiterfxe i s necessary to evoke addi~~onalener~y-absorbing mechanisms for iniproved toughness. The lack of chemical compatibility at the rcinforc~~entllnatrix interface promo tes adverse interfacial reactions leading to brittle phase formation. The mismatch in CTE of the reinforcement and matrix materials leads to the generation of the residual stresses and cracking at, or near, the interface. The other important issue i s the environmental protection of both tlie reinforcement and fibres. The introduction of reinforcement coatings andlor complaint layers at the reinforceinent/matrix interface eliininatcs or greatly reduces these problems and facilitates the design of optiinized interfaces. The d i ~ c u l t ylies in the selection of the coatings. As in other areas of materials developnieiit the approach is to use theoretical analysis, e~perime~ital determination and modelling. For the prediction of the rei~forceincn t/rnatrix chemical interactions, a key issue 111 selecting the reinforcement coatings, recourse is made to both t~ermodynarnicand kinetic data. In many cases the lack of informatio~ion ~hermo~ynamic data, phase relations and diffusion coefiicients hinders this approach. However, great efTorts are being made to generate and use appropriate thermodyiiamic data in predicting chemical compatibility at the interface. In most cases the interEacia1 chemical interactions between the reinfor~ementsand fibres are diffusioncontrolled, showing the expected linear relationship between the thickness of reaction zone and time. However, it also appears that the interfacial interactions are much more complex in ordered alloys and occur more slowly than in disordered alloys as indicated by the higher activation energy. Act~vation energies for a series of SCS-61 fibre-reinforced Ti-A1 intermetallics were found to range from 200-300 kJ/ mol (Yang and Jeng, 1989). It has been reported (Yang rewer and Unnam, 1983) that a TiAl coating on SCS-6 fibre in Ti-Al-based intermetallics reduces the transport of Ti towards the SCS-6 fibres. This reduced diffusivity, together with the lower 'Trade name - Textroii
583
Cou:ting Techriu logy
verification, together with ~ ~ i d e p e n ~ e nx t~ e r ~ ~ e n t a tion are required to refine the data given in Table 2. Similar data have been used in d e s ~ ~ ~ ~ n composites with refractory metal and i ~ ~ ” t eturc m ~~ e ~r ~stren~th ~ e ~ ctory metal fibres clearly need otectioii a ~ a i n h~ it ~ h - t e ~ ~ e r a the brittle ceramic fibres may
generally found in ordered alloys, lowers e chemical reactions at the interface. Thermody~iamiccalculation has been used to adopt a novel approach to desi ning interfttces in composites by showing how p r e - o x i ~ i ~ ~the n g reinforcing phase
for iinprovcd toughnes have been used include:
ac=eAGIRTallows calcu~ationof the activity of carbon. impart improved toughness t delamination and fibre pull-out.
An alternative approach is to form a layer of the surface of the reinforcement by externall 0. Fickian calculations can be used to ascertain the thicknes$ o f the surface layer. The coatings produced by both methods have been found to be very effective at higher temperature (>> 1200 “Cj. ertheless this a~proach, based on theoretical calculations, is conceptionally so~xnd. Such therniodynainic calculations tog esperinientall~ determined kinetic data Abb~schian,1992) have allowed determination of the tentative compatibility data shown in Table 2 (Vasudevan and Petrovic, 1992). The data in Table 2 need to be considered in conju~ctionwith information on CTE to provide a cornplete picture (Xiao and Abbascliiaii, 1992). ~ a l c ~ l a t i o n sbacked . up by experime~tal
isms, In selecting the coating systems, use can be made of numerical analysis as shown below. Such ail analysis, using the ~eneral f i n i t ~ - e l e ~ e npackage t ABAQUS, can be applied to a two-d~mensioiial model of a fibre/coating/~natrix system allawiiig d e t e ~ ~ i n a t i oof n the e~ectivenessof the c o a t ~ n ~ins various energy-absor~ingniodes during crack propagation (Griffin et uE., 2000j. This analysis was applied to two composites: 1. Fe& matrix reinforced with Al@, fibres coated with Ti,Al.
Fibre coatings for various reinforceme~tsin oSi, and other matrices (Shah et al., 1990; ~ a s ~ and ~ ~~~ t~r o~v i ~c ~ a, n 1992). Symbols shown are defined as follows: C = compatible; R = reactive; WR = weakly reactive
Matrix
MoSi, GoSi, Cr,Si Ti,Si3 NiAl Ni~Al TiAl Ti,Al SIC A1,Ta Nb,AI Nb,Al Co,Nb Cr,Nb Fe,Nb (Nb .Ti), AI (Nb,Ti)Al
C o ~ ~ a t i b i l i with t y the following rel~rorceme~t i~ater~als
C C C R CjWR R R R C
R R R R R R R C
c c C
c C C
R
c
CjWR
C C CjWR CjWR
R? R? R? C C/WR CjWR CjWR CjWR CjWR CjW R CjWR CjWR
c C
C C C C CjWR R
R R C R R R
C C C
R
c
R R R
C
R
C C C C CjWR R CjWR R
c 9 7
7
C
App licu t ions
584
e,Al matrix reinforced with A120, fibres coated with TiAl.
In tlie analysi~a typical cell was chosen of five fibres situated in a r e c t a n ~ ~ ~ matrix lar containing a fibre volume fraction of 30%. A seed crack of 0 . 2 7 ~ length ~1 was placed in the matnx away from the fibre~coating interface and the composite strai~ed.The relevant fibre dimensions arc the fibre radius (including coating) rf =6pm giving a matrix radius rm=9.7pm, leaving a matrix-to-fibre distance of d= r,v = r 3.7 pm. Taking the tensile properties of the fibre/coating and coating/ matrix interfxes as half the tensile strength of the weaker material; i.e., for the coating/matrix interface, the tensile strength ) of the interface as that of tlie matrix, TS =200 M
f,
ith a Ti3AI coating the crack propagates through the matrix, reaches the matrix/coating interface aiid causes d e l a ~ i n ~ ~ tat i othis n interface preventing catastrophic fracture. 2. With a TiAl coating the crack propagates in the matrix and continues to propagate through the fibre and coating ~ a u s i n gcatastrophic fracture, although a small amount of delaminatioii occurs. Titanium aluniinide~b~sedcomposites are distin&-based materials in two aspects:
(i) Their low capability at high temperatures requiring low processing temperatures. (ii)Their high reactivity. The scheme to reinforce titanium aluminide-based 1, ~ i ~ ~ and b ATiAl 1 - (Table 2) with SCS fibres s from problems due to interfacial chemical interactions and CTE inismatch (Smith et al., 1990; odes, 1992; ~kowroneket ul., 1988). The solution lies in the use of a diRusion barrier or mterlayer. The selection o f such coatings needs to be based on data relevant to the formation and growth of the reaction zone(s). 0th thermodynamic calculation determined kinetic data indicate de-based niaterials lead to a Tiatrix (Smith et al., 1990; Brindley, nterfacial reaction becomes dominated by iiiterdi~usioli:Ti (fm. matrix)-+fibre and that of Si and C (fm. ~ b r e ) ~ n i a t r iTi-A1 x . quickly becomes saturated with C allowing TIC precipitation near the fibre. Faster h S i c allows Ti silicide to be formed sion through the Tic, reacting with di~usingC, leads to the growth of TIC. Ti aluminide matrix initially reacting witli T i c causes Al-enrichment in the matrix allowing the formatio~of AlTi3C.
As in other intermetalliG composites, matcliing of fibre/matrix CTEs IS necessary to avoid residual stresses at the fibrelmatrix interface. TiAl/SCS composites have been observed to enerate significant residunl stresses due to the lack o CTE compatibility (CTE for matrix 4.9 x iO--G/"C, CTE for fibre 10.9 x 10-6/"C). Te~~sile stresses are generated in axial and hoop directions; the compressive stress iii the radial direction secures t faeilitating load transfer, ama age in the matrix w during fatigue cycling (Boss and Yang, 1990). The results from previous studies and calculations indicate therniodyn AI@, and TiB, monofilameiits and matching CTE in ~f-TiAl-basedcomposites. ased on this information several fibre coatings for 47 A1-2Ta (at.%) have been examined (Boss and Yang, 1990). An inte~estingapproach to selection of diRusion bar~erslinterlayersfor the iC/TiAl composites has been based on the objectives: 1. To minimize the interaction between the carbon coating on the surface of the S i c and the TiA1. 2. To prevent diRi,ision of Ti, A1 and Si ( aE., 1990).
Noble metals Pt/Pd were identifiedas caiididate coating materials. Thermodynainic data indicate chemical stability of PtjPd in relation to carbon, and the high melting points of tliese metals imply their ability to reduce dieusion. These observationswere used to study a number of coated composites. T i - 4 8 ~ 1 - 1powder ~ with Pd foil between the matrix and Sic fibre at the consolidation temperature still allowed Ti and A1 difhsion through Pd towards Sic producing extensive interaction and generating fibre reaction layers. The layers nearest to the fibre were characterized by the presence of silicide and siliconcontaining compounds while the layers nearest to the matrix, by the Al-Pd and Pd-contain in^ compounds. The reaction layers from thc fibre to the matrix formed in sequence: Fibre/TiAl-Pd silicide/TiAl~Pdcarbide (containing Si)/a,TiAl with Pd/A1, d,/Pd-containing y-TiAl/ Matrix (Norman et a!., 1990). Because of the enormous problems that occur in ~ i C / T idue ~ l to the lack of chemical compatibility and CTE match at the fibrelmatrix interfxe, A1,0, fibres are considered to bc a better reinforcei~entfor TiA1. A report from Das and ~ r i s ~ ~ a i ~ u(1r 992) t h y on A1203 (mono~1anient)-reinforced Ti-48A1-3V, Ti-48AI- 1 Ta, 'Trade name - DERA
585 -0.3Ta consolidated by "G indicates the absence reaction at the interface, although TEN1 examination us reaction zone at ty in the form of the matrix nnicrostructu~e,single phase ? + et2, being prone to d e v e l o ~ i ~ g interface instability. In the sapphire/TiAl system, coatings have been used to promote interface deboiiding and sliding with a low shear resistance relative to fibre strengths. Small z/o gives larger fibre pull-out length to increase the work of fracture. uitably designed fibre coatings, using as a basis de IT interfacial tougl~ness $ reinfor~menttoughness, allow fibre debonding and sliding. The use of double-layer fibre coatings - the first layer to achieve debonding and the second layer to prevent difksion reactions - in sapphirely-TiAl demonstrates the construction of so-called 'designer coatings'. Previous work ackin and Uang, 1992) has considered three types of inner coatings - carbon black (deposited by an acetone flame), colloidal graphite (slurry dipping) and a mixture of graphite with lO%Al,O, (sol process) and an outer (sol process). The results from fibre push-out demonstrate debonding aiid sliding for each coating type and the eEciency of the outer coating as a digusion bariier.
s
I n t e r ~ e ~ a l l iand c Ceramic ~ o a ~ ~ Dahotra, g s , N. Sudarshan, T. S., eds., Marcel Dekker, (1999). O x ~ ~ a t i oand n Corrosion of ~ ~ ~ e r m e tAlloys, a ~ l ~ cWelsch, 6. and Desai, P. D., eds., Purdue U. Oftice of Publ. (1996). Growth Kinetics of C h ~ ~ ~C~oc~ na ~l o u nLayers, d Dybkov, V I., Cambridge h t l . Sci. bubl.
Three recent conferences also contain nunierous papers on the preparation and properties of coatin for IMCs and on their oxidation behaviour. $ ~ r ~ ~ c t ~I nr at l~ r ~ ~ t ~Darolia. l l i ~ , ~R., , Lewandowsk~,J. J., Liu, @. T., Mutin, P L., Miracle, D. B., aiid Nathal, M. V., eds., TMS (1993). S ~ ~ u ~ I~n ~u err ~ae~t a l l ~1997, c s ~ ~ t iM.~ V., ~ lDaroliia, , R., Liu, C. T., Martin, P. L., Miracle, D. B., Wagner, R. and Yarnaguchi, M., eds., TMS (1997). 5th ~nternatio~al C o n ~ e r e ~onc ~S t r ~ ~ c t u rand ~l F ~ ~ ~ ~ c t i ~ n a l I n ~ e r ~ ~ ~ f f Vancouver, ~ljcs, Canada, ( ~ 0 0 ~ )(abstracts . available on the Tnteriiet at www.tnis.org/meetiagY/ speciality/ICSF~-2000).
s Aiken, R. M. Jr. (1992). Scr Metall. JMatc2r.. Baiiovic, S. W., DuPont, J. N. and Proc. 11th Annual Con$ on Fossil Energy ~ a t e r i u l s , ORNL/FMP-97/ I, Oak Ridge National Laboratory. p. 279. and Gage, P R. (1965). J.
The reader's ~~ttention is also directed to the following.
, I) E., and Felten, E. J. (1965). Trans. T M S - ~ ~ 233, M ~1093. , Ber~owitz-Mattuck,J., Rossetti, M., and Lee, D. W. (1970). Met. Tram., 1, 479. Berztiss, D. A., Cherchiara, R. R., Gulbransen, E. A., Pettit, , and Meier, C. €3. (1992). at er SCL and Eng., , 165. d, J. R., Buckley, R. A., .Tones, H., Sellers, C. M., McCartney, D. C., and Horlock, A. J. (1998). J . er Sci.,33, 4417. Bordeau, R. G. (1987). Development of Iron Aluiiiinides. AFWAL-TR-87-4009, Wright Patterson Air Force Base, OH. Bose, S. (1992). Nater. Sci. Eng., Boss, D. E., Uang, J. M. (1990 Composrtes, eds., Anton, D., Martin, P. L., Miracle, D,, McMeekiiig, R. MRS Syrnp. Proceedings vol. 436. Bowman, K. 9. (1 992). Refractory iiietal disilicide research, ed. Desai P D., NIAC report 2, West Lafayette, IN, p. 9. mialek, J. L., Locci, 1. E.
O.widution oJ' ~ i ~ e r m ~ t ~ ~Schiiltze, l l i c , ~ ,NE. and Grabke, H. J., eds., Wiley, 366 pp., (1998).
Brady, M. P., Smialek, J. L., and Brindley, W. J. (1996b). Sub~ittedto U S . patent office.
It is clear that further research and development will be required to achieve better coatings of and for IMGs and to better understand their behaviours. Particular developments are anticipated in three areas: processes (tailored microstructures by growth from seeded melts, utilization of naaao-particles,powder injection molding and plasma spraying); non-destructive evaluation (nano-indentatio~to study mechanical behaviour of thin films and quantified TEM to monitor processes and properties); and computer sirnulations (to study the behaviour of both coating, substrate and the combination thereof).
A ~ p lt ions ~ c ~
584
Fish, J. and Du~L~ette, D. (1993). Proc. 3rd fnt. S y ~ p ~ ~ s on ium High Teinyernture Corrosion m d Protection qf Nuterials,
Behaviour of ~ ~ Matrix ~ Composites, a l eds. Hack, J. E. and Arrateau M. F., TMS, ~ a r r e n d a ~PA, e , p. 39-50. High Te~perature Ordered I, eds. StoloE, N., Kocli, C. C., , MRS, Pittsburgh, 419-424. rindley, W. J,, ~ ~ ~3 . L., ~ and ~ Cedwi1~, e k , M. A. (1992). ~ ~ T Review E ~ - 1992, P NASA CP-10104, vol. I 4s. aebs, V. A. (1990). J. Vac. Sci.
urnell-Gray, J. S. (1998). An assessment of the oxidation resistance of an iridium and an iridiuin/platinum low activity aIuminide/MAR~M002 system at 1100 "C. Conf Proc. 5th Kn~ern~tional Conference on Advances in Surface E~gineering,SBo Paulo, Brazil, 1113 November. H., and Schlichting, J. (1973). Werkst.
Fitzer, E. (I 956). Plansee Proc oJ 2nd Seminar, Reutte/Tyrol, Pergamon Press, p. 56. Furukawa, H. (1991). 86th Japan Society qf Civil E?i~~neer.s, Preprint 54. y, J. S., Fisher, G., atid Datta, P. K. Goodwin, G. M. (1991). Proc. 11th Annual Conf on Fossil (1 997). The hot corrosion resistance of p ~ ~ t i n u ~ - r h o d i u i ~ ~~ n e r~~a "~ ~e r ~ aORNL/FMP-97/ ls, 1, Oak Ridge National Laboratory, p. 217. modified diEusion coatings on directionally solidified 002 superalloy at 900°C. Conf Proc 3rd Int Goward, G. W. and Boone, D. H. (1971). Oxid. Met., 3,475Conf Righ Temperature Mimrials, Edinburgh, 23-25 95. September. Griffin, D., Daadbin, A., and Datta, P. K. (2000). Surface and dett, C. (1987). J. Mater. Sci., 2 Coating T ~ ~ l ~ n o l o126, g y , 142. Rapp, R. A. (1995a). Mater. nd Pelletier, J. M. (1998). Mater. Scr. und Eng., pp, R. A. (1995b). Met. Tram., 2 6 ~ ,
777 c:oc1teran1,
B.
v.,
and Rapp, R. A. (1995).
. A. (1996). Oxid. Met., 4 (1986). Developinent of AFWAL-TR-$~~4155, Wright Patterson Iron Al~~~ninides, Air Force Base, OH. Danielewski, M. and Filipek, R. (1996). J C o i p . C h m . , 17, 1497. anielewski, NI., Datta, . and Filipek, R., Bachorczyk, R. (2000). .Journal of C U ~ U L i q ~ d d86, ~ , 1-3. anielewski, M., Filipek, R., Holly, M.,and Bozek, B. (1994).
S. (1992). Ppoc. Materials
Holly, K. and D a n i ~ ~ e w skM, ~ , (1994). Phys. Rev. B, 13336. Hiifner, S. (1996). Photoelecfron Spectroscopy: Principles and ~ p p l i c a ~ ~ o2nd n s , edn., Sprin~er-Verlag,144. Klansky, J. L., Nic, J. P., and Mikkola, D. E. (1994). J , Mater. Res., 9, 255. Krisbiian, P. and Kaufnnann, M. J. (1994). ~ e ~ a l l u r ~ zand cal Lee, Li-Lien, Laughlin, D. E., and Lambeth, D. N. (1995). l.E.E,E. Tram. Mogn., 31 . A. (1996). S c r i ~ t a
. R. (1996). Oxid. Metals, 45,267. Liu, Z.? Gao, W., and Wang, F. (1998). Scripta Mater., 39, 1497. Ma, Q. and Clarke, D. R. (1994). J. Am. Cerani. Soc., 77 (2), 298-302. ai, T., and Nakayama, U. (1989). Scripta
in Science and Technology,
ionne, S. and Lo, J. (1996). In Proce.s~si?~g and Fa~ricationof Adwnced Materials Jf, cds,, Srivatsan, T. S. and Moore, J. J., The Metallur~icalSociety, p. 469.
Mackiii, T. and Yang, J. (1992). lnter~etullic Matrix Composites I I , MRS Syni~osiumProceedings 273, eds. Miracle, D. B., Anton, D. L., and Graves, J. A., p.343349. , M. J. and Weclit, R. T. (1992). Mater. Scr. Eng.? 19-31. Maziasz, P. J., Liu, C. T., and Goodwin, C. M. (1995). Heat Ressstant M ~ t ~ r i aII, l s eds. Natesan, K., Ganesan, P., and Lai, G., ASM International, Materials Park, OH, p. 555. McCarron, R. L., et al. (1992). ~ i ~ a ~ ieds. u ~ Froes, z, F. H. and Caplan, I, TMS, Warrendale, PA, p. 1971. McKamey, C. G., 5 e Van, J. H., V K. (1991). J Mater Res.,
Coating Technology
587
Sands, T., Palrnstrom, C. J., Harbison, J. P. (1990). at er. McKee, D. W and Luthra, K. L. (1993). Surface and Scr. Rep., 5, 99 Coatings Teclinolog,y, 56, 109. SauthoR, 6. (1986). Z. et all^^., 77, 554-566. Meier, C. and Pettit, E;. (1992a). Muler. Sci. and Techizology, Scofield, J. H. (1978). J. ~ l e c ~ r Spect. o n and Rel. Phen,, , 331. Shah, D. M., Anton, D. L., and Musson, C. W. Meier, G. and Pettit, F. (1992’0). Mater Sci. Eng., A153, 548. Intermetallic Matrix Composites, MRS Proceedings, Vol. Meschtcr, I’ J. (1992). Metall. Trans. A, , eds. Anton, D.? Martin, P., Miracle, D., and Munroe, T. C. and Gleeson, B. (1996). Met. Trans., ~ ~ A , McMeeking, R., p. 333-340. 3761. Shiiiiizu, T., Iikubo, T., and lsobe, S. (1992). Mat. Set. and Natesan, K. and Cho, VV. D. (1994). In Proc. 6th Annual Con6 Eng., AL53, 602. Fossil Energy Materzds, O~NL/FMP-94/I1p. 227 Shirley, D. A. (1972). Phys. Rev., Natesan, K. and Johnson, R. N. (1995a). Proc. 2nd Int. Con$ Siklta, V K., McKaniey, C. G.. Howell, C. R., and Baldwin, on Heat ~ e ~ ~ i s t ~aantte r i a l ~Gatlinburg, s, TN, September R. H. (1990). Fabrication and inechanical properties of 11-14, 1995, ASM International, Materials Park, OH, Fe,Al-based alurniiiides, ORNL/TM- 1 1465, Oak Ridge Nalional Laboratory, Oak Ridge, TN. and Tortorelli, P. F (1997). Proc. Int. Symp. on Skowronek, C. J,, Pattiiaik, A., and Everett, R. K. (1988). ncl Iron Alumintdes: Processing. Properties arid Naval Research Laboratory technical memorandum 1038, Application,i;, Cinciniiati, October 7-9 1996, ASM November 1988. International, Materials Park, OH, p. 265. Smialek, J. L. (1993). Corrosion Scze~zce,35 (5-8), 1199. Natesan, IS. (1993a). In Proc. 7th Annual Con6 Fossil Energy Smialek, J. L. and Humplirey, D. L. (1992). Scripfu. Mefall. Materials, ORNL/FMP-93/1, p. 249. at er., 26, 1763. Natesan, K., Klug, K., Renusch, D., Veal, B. W., and Brindley, P. K. (1990). Criimditch, M. (1996). Micro‘~tructura2aid ~ e c h a n ~ c a l Smialek, J. L., Gedwill, M. Scripta. Mct. et Mater., characterizutioz of alzmina scales thermally developed on Smith, P. R., Rhodes, C. G., los, (1990). ~ n t e r ~ a czrz es iron a l u ~ z n z ~alloys, e Argonne National Laboratory Report ANL/FE-96/0 I. ~ ~ t a l - ~ e r a Composites, mIc ed. Lim, J. et al., T M ~ ~ A I M E , Warreiidale, PA, 35-38. Natesan, K. (1993). Mater. High Temp., Natesan, It., Richier, C., Veal, B. W., Grimsditch, M., Streiff, R. (1993). Journal de Ph!?siqw f V Collogue C9, Renusch, D., and Paulikas, A. P. (1995)- C h e ~ i c a land StreiE, R. and Poize, S. (1996). Met. ~ a ~ eTrans. r . A, m ~ c ~ o s t r u c t ~characteri~a~ion r~~l of t h e r ~ a l l y grown 3761. cllzivrzina scules, Argonne National Laboratory Report Streiff, R. and Poize, S. (1983). High Tempercrture Corro,i;ion, ANL/FE-95/02. ed. Rapp, R. A., NACE: Houston, TX, p. 591. Norman, J. H., Reynolds, G. H., and Brewer, L. (1990). Iii Takei, A. and Ishida, A. (1990). Rep. 123rd C o ~ ~ i t ~ e e In~ermetull~c Matrix Composites, eds. Aiiton, ID., Ma Jirpanese Societ~for the Promotion of Science, 31, p. 327. P. L., Miracle, I).B., McMceking, MRS, Proceedings Takei, A. and Ishida, A. (1992). Zn High Te?ri~~erat~re p. 369-377. Corrosion of A~vanced Ma~erials and Coating~s, eds. Perkins, R. A. aiid Meier, C. H. (1991). Proceedirzgs qf the Saito, Y., Onay, B., and ~ a r u y a n i aT., , 317-324, Elsevier. I i z d ~ ~ ~ ~ t r ~ ~ - l i i ~Advanced i v e r , ~ ~ tM y a ~ e r i ~Conference, ~I~ ed. Taniguchi, S. (1994). M R S Bulletin, p. 3 I , October. Smith, F. W., Denver CO, 92. Taniguchi, S. and Shibata, T. (1992). Is1 High ~ e ~ i p e r ~ z f ~ r e Perkiris, R. A. aiid Meier, G. H. (1989). Procee~ingsof the ~ o r r o s o~ f oAdvcl~ced ~~ ~ ~ t e r z aand l s Co~tings,eds. Saito, hdustry-lirziversity Advanced ~ c l ~ e r i aConference l~s I I , ed. Y., Onay, B., and Maruyania, T., 309-31 5 , Elsevier. Smith, I; (Golden, CO: Advanced Materials Institute), Taiiiguchi, S., Shibata, T., Asanurn, N., LOU,H., and Wu, W. p. 92. (1993). Oxid. Mef., 39, 457-473. Regan, R. E., Baginski, W A., and Krier, C. A. (1967). Tortorelli, P. F and De Van J. H. (1992). ~ a ~ Sci. e r Eng., Ceram. Bull., 46, 502-509. Renusch, D., Veal, B. W., Koshelev, I., Natesan, K., Van Loo, F J. J. and Rieck, 6. D. (1973). Acta Mefall., Grii~sditch, M., and Hou, P. U (1996). Stress 61-71, 73--84. d e t e ~ i n a ~ in i ~thermally n grown alumina scales using Vasudevan, A. K. and Petrovic, J. J. (1992). Mater Sri. Eng., ruby luminescence. In Proc. A155, 1-17. Society, San Francisco, April Wawner, F. E. (1987). Final report, USAF contract F33615Rhodes, C. G., Brampton, C. C., 86-C-5151. Intcrmetallic Matrix Composites eds. Anton, D., Martin, Weisch, G. and Kaveci, A. 1. (1989). Oxidation of High P., Miraclc, D., and McMeeksng, R. MRS Proceedings T e ~ ~ e r ~ t u~r ze ~ e r ~ e ~ eds. a ~ ~ Grobstem, ~cs, T. and Doycbak, J., TMS, Warrelidale PA, 207. Rhodes, C. C. (1992). l n Inter~etallic,sMatrix Composites, I1 Welsch, G., Freidman, S., and Kahveci, A. (1991). , J., and Anton, D. MKS Microscopjj of Oxidutiorz, eds. Beiinett, M. J. and Lorimer, G. W., Institute o f Materials, London, 193.
588
Applicn t ions
Wirkus, C. D. and Wilder, D. R. (1966). J . Am. Ceram. Soc., , 173-177 Xiang, Z , D., Rurnell-Gray, J. S., and Datta, P I(.(2OOOa). ~ o r ~ a t ~o fo almninide n coatings by pack cementation process. Con$ Proc. E~roCorr.2000, London September. Xiang, Z. D., Datta, P. K., and Burnell-Gray, J. S. (2000b). Powder forinulation design for the co-deposition of At and Gr and o f A1 and Si on metal substrates by the pack cenientation process, S u r f ~ ~c ~~ g i ~ ~ e17(4), r i n g287-294.
Xiao, L. and Abbaschian, R. (1992). Mater. Set. Eng., 135-1 45. Yang, J. M. and Jeng, S. M. (1989). Scriptu Metall., 23 (9j, 1559. Yoshihara, NI., Suzuki, T., and Tanaka, R. (1991). ISIJ I ? z t e r n a t i ~ ~31~ l(1, 0), 120I. Zheng, M., Hc, Y., and Rapp, R. A. (1997). 11th Annual Car$ on Fmsil Energy ~ ~ ~ ~O R ~ ~ ~~ I i~ M # PZ- 9,~7Oak 1~1 , Ridge National Laboratory, p. 33 1.
the ~ l a ~ k s ~ati tt h the skin of a croc the roe of a fish. have not seen a ~ ~ a c ~ son ~ ia t h c o i ~ ~ i s s i oan ,foun
vinced of these trut e role of never pr nown to the u n ~ n o w n ,of ot derived directly fr n, and o f a r r a ~ g i nthe ~ facts the order most a ing of them. by b
is too w o ~ d e r f ~ tol e true, if it be ~ o ~ § ~ s t e n t laws of nature, an in such things as these, ~ x p e r i ~ise the ~ t best test of such consistency.
Significant progress has been achieved iii creating and characterizing intermetallic alloys that offer revolutionary enhanceinents in performance for many structural applications (Westbrook, 1993; Williams, 1997). However, the success in transitioiiiiig these opportunities into commerc~allyavailable products has been limited by both technical and non~~echnicaliinpedinients. Many of the non-technical impediments have been discussed by Williains (1997) and Dimiduk (1999). From a technical perspective, many of the unique properties of intermetallic alloys that make them attractive for hi~h-te~nperature structura~applications, such as high melt in^ and disordering temperatures, high stiffness, low diffusivity, etc., also make these alloys a challenge to process into useful products. In particular, consider the intermet~lliccoinpound alloys of TiAl (Ll", tP4), Fe,Al (DO3,cF16), FeAI (B2, cP2), Ni3Al (L12, cP4) and Ni cP2). all of which have relatively high melting atures, aiid retain their ordered structures to or nearly to their respective ost of these alloys have relatively high ductile-~o-brittletransitioii temperatures, often making processing, machining and handling difficult. This lower ductility, coupled with high stiffness and low-to-moderate strength, can lead to hot tearing and cracking if cooling rates during processing are not adequately controlled. Due to the fnct that these alloys coiitaiii high fractions of Al, and the TiAl alloys contain Ti, their
reactivity can pose challenges to traditional melting and casting technology, and also require greater care compared with traditional metallic alloys in other high-temperature processing. Additionally, the high fraction of A1 in these alloys leads to ~ifficultyin formulating and controlling alloy composition, due to the relatively low melting temperature of Al, and the volatility of A1 at elevated processing temperatures, especially in vacuum. In spite of these challenges, some success has been achieved in the casting of some i~ternietallicalloys utilizing existing or modified cast-process technology, when exercised with care. This chapter focuses primarily on three classes of cast aluminide intermetallic alloys that have received consi~erableR& that have acliieved at least limited commercial success: near-gamma titanium aluminide alloys; nickel aluminide alloys; and iron aluminide alloys.
ear-
Y
The development of casting technology for gamma TiAl alloys has been aided by the ability to use conventional Ti melting and casting technology, with only minor modifications. any barriers to the successful cast processing of gamma alloys have been largely overconie, such as component fill, cracking and hot tearing, aiid surhce-connected porosity (Ram and Barrett, 1996, McQuay et al., 1999). Soine significant remaining barriers include the cost of ingot, the ability
~ ~ ~ e ~ m eCt oa ?~ ~ ipc ~Vol. u ~3,~ Principies ~ . ~ ~ and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 0 2 0 0 2 John Wiley & Sons, Ltd.
to control d i ~ e n s i o n sand ~ the estab~ishinentof a vendor base. The vendor base issue extends beyond vendors, to include capable machining sources for gamma alloys.
Conventional titanium alloy casting processes are manufxture of gamma alloy ause of the reactivity of molten ic crucibles, cold-wall crucible melting IS required ng aerospace-grade ingots and castings. 11-diameter ingots, generally from 100-200 mm, are required for induction-skull melting (ISM), investment casting, and typically from 100-350 mm for vacuL~m-arc reme~ting investment casting. To take advanPage of ec of size, conventional titaiiirini casting ingots are generally converted froin large-diameter ingots ( 3 6 0 - ~ ~mm) 0 via the~~al-mechanical processing. These processes are currently not cost-eRective Tor gainnia ingots because of their poor workability, so sually melted and cast directly The three most Common cold-wall crucible melting a ~ ~ i a ~ amelting rc
), and ISM. The VAR
ingot size a i d shape, more control over thermal s t r ~ s ~ ewhich s, can cause c r a c ~ ~ nofg the ingot, and revert scrap easily. It is significant ss operates in an inert atmosphere, improves the conipositional control of volatile elements such as A1 and Mn (Dowson et furnaces are limited to melt sizes e to their high relative power , 1995). Therefore, ISM is usually used for sinall development heats, or where expensive ingot-heat-qualificatiori testing is not required. Ingotproduction processes have matured to where the typical alloy ingot cost is in the range from $35-60/ elly, 1995; McQuay and Larsen, 1997). Additional alloy ingot cost reductions should be realized with higher volunie production. Sli
VAR-pour is perhaps the most conimon iiivestiiient melt practice used for g a i ~ m aalloys, again due
to its ava~lability,and it remains the only practical technique for castings with pour wcights in excess of 60 kg. One li~nitationof consumable-electrode VAR casting is the limited control the caster has over the melt: When the target amount of alloy is determined to have been melted from the consumable electrode, the arc is turned OR, the electrode is withdrawn, and the crucible must be immediately poured. ISM is gaining in popularity in Ti alloy castings, i~cludinggamma TiAl alloys, because of its increased flexibility and higher productivity in casting compared with VAR-pour (Larseii and Govern, 1995). Recent
be taken in contro~~ingthe amounts of volatile 1999). Cold-wall crucible melting limits the available superheat of the melt, generally in the range of 1060 "C, with the amount of available superheat being somewhat a function of the crucible and coil design, and the frequency and available rength of the power supply. However, Yamada and ernukai (1998) have reported higlicr inelt superheat temperatures for gamma TiAl alloys melted in an ISM furnace with a crucible designed to maximize levitation. This limited superheat can restrict the ability to fill thin-wall sections ( < 2.5 mm) iii conventional gravity casting. Two pressure-assist processes have been developed which improve fill of thin sections: centrifLiga1 casting and counter-gravity casting. Centrifugal casting, as practiced for Ti and TiAl alloys, usually involves spinning one or more molds on a turntable. The molten alloy i s usually delivered from the crucible to the molds via a tundish or piping system, usually niade of steel. The tuiidish further cools the alloy being cast, and lowers the casting-alloy yield, as a skull of alloy is solidified on the tundish. Additionally, high scrap rates can be associated with c e n t r i f u ~casting ~l due to mold failure, and due to the highly turbulent nature of the alloy under centrifugal pressure, which can lead to gas porosity and cera~ic-shellinclusions. Another niethod of pressure-assisted investment casting that is gaining in popularity i s the Countergravity Low-pressure ~ n e r t - ~ t m o s ~ h (CLI) e r e investment casting process developed by the Hitchiner Corporation (Chandley, 1991). Other variants of the ~itchiiierprocess include CLV for Vacuum casting, and CLA for Air casting, All three of these processes utilizc a gas-pressure differential between the melt and mold chambers to drive the molten alloy up through a
Casting
fill tube and into the rnold cavity. The rate of rise of the metal is specified by a computer-controlled vacuum system, which allows for bottom feeding with a greater degree of control over metal velocity. A schematic of the CLI casting process is shown in Figure 1. Properly practiced, this technology can both improve fill of di~cult-to-castparts, and dramatically reduce defects. itchiner has developed a version of this technology for casting Ti and gamma TiAl alloys utilizing a ceramic crucible VIM system (~handleyand Flemings, 1994). ~ o r m a l l y ,melting Ti and TiAl alloys in a ceramic crucible is not possible due to the high level of reactivity of the molten alloy with the crucible, which leads to c o n t a ~ i n a tof~ ~ the~ melt. itchi in er has addressed this problem by melting scrap or elerneiital Ti, AI and master alloys in the VIM crucible, making use of a split induction coil and the exothermic heat of the Ti and A1 to reduce melt time, thus limiting the amount of oxygen pickup to around 1500 ppm. While this level of oxygen is nearly twice that allowed for most aerospace gannm alloy specifications, it may be acceptable for automotive or commercial ap Daido ~ r e ~ i s i oCasting, ii a licensee of the counter-gravity casting technology, has developed a version of the Hitchiner technology which couples counter-gravity casting with an induction skull-melt system, named Levi-Cast, to produce gamma TiAl alloy turbocharger wheels (Isobe and Noda, 1997; Noda, 1998). This system has the benefit of clean, controlled induction-skull melting, with the superior 611 control of counter-gravity casting.
uie 1
Sche~aticof the H i t c ~ i i i ~ CLI r casting process
593
While the majority of casting processes employed for gamrna TiAl have employed investment casting, several conipanies have developed novel versions of metallic-mold castin processes to produce simple c o m p o ~ e ~ t such s, a a u t o ~ o t i v eexhaust valves, at costs potentially lower than that which can be achieved by investment casting. The Howmet Corporation has developed a gr metal inold (~~~~ process which utilizes an system and a permanent metallic niold to form casting (Colvin, 1995). This process appears advantageous for simple c o ~ p o n e n tshapes in fairly high^ volume applications. Because this process eliminates many of the process steps related to creating and removing invested ceramic molds, it should lead to cost reductions of betwcen 15 4 0 % ) versus investment casting, especially for components which require sigiiificant machining, such as auton~otive exhaust valves. Additionally, because of the rapid solidification and cooling in the inetallic mold, finer i~icrostrL~ct~~res, and improved tensile properties versus investi~e~it cast valves have been reported (Jones et al., 1995). A Cernian government and i by ALL) ~ a c u Technologies u ~ a centrifugal permanent mold utilizes heated refractory metal molds to produce automotive valves ( lum et al., 1999). The goal of this project is to develop and scale-up a g a ~ Ti~ a valve-casting process with costs that are low enough to *
Processes artd Phenomena
594
Com~arisonof casting processes for gamma TiAl alloys ~~~-
Process
Advantages
Disadvantages
VAR Gravity Pour
Widely available Large casting capacity
ISR Gravity
Flexible melt control Direct alloy formulation Rapid cycle time improved fill Large casting capacity
VAR Centrifugal
ISR Ceiitrifugal
Improved fill Flexible melt control Direct alloy formulation
ISR Counter-~ravity
Improved fill Flexible melt control Direct alloy forinulation Rapid cycle time Improved fill Rapid cycle time Melt formulation Higher superheat Low cost Rapid cycle time Refined microstructure Improved strength Low cost Refined microstructure Improved strength Reduction in porosity
VIM Counter-~ravity
ISR Gravity Metal Mold
Centrifugd Permanent Mold
be attractive for mass-produced automobile applications. ~ e n t r i f u ~force a l is utilized to reduce porosity in the valves, in order to reduce the need for hot isostatic pressing (HIP) processing, and thus reduce cost. Although most of the macroporosity is eliminated froin the head, there is soiiie porosity reported in the center of the valve stem. lum et al. have also reported i m p r o v e ~ e n ~in s their valves relative to P’ed data found in the literature, although a reduction in ductility was reported. As can be expected, all of these melt~c~sting systems have advantages and limitations, and some are better suited than others for certain products and markets. A brief summary of the advantages and disadvantages of each of these systems is given in Table 1. Gt
rocess~
Huang and Chesnutt (1995) have presented a review of the solidi~cat~on and microstructural development of
Lirnited me1t cont r ol Slow throughput May require tundish Limited melt capacity
Limited melt control Slow throughput May require tundish Lower casting quality Higher potential inisrun Lower casting quality Higher potential for inisrun Lower casting quality Limited melt capacity Limited melt capacity
High interstitial content Limited control over melt chemistry
Limited to simple part configurations
Limited to simple geometry Higher tooling cost and life High furnace/tooling cycle time
near-gamma alloys. The amount of A1 in the alloy can have a dramatic effect on the grain structure, and subsequent niicrostructure and properties of castings. Generally, higher A1 promotes coarse dendritic cast structures and improved creep and oxidation resistance, while lower Al pronlotes finer grained equiaxed i~icrostrL~~tures and improved strength (McCullough et al., 1989).Recent studies have also demonstrate^ that the cooling rate following casting also has an effect on the macrostructure, niicrostructure~ texture and mechanical properties of the casting ~ ~ u r a l e e d ~ a r a n et crl., 1997; Rishell et nE,, 1999; De Graef et al., 1999). The successful casting o f gamma-alloy components requires careful attention to casting parameters, such as mold materials, mold preheat temperatures, and gating design (Larsen and Govern, 1995; Ram atid Barrett, 1996; McQuay et aE., 1999). In parti~ular,investment casting mold systems utilized in the casting of other structural alloys, such as titanium and nickel-based superalloys, may be too strong for the lower ductility
casting
and strength of gamma TiAl alloys, which can lead to component hot tearing and cracking. Hence, niodifications are often made to these mold systems to make them more suitable for gamma TiAl alloy casting. Mold preheat temperature is one of the most important, controllable, casting variables at the disposal of the caster. Higher preheat temperatures not only improve fill and feeding, but also reduce the thermal gradients and the cooling rate in a given casting. The reduction of thermal gradients and cooling rates can be critical in producing tear- and crack-free components. However, higher preheats can also lead to severe metal/mold reactions, and can increase the propensity for surface-connected porosity (Larsen and Govern, 1995; Rani and Barrett, 1996; Rishell et al., 1997). Slower cooling rates can also lead to coarser microstructures and inferior mechanical properties. Therefore, a balance must be found in order to achieve a component which meets the customer's design requirements. The successful adaptation of foundry practice to gamma alloys has also been aided by computational modeling, both for improved component fill and feeding, and for thermal stress modeling to reduce component cracking or hot tearing (Larsen, 1996). However, further work is required to improve the thermo-physical and thermo-mechanical databases for the casting alloys, in order to improve the model's predictive capabilities for fill and for thermal stresses during solidification and cooling of the casting.
595
In the as-cast form, most two-phase g a m ~ aalloys exhibit coarse, columnar, primary alpha colonies, with some amount of interdendritic gamma. The alloys also exhibit interdendritic-connected porosity and centerline o ~ s hot iso&dtic porosity; so iiearly all a ~ p l i c a ~ i require pressing (HIP) to ensure soundness, just as with conventional Ti-alloy castings. Cominon tures are between 1165 to 1260"C, at pre 125 to 175 MPa for up to 4 hours. The columnar grains generally form perpendicular to the mold surface, growing inward toward the center of the casting, leading to a cast texture (De Graef et al., 1999). In addition to closing porosity, HIP processing promotes the fomation of equiaxed gamma grains, and heat treatments can be used to further modify the microstructure. The most common microstructure for HIP+heat treated, two-phase, gamma alloys i s the cast-duplex microstructure, which consists of equiaxed gamma grains (typically 50-100 pm) and ?/a2 (typically 100-250 pm). A typical treated duplex 48-2-2 alloy microstructure is shown m Figure 2. Some alloys, such as t are heat treated near the alpha transus to produce a nearly lamellar microstructure (typically 100-400 pm) in order to maximize creep resistance (Lupinc et al., 1999). Figure 3 illustrates a heat-treated, nearly lamellar ABB-2 alloy microstructure. Another class of cast gamma alloys conpain boron as a grain refiner, and are known as the XDTMallays. The ~re addition of B transforms the as-cast i ~ i c r o s t r u c ~to
Figure 2 Macrograph and micrograph of the GE 48-2-2 alloy in the HIP and heat-treated duplex condition
Processes and P h e n o ~ i ~ n a
rograph and ~ ~ c ~ o ~ r a p h lloy in the HIP and heattreated, ncarly lamellar condition
onies and g a ~ ~ non a co~tainingalloys. A typical asmic tructure i s presented in Figure rain sizes for the alloy are ~ p p ~ o ~ i u m a t50-100 ~ l y pum, and alloy (Larsen, s r a ~ n ~ ~ e effect ~ n i ntends ~ to reduce the sensitivity of ~ i c ~ ~ ~ tand r ~properties ~ c t u ~to ~castThe alloys of prime interest are s u b ~ s ~ o ~ c l i i o i ~ e t ~ c ~ o ~ ~(e.g. t r section y size) c ~ ~ ~with a r egamma ~ Ti 1 alloys, with typical ~ o m ~ o s i t i ~ o ~ ~s ~ i n g
Macrograph and micrograph of the 47XDTMalloy in the HIP and heat-treate~duplex condition
597
Castiiig
alum~numin the range of 45-47 atomic %, and various additional alloyiizg elements. A list of common alloying elements is given in Table 2. A number of alloys have been developed over the last 10 years, some developed principally for wrought applications and later applied as cast alloys, and others which were developed later speci~callyfor casting. It is not practical to list all of the alloys undergoiiig development for cast applications, but Table 3 summarizes alloys which have received serious development and study and which are reported in, the literature. One of the first generation wrought alloys, the GE 48-2-2 alloy, was developed to maximize roomtemperature ductility and toughness. The Lockheed1 45XDTM and 47x11~~ alloys were developed to maximize strength, and to promote homogeneous cast microstructures through the use of titanium-boride as a grain refiner. Later, some alloys -Alstom ABB-2 and the Honeywell ~~S alloys) were developed to improve creep and ility. The Steel and IHI alloys were developed primarily as second-generation alloys for iiiiproved castability and a balaiice of properties. A more thorough summary of these and other alloys can be found in a review by Dimiduk et al. (1999), and in the references cited in Table 3. Common alloying elements for gamma TiAl alloys Common elemental additions Cr, Mn, Zr Cr, Zr, W, Si, C, €3, 0 W, Ta, MO, Si, C Nb, W, Ta, Si B
Effect ~ ~ p r o ductility v~d Improved tensile strength Improved creep strength Iinpraved oxidation resistance Grain size control
Not all alloys have equal casta~ility,The term castability is often assumed to refer principally to the fluidity and mold filling characteristics of the alloy. However, there are other attributes that can affect the caster's ability to produce a ~ o ~ ~ o that n e meets ~ t customer specifications and requirements. They can include the sensitivity of an alloy to cooling rate and section size, and the susceptibility of the alloy to form surface-connected porosity which is resistant to consolidation. Cast gamma loys which do not contain at least 0.5 atomic % , typical solidify in a coluinnar dendritic manner, wi shrinkage porosity in the interdendritic region. ishell et all, (1997) demonstrated that this interdendrit~cshrinkage (under cast conditions which do not fav casting surface, e.g. higher m can lead to a network of porosity opening up to the surface. Hence, even with HIP processing, shrinkage may persist. Additiona~ly, the m o ~ o l i t ~ i alloys c usually exhibit a casting texture, as has been noted by Naka et al. (1997) and De Graef et al. (1999). Conversely, alloys which contain in excess of 0.5 atomic % B solidify with more of' an equiaxed structure, and do not e x ~ i ~ castin it interdendritic shrinkage under normal casting coiiditions (Larsen et al., 199 cont~iningalloys with which can be optimized (Larsen and Govern, Additionally, the boron acts to refine the grain size, thus iniproving tensile properties, arid reducing microstructural sensitivity to section thickness and processing variables. , whether added as T The additions elemental B, or via ntaining master alloys, re in the fo~mation var~ety of meta~lic borides, (Wyman et al., 1989). In
able 3 Cast gaiima engineering alloys and their attributes
Alloy name
~ o i n p o s i t i o(at. ~ %)
GE 48-2-2 L-M 45SDTM
Ti-47A1-2Nb-2Cr Ti-45A1-2~-2~n-0.8~01% TiB,
L-M 47XDTM
T~-47Al-2~b-2Mn-O,8vol~~ TiB,
Honeywell WMS AB~-~lstom ABB-2 , CKSS TAB Daido Steel IHI
Ti-47A1-2Mb-1Mn-0.5W-O.SMo-0.2Si Ti-47Al-2~-O.~Si Ti-47Al-1.5Nb-l~n-1Cr-0.2Si-0.5~ Ti-48Al-ZN b-0.7Cr-0.3Si
Ti-45A1-1.3Fe-l.lV-02.5B
Attributes Ductility, fracture toughness Tensile & €atigue strength, castab~lit~ Elevated teniperature strength, castability Creep resistance Creep & oxidation res~stance Castability, property balance Ductility Castability
ef.
Austin et al., 1997 Larsen and Govern, 1995 Larsen and Govern, 1995 Seo et al., 1995 Lupnic et al., 1999 W a ~ n e ret al., 1995 Noda, 1998 Nishikion et al., 1999
598
Processes and Phenomena
refractory metal-containing alloys, the B may also form refractory borides, which may reduce the grainrefining effect at a given 1E3 level (Cheng, 1999). Although the grain-refining benefits of B additions have been adequately demonstrated (Larsen et al., 1990: Huang and Hall, 1991; De Graef et al., 1992; Cheng, 1999), there remains some controversy regarding the mechanism by which the refinement occurs. Recently, Gheng (1999) reviewed the earlier proposed mechanisms and put forth a new proposed mechanism.
The successful alloy conversion from nickel-superalloys to gainma TiAl alloys has been aided by the development of casting~designguidelines that incorporate into the component design the unique physical and mechanical properties of the alloys. In addition to the mechanical- and physical-property requirements of the Component, the component design must be 4castable” Some of the key geometrical considerations in superalloy~to-gamn~a conversions are fillet radii, minimum section thickness, and (in components such as airfoils) taper from the airfoil root to the tip or shroud (Ram and Barrett, 1996; McQuay et al., 1999). Additionally, increasing the thickness of the component, which is later removed via chemical milling, may be required for fill and component integrity. In this respect, the ~ a m m aalloys behave more like titanium than nickel alloys. While direct alloy substitutions are usually only successful 011 the simplest of components, a cQncurrent design approach whicb balances weight, cost and performance has proven very successful, even for diffi~~~lt-to-cas~, low-pressure-turbine blades in excess of 3 0 0 ~ m in length. An example of a successful superalloy”to-gamma alloy conversion is the PW4084 Low-Pressure-Turbine
Finished 47XDTMalloy MTU PW4084 LPT 9-stage blade
(LPT) stage-9 blade developed concurrently by MTU and Wowniet for an engine-test program that is shown in Figure 5 in the finished condition, The production PW4084 9th-stage superalloy blade i s an equiaxed, cored (hollow) casting. MTU, utilizing input from Howmet, redesigned the airfoil as a solid airfoil, with slightly increased fillet radii, airfoil taper, etc. Approximately OSmm of chemical milling stock was also added to each surface of the blade to assist fill. This concurrent design ap~roachhas led to a casting process which is approaching an acceptable yield through final X-ray and fluorescent-penetrant inspection. Reportedly, GE and PCC, following a similar component redesign a p ~ r o a ~ have h , achieved similar results on a CE90 LPT blade (Austin et al., 1997). ~imensional capabilities of investment-cast, gamma-alloy components are similar to capabilities for other investment-cast products, However, due to the limited room-temperature ductility of gamma alloys, cold straightening of components to achieve dimensions is not practical. To solve this problem, Howmet and PCC have developed non“~traightening, and ~ot-straighten~ng techniques that meet customer dimensional requirements (Ram and Barrett, 1996; McQuay et al., 1999). Fundamental developments and the use of realistic numerical models for casting are anticipated to have a growing impact on the diversity and cost of specific gamma components. Progress has been made for several cast gamma alloys in generating the thermophysical data required for numerical heat transfer, solidification and stress modeling. This modeling is becoming a practical tool fop. gating and mold design, and has led to the successful casting of several difficult” to-cast components. 2.7 ~ e c ~ n o l o ~ y ~~~
The selection of a revolutionary emerging material, like gamma TiAl alloys, entails considerable cost and risk, regardless of the niagnitude of the potential benefit of its selection over a mature material technology. In non-military aerospace applications, it has become exceedingly difficult for a single engine program to shoulder the risk and cost of technology development and productio~readiness, ~ a t e r i a quall ification and production inception. Over the past decade, significant government and corporate resources have been expended in order to bring gamma TiAl alloys to a state of production readiness for a variety of commercial and aerospace components. These efforts have made progress towards
Casting demonstrating technology and production readiness for cast gamma TiAl alloy applications. However, some significant work remains for some of the most challenging applications, such as turbine blades; while other applications, such as turbocharger turbine wheels, are finally emerging. Technology readiness generally includes establishment of process, structure and property specifications for specific applications. Requisite ineclianical-property data and design methodologies for many of the t implementation of alloys in Table 3 are s u ~ c i e n for many applications. Although cast processing has advanced considerably for gamma alloys, many important component~processingtechnologies require additioizal effort: joining and repair, surface treatments and coatings, and machining and finishing technologies. Technology readiness has been demonstrated for a number of components in recent years via engine testing. These tests are crucial in demonstrating the benefits of gamma alloys, and in assuring the design community that these low-ductility alloys can perform in predictable ways in service. Additionally, if planned well, these engine-de~onstrationtests can facilitate some level of production readiness that is not generally achieved in development programs. Finally, any successful production application, including automotive, will help to lower the risk and cost of subsequent
599
applications, even for different c o i ~ ~ o n e ntypes t in different markets. A summary of published engine tests of cast gamma alloys is given below for turbocha~gerwheels, automotive intake and exhaust valves, and turbine engine components. 2.7.1 ~ ~ r b o Twhine c ~ ~Whech ~ ~ ~ r
At the TMS spring meeting in 1999, Tetsui (1999) from MHI announced that ~ i t s u b i sMotors ~i had launched the production of a turbo-charged production autamobile which features a gamma TiAl alloy turbine wheel - the Lancer Evolution \I1 (also known as the Evo VI). The Evo VE, based on Illitsubishi’s world championship-winning rally car, is one of the hottest production cars in Japan. The gamma turbine wheel shown in Figure 6 is produced by Daido Precision Casting by the Levi-Cast process. Due in part to its advanced turbocharger, the EVCI VX can accelerate from 0 to 60 mph in 4.4 seconds. Although this application represents a small volume production launch, the importance of a commercial production success for the development and design community shouldn’t be understated. Nearly every major turb~chargerand diesel engine manufacturer have successfully tested cast gamma TiAl alloy turbocharger turbine wheels, some of which
Figure 6 Daido Levi-Cast MHl alloy turbo~h~rger turbine rotor for Mitsubishi Motors (courtesy of MHI)
600
Processes and ~ h ~ n o ~ ~ n u
tests have been reported at conferences and in the open literature: ABB (Naziny, 1998), Honeywell Garrett HI (Tetsui, 1999), and Toyota (Nishino et al., 2000). These wheels range in size from 50mm diameter for the Mitsubishi gasoline engine turbocharger (Figure 6) to 250 mm diameter for an industrial diesel turbocharger turbine for ABB (Figure 7). Nazmy (1998) has reported a nificant reduction in 'turbo lag', which i s the delay ti required to spin the turbocharger rotor up to speed where it can eflectively provide more complete combustion for the engine. The more responsive light-weight gamma wheel resulted in a dramatic reduction in particulate emissions for a large turbocharged indus trial diesel engine. Baker (1998) reported that the commercial diesel-truck-size gamma wheels tested have more than doubled the fatigue life of the production IN713LC. In fact, there were no direct failures of the gamma turbine wheels; all failures were initiated by primary failure of the A1 compressor wheel. Tetsui (1999) reported that Mitsubishi has conducted endurance testing on a diesel automotive gamma turbocharger turbine wheel in two alloys, a high Nb alloy and a low b alloy (actual alloy chemistry was not reported). Both alloys performed
ure 7 ABB TPS 57 ABB-2 alloy turbocharger turbine wheel
well, with less environmenta~degradation than found in coniparable static air oxidation experiments, although some blade tip erosion in the low Nb alloy was seen. Nishino et al. (200~)reported the a~plication of a gamma turbine wheel, in combination with an improved stiffness shaft, can reduce the moment of inertia for the turbocharger wheel assembly by 42%, and increase the resonance frequency of the system by 44%. Together, these im~rov~ments allow a significantly improved responsiveness and an increased niaxirnum rotational speed. ~ ~ t o ~ o t gasoline ive tu~bochar~er turbine wheels operate at temperatures exceeding 1000 "C, which is up to 200-250 "G higher than diesel t urboch~r~er turbine wheels. This more aggressive thermal eiivironment is probably too excessive for ~ a m m aalloys with low refractory metal content. However, Nb alloy, and ABB's W- and S~-~ont ai nialloys i i ~ may be suitable for the higher t e ~ p e ~ a t u r gasoline e turbocharger ap~lication.
2.7.2~ u t ~ ~ o~ t ~ vt and ea ~ ~e h a V~~ lsv et s Several automotive and racing engine m a ~ ~ f a c t u r e ~ s have tested cast gamma valves because they can
60 1 provide benefits such as increased horsepower, eEciency and possibly lower emissions. titanium-alu~inideexhaust valves have been success-
1999) and others. Mass production of automotive engine valves requires prices dramatically lower than those available today that utilize ~ r o t o t y ~metal~mold e and invest~ e n t - ~ ~ stec~nology t i n ~ . How~ver,racing valves offer an opportunity where a premium will be paid for improved ~ e r f o r ~ a n cand e , where reliabil~tyis crucial. 2.7.3 ~ e r o s ~ u c~ e
~ ~~~i~~ r ~ o~ ~ ~ ~o ~ ~e ~ te s
~ i t h i nturbine engine a~plications, a number of successful engine tests have been reported. the most publicized engine test was the CE CFC-8OC low-~ressure-t~rbine blade test ( A u s t i ~and Kelly, 1995). Figure 8 is a picture of the assembled gammabladed rotor. Additional tests include: the Volvo highpressure t u r b ~ n e ~ ~ l adampers de
The GE 48-2-2 alloy CFC-8OC lovv-pressureturbine rotor prior to testing (courtesy of the rotor is about 12Ocrn 111 diaineter
compressor blades (see Figure 9), c~mpressorinner diainctcr shroud segments and exhaust nozzle tiles. The blade and sliroud tried in an F-119 deri 1998, followed by full engine testing for all thrcce gamma c o ~ ~ o n e n t s .
S
Fl19 derivativ et ul., 1997; Davidson, 1996). Tlre gamma alloy
Although progress has been ade in nearly every important step in tlic transition o f the gainnia alloys e r from development to p r o ~ u c t i o a~ ,~ ~ n i of~ ~ssues
blades (of 3.8cm height) for the F119 derivative CAESER t engine (courtesy of Pratt ~ ~ i t n ~ y )
602
Processes and P ~ e n o ~ e ~ a
remain which require attention. The following is not a coinpreheiisive list, but points out some of the most serious barriers to commercial production. Alloy cost: Although cost has steadily decreased over the last few years, continued improvements are required in the development of low-cost aerospace and automotive-grade melt-stock. Low-cost machining technology and vendor base: Progress has been made in nearly every type of machining, but additional improvements are required in low-cost conventional machining, especially in turning, grinding and drilling; additionally a competent vendor base needs to be established that can handle low-cost production. Welding and joining: Joining technology will continue to be required in nearly all gamma applications on the horizon, and weld-repair technology will be required for m e d i u ~and large structural castings. Production readiness: Product development cycles coiitinue to shorten, making it even more difficult to develop processes and to demonstrate production readiness within the given schedules. Additionally, although much activity is currently underway to produce engine test components, these activities do not always lead to the development and maturation of low-cost production processes.
The interest in cast, single-crystalNiAl alloys lies in the combination of high thermal conductivity and the potential for improved operating temperature versus single crystal superalloy s. However, state-of-the-art sing~e-crystalWiA1 allays lack the balance of properties available in conventional superalloys. In particular, no single alloy has both high-temperature creep resistance, and adequate ductility and toughness. Additionally,due to the higher melting temperature of the NiAl alloys, only an expensive fabricated blade and vane process has been developed, the cost of which might preclude or ecause of these key barriers, development activity for single-crystal NiAl alloys for turbine engine use has recently been dramatically reduced.
Single-crysta~casting of NiAl alloys has been demonstrated in a number of modified processes. The man techniques utilizing chill plates modified and pigtail or spiral crystal selectors are preferred (Oti
and Yu, 1993; YU ef al., 1993; Darolia, 1993; Goldman, 1993). The melting temperature of NiAl (1638 "C) is 300 "C higher than that of conventional nickel-based superalloys. This te~perature posed challenges for furnace, shell and core system design. In order to accommodate the higher melt temperatures and higher thermal conductivity of the NiAl alloys, changes were made to the heating controls, furnace insulation, and baffles of the casting furnaces. Conventional invested ceramic mold and core materials commonly utilize silica-based binders which have proven inadequate for casting NiAl alloys (Oti and Yu, 1993; Darolia, 1993). At appropriate NiAl mold preheat and casting temperatures ( > 1638 "C), the silica volatizes and sinters excessively. Due to their inadequate strength at temperature, the shell system and cores would also slump or creep excessively. If cast with a core, parts would often crack during cooling, due to the thermal expansion mismatch between the core and alloy, and due to the lower ductility of the alloys. Additionally, there were often unacceptable metal-mold reactions. In order to alleviate these problems, alternative binders and shell materials were developed with higher temperature capabilities, with a better thermal expansion match, and lower reactivity facecoats (Oti and Yu, 1993;Darolia, 1993;Price, 1993). Due to the difficulty in casting the candidate high pressure turbine (WPT) blades and vanes with cores, an alternative casting and fabrication process was developed (Darolia, 1993). A schematic of this process is presented in Figure 10. Utilizing the modified casting techniques, an oversize, solid, single-crystal blade or vane could be cast. Because of the low ductility of the NiAl alloys, low-stress machining techniques were developed (such as wire EDM) to split the airfoil, and p l ~ g e to - ~m a~c ~~n e the inner-core passages. Following machining, the airfoil halves would then be joined via activated diffusion bonding. Electrostream drilling and abrasive-waterjet cutting have been successfully developed for cooling-hole drilling and t r ~ ~ i n g , respectively. A finished vane is shown in Figure 11. Utilizing these processes, GE and its casting vendors PCC and Howmet succeeded in producing cast-andfabricated HPT vanes and blades suitable for engine testing, although the yields through engine-test~ready components were low,
~ i Alloys A ~ As is found in directionally solidified (DS) and singlecrystal (SX) superalloy casting, mold preheat
603
Casting
9
Figure 10 A scheinatic of the fabricated blade processing sequence (Darolia, 1993), reproduced with permission of The Minerals, Metals and Materials Society
temperature, alloy superheat temperature, and niold withdrawal rate are the most important parameters in controlling the grain structure and dendrite-arm spacing (DAS) in single-crystal NiAl alloys. The required mold preheat temperatures were found to be at least 28 "C and alloy~superheattei~peraturesat least 56°C above the alloy liquidus temperature (Oti and Yu, 1993). In general, the higher the niold preheat and alloy superheat, the higher the probability for producing single-crystals. The tliennal conductivity of NiAl alloys can be froin three to eight times that of conventional nickel-based superalloys, depending on alloying. Although this high thermal conductivity is attractive in turbine engine applications, as it leads to lower metal temperatures, it poses some unique challenges for single-crystal growth. Even with modifications to the Bridgman furnace, the higher thermal conductivity of the NiAl alloys, made achieving high thermal gradients and ~aintaininga small mushy zone in the baMe region of the furnace more digcult. In turn this difficulty has led to higher defect and scrap rates. Additionally, computational
modeling and experiments confirmed that the higher thermal conductivity of the alloys also increased the difficulty of achieving a single crystal through the grain selector, due to the expanded size of the equiaxed region next to the chill zone, and the reduced DS region in the grain selector (Uu et al., 1993).
Figure 11 Fabricated and ~nish-machinedNiAl alloy HPT vane, approxmately 4 cm in height (Noebe and Walston, 1997), reproduced with permisslon of The Minerals, Metals and Materials Society
604
Processes and P h e ~ ~ ~ e n a
first are a family of equiaxed alloys developed principally by the US nt of Energy at the Oak Ridge National L (ORNL) which are finding use in the metals processing industry. The second class is a DS alloy developed by the researchers at the Beijing Institute of Aerospace Materials (BIAM), which is being applied to aeroengine turbine vanes and blades in China. The development of casting technology for equiaxed Ni,Al-based alloys has been aided with the development of an air-melting process that allows m i n i ~ u m pick-up of impurities such as oxygen and nitrogen. The air-melting process known as Exoevi and Sikka, 1997) uses air(AIM) furnaces available at most Noebe and Walston (1997) reported on the manufacture, iron- and nickel-based alloy foundries. The concept of rig testing and successful engine testing of NiAl HPT ~ ~ o - ~ which e l t is ~ described ~ , in detail later, is easy vanes. These vanes were produced by the fabricated to adapt for most of the foundry melters. Other airfoil method from cast, oversized SX slabs. A floatiiigbarriers to successful casting of Ni,Al-based alloys vane design alleviated thermally induced stresses on the have been overcome through the development of data vane at~chments. Although technical feasibility was on physical properties and casting soli~ification demonstrated with the successful fabrication, installation niodeling (Viswanathaii et al., 1997) and the study of and engine test of the NiAl HPT vanes, several significant parameters controlling porosity (Ho et al., 1991; b r e a k t ~ o u g hare ~ required for the full-scale production Cheng, 1992; Sekhar et al., 19911, Also, for successful of cast NiAl alloy co~ponents. transition to commercial applications, ideiitification of One of the inaiii barriers to implenientation of SX NiAt alloys i s the need to balance high~temperatureproperties alloy co~positionswith appropriate combinations of and durabi~ty(fracture toughness and ductility). Even ductility, creep strength, and weldability (Aoki and with the technical feats achieved on the NiAl HPT vaneIzunii, 1979; Liu et al., 1985, 1988; Liu aiid Sikka, test program, some target components, such as HPT 1986) is important. Further advances to successful blades, cannot be produced without significant improvetransition to commercial applications have occurred ments in durability. Development eff'orts to produce a through cooperative casting and in-service testing NiAl alloy HPT blade, which were to parallel the HPT programs (Cooperative Research and ~ e v e l ~ ~ m e n t vanes, were abandoned due to the low impact resistance ~greementsbetween Oak idge ~ a t i o n a ~l a b ~ r a t o r y of the SX NiAl alloys (Walston and Darolia, 1997). and Delphi Saginaw 1998 and Sandusky Internat~onal Noebe and Walston (1997) proposed that achieving a (Sandusky) and Bethlehem Steel combination of room-temperat~eductility and toughogy transfer efforts for the Ex0 ness comparable to the two-phase gamma+al~ha-ZTiAl compositions, property database, and operating experialloys is an appropriate goal. ence to castiiig foundries (both static and ce toughen in^ of NiAl has been attempted using a and to potential users have further aided in n u m ~ e rof di~erentin~rin~ic and extrinsic ~ ~ c h a n i ~the ~ current s com~ercia~zation status ( ~ i ~ ~ a , (Darolia et al., 1996; Noebe and Walston, 1997). Ni,Al-based alloys. Some remaining barners to broad owever, those which successfully improve the toughness commercialization include cost of inv~riablyhave led to reductions in creep s t r ~ ~ g t h , as compared to currently used alloys, a roade er vendor making them less attractive compared with SX superbase (currently there are but three casting houses), lack alloys, Hence, the balance of properties problem remains of acceptable ~nachiningprocesses, lack of welding the most s i ~ i ~ c abarrier n t to NiAl alloy implementation. experience using c o ~ m o n l yused welding methods, and limited industrial operating experience.
It was also found that the DAS for NiAl castings is much larger, and the arms much coarser compared with nickel-based SX alloys. Typical NiAl DAS were reported to be between 125 and 2 0 0 ~ m(Qti and Yu, 1993; Yu et al., 1993). Oti et al. noted that while the secondary dendrite structure is clearly defined in the longitudinal view, the primary deiidrite structure is difficult to see in transverse section. The coarser dendritic structure was attributed by Yu et al. to the higher alloy superheat temperatures that would cause higher diffusion rates and dendritic coarsening during cooling.
There are two classes o f cast Ni,Al-based alloys currently under development or in production. The
For a long period, the lack of comnierc~alizationof Ni,Al-based alloys was caused by the concern that the
Casliag
conventional practice of AI will cause the overheating of liquid aluminurn, because of its lower melting point, prior to the start of melting of nickel. Further concern was that the super~eateda l u m i n u ~would seep through cracks in the furnace lining and attack the induction coil. If tlzat happened, there could be an explosion. However, a closer look at the tbermodynamic data ( ~ e ~ m a1992) n , suggested a totally different possible concern (see Figure 12). These data show that the $ is s ol ~ t i o~ of elements such as nickel, iron, cobalt, and copper in a i u m i n ~ is n ~e x o t ~ e ~The ~ cextent . of the exotlierm is maximum for the atomic ratio of nickel to aluminurn of 1:1. The E xo- ~ e l t T Mprocess takes advantage of this exotherm and uses the following concept for melting Ni,Al-based alloys. In the Exo-MeltT~process the Ni,Al is thought of as 2 ~ i + ~ i~ ~i tl this h. concept in mind, the total nickel content of the Ni,Al-based alloy is divided into three parts: two parts of the nickel are placed at the bottom of the melt crucible and the rernaining nickel and aluminurn are alternately stacked on the crucible top. All of the alloyxiig elements iiicludiiig boron, chromium, z i r c o ~ i u ~and , molybdenum are stacked between the top and bottom layers. A schematic of the stacking sequence is shown in Figure 13. As thc
605
Figure 12 Extent of exothermic reaction for formation of various aluminides
Schematic of furnaceloading sequence employed for the Exo-MeltT" process in. melting nickel al~~ini~es
Processes and P ~ e ~ o ~ e ~ a
606
induction power is turned on, all of the elements couple with the coil. The extent of heating of each element is a f~~nction of its electrical resistivity. Because of resistivity differences, nickel tends to heat faster than aluminum. aluminwn heats both by coupling with the coil and by conduction of heat from the heated nickel. The aluminum becomes liquid, it then reacts with the heated nickel surrounding it, and reacts to form superheated NiAl liquid (above ~ 1 6 4 0 ° C )The . NiAl liquid travels down because of gravity and shares its energy with the heated nickel at the bottom of the crucible. The NiAl liquid also shares its energy with the alloying elements on its way down and dissolves some of them in the process. The net result of the Exo-melt stacking is that the exotlierm is created in a controlled manner, and its energy is used to promote the melting of the r~mainingmelt charge. A non-Exo-MeltTM or conventional process where nickel is melted and aluminum added causes the bath temperature to rise from 1400 to 2300 "C, with a plume over the bath of some of the vaporized elements including aluminurn. Such a process is not only unsafe for the melter but is not able to achieve controlled chemical coniposition of the melt and overheats the crucible by several hundred degrees above the needed use temperature. Thus, in contrast to the conventional process, the Exo-MeitTM process offers advantages listed in Table 4. Since the successful demonstration of the ExoMeltTM process in the laboratory, its details have been transferred to three foundries: Alloy Engineering & Casting Company (AEC) in Champaign, Illinois; andusky CO; and Alcon Industries, Inc. (Alcon) in Cleveland, Ohio. All of the foundries have successfully melted large numbers of heats using the Exo-MeltTM
process. A comparison of nominal chemical analysis of 1C-221M, one of the castable Ni&-based alloys, with the range observed for heats made using virgin and revert stocks in a pilot commercial run of 94 heats carried out at AEC shows excellent agreement.
4.2.1 Recovery of Vmious Elements
The oxidation of ~luminum,zirconium, and boron during air melting is considered a concern for their retention. However, melting of a large number of heats in laboratory and foundry environ~entshas shown that the use of the Exo-MeltTMprocess results in zero loss of alloying elements except zirconium, which requires a loading of 120% in the melt stock. t i ~ l 4.2.2 Pick-up from ~ o ~ ~ e~ ~e l cStock of Elements Other Than Those Specified in Nominal Composition
The commercial grade melt stock tends to introduce impurities, such as carbon and sulfur. A systematic study (Sikka and Santella, 1997) of a large number of heats to determine the effect of impurity elements on weldability has allowed the authors to set acceptable limits of carbon, sulfur, silicon, and boron. The pickup of silicon and iron during foundry practices is described in Section 4.2.3. Since limited opportunity occurs to remove these impurities during air-induction melting, carbon and sulfur contents can only be kept to a ~ i n i m u mor within acceptable limits by proper selection of the melt stock. If pick-up of these elements from previous melts is of concern, it can also be controlled by
Benefits of ~xo-MeltTM relative to conventional casting Melting process Feature Power use Time to melt Cost Safety
Conventional
Unit power Unit time Unit cost Safety issue due to excessive heat Safety issue due to excessive wear of the crucible Melt te~perature No control Crucible life Limited due to overheating and excessive wear o n control Melt c o ~ ~ o s i ~ ~ No Excessive inclusions due to overheating Oxide inclusions Industrial acceptance Not inany companies interested due to safety concerns
Exo-MeltTM One-half the power of the conventional method One-half the melting time of the conventional method One-half the cost of the conventional method No safety issue because of controlled process No overheating of crucible Real control Extended life due to temperature control Real control Very low due to good control of temperature and melt time Used successfully by several companies
607
Casting starting with a new furnace lining or using a pure nickel wash-heat prior to melting nickel-aluminide alloys.
4.2.3 ~ i c k - o~f p~ Z e ~ e P zfrom t s ~ o ~ n Practice ~ r y Silicon and iron are commonly picked up when melting nickel alurninide in an iron-based alloy foundry. The silicon is picked up from the fact that nickel aluminide tends to react with the sand (SO,), especially when extra metal is poured into sand pig molds. If this pig is then used in melting revert heats without removing the sand, the aluminum and zirconiuin contents of the nickel aluminide reduce SiO, to silicon, which is picked up by the alloy. The silicon from this source can be reduced by two approaches: (1) pour the extra metal in Zr0,-wash-coated pig molds (such a wash eliminates the reaction of molten IC-221M with sand); and (2) remove sand that is stuck to the pigs by grit blasting prior to melting. Another source of silicon pick-up is by the attack of molten nickel alurninide on SiQ, that is present in zircon crucibles. Silicon pick-up from this source can be minimized by two steps: (1) use A1,0, as a furnace crucible or lining rather than zircon, and (2) minimize the time that molten IC-221M stays iii contact with the crucible. Proper scheduling of melting and casting of heats so that the molten metal does not remain in the furnace for long periods before pouring can reduce the contact time. The iron pick-up can occur in at least two different ways. First, being in an iron foundry, there is always a chance of a small piece of iron or steel from other heats getting into the nic~el-aluminide melt stock. Pick-up from this source can be minimized by careful controls in the foundry practice. The second source for the pick-up of iron is the steel liner that is typically used for a new furnace. The best method to reduce such a pick-up is to run a wash heat of nickel after setting the new lining and prior to melting Ni,Al-based alloy IC221M. Any pick-up of iron over 1 wt,% has the tendency to precipitate the NiAl beta phase, which lowers the high-temperature strength of IC-221M.
4.2.4 Melting of Foundry Revert Stock The foundry revert stock consists of several items: (1) pigs cast from metal left over after casting, (2) runners and risers removed froin the castings, and (3) any defective castings that are beyond repair. Any of these three stocks may add impuri~iesto the alloy. The runners and risers can also add silicon if some of the sand is stuck to the castings and can also add iron
impurities if some of the steel beads become trapped in the defective area. Proper removal of any sand from the runners and risers and removal of trapped steel beads can minimize the pick-up of silicon or iron from the foundry revert stack.
4.2.5 idat at ion of MoltePz Metal If molten IC-221M is exposed to air for long periods of time, zirconium is the first element to oxidize, followed by aluminurn. Such oxidation can be minimized by using an argon cover during AIM and a Kaowool cover over the melt transfer ladle while moving from the furnace to the mold. Similarly, flushing the molds with argon prior to casting can minimize the oxidation of molten metal flowing through the sand molds.
4.3 Sand Casting ~ e c ~ ~ i ~ o l o g ~ ? Sand casting is the most common method used for the production of Ni,Al-based alloy (IC-221M) components (Sikka et al., 1995; Viswanathan et al., 1997; Deevi and Sikka, 1997). The finished sand-cast components so far produced have weighed from 50 to 200 kg (100 to 400 lb). The sand-cast components have varied in range of complexity from uniform thickness sections across the entire casting to those with significantly varying thickness across the casting (Figure 14). Typical foundry sands and molding procedures are acceptable for casting of nickel aluminides. The flushing of argon through the sand molds minimizes the formation of oxide skin on the cast surfaces. There is some indication that nickel aluminide does shrink more than the commercial alloys (HU, HT, etc.) that it replaces. However, for most applications, the casting patterns for current alloys can be used for nickel aluminides without concern for dimensional tolerances. A computer modeling of the gating system and mold-filling orientations for nickel aluminide has been developed and experimentally verified (Viswaiiatha~et al., 1997). Sand castings of nickel-aluminide alloys are commercially produced in the United States at AEC and Alcon. These companies use the Exo-MeltTMprocess for melting both virgin and revert stock prior to filling sand-casting molds. Each of the companies has developed their niche markets and many applications of nickel aluminides are currently underway. The inherent difficulty in machining nickel aluminides, lack of suitable machining tool materials, and lack of experience at the foundries makes the cutting of
608
Processes and P h e ~ o ~ ~ n a
~ ~ ~ t o g~r ~a o~ w h ai complex ~ g 60x 6Ox 3 crn sand-cast tray from n i c k e l ~ a I ~ ~ i alloy n i d ~IC-221M
runners and risers, ~rinding, machining, and weld of continuing concern for all a1 research efTort is needed in the inding, drilling, and single-point machi~ing.
ecause many component parts contain thin sections, centrifugal casting is the most economical method for the production of nickel altiminide for applications such as thin-wall (6-mm) radiant burner tubes for efficient heat transfer, steel plate ~ustenitiza~ion rolls ( 3 5 ~ 2 ~ - m ODx25-mm m wallx4-m length), steel continuous~castiiig guide rolls (75-1 50-mm ~ ~ x 2 ~ - m wall), m and steel strip hydrogenannealing-furnace seals and rolls (1 50-200-mm OD x 1&25-nim wall). Centrifugal castings of nickeles listed above have a~umini~ alloys e of all the size conv~ntionalmold been commercially cast by u washes and parameters similar to those for commoniy catio~~s. It is remarkab~eto ality centrifugal castings of in the smallest size tubes of 67-mm Or>x 6-mm wall thickness. The surface quality is similar to that of of n i ~ ~ e l - a l u ~ i n icastings de ntional alloys. However, the inner surfaces of -based tubes tend to contain oxide particles of 3. The appearance of such oxides in i~Al-basedalloy tubes results from the presence of stable oxide-forn~i~ig elements such as zirconium and a l u ~ i n u i which ~, are not present in the conventional
alloys. The presence of Zr inner surface has no influence on the performalice of centrifu~~lly cast tubes ~ i 3 A l - ~ a s ealloys d for furnace roll applications. owever, the presence of ZrOz and Al2O3particles does cause some difficulty in machining of inner surfaces, as req~iredfor assembly of components. i e ~ and machining, the Other than ~ ~ c u ~in tcutting centrifugal casting process for nickel-aluminide alloys is well established for co~mercialproduction. The C primary m ~ n u ~ ~ c t UOf r eCentriftlgally ~S pipes in the United States are AEC and
4.5.I DS ~ ~ v eC~'~ti~g ~ t ~ e ~ ~
Han and Xing (1997) have reported the successful alloy, named Alloy IC6, which reduction versus conipetitive nickel superalloys, and has exccptioiial strength and creep resistance without the use elements such as Ta, Hf and composition is Ni-7.8Al-14Mocent). Three major phases are present in this alloy: gamma, gamma prime and borides. The ingot production and DS casting processes used for the IG6 alloy are very similar to the processes used for conve~tionalD between 1570-1600 "C for 15-20 minutes. DS components are produced using the power reduction method
Casting
in a VIM furnace. Alumina molds are used, connected to a water-cooled copper chill plate, utilizing a mold withdrawal rate of ci-8mm per min. Han and Xing (1997) also report that a solut~oningheat treatment followed by rapid cooling is applied to components in order to obtain optimum mechanical properties. The focus of recent develop~enthas been on turbine vanes, with engine and flight tests being
that the alloy has been selected for production for advanced aeroengines, including vanes with complex cooling passages. 3.2 ~
~
~ ~ z~ v ~e st~Cast t ~e ivlg e ~d
I n v ~ s t ~ e cnat ~ t i nis~not a very common method for applications identified for equiaxed N i ~ ~ l - b a s ecornd positions, However, for potential turbocharger applications several investment casting parameters were tested (Sikka et al,, 1991). included alloy-pouring temperature, shell temperature, and grain refiner coating. Test bars were cast for d i ~ e r ~ sets n t of condi~ionsand evalua~edfor micro-
609
structure, microporosity, and mechan~cal pro~erties. The best pouring and shell temperatures were 1482 and 1038"C, respectively. A11 of the casting trials were carried out in a vacuum furnace. One of the problems with the investment-cast microstructure of IC-221M was the reaction between zirconium in the alloy and the shell material. The reaction product was stringers that progressed from the surface inward (Sikka, 1996). The surface reaction depths were 0.0, 0.051, and 0.076mm for alloys c o n t a ~ ~ i n0.85, g 1.28, and 1.70 wt.% 22, respectively. In most cases, stringers would be ground out from the surface of the investment casting. However, for critical applications, replacing zirconium by ha~niumcould elim~natethe stringers. Subsequent to investiiient casting trials of turbocharger parts in a vacuum furnace, a limited number of investment castings were also made in air (Sikka, 1998). The alloy was melted by the E x o - ~ e l t T ~ process, filtered through ceramic filters, and poured in air into preheated ceramic shells. There is currently no foundry that coinmercially produces investmentcast components of equiaxed Ni~Al-basedalloys.
Effects of cooling rate, co~position,inoculation, and air-versus-vacuum melting on Ni,Al-based alloy cast microstructure have been studied ex~ensively( d., 1991; Cheng, 1992; Sekhar et al., 1991). The grain size increased sharply with decrea (Figure 16). For the same cooli inoculation is an effective method of reducing the grain size. Figure 16 also shows that for the same cooling rate, grain size is the same for casting in air or
DS cast turbine vanes made of Alloy IC6 (23 pieces) and other DS superalloy (1 5 pieces) in an advanced aero-engine after 25 hours of engine testing (Han and Xing, 1997),reproduced with perinission from The Minerals, Metals and Materials Society
Figure 16 The variatlon of mean grain radius of IC-221M alloy with the imposed cooling rate
Processes and P ~ e n ~ ~ z e ~ a
610
in vacuum. The other cast m~crostructuralfeatures such as radius of macroporosity ( ~ ~ a ~ ~ radius o ~ o of ~ ~ niicroporosity ( ~ ~ ~ secondary-dendrite-arin ~ ~ ~ * o ~ ~ j , spacing (&), radius of y’ particles (Ryt),and radius of y‘ cells (Ry#cell) as a function of cooling rate for Ni,Albased alloy IC-221M are shown in Figure 17. The volume fraction of the macroporosity as a function of cooling rate for IC-221M is shown in Figure 18. Other data show that the volume fraction of macroporosity is nearly two orders of magnitude more than that of m~croporosityand, thus, its control will have more influence on properties of the castings. Data in Figures 16 through 18 indicate that increasing the cooling rate during solidification (to suppress hydrogen evolution, see Section 5) is one method for improving the overall cast ~icrostructure of IC-221M. Further improvements in casting quality can be obtained through guidance provided by solidification modeling (Viswanathan et al., 1997). Such modeling optimizes gating system design, pouring angle to feed the casting, pouring temperature, and even the selection of proper type of sand for sand castings. No specific heat treatments have been identified to improve the microstructure and properties of Ni&based alloy castings. The only heat treatment often considered is the preoxidation of a casting at 1050 to 1150 “C in air prior to service. Such treatment is used to form Al@, on the surface, which can offer benefits of reduced carburi~ation,improved wear resistance and a potentially non-sticking property for certain types of contact.
) ,
0.1
o in air tl in v 0.01 0.01
0.1
1
Ni,Al-based nickel aluminides are attractive for hightemperature stru~turalapplications because of certain attributes including: (a) resistance to high-temperatu~e oxidation, (b) resistance to hi~h-temper~ture carburization, (cj increase in yield strength with temperature^ (d) good creep strength at high temperature, and (e) good fatigue resistance. These alloys also provide the benefit of improved service perforn~a~ce by forming A1203on the surface through p~e~xidation in air at approximately 1050 to 1100 OC. The Ni,Al-based alloy, without chroniium, can be pre-oxidized to create electrically nonconducting surfaces, yet with bulk metallic characteristics of being machinable and having high toughness. The Ni,Al-based alloys of interest are substochiometric y with alumi~umor aluminum equivalent of 21 to 24 at.%. Tke aluminum equivalent is defined as
Aleql,lvalent (at.%) = Al(at.%) Zr(at.%j + Mo(at.%)
h
1
0.01
0.1
1
10
1m
Figure 17 The variation of microstructural features with the imposed cooling rate for nickel aluminide alloy, IC-221M, equiaxed solidified in air with no inoculant addition
1
Figure 18 The volume fraction of macroporosity decreases with increasing cooling rate for nicke1”al~~nide alloy, IC221M.
%g 0
10
+ 1/2 Cr(at.%)+
Although the aluininum equivalent is the key to maintaining the alloys at high volume fractions of y ’ , various alloying elements are added to yield specific benefits. A list of common alloying elements and their benefits is given in Table 5. A number of ~i3Al-basedalloy com~ositionshave been identified over the past 15 years. However, the compositions of mast practical interest and those undergoing commercialization effort are limited to
61 1
Casting Common alloying elements for Ni,Al-based alloys ~
Alloying element
Effect
Boron with -24 at.% Improved room temperature ductility alummum through boron segregation at grain boundaries Chromium to Improved intermediate temperature (600 to SSOOC) oxygen partially replace aluminum embrittlenieiit through protective Cr,O, formation Improved room- and highM 01y bdenum tempe~dturestrength through solid-solution strengthening Zirconium Improved high-temperature strength through solid-solution strengthening, improved oxide spallation resistance, reduced macro- and rnicroporosity from solidi~c~tion, and improved weldability
those listed below, along with key attributes of each of these compositions: Castable cornposition with nearly 100% y' phase and limited intermediate temperature (600 to 850 "C) ductility. IC-396LZr: Composition with window of temperature and strain rate for its hot workability and contains nearly 85% y' and 15% y phase. IC-22 1M: Castable composition with nearly 85% y' and 15% y with presence of eutectic phase Ni,Zr,. The eutectic phase has a melting point of 1172°C and that is what limits the upper use temperature for certain applications. Most of the experience is with the casting, machining, welding, and in-service operation of this alloy. The alloy has the most mechanical property data available of any castable Ni,Al alloy. Castable composition with balance of 26-438: molybdenum, chromium, and zirconium specified to eliminate the JSi5Zr3eutectic phase. The elimination of the eutectic phase increases the use temperature of the alloy but reduces its weldability and oxidation resistancc. The alloy has been statically and centrifugally cast but has only limited casting and in-service operating experience. An affordable and castable DS alloy ZC6: developed in China by ~ I A MThe . alloy
IC-so:
is attractive as a S turbine vane or blade alloy, with lower density, lower cost and better creep performance relative to other DS alloys. Efforts at QRNL, in close collaboration with the licensees and users, are continuing to expand current applications and develop new applications for Ni,Albased alloys. Some of the current and potential applications of the cast ~ i , ~ l - b a s e dalloys are described in Table 6.
Y
The development of casting technology for Fe,Al- and FeAl-based alloys has been aided with the development of the Exo-MeltTMprocess (Sikka ef OZ., 1995; Sikka and Deevi, 1995) for their air melting. The casting technology has also been aided ( S i ~ k a1996) ~ by recogiiiziiig that moisture control of the melt stock is essential in order to avoid extensive porosity that can occur in the castings. Any moisture associated with either the iron or aluminum melt-stock produces large amounts of hydrogen through the following reaction:
2 A1 + 3 H20 + A1203
+6
The hydrogen is soluble in molten iron aluminides and is rejected on cooling, resulting in both micro- and macroporosity in ingots and castings. C i r ~ u i ~ v e i ~ t i ~ n of other barriers to success in casting of iron aluminid~s has come through the develop~entof physical property data and casting solidification inodeling (Viswanathan et al., 1997). The low room-temperatur~ductility and high brittleto-ductile transition ( TDT) temperature for coarsegrain-size cast materials have limited the use of iron aluminides. For most applications, iron aluminides are being considered as worked cast ingots or by consolidation arid processing of powders. Remaining barriers to coimnercial develop~entof iron aluminides include identi~cdtionof compositions with higher room-temperature ductility, lower BTDT temperature, and resistance to environmental effects; insufficient machining data; weld process parameters and filler metals; and property data for castings. Although limited trial castings of iron aluininides have been done at commercial fou~dries,there are currently no vendors for the production of iron aluniinide ingots or castings iii the United States or overseas.
Processes and P ~ e ~ o ~ e ~ a
612
Current and potential applications of Ni,Atbased alloys Applicat ~ o n ~ ~ b j e c t ~ v e
Component Thin wall tubes and statically cast return bends Thick-wall tubes and pipe and statically cast trunions
Statically cast trays and fxtures
Statically cast die blocks
Statically cast tube Iyangers
Statically cast coniponents
Radiant burner tubes in various heattreating furnaces To increase life from higher creep, oxidation, and carburization resistance Furnace rolls in austenitizing and hydrogen annealing furnaces To increase life through elimination of blisters, nonstickiiig of iron oxide to roll and higher creep and hightemperature wear resistance Trays and fixtures for holding coniponents during carburization and annealing processes To increase life from higher creep st~ength and resistance to carburization and Oxidation Dies for hot-forging process To increase life from its peak in yield strength at 850 "C and good oxidation resistance Hangers to support tubes operating at high t e ~ p e ~ a t u r in e schemic~lplants To increase life from its high creep strength and oxidation resistance Variety of applications for heat-treating furnaces To increase life from its higher creep strength, oxidation, and carburization resistance weess
The fo~mationof iron alu~inides(Fe3Al and FeAl) from the elen~entsis exothermic similar to that for nickel a~ui~iiiides(see Figure 12). Although the exothemic energy for the iron aluminides is significantly less than for nickel aluminides, the Exo-MeltTM process furnace-loading scheme, similar to that for nickel al~iiiinide(Figure 13), has been found to be extremely useful in lowering the hydrogen and oxygen contents, and improving the recovery of alloying elements in casting FeAl (Sikka et al., 1998).
have been reported by r et al. (1998). There are and VIM and secondary several other reports o methods such as TAR and ESR (Sikka, 1991a,b, 1994; al~gidadet al., 1997). The crucible and the rernelting process have been
Replacing Cast stainless steels such as HU, HT, HK, etc. Cast stainless steels such as HU and HP ~ o d ~ ~ e ~
Cast stainless steels such as MU, HT, HK, etc. H-13 and other steel-baseddie materials materials
S ~ ~ ~ R T H and E ~other ~ ThighM temperat~irealloys Alloy steel and cast stainless stcels
of VIM ingots incre~sesthe from 0,0010 wt.OA to 0.0020 to 0.0040 we.%. Th agnesium content of 0.0020 wt.% had no detri~entaleffect on the processing of Fe,Al-based alloy ingots. magnesium content of 0,004 wt.% resulted in hotshortness in ingots during hot forging and hot rolling. The nitrogen and oxygen contents of VIM alloy ingots were extremely low. There were no further reductions in their values noted for the ESR process (Sikka et al., 1991). Sunder et al. (1998) reported reductions in nitrogen, oxygen, and sulfur contents by AIM with a flux cover and by ES of the AIM ingots. The flux used during melting was a mixture of Al,O,, CaO, MgO, and CaF,. The use of flux during AIM also resulted in the elimination of hydrogen-induced porosity (Sunder et al., 1998).
xo-MeltTMprocess i ical process for melt in^ ingots. 5.
increases the ~agnesiLimcontent to 0.0020. The ESR
Limited work has been devoted to sand castings of Fe,Al- and FeAl-based alloys. Causes for limited effort
Casting
in this area include low room-temperat~reductility, which is further reduced by the coarse grain size that results from sand casting. In spite of the limitations, FeAl-based alloys were successfully cast into large and complex-shaped components (Sikka et al., 1998). The specific components made are known as skirts, grate bars, pallet tips, and return bends for radiant burnertube assemblies. The castings varied from 5 to 150 kg. All had coarse grain size. Limited trials of sand castings with 0.75 wt.% Ti showed a very significant reduction in grain size. Although the sand castings listed above were made at two different commercial foundries, neither of them FeAl-~asedalloy castings. Reasons for lack of commercial applications include: low hightemperature strength of FeAl-based alloys (which limits their use for many a p ~ l i ~ ~ t i owhere n s these alloys would supply exceptional corrosion properties), in long tern tests alloys did not perform as well as they did in shorter tern tests, and limited ability for cutting and ~n a c~in inofg these alloys.
The centrifugal casting process has been used successand thick-wall tubes of FeAlz et a!., 1997; Sikka, 20~0). s yet no applications of FeAl that they will occur within the next two to three years. Applications under investigation are guide rolls for continuously-cast copper billets for ethylene cracking. In the latter case, FeAl may have to be used as part of a duplex tube, where FeAl wiil be on the inside to provide carburization/~okingresistance. Other ~pplicationsof the tubing will be in environment^ contain~nghigh gaseous sulfur d/or oxygen or molten carbonate salts. Although the cost of FeAI-based alloys is re able, the di~cultiesin cutting, machining~and we1 continue to be de~errentsto their commercial use in -temperature strength limits the use of Fe,Al- and FeAl alloys for envi r o~ m ~ nt where s they show ~nmatchedc o r ~ ~ s ~resistance. on
Fe,Al- and FeAl-based ystematic studies on the crostructure have been reported. V~sw~nathanet al. (1997) conducted a cation behavior of a Fe,Al-based were exam~nedafter arc melting
613
followed by chill casting into water-cooled copper molds, air melting followed by casting into graphite molds, and electroslag melting and casting into large ingots. The microst~ucturesof all three types revealed the presence of two distinct second-phase particles, the first distributed within the grain and possessing a rod ~orphology,and the second present at the ~s-cast grain boundaries in a globular interdendritic morphology. The globular particles, based on their location, were suggested to have preci~itatedfrom the last liquid to freeze. Although second-phase particles of each class are observed in all three types of castings, both the particle size and spa between the particles increased with the solidi^ Viswanathan et al. (1997) also conducted homogenization studies on samples from slab-rnold in The homogeniz~tion treatments were at 1200 and 1300 "6: for periods of I to 32 h. The homo~enizatjon at 1300 "C caused immediate dissolut~onof the interdendritic phase and reprecipitation of large particles at the grain boundary. Rod particles also seemed to partially dissolve, but the effect of homogenization was less dramatic.
The Fe,Al- and FeAl-based alloys are attractive (McKamey et al., 1991) because of their many attributes including: (a) low cost because of inexpensive constituents of basic elements iron and aluminurn, (b) excellent resistance to oxidation, (c) excellent resistance to h~gh-temperaturesulfidation, and (d) excellent resistance to carbur~zation,The FeAl-based alloys offer additional benefits over Fe3Al-base respect to their resistance to molten salts Tortorelli, 1997; Tortorelli and Natesan san, 1998), higher strengths, and lower d Fe,Al- and FeAI-based alloys suffer from environmental embrittlement caused by (Stoloff and McKamey, 1997; §to Maziasz et al., 1997). In addition, both Fe,Al- and FeAl-based alloys possess low room-temperature ductility ( ~ c ~ a m eety al., 1991; Liu et al., 1998; Baker and George, 1997). Many alloy design studies amey et al., 1991; Sikk 1997; Kumar and Pang, 1998; have been done to improve room-te~peratureductility and reduce environmental effects. These studies r~sultedin several modi~edcomposi~ionsfor Fe,Aland FeAl-based alloys. The most commonly explored Fe,Al-based compositions are FAS, FAL, and FA-129 (Sikka, 1997; McKamey et at., 1991).
614
Processes and P h e ~ o ~ e n a
Chroniiui~ is added to Fe,Al-based alloys to improve their resistance to aqueous corrosion and to reduce the environmeiital effect. Boron is added to improve room-temperature ductility. Zirconium is added to improve the resistance to oxide spalling under cyclic oxidation conditions. Niobium and carbon are added to improve the creep strength through the precipitation of niobium-carbide particles. Effects o f other elements in an experiinental alloy are described by McKamey et aE. (1991). There is a broad range of aluminum in the FeAlbased alloys that are being tested by different investigators. Alui~inumvaries from 21 to 26 wt.%. ~ i t h i nthis range, the most commonly added elements include 0.4 MO, 0.05 C, 0.15 Zr, and 0.008% B (Maziasz et al., 1997). The effects of various elements on alloy prop~rtiesare described by Liu et al. (1998). mar and Pang (1998) and Pang and Kurnar (1998) cribe the effect of carbon on FeAl-based conipositions. Fraczkiewicz et al. (1998) and Baker and George (1997) describe the effect of boron 011 FeAl-based compositions. A specific FeAl-based composition used trials by Sikka et al. (1998) is Fe (balance), I%), Zr (0.15-0.2%), C (0.074.10%), B (0.005), and MO (0.4-0.45%).
Aoki, K., and Izumi, 0. (1979). Nippon Kinzoku Gakkais~i, 43, 1190-250. Austin, C. M., and Kelly, T. J. (1995). In Gamma Titanium ~ l u m i n ~ d e(eds. . ~ Y.-W. Kim, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, p. 413. Austin, C . M., Kelly, T. J., McAllister, K. G., and Chesiiutt, J. C. (1997). S ~ r ~ c t Inte~me~all~cs ~~al 1997 (eds M. V Nathal, et al.). Seven Springs, Champion, PA, p.413. Baker, C, (1998). Oral presentation at High Temperature Materials Seminar at AES, Torrance, CA, 26 February 1998. Baker, I., and George, E. P, (1997). In International Symposiunz on Nickel uiad Iron Alurwinides: Processing, Properties, and A~pl~ca~ions (eds Seetharama C. Deevi, et al.). ASM International, Materials Park, OH, pp. 145-156. Baligidad, R. G., Prakash, U., Radhakrishna, A., Rao, V. R., Rao, P. K., and Ballal, N. B. (1997). In rnternationul S y ~ p o ~ ~ 7on u mNickel and Iron Aluminzd~s: Processing, ~ r o p e r ~ iand e ~ ~A~plications , (eds Seetharama C. Deevi, et al.). ASM International, Materials Park, OH, p. 177 BIuni, M.,Choudhury, A., Scholz, H., Jarczyk, G., Pleier, S., Busse, P., Frommeyer, G., and Knippscheer, S. (1999). In Gamma T~~anium A l u ~ n ~ n i 1999 ~ e s (eds Y.-W. Kim, D. M. Dimiduk, and M, H. Loretto). TMS, Warrendale, PA, p. 35.
Chandley, 6 . D. (1991). US. Patent 5,042,561, August 27, 1991. Chandley, G. D., and Flemings, M. C. (1994). U.S. Patent 5,299,619, April 5, 1994. Cheng, C. T. (1992). ‘Microstructure and Porosity Evolution in Equiaxed Solidified Multi-component Nickel Aluminide’, Ph.D. dissertation, University of Cincinnati, Cincinnati, OH. Cheng, T. T. (1999). In Gamma Titanium Alunzinides 1999 (eds Y.-W. Kim, D. M. Dimiduk, and M. H. Loretto). TMS, Warrendale, PA, p. 389. Colvin, G. (1995). In Titaniu~z ’95, Vol. 1 (eds P. A. Blenkinsop, W. J. Evans, and H. M. Flower). Birmingha~,UK, Institute of Metals, p. 691. Darolia, R. (1993). In Structural Interme~allic,~ 1993 (eds R. Darolia, et a/.). Seven Springs, Champion, PA, p. 495. Darolia, R., Walston, W. S., and Nathal, M. V. (1996). In Superalloys 1996 (eds R. D. Kissinger, et al.), TMS, Warrendale, PA, p. 561. Davidson, D. E. (1996). In Processing of’SziperalEoys 1996 (eds R. D. Kissenger, et al.). TMS, Warrendale, PA, p. 539. De Graef, M., Lofvander, J. P. A., McCullough, C., and Levi, C, 6. (1992). Acta Met., 40,3395. De Graef, M., Biery, N., Rishel, L., Pollock, T. M., and Crarnb, A. (1999). In G a ~ T~~ a~ a n Al~min~des zu~~ 1999 (eds Y.-W. Kim, D. M. Dimiduk, and M. €3. Loretto). TMS, Warrendale, PA, p. 247. Deevi, S. C., and Sikka, V K. (1996). ~ ~ t ~ r ~ e t a l l ~ c s , 357-375. Deevi, S. C., and Sikka, V, I(.(1997). Inter~ietall~~s, 5, 17-27. Dirniduk, D. M (1999). Gamma titanium a~uminidealloys an assessment within the competition of aerospace structural materials, Mater. Scr. Eng. A , ~ ~ 62831. , Dimiduk, D. M., McQuay, P. A., and Rim, Y.-W (1999). In ~ ~ t a n ’99, i u St. ~ Petersburg, Russia, in press. Dowling Jr., W. E., Allison, J. E., Swank, L. R,, and Sherman, A. M. (1993) SAE Tec~nicalPaper Series, No 930620, p. 3 1. Dowson, A. L., Johnson, T. P., Young, J. M., and Jacobs, M. €3. (1995). In Gamma T ~ ~ Aluminides a n ~ (eds ~ ~Y.-W. Kim, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, p.467. Fraczkiewicz, A., Gay, A.-S., and Biscondi, M. (1998). Mater. SCL.h E H ~ A258, ,, 108-114. German, R. M. (1992). In T ~ e ~ mAnalysis al in ~ e ~ a l l ~ {(eds rgy R. D. Shull, and A. Joshj). TMS, Warrendale, PA, p. 205231. Goldman, E, H. (1993). In High Temperature Intermetallic Alloys V (eds I. Baker, et al.). MRS, Pittsburgh, PA, p. 83. Han, Y. F.? and Xing, 2. P (1997). S ~ r u ~ I ~n t~e r~~ ea~la ~ Z ~ c s 1997 (eds M. V. Nathal et al.). Seven Springs, Champion, FA, p. 713, ~art~eld-Wunsch, S. E., Sperling, A., Morrison, R. S., Dowling Jr., W. E,, and Allison, J. E. (1995). In Gumma Titamurn A l ~ m i i a i ~(eds s Y.-W. Kini, R. Wagner, and M. Yarnaguchi). TMS, Warrendale, PA, p. 41.
Casting
615
Naka, S., Thomas, M., Sanchez, C., and Khan, T. (1997). In Structiaal I n ~ e r ~ e ~ a1997 l l ~ (eds ~ s M. V. ~ a t ~ aetl a/.). , Seven Springs, Champion, PA, p. 313. Huang, S. C., and Chesnutt, J. C. (1995). In Intermetallic Natesan, K. (1998). Mater. Sci. & Eng., ~ompounds:Vol. 2, Practice (eds J. H. Westbrook, and Natesan, K., and Tortorelli, P. F. (19 R. L. Fleischer). Wiley, p. 80. Symposium on Nickel and Iron A l ~ m i n i d ~Processirzg, ~~: Huaiig, S. C., and Hall, E. L. (1991). M X S Synp. Proc. 213, 827. Properties, and Applications (eds Seetharama C. Deevi, et Hyman, M. E., McCullough, C., Valencia, J. J., Levi, C. G., al.). ASM International, Materials Park, OH, p. 265. and Mehrabian, R. (1989). Metal. Tram., 20A, 1847. Naziiiy, M. (1998). Oral presentation at Aeromat '98, Isobe, S., and Noda, T. (1997). In Structural Intermetallics Washington DC, 15-18 June 1998, ASM Internatio~al. 1997 (eds M. V. Nathal, et al.). Seven Springs, Champion, Nishikiori, S., Takahashi, S., and Tanaka, T. (1999). In PA, p.427. Gamma ~ ~ t a n i uAm l ~ & ~ i n i d1999 e s (eds Y.-W Kim, D. M. Jones, P E., Porter, W. J. 111, Eylon, D., and Colvin, G. Dimiduk, and M. H. Loretto). TMS, Warrendale, PA, (1995). In Gamma T ~ t a n ~ uAluminides m (eds Y.-W. Kim, p. 357. R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, Nishino, K., Kawaura, H.,Tanaka, K., Horie, T. p. 53. Uchida, H. (2000). J. Gm Turbmc Soc. Jap ., Kumar, K. S., and Pang, L. (1998). Mater. Sci. & Eng., A258, Noda, T, (1998). Application of cast gamma TiAl for 153-1 60. automobiles, In~ermet~llics, 6, 709. Larsen, D. E. (1996). Ma~erialsScience and Engineering A, Noebe, R. D., and Walston, W. S. (1997). In Structural A213, 128. Intermetallics 1997 (eds M. V. Nathal, et at.). Seven Larsen, D., and Govern, C. (1995). In Gumma Titanium Springs, Champion, PA, p. 573. Aluminides (eds Y-W. Kim, R. Wagner, and M. Oti, J. A. and Yu, K. 0. (1993). In Structural ~rzternze~allics Yainaguchi). TMS, Warrendale, PA, p. 405. IQ93 (eds R. Darolia, et al.). Seven Springs, Champion, Larsen, D. E., Kampe, S., and Christodolou, L. (1990) MRS PA, p. 505. Pang, L., and Kumar, K. S. (1998). ~ u t e rScz. . I& Eng., Lm, C. T., and Sikka, V. K. (1986). J . Met., 38, 19-21. 161-166. Lm, C. T., White, C, L., and Hortoii, J. A. (1985). '4cta Price, A. (1 993). Research, Howmet Research Corp., Metall., 33, 213-219 Whitehall, MI, unpublished. Liu, C. T., White, C. L., and Lee, E. H. (1988). Scr. ~ e ~ ~ l ~ . , Ram, S. V., and Barrett, J, R. (1996). ~ ~ ~ '95. a Vol. n II ~ u ~ 19, 1247-1250. (eds P. A. Blenkinsop, W. J. Evans, and H. M. Flower). Liu, C. T., George, E. P., Maziasz, P. J., and Schneibel, J. H. Institute of Materials, B i r m i ~ g h a UR, ~ , p. 88. (1998). Mater. Sci. & Eng., ~ 284-98. ~ ~ , Reed, S. (1995). In G a m ~ a T i ~ a n i u ~ ~ 1 ~ 7 ~ ~(eds ~ i Y.-W des Lupinc, V., Marchionni, M., Onofrio, G., Nazmy, M., and Staubli, M. (1999). In Gamma Titaniun2 A l ~ ~ ~I999 ~ ~ ~ ~ iKim, ~ e R.s Wabqer, and M. Yamaguclii). TMS, Warrendale, PA, p.475. (eds Y.-W. Kim, D. M. Dimiduk, and M. H. Loretto). Rishe1, L. L., Pollock, T. M., Cramb, A. W., and Larsen, D. TMS, ~ a r r e n ~ aPA, ~ e ,p. 349. E. (1997). Proc. of tlze 1997 Interna~io~ial S ~ ~ ~ on~ o ~ s Maziasz, P. J., Goodwin, G. M., Alexander, D. J., and Liquid Metal Processing and Casting (eds A. Mitchell, and Viswanathan, S. (1997). In rrzternational Symposium on P. Aub~rtin).Santa&Fe, New Mexico, p. 214. ~ i c ~ eand l Iron ~ ~ u ~ i n i d Processing, es: Properties, and Rishel, L. L., Pollock, T. M., and Cramb, A. W. (1999). Proc. Applications (eds Seetkarama C. Deevi, et al.). ASM of the 1999 I~ternationul Sj~mposizmmon Liquid M c t d Intermtional, Materials Park, OH, pp. 157-176. Processzng and Casting (eds A, Mitchell, L. Ridgway, and McCullough, C., Valencia, J. J., Levi, C. G., and Mehrabian, M. Baldwin). Santa Fe, New Mexico, p. 287 R. (1989). Acta Metal., 37, 1321. Roberts, R. J, (1996). iritanizrm '95. Vol. II (eds P. A. McKamey, C. G., DeVan, J. H, Tortorelli, P. F., and Sikka, Blenkinsop, W. J. Evans, and H. M. Flower). Institute of V. E(. (1991). J. Mater. Res., 6(8), 1779-1805. McQuay, P. A., and Larsen, D. (1997). In S t r u ~ ~ ~ r a l Materials, Birmingh~rn,UK, p. 1462. Sekhar, J. A., Lin, C. S., and Cheng, C. J. (1991). In Nature In~ermetallic~~ 1997 (eds M. V. Nathal, et al.). Seven and Properties of Semi-Solids ~ ~ ~ e r(eds ~ aJ. l A. s Sekhar, Springs, Champion, PA, p. 523. and J. Dantzig). TMS, Warrendale, PA, pp. 267-290. McQuay, P. A., Simpkins, R., Seo, D. Y., and Bieler, T. R. Seo, D. Y., An, S . U., Bieler, T. R., Larsen, D. E., Bhowal, P., (1999). In Gamma iritanium ~luminides1999 (eds Y.-W. and Merrick, €3. (1995). Gamma. TifanziimA1unmide.s (eds Kim, D. M. Dirniduk, and M. H. Loretto). TMS, Y.-W. Kim, R. Wagner, and M. Yamaguchij. TMS, Warrendale, PA, p. 349. Warrendale, PA, p. 745. Muraleedhara~,K., Rishel, L. L., De Graef, M., Cramb, A. Sikka, V. K. (1991a). In ~ i ~ ~ - i r e m ~ e r a tOurrdee r c ~ W., Pollock, T. M., and Gray 111, G. T. (1997). In Structural Intermetallics I997 (eds M. V Nathal, et al.), Intermetallic Alloys IV, 213 (eds L. A. Johnson, et at.). Seven Springs, Champion, PA , p . 215. Materials Research Society, ~ittsburgh,PA, pp. 907-9 12.
€30, C. T., Cheng, C. T., and Sekhar, J. A. (1991). Metall. Trans., 224 225-234.
616
Processes and P h e n o ~ ~ n ~
Sikka, V. K. (1991b). In ~ e a ~ - R e . s ~ s~~aa~n e~ r ~ (eds a l s R. Natesan, and D. J. Tillack). ASM International, Materials Park, OH, pp. 141-147. Sikka, V. K. (1994)- In Processing, Properties, and A p p 1 ~ ~ ~ ~ iuof nIron . s A l u ~ z ~ n ~ d(eds e s 3. H. Schneibel, and M. A. Crimp). TMS, Warrend~~e, PA, pp. 3-18. Sikka, V. K. (1996). In ~ h y s i c a~l e ~ a l l u r and g y Processing of I ~ t e r ~ e t a l l i c~so m p o ~ n(eds d ~ N. S. Stoloff, and V. K. Sikka). Chapnian & Hall, New York, pp. 561404. Sikka, V. K. (1997). In r~~te~nationai Syrnpo,si~rnon ~ i c k and ~l Iron A i u ~ ~ n ~ Proces'sing, d e ~ ~ : P r o p ~ r t i ~and s , Applica~io~s (eds Seetharama C. Deevi, et al.)" ASM International, Materials Park, OH, p. 361. Sikka, V. E=. (1998). Oak Ridge National Laboratory, Oak Ridge, TN, unpublished research. Sikka, V. K. (2000). Oak Ridge National Laboratory, Oak Ridge, TN, unpublished research. Sikka, V. K., and Deevi, S. C. (1995). at er. Technol., 10(5/ 6), 97-100. Sikka, V. JS., and Santella, M. L. (1997). Oak Ridge National Laboratory, Oak Ridge, TN, unpublished research. Sikka, V. IS.,McKamey, C. G., Howell, C. R., and Baldwin, R. N.(1991). Properties of Large Heats of Fe3Al-Based Alloys, ORN~/TM-11796, Oak Ridge National Laboratory, Oak Ridge, TN. Sikke, V. K., Deevi, S. C., and Vought, J. D. (1995). Adv. etrau, J., and Mackey, B. 2 229-235. ~ ~ , (1998). Mater. Sci. h Eng., ~ Stewart, D. C. (1994). Oral presentation at Aeromat '94, Anaheirn, CA, 6-9 June, 1994, ASM International. StoloR, N.S. (1996). In ~ ~ y s i e~aelt a l l u r g yand Processing of ~ ~ t e r ~ e ~ a~ lol~i pc o~u n d(eds s N. S. Stoloff, and V. K. Sikka). Chapmaii & Hall, New York, 479-516. StoloR, N. S. (1998). at er. Sci. & Eng., A258, 1-14.
StoloE, N. S., and Mclkamey, C. 6 . (1997). In I n ~ e ~ ~ a ~ Syrnposiu~un Nickel- and rron"A1urninides: Froc~s.singi (eds Seet~aramaC. Deevi, et Properties, and ~~plicatio~s al.), ASM International, Materials Park, OH, pp. 65-72. Sundar, R. S., Baligidad, 73. G., Priisad, Y. V. R. K., and Sastry, D. H. (1998). ~ ~ t eSci. r . Tetsui, T. (1999). In G a ~ Titanium ~ a ~ ~ u ~ ~ 1999 n ~ (eds d e s Y.-W. Kim, D. M. Dirniduk, and M. H. Loretto). TMS, Warrendale, PA, p. 15. Tortorelli, P. F., and Natesan, K. (1998). ~ ~ t eSci. r . di Eng., ~ 2 115-125 ~ ~ , Viswanathan, S . , Duncan, A. J., Porter, W. D., and Webb, D. S. (1997). In International Syrnposiu~on ~ i c l ~and e ~ Iron~ l ~ ~ i n i d~erso:c e s ~ iProperties, n~, and Appii~at~on~s (eds. Seetharama C. Deevi, et a/.). ASM International, Materials Park, ON, p. 243. Wagner, a.,Appel, F., Dogan, B., Ennis, P. J., Lorem, U., Mullauer, J., Nicolai, W. P., Quadakkers, W., Singheiser, I;.,Sniarsly, W., Vaidya, W., and Wurzwiilher, K. (1995). e ~ Y.-W. Kim, R. In G a ~ m aTitaniurn ~ l u m ~ n i d(eds Wagner, and M. Y a m a ~ ~ ~TMS, i ) , ~ a r r e n d a ~ ePA, , p. 387. Wa~ston, W. S., and Darolia, R. (199'7). In Structural ~ n t ~ r r n e ~ a i 1997 i i c ~ (eds M. V. Nathal, et al.). Seven Springs, Champion, PA, p. 613. Westbrook, J. H. (1993). In Structural Intermetailic~~ I993 (eds R, Darolia, et al.). Seven rings, hamp pion, PA, p. 1. I ~ t ~ r ~ e ~ a l i1997 i c s (eds Williams, 5. C, (1997). In Str~~cturai M. V. Nathal, et a,.), Seven Springs, Champion, PA, p. 3. Yamada, J., and Demukai, N. (1998). U.S. Patent 5,837,055, N o v e ~ b e r17, 1998. Yu, K. O., Oti, J. A., and Walston, W. S. (1993). In Nigh ~ e ~ p e r a t u~ ~ e~ t e r ~ e Alloys t ~ l l iV c (eds 1. Baker, et d). NRS, Pittsburgh, PA, p. 915.
-
I n t e ~ ~ etallic alloys are an emer~ingclass of materials that exhibit attractive thermophysical properties and thus oRer interesting perspectives for applications in
major concern for the fabrication and reliability of components. These involve shrinkage porosity, segregation of alloying elements, texture, and coarse ~icro$t~ucture. ~ttain in gchemical ho~ogeneityand re~nementof the microstructure are therefore the most im~ortantprere~uisitesfor the engineering application of intermetallic compound$. To this end, a large effort has been expended to establish procedures for wrought process~ngfor various intermetallic alloys, In broad terms, the techniques applied bear a number of similai.i~i~~ to the processing of conventional materials; however, the processing routes have to be adjusted to the particular properties of ordered intermetallic phases. These involve:
limited ductility and susceptibility to cleavage fracture, which often determine the failure modes under hot-working conditions.
In this chapter the current status of thermo~echanical processing of intermetallic alloys will be reviewed using y(TiA1) as the primary example. Int e~et al l i c alloys of technical significance are m~ltiphaseassemblies, which during hot working can undergo complex phase t r a n s f ~ ~ a t i o n The s . evolution o f the microstructure strongly depends on the pathway and kinetics of these transformations. Thus, there is no hard-andfast rule for the identi~cationof processing windows for intermetallic phases. For this reason no attempt will be made to reiterate all that has been said on hot working of a variety of intermetallic phases. Instead, attention is centred on the w roug~tprocessing of titanium aluminide alloys, a subject which has been documented in reasonable detail. Thus, the data can be used to typify the processes and illustrate the diflficulties involved in the wrought processing of intermetallic phases iii general. Special conside~ation is given to areas where relatively recent work has in some way changed the perspective. only brief mention is made on the hot working of other intennetallics in section 4.
- a s i g ~ i ~ c aanisotr~py nt in deformed material due to -
the lack of independent slip systems; low dislocation mobilities and the difficulty of the dislocations to cross lide and climb, both of which impede recovery; obility, which retards recryst allization;
Titanium aluminides exhibit several desired properties for high-temperature technologies, among them: excellent strength and elastic stiffness, and good resistance
r ~ ~ e r ~C ~ ~ ao~ ~ ~ iVol. c ~3, Prtnciples ~ ~ and~Practice. ~ Edited ~ :by J. €4. Westbrook and R. L. Fleischer.
~ ~ 0 John 0 2Wiley & Sons, Ltd.
618
Processes and Pheno~eria
against oxidation and ignition. Combined with the low density of about 4g/cm3 these properties are particularly attractive, because they provide an opportunity to attain significant energy savings while meeting current environmental pollution regulations (Huang and Chesnutt, 1995 and Lipsitt et al. in this volume). Thus, the materials have a promising potential for extensive use as light-weight structural components in a wide range of coinniercial applications including the a u t o ~ o t i v eand aerospace industries,
Currently, most efforts have been focused on alloys with the general composition (Kim, 1995, ELim and imiduk, 1997; Appel and Wagner, 1998): Ti-(45-49) A1 + (0.3-5) X with X design at in^ modest amounts of several other elements. Except where explicitly noted, all compositions throughout the chapter are quoted in atomic percent. Additions of Cr, Mn and V up to a level of 2% for each element have been shown to enhance ductility. The role of various other third alloying elements is to improve other desired properties such as oxidation resistance (Nb, Ta, Wj and creep strength 1993). Boron additions greater ve in refining the grain size and stabili~ingthe micros~ructure (Hyman et al., 1991). emands for higher strength coupled with good oxidation resistaiice have led to the development of a y(TiA1) alloys with the base-line ang, 1993; Chen et al., 1993; Paul et al., 1998; Appel e f al., I999a, 2000a): Ti~45AI-(5-10)Nb + X ial attcntion will be given to these alloys, because have the potential to extend the service range of ~onventionalt~taniumaluininide alloys. The addition of ternary and higher elements not only changes the relative stability and t r a n s f o r ~ ~ t i o n phases, but also brings new phases owever, the general trends reported for binary Ti-A1 alloys also extend to more coniplex Thus, the binary phase diagram proCullough et al. (1989) will be used as rcference for illustrating the phase transformations taking place. Figure 1 shows the central part of this diagram and indicates the base for engi~ieeringalloys (45-49 at.% Al). The four stable solid phases existing in this composition range are the disordered solution
phases hcp-a(Ti), bcc-P(Ti) and the ordered internietallic phases y(TiA1) with L1, structure and a2(Ti3Al)with DOl9structure. As demonstrated in Figure 1, peritectic solidification and eutectoid reactions occur in alloys in the composition range Th(45-49)Al. When these alloys are produced by conventional ingot metallurgy, under relatively slow cooling, a lamellar morphology may evolve which consists of thin parallel a2(Ti3Alj and y(TiA1) platelets. After solidification the y platelets grow from the prior CI grains with crystallographic alignment according to ( ~ l a c k b u 1970) ~,
The a-phase lamellae that remain, subsequently transform during cooling to the C I ~phase at temperatures below the eutectoid temperature. This solidification pathway leads to the formation of a dendritic structure of a2 and y lamellae and interdendritic regions of nominally single-phase y grains which are the last to solidify from the melt. Figure 2 shows the lamellar microstructure of a Ti-45AI-lONb alloy as observed by optical microscopy. Boron additions affect the kinetics of the hightemperature phase transformations in that the borides
1600
1400
1200
1000
800 ---
Ti
30
40
50
60 att.%AI
Figure ]I Central part of the binary Ti-A1 phase diagram (McCullough et al., 1989). The base for engineering alloys ranges roughly from 45 to 49 at.%. The temperature ranges for important thermomechanical treatments are also indicated with: I - thermal treatments, 2 - forging and extrusion, 3 hot isostatic pressing
Optical micrograph of a Ti-45Al-lO~balloy with a nearly lamellar niicrostructure
may provide nucleation sites for the high temperature man et al., 1991; Bryant et al., containing j-phase-forming elements, such as Cr, MOand W, under fast cooling often exhibit alternative decomposition paths leading to more complex microstructures (~cCulloughet al., 1989; Dimiduk and Vasudevan, 1999). Ingot production of TiAl alloys closely follows that for conventional t i t a n i u ~alloys alterations. Vacuum arc melting (V most widely used practice for preparing ingots from elemental or inaster alloying additions. In order to ensure a reasonable chemical ho~ogeneitythroughout ingots of 200 to 300niin diameter the melt-stocks are usually double- or triple-melted. Plasma arc melting and induction skull melting techniques are currently being developed as cost-eEective technologies for clean melting and the production of large-scale ingots (Reed 1995; Dimiduk et al., 1998; McQuay et al., 1999; ~ c ~ u and a y Sikka, 2001). Loiigitudiiial macro-sections of the as-melted ingots are usually characterized by large columnar grains growing inwards and upwards along the direction of heat extraction. The size of lamellar grains in as-cast alloys with 46 to 48 at.% A1 is typically 100 to 500 pm de~endingon cooling rate and, thus, on ingot size. The nornials to the lamellar platelets existing within the columnar grains have an orientation parallel to the long axis of these grains. Thus, the majority of individual columnar grains have lamellae arranged in a similar orientation, which gives rise to a significant casting texture.
The solidi~cationreactions described above lead to an unavoidable micro-segregation, the extent of which depends on the nominal A1 level and the content of refractory elements (Martin et al., 1993). A1 is rejected to the interdendritic region, while refractory elements, in particular those stabilizing the j phase, are concentrated in the dendritic cores. Figure 3 demonstrates these features as observed on a 150 kg ingot of nominal composition Ti-45A1-1ONb (Appel et al., 2000b). The ~icrostructureconsists of lamellar colonies formed at the prior a dendrites and interdendritic y grains. Elemental EDX (energy dispersive X-ray) mapping (Figure 3a-c) and quantitative analysis (Figure 3d-f+) show that the interdendritic regions are rich in A1 (49 at.%) and depleted in Nb (7.5 at.%). These values compare with the values of 45 at.% Al and 10 at.% Nb determined in the dendritic cores. These concentrations vary on a length scale of about lmm. It should be noted, however, that the segregation pattern may strongly change with the A1 level and the nature of the refractory elements. The chemical gradients can be mitigated by isothermal homogenization treatments at temperatures in the (a + y), single phase a, and (a + /3) phase fields. Heat treatments in the ( a + y) field are usually insufficient to dissolve the dendritic AI segregation. Not surprisingly? annealing in the a field leads to signi~caiitlyfaster homogeiiization kinetics; liowevcr, the debits are rapid grain growth and reduced hot workability. Apparently, a suitable comp~omiseconsists of homogenization at temperatures just below the a-transus temperature, Ta,for several tens of minutes (Martin et al., 1993; Semiatin, 1995; ~ i m i d u ket al., 1998). For three-phase (a + /?+ y) alloys, homogenization inay be carried out at even higher temperatures in the (a + j) field, utilizing the jphase as a grain-size controlling agent (Martin et al., 1993). These treatments require tight control of the content of j-phase-forming elements. However, in ingots weighing over 100 kg, it is difficult to attain c h e ~ i c a lhomogeneity even after annealing for several hours at 1400 "C. Apparently, Nb, Ta and W are the most difficult alloying additions to redistribute artin in et al., 1993; Reed, 1995; Dimiduk et al., 1998; McQuay et al., 1999). VAR ingots typically contain 100 to 300ppm nitrogen and 500 to 800 ppm oxygen by weight. Ingots with oxygen levels in excess of 1200ppm are generally unacceptable for subsequent hot-working. Before subsequent processing, the ingots are usually hot isostatically pressed (HIP) in the ( a + ? ) phase field at about 200MPa for several hours in order to seal casting porosity.
620
52
Figure 3 ~ ~ ~ r e g apattern t j ~ n o f a 150kg ingot with the nominal composition T i - 4 5 A l - 1 0 ~(a-c) ~ ~ X-ray maps showing the elemental distribution o f Ti, Al and Nb, respectively, utilizin Ti-K,, AI-& and Nb-L, peaks; (d) Back-scattered electron image of the area shown in fa+), the line drawn indicates an X-ray line scan ~ C ~ Othe S S in~erdendriticregion; (e,f) variation of the Al, Ti and Nb concentrations alon the line indicated in (d). R e p r ~ d u ~ ewith d per~iss~oii of Wiley-VGH
s
~tructuralrefinement due to recovery and recrystallization is triggered by the imparted strain energy and, thus, depends on the nature of the deformed stale. In the present section a brief account will be given on deformation phenomena in ~ ~ ( T i ~ l ) - ba alloys. s e Deformation of two-phase titaniL~m aluminides is very complex due to the heterogeneity in crystal structure, interfaces and microstructure ( ~ a u t h o 1995). ~, At low m~eratures deformation is mainly confined to the y phase, as activation of deformation in the a2 phase i s difficult. Under most conditions slip of the y(TiA1) takes place by gl dislocations with the Burgers vector islocations with the 1/2(1121, respectively. In addition, order along 1 /6(1121 { 11 1) occurs (Yamaguchi and Umakoshi, 1990; Yoo et al., 1994; Appel and ~ a g n e r 1998; , Yoo, 2001). The tendency to deform by y increases with te~perature.Howmetals, there is only one rection per { 111) plane that red L1, structure. There is growing evidence that the activation of the individual deformation mechanisms requires signi~cantlydifferent shear stresses (Appel and Wagner, 1998). Thus, in terms of the von Mises criterion for plastic deformation of polycrystalline materials, there is probably a lack of enough independent slip systems, which can operate at compdrable stresses, in order to allow uniform d e f o ~ ~ a t i oIn n . unfavourably oriented grains or lamelh constraint stresses may therefore develop soon after yielding, which lead to rem mature failure. There is a marked influence of the aluminium concentration on the deformat~onmechanisms. Alumi~ium-richsingle phase y alloys preferential~ydeform by supe~dis~ocatio~~s, whereas the y phase in titaniumrich (az+?) alloys deforms mainly by ordinary dislocations and order twinning ( ~ a ~ a g u c hand i Umakoshi, 1990). The dendritic segregation of aluminium in cast microstructures (described in the previous section) may therefore lead to an inhomogeneous deformation, in that different dislocation mechanisms are operative in the dendrite cores and interdendritic regions. This leads to a signi~cantlocal variation in the imparted strain energy, which drives recrystallization, and thus imposes severe constraints on hot-working operations. The nature of the deformation mechanisms strongly depends on temperature (Appel and Wagner, 1998). D ~ ~ ~ s i o n - a s s ~climb ~ t e d of ordinary dislocatjons
becomes predominant at temperatures above 700 "6. Climb of superdislocations is apparently diffic-txlt, because these di~locations are widely ~issoc~ated (Veysi6rre and Dounin, 1995). 800°C is characterized by dynamic recovery and recrystallization, with the relative contributions depending on temperature and strain rate, As with many ~onventionalmetals, the basic ~ e c h a n i s ~o fs d y n a ~ ~ irecovery c are dislocation which results in the f o r m ~ t ~ oofn aries, At temperatures over 1000 'C Burgers vector b = (1001 may ~ o n t r i b u t to ~ these processes (Whang and Hahn, 1990). A particular feature of two-phase t i t a n i u ~ aluminides is that twins are also involved in recovery and ~ e c r y s t a l l i ~ ~ tion (Appel and Wagner, 1998; Appel, 1999). Twins generated at elevat~dte~peraturesare often fra ted and exhibit rough interfaces (Figure 4a) morphology is strongly distinguished from that observed after room-temperat~re d~formation and probably arises from reactions between twin~ing partial dislocations and matrix d~slocat~ons that have been incorporated into the twin/~atrix interface. Cl~mbof the reactant dislo~ationscan observed complex interfacial dislocatioii networks and formation of sub-boundaries. Dynamic recrystalliza~ tion is often associated with blocked slip or twinning. an In this respect the lamellar m o r p h o l o ~plays ~ important role. Work carried out by (1990) and Umakoshi et al. (1992) demonstrat~dthat the flow stress of the lamellar ~ i c r o s t r u ~ t u is r e very sensitive to lamellae orientation. The Bow stress for shear d e f o ~ a t i o nacross the lamellae is almost one order of magnitude higher than for d e f o ~ a t i o alon i~ the lamellae. The lamellar boundaries were found to be strong barriers imp~dingdislocation glide and twinh i al., 1992; ning (Fujiwara et al., 1990; ~ m a ~ o s et , Thus, the slip path of dislocations Appel et ~ l . 1993). and twins is essentially limited by the width of the lamellae. High constraint stresses occur in front of the immobilized shear bands, and more slip systems are activated than in the lamellae centres (Appel and Wagner, 1998). These stresses often give rise to recrystallization, an example of which is shown in Figure 4b. Work hardening of ?(TiAl)-bas~ alloys at low temperatures is ascribed to long-range elastic dislocation interactions, which often result in the formation of junctions and sessile multi-poles (Appel et al., 1999b). These processes certainly increase the amount of stored elastic energy and are beneficial to dynamic recrystallization. However, under these conditions a substantial
622
Processes and Phenomena
Recovery and recrystailiz~tion during high temperature (800 "C) tensile deformation of a Ti-48AI-2Cr alloy. (a) Structure of defoi-mittion twins, note the forrnatroii of dislocation networks (arrow 1) at the twinlniatrix interfaces and the emission of dislocation loops (arrow 2); deformation to failure at E ~ = 10.2%; (b) Interaction of deformation twins with lamellar interfaces. Note the i~mobilizationof the twins at the interfaces. The region designated by the arrow is shown 111 the insert at a bigher ~agnific~ttion and demonstrates an earlier stage of recrystallization; deformation to strain E =8.9%. Reproduced with permission from Elsevier Science
contribution to work hardening is also provided by dislocation dipoles and debris, which can easily be annealed out. At temperatures above 700 "C, the workhardening cliaracteristics become strongly rate dependent. The behaviour implies that diffusion-assisted climb processes are involved, which probably play an important role in dynamic recovery and subgrain ~or~~tion. In y(TiAl) dislocati~n~ o b i l i t yis impeded by a strong glide ~esistance(Appel and Wagner, 1998). The Ti-AI directional bonding is one of the most important factors ~overningthe velocity of ordinary and superdislocations, and results in a high Peierls stress (Greenberg and Gornostirev, 1988). Additional glide resistance arises from localized point obstacles, jog
dragging and defect atmospheres. The non-conservative climb processes occurring at elevated temperature^ are impeded by the relatively low di~usivityof y(TiA1) (Herzig et al., 1999). Thus, the low mobility of dislocations persists over a wide temperature range. This may influence the deformation behaviour at high temperatures where competitive deformation mechanisms, such as recovery and dynamic recrystallization are operative. For exa ple, if dynamic recrystallization refines the microstructure, then subsequent deformation may preferentially occur by grainboundary sliding. As will be discussed in section 2.3, this local change of the deformation mechanism may result in very inhoniogeneous recrystallization during hot working.
623 Little information is available on the deformation behaviour of the a2 phase in (a2+ y ) alloys. After roomtemperature deformation, local plasticity by glide of 1/3(1120) dislocations on prism planes has been recognized, whereas basal or pyramidal slip appears to be quite difficult. The high brittleness of the a2 phase at low and ambient teniperatures has therefore been attributed to there being insufficient independent slip systems. At temperatures above 800 "C, more homogeneous activation of prismatic glide and the occurrence of c-component dislocations were observed, which were thought to reduce the strong plastic anisotropy of a2(Ti3A1)(Wiezorek et al., 1997). In short, there are inany more restrictions upon possible deformatioii modes iii y(TiA1)-base alloys than for disordered metals. Such behavior influences decisions on appropriate hot-working conditions.
The non-uniform ingot microstructure and the poor failure resistance of TiAl alloys make hot working of ingot material difficult. Research in this field has made significant progress in the last several years, and the problems associated with large-scale wrought processing are now being overcome. For additional background and details, see the excellent reviews of Semiatin (1995); linayev et al., (1995); Semiatin et al., (1998). For the fabrication of structural components multistep forging with intermediate heat treatments is needed in most cases. Apparently, the most critical step is to convert the coarse-grained, textured and segregated microstructure into a more homogeneous and workable structure that is suitable for secondary processing. The range of potential temperatures and strain rates for hot-working operations of ingot material is usually evaluated through compression testing of cylinders with volumes of a few cubic centimeters followed by rnetallographic inspection. Flow curves determined on a Ti-47A1-4(Nb7Cr, Mn, Si, S ) alloy at different temperatures and strain rates are shown in Figure 5. Uniform d~formation is characterized by a cylindrical specimen maintaining a cylindrical shape, with little or no bulging. Hotworking defects include cavities, internal wedge cracks, and surface-connected cracks, any of which may lead to porous and cracked forgings. In this way workability maps for isothermal deformation can be established that define a domain of uniform deformation by the absence of the failure modes mentioned above (Nobuki et al., 1990; Davey et al., 1995; Singh et
I$=
In (E + I )
Figure 5 Flow curves of cylindrical compression samples of 18mm diameter and 30mm height tested at the conditions indicated; Ti-47A1-4(Nb, Cr, Mn, Si, B). The true stress K f was calculated from the cross-sectional cliauge under the assumption that homogeneous deformation occurred throughout the whole volume. +=ln (et- 1) 1s the true strain
ul., 1995). Accordingly, forging operations can be carried out near the eutectoid temperature with strain rates up to 10-2 s-l. The flow-stress response observed in this domain reflects the effect of dynamic recrystallization in that the flow curves exhibit a broad peak at low strains ( 8 % lO%), followed by flow softening to an ostensibly coiistant stress level at strains E = G O to 90% (Figure 5). Under these conditions the evolution of the microstructure occurs by thermally activated deformation and recovery processes, respectively, and thus depends on temperature, strain rate and strain. Likewise, the peak stress op exhibits a systematic variation with testing conditions. A detailed study of these effects was performed on a Ti-45.5A1-2.2Cr-2Nb alloy in the temperature range 1093 to 1320°C (Seetharaman and Semiatin, 1996). An average value of the strain rate sensitivity iFz = (aln a,/dln i)*,& was found to be 0.28, the apparent activation energy was Q=417kJ/mol. The effects of strain rate and temperature are often incorporated into the ZenerHollomon parameter Z , which is defined as (Humphreys and Hatherly, 1995)
Processes and P h ~ n o ~ e ~ a
624
where li is the universal gas constant. For the range of testing conditions mentioned above, or, was found to be u~iquelyrelated to . Using this formulation the peak stress data have been described (~eetharaman and ~emiatin,1995) as: up =
=:
C (E exp (Q/liT))"
or isotheri~altests the magnitude of the coefficient a, The parameters so estiinated are with the a s s u ~ p t i othat ~ diffusi~nassisted, non-conservat~ve dislocation processes are d In hot working of y-base alloys. the conditions mentioned above refinement in the microstructure by dynamic recr~stalliza~ion. Although there is a vast body of literature going back 10 years and a collection of reviews, which are d~tailedabove, the exact nature of the recrystallization and phase transformation ) aluminides is not yet processes in ( ~ ~ - 1 - ytitanium crystallizati~n of ordered structures is expected to be di It, mainly for two reasons stly, the ordered state has to ndly, there is a drastic reduction obility compared with disor~ered metals. It is only recently that inforniation on the atomic processes involved in recrystalli~ation and phase transfo~mationof TiAl: alloys has been obtained ( ~ e l ~ r i netg al., 1999; Appel et al., 1999~).This work used a high resolution electron microscopic study of crept samples that had been deformed almost eight orders of magnitude inore slowly than usual in hot working. everth he less, some qualitative information on the con~ersionof the l a i ~ e1l ~toe a spheroid~~ed microstructLir~under hot-working conditions may be deduced. The phase transformation and recrystallization processes were found to be closely related to the m i s ~ a t c hstructures of the interfaces. A prom~nent ledges feature is the formation of niu~ti~~e-heig1~t which often had ~erpendicularto the y / y i~ter~aces, rown into zones of over lOnm in width. The atomic arrangement in these zones is reminiscent of the 9R ~ ~ ~ u ~a tphase ~ r e~ r, o ~ a b having ly a slightly higher energy than the Ll0 ground state (Ernst et al., 1992). As the slabs grow further, it might become energetnucleate a new y grain. The newly ains of 10 to 20nm size are , giving the impression that the ordered state is iminediately established after grain nucleation or that nu~leationoccurred in the ordered state. The li12+y phase transfor~ationis more complex; not only must the stacking sequence be changed, but also the local chemical composition has to be
adju$ted by 1on~"ranged i ~ ~ s i o n . is ample evidence that these processes are as d with the propagation of ledges and enhanced self-diffusion along the cores of misfit dislocations (Appel et al,, 1999~). The as-forged structure is often banded, consisting of stringers of a2 particles in a finetallized y matrix (~emiatin,1995; Davey et al., 1995; Iniayev et al., 1999). In two-plnase alloys it is also c ~ m m to o~ observe l ~ ~ e lcolonies ~ a r lying in the plane of the forging and associated with shear bands (Figure 6). The m i c r o ~ t r u c t ~ evolution. ~~al has been systematically studied on a series o f binary and technical alloys with aluminium contents ranging between 45 and 54 at.% (Imayev et al., 1999). The samples were compressed at T==1000 "C to different strains ( E = 10 to 7 5 O / 0 ) , and the microst~L~cturewas assessed by quantitative metallography. The volume fraction of recrystallized grains was taken as a measure of the onv version of the ingot $ ~ r ~ c t u r eThe . d ~ ~ r eofe dynamic recrystalli~ation increases with strain, howlization occurred before ever, 110 substantial r the flow stress peak. obse~ationshave ) on a Ti-49.2~41-2. reported by Davey et 2Mn alloy, deformed at temperatures between 900 "C and 1140°C. There is also a nzarked effect o f the a ~ ~ m i n i u m concentration on the recryst~lliza~~on behaviour (Figure 7) (Imayev et al., 1999). The largest volume fractions of recrystallized grains were in allays with aluminium contents of 48, 49 and 50 at.%. As-cast,
Figure 6 Back-scattered electron ~ ~ c r o g rof ~ a~ forged ~ 1 i Ti45A1- 1 0Nb alloy, which bad been s~bjectedto single-step forging at T= 1100"C,&= 10-3s-1 to strain e=65%. Note the laineUar colony lying in the ptaiie of the forging that probably ~ e dhot existed M the casting and was not ~ e f ~ during working. Forging direction vertical
100
w e 7 ~ e p ~ n d e n of c ethe volume fraction of recrystallized grains on the alum~niumcontent of binary and coniplex developmental alloys. ~ ~ f o r i ~ a t i oati i T= 1000 "C and i.= 5 x ~ O - ' S - ~ to strain E = 75%. Reproduced with ~rmissioiiof The Minerals, Metals and Materials Society
these alloys had a duplex or nea ture with relatively small grain s Deformation was by mechanic climb of ordinary dislocations lion was mostly initiated boundaries. This combination of fine as-cast grain size and deformation processes is apparently a good neous refinement of the inicrclstruc ture. for the slow recrystalliza-
deformation may therefore ~referentially occur by grain-boundary sliding. Thus, outside the shear bands the amount of imparted strain energy i s relatively low, which makes recrystallization s l ~ ~ g i s These h. mechanisms not only result in an inhomogeneous microstr~cture,but often lead to premature failure of the work-piece. Strain localizat~onand hear-ba~d formation are therefore critical issues in hot working y(TiA1)-base alloys. Alth these topics are frequently addressed in the li re, many details of the mechanis~are not yet clear. The small fraction of in Ti-54A1 is probably a consequence r~rystalliz~tion of the particular deformation mode of As mentioned in section 2.2, deformation in these
However, the restricted ability of the s~perdislocatio~s to cross-glide and climb, described in section 2.2, apparently makes formation of sub-boundaries and r~crystalli~ation difficult. y(TiA1)-base alloys can be more easily recrystallized when small particles such as borides and silicides are 8, wher present. This is demonstrated in devel o~i ~ent of the recr~stallized fraction strain in boron- and s~l~coii-conta~niiig alloys is compared with that of the equiv (Imayev et al., 1999). It must be adm~ttedthat the relative effectiveness of the various allaying elements in
mechanisms are almost the same as in the alloys with iiear~stoichio~etric compositions. The Ti-rich alloys have a coarse-gr~ined,lamellar stru~turewith colony sizes up to 2000 pm. In these materials highly localized shear bands are often formed, which apparently
almost absent in the 45" orientation, where slip propagates along the lamellae. These observations reflect the strong plastic anisotropy of lamellar ~icrostructuresmentioned in section 2.2 and lead to the impression that the lamellar colonies observed in forgings are remnants from the lamellar cast structure that were not d ~ f o r m eduring ~ hot working, but were probably rotated into a favourable orientation. The shear bands consist of very fine, equiaxed grains and often completely traverse the work piece. Subsequent
n, o/a igure 8 Effect o f boron additions on the recrystallization behavior of a,(Ti,Al) -ty(TiA1) alloys. Dependence of the volume fraction of recrystalli~edgrains on strain E for boroncontaining alloys, complex develop~entalalloys and binary alloys with the same AL content. Deformation at T= 1000°C and i = 5 x 10-4s-1. Reproduced with permission of The Minerals, Metals and M a t e ~ ~ Society a~s
626
Processes and Phenomenu
the deve~opment of uniform microstructures during hot working is not yet clear. The beneficial effect of the boride particles may arise for two reasons. Boron is known to signi~cantlyrefine the as-cast microstructure (section 2. l), which is generally a good precondition for homogeneous hot working and recrystallization. However, it might also be speculated that particlestimulated, dynamic recrystallization occurs. This is expected when dislocations are accuiiiulated at the boride particles during deformation. At high temperatures the dislocations may be able to overcome the particles with the aid of therinal activation without forming pile-up structures. Thus, particle-stimulated recrystallizat~onwill only occur for larger particles, lower temperatures and higher strain rates. In this view, optin~izationof particles sizes and hot-working conditions are of major concern for ingot break-do~n of boron“c0ntaining alloys. The failure criteria and hot-working limits of TiAl alloys seem to be closely related to the deformation ~echanismsdescribed in section 2.2. In y(TiA1) the (1 1 1) planes serve as dislocation glide planes and twinhabit planes. Tt is now fairly well established that the cohesive energy of these planes is relatively low (Yoo et al., 1994; Appel et d.,1995; Yoo and Fu, 1998) making TiAl prone to cleavage fracture on (131) planes. Thus, blocked slip or twinning may lead to crack nucleation. Unfavourably oriented grains or laniellar colonies may therefore provide easy crack paths, so that the cracks can rapidly grow to a critical length. Stable crack growth requires that the plastic zone follows the cleavage crack, which with low dislocation ~ o b i l i t yappears d i ~ c u l t This . combination of low dislocation mobility and susceptibility to cleavage fracture limits the ability of the material to a c c o ~ ~ ~ o d constraint ate stresses and thus severely represses the hot working of y(TiA1)-base alloys.
Primary ingot break-down can be accomplished on an industrial scale by forging and/or extrusion. The technological standard currently achieved is as follows.
2.4.1 Forging Typical conditions for large-scale isothermal forging are T = 1000 to 1200°C and i= iO-3 to 10-2s--i ( ~ e ~ i a ~1995; i n , Emayev et al., 1995; Semiatin et aE., 1998; ~ i m i d u ket al., 1998; Clemens et al., 1999a; Appel et al., 20~0a;Cleinens and Kestler, 2000). As demonst~at~d in Figure 9, 5Okg billets have been
successfully forged within this processing window to height reductions of 5:1 (Appel ct al., 2000b). Edge cracking was usually minimal and surface appearance was good in all cases. The microstr~cture,which is developed under these conditions, typically appears as a partially recrystallized, lamellar structure with all the
re 9 so thermal forging for ingot break-down of a twophase TiAl alloy of co~positionTi-47A1-4(Nb, Cr, Mn, Si, B). (a) The largest pancake of 580mm diameter and 50 mm height shown 111 the figure was obtained from a cylindrical billet (270 mm diameter, 250 mm height) by single-step isothermal forging in the (a2+y) phase field under a nitrogen atmosphere (Thyssen Urnformtechnik Turbinenkomponenten CmbH, GKSS Research Center), (11) Optical micrograph showing fine”grained ~ecrystallized regions and remnant lamellar colonies with a preferred orientation of the interfaces. The forging direction is vertical. The fracture toughnesses determined for crack propagatioll parallel and perpendicular to the forging direction are indicated
Forming
627
structural inhomogeneities described above (Figure 9b) (Appcl et: al., 2000b). In an attempt to further improve the structural homogeneity, isothermal forging has been modified in several different ways. Canning and thermal insulation of the work-piece is very effective in avoiding surface chilling and cracking (~emiatin,1995). This technique expands processing windows by decreasing the minimum teiii~erature,increasing the highest strain rate, and increasing the maximum strain under which deformation without observable macroscopic failure occurs. Thus, by canned forging, a larger amount of strain energy can be imparted, which is certainly beneficial for homogeneous dynamic recrystallization. Canning results in a significant refinement of the microstructure, when compared with uncanned isothermal forging, as de~nonstratedin Figure 10b vs. Figure 10a (Appel et al., 2000a,b). However, even under these conditions recrystallization of the lamellar structure is i~complete.A more ho~ogeneousrefinecan be achieved by twoment of cast n~ic~ostructures step isothermal forging which involves an increment of static recrystallization due to an inter~ediateheat treatm~nt(Figure l0c). The so-called a-forging process consists basically of ing a metastable a microstructure practice ~omprisesbillet preheati ture high in the (a+?) phase field, followed by rapid cooling to a temperature low in the (a+?) field and subsequent forging. The microstructure obtained is fully recrystallized; however, the application of the technique i s restricted to relatively small ingots, because rapid cooling is required. 2.4.2 ~ x t r ~ ~ i o n Extrusion of TiAl alloys is usually carried out at temperatures around the a-transus temperature (typically 1250 to 1380°C). Under these conditions severe oxidation and corrosion occur; thus, the work-piece has to be encapsulated. Conventional Ti alloys or austenitic steels are usually used as can material. At the extrusion temperature the can materials have significantly lower flow stresses than the TiAl billet. This flow stress mismatch is often as high as 300NIPa and leads to i n h o ~ o g e ~ e extrusion o~s and cracking. These Problems can largely be OverCOme by an effective thermal insulation, which reduces the heat transfer from the work-piece to the can and enables contro~led dwell periods between preheating and extrusion (Semiatin, 1995). For extrusion temperatures above 1000 "C, heat losses are mainly caused by radiation;
r a pthe ~s @re 10 Back-scattered electron ~ ~ ~ ~ r o g of microstructure of a Ti-45Al-IONb alloy observed after ingot break-down utilizing different variants OC isothernil forging. The forging direction is vertical in the ~ ~ c r o g r a p (a) ~s. ~ s o ~ h eforging ~ a l at 1100"C to strain E = 65%, (b) canned isothermal forging at 1000 "C to strain E = 75%; (6)two-step isothermal forging at 1150 nc to a total strain E = 880/;,. ~ e p r o d ~ with c e ~p e ~ s s i o nfrom Elsevier Science
628
Processes and P h ~ ~ ~ ~ ~ e n a
however, by a novel can design involving radiation shields, this radiation loss can effectively be prevented (Appel et al., 1997a). Taking advantage of these concepts, extrusion processes have been widely utilized for TiAl alloy ingot break-down. The high hydrostatic pressures involved should allow forming of virtually any composit~ondesired. For example, 80 k composition T i - ~ 5 ~ l - ( ~ - l O+)X~ bwere uniformly extruded into a rectangular shape with a reduction of the cross-s~ctionof 1O:l (Appel et al., 2000a). The ingots, originally of 192 mrn diameter and 700 mm height were canned using austenitic steel, sealed in vac~~um, soaked at the intended extrusion temperature for 2 hours and then extruded. Extrusion was performed at temperatures below and above the atransus temperature at a rain speed of about 15niin/s. TiAl extrusions had a cross-section of x 30mm2 and a length of 6x1. For the can design see Appel et al. (1997a). Typical mi~rostructL~res that can be produced by canned extrusion are shown in Figure 11 on a Ti-45A11ONb alloy. The composition is representative of a new family of higli-stre~tli y-base alloys that have the potential to extend the service range of titan~um alurninides. The a-transus temperature of this alloy by differential scanning calorimetry as Ta= 1322 "C. Extrusion above I;, resulted in a refined, nearly lamellar microstructure colony size of 30 to SOyin as shown in Figure xtrusion below T, led to duplex mi~rostructures oarse- and ~ ~ e - g r ~ i banded n e d regions (Figure 11b). These structural inhomogeneities are associated with signi~cant variations in the local chemical composition, which are manifested at a length scale comparable to, or slightly smaller than, that of the ase 12). This observatioii provides that the dynamic recrystallization ng is strongly affected by local composition. The coarse-grained bands probably originate from the prior Al-rich interden~riticregions, where no a2 phase was present. Thus, grain growth llization is not impeded by particles contrary, the fined-grained bands or ies are formed in Al-depleted core regions of the dendrites, The structural and chemical homogeneity of extruded products can be improved by utilizing higher reductioii ratios or multi-step processing, the latter, constraints on the geometry have been made to ovcrcome ual-channel angular extrusion 1981). In this method the
ure 11 Back-scattered-electronimages of a Ti-45A1- lONb alloy extruded to a 7:3 reduction. (a) early ~ a r n e ~ ~ a r inicrostructure observed after extrusion at 7.,+ AT; (b) Duplex structure with a banded morphology observed after extrusion at T, - AT. Reproduced with ~ e r ~ i s s i oofnWileyVCH
work-piece is extruded through an angular channel, which provides the opportunity to use the channel angle to adjust the imparted strain energy. Other advantages of the technique are moderate working pressures and tlie ability to control crystallographic texture and meclianical anisotropy during multi-pass processing by judicious rotation of the work-piece between the passes. Large-scale processing utilizin ECAE technique, however, is still in its inf'mcy. Hot-working operations are usua~ly followed by thermal treatments to reduce internal stresses, to eliminate the dead-metal zone in forged products and to stabilize the microstructure. Much effort has also been made to produce ~ n e - g r a i n efully ~ ~ lamellar microstructures which exhibit the best balance of niechanical properties. A coi~pre~ensive review of
629
ure 12 Structural and chemical i~ihomogenel~ies of an Ti-45Al-1ONballoy extruded at Ta-Al' to a 7:l reduction. (a*} X-ray maps showing the elemental distribution of Ti, A1 and Nb. (d) Back-scattered electron image of the duplex microstructure with a banded morphology of fine- and coarse-grained regions; (e, t) variation of the Ti, A1 and Nb coiicentrations along the line indicated in (d). It should be noted that the chemical inhomogeneities occur at a length scale comparable to those of the cast material, see Figure 3
630
Processes and P h e ~ o ~ e n a
the recent advances in this field has been given by imiduk et d.(1998).
The refined microstructure after hot working generally results in significant strengthening compared to cast material (Kiln, 1995; Iinayev et al., 1995; Martin et aZ., 1995). The increase in yield stress can be rationalized in terms of dislocation/gr~in boundary interactions, although quantitative descriptions by Hall-Petch relations are often difficult due to the complexity of the microstruct~res.Figure 13 shows the dependence on temperature of density-compensated yield stress for forged and extruded y(TiA1) alloys. The specimens were annealed at 1050°C for 2 h in air, to release internal stresses and stabilize the microstructure, and were then furnace cooled. For comparison the diagram also contains data for nickel-base superalloys and a conventional t i t a n i u ~alloy. Extremely high tensile yield stresses in excess of 10OOMPa were obtained on Ti-45A1-(5-l0)Wb derivative alloys after extrusion with a reduction of 7:l (Appel et al., 2000a,b). The main advnntage of wrought processing is the improvement of the ductility and reliability of the material. In duplex materials a finer overall grain size, irrespective of the volume fractions of y grains and colonies, generally leads to a higher ductility. Despite the structural and chemical inhomogeneities of materials processed as described above, plastic tensile strains generally remain above 2 to 3% with good reliability. For example on the alloy variant TNB-V (Figure 13) a plastic tensile elongation at room temperature of r=2.5% was determined with a fracture stress of about 1 1 0 0 ~ P a(Appel et al., 2000a,b), This combination of roo~-temperaturestrength and ductility is the best ever reported on y(TiA1) base alloys. Thus, wrought alloys of this type can be attractive alternatives to the heavier, nickel-base superalloys in certain ranges of stress and temperature. Although little documented in the literature, forgings or extrusions of y(TiA1) alloys often exhibit significant mechanical anisotropy . The largest pancake shown in Figure 9a was assessed by comparing the strength properties in the axial and radial directions, respectively. Within the pancake the banded structure is radially oriented in correspondence to the material flow during forging. The tensile strengths determined for ~e fo r mat~oinn the axial direction are almost 10% higher than those in the radial direction. However, the larger tensile elongations are generally found for
300
50 0
200
400
600
800
1000
1200
q- W) Figure 13 Temperature dependence of density-adjusted yield stresses for forged and extruded ~(TiAl)-base titanium aluminide alloys. (1) Forged Ti-47A1-2Cr-0.2Si,n e a r ~ ~ a r n ~ ~ microstructure; (2)Extruded Tii-45Al-(5-1O)Nb, duplex microstructure; (3) TNB-V, Ti-45A1-(5-1O)Nb-t- X, duplex rnicrostructure. For comparison the values for nickel-base superalloys and a conventio~al t i t a n i u ~alloys are given, with (4) IMI 834, (5) R6ne 95, (6) Inconel718, (7) IN 713 LC
samples with a radial orientation. The mechanical anisotropy is more pronounced with respect to fracture toughness. Relatively low values of KIc= 10 to 12 MPadm were determined for crack propagation in the radial direction parallel to the plane of the pancake. This compares with values of KIc= 16 to 20MPaJm for crack propagation in the axial direction (Figure 9b). The effect was attributed to a relatively high volume fraction of ~ e ~ n a lna~el t l ae that had a preferred interface orientation parallel to the plane of the work-piece. In view of the susceptibility of y(TiA1) to cleavage, these colonies may provide easy paths for crack propagation in the radial direction. A similar anisotropy of the mechanical properties has been recognized in extruded material (Appel et al., 2000a).
The considerable effort that has been expanded in modeling of wrought processing has provided valuable insight into identification of' process variables, effects of heat losses and development of secondary tensile stresses in the peripheral regions of the work-piece (Semiatin and Seetharaman, 1997). The authors claimed that their constitutive relationships describe large-scale forging sufficiently well, for relatively slow isothermal forging. For instance, it has been predicted
Forming
that for crack-free forging the peripheral tensile stresses developed in the work-piece should be kept below 200 MPa, depending on alloy composition and microstructure. Unfortunately, it is often the case that the particular parameters that are being controlled in a given experiment represent only a fraction of those iiifluencing hot working; thus, comparison of various sets of experimental results is very difficult. In this category fall the effects of lubricants, surface finish of dies and work-piece, axial temperature gradients, and can design.
roc~ss~s As mentioned above, improved ingot-breakdown procedures lead to refined microstructures. As pointed out by Semiatin (Semiatin, 1995), fine, uniform microstructures lead to improved workability due to reduced flow stresses and increased fracture resistance and, therefore, are beneficial for sheet rolling, superplastic forming and isothermal closed-die forging. ~en.erally,for sound processing of medium-sized or small components the require~~ents on the specific forming process variables, e.g. strain rate, temperature, maximum strain etc., are more demanding compared to the larger-scale wrought-processing methods mentioned above, thus leading to a relatively narrow processing window. These demands arise predominantly from the strong dependence of flow stress on strain rate and temperature of y(TiA1) base alloys. The temperature sensitivity of these alloys gains particular importance for components with thin sections, where heat loss due to radiation during transfer operations and chilling due to heat conduction into the tools d u r i n ~n o n ~ ~ s o t h e ~ora l near-isothermal forming operations is unavoidable aiid, thus can lead to inhomogeneous flow behavior and damage on a microscopic scale, or gross failure. Another important parameter is texture, which influences both the forming behavior and the mechanical properties of a final component. Only limited work on. texture evolution during tliermomechanical processing of y(TiA1)-base alloys has been reported. Since most of the texture analysis has been done on y(TiA1) sheet material, texture-related topics will be discussed mainly in section 3.2. 1x1 the following sections secondary forming processes for y(Ti~l)-basealloys are described with special emphasis on rolling of sheet and foil, subsequent sheetforming techniques, as well as on isotherinal and nearisothermal forging.
63 I
Especially in Japan and the U programs have been in.itiated in focusing on rolling techniques, such as so thermal rolling in the former instance (Fujits~~na et al., 1993; Morita et al., 1993) and conventional hot-pack rolling in the latter instance ( 1995). Undoubtedly, the fuiidai~ent~l research carried out by Semiatin and co-workers (Se~iatinet al., 1990, 1991, 1994; S e ~ i a t i nand Seetharaman, 1994, 1995) established a sound understandiiig of the iiiteraction of process variables and the evolution of ~icrostructures as well as an insight into failure mechan hot-pack rolling. In Europe, driven by the the Sanger programs, y(TiA1) sheet-rolling activities were intensified in the early nineties with the aim of establishing a process feasible for industrial equipment and of providing scale-up potential. In the course of these activities Clernens and co-workers developed a so-called advanced sheet rolling process (ASRP) which allows processing of large y(TiAl) sheets on a coiiventional hot-rolling mill (Glemens et al., 1993, 1995). In general, the requirements on the rolling process for y(TiA1)-base alloys are niainly linked to an accurate choice of the processing variables, which have to be adapted to the specific properties of the rolling stock (the alloy and its tliermomechaii~calhistory) as well as to the rolling equipment used. In the layout of rolling schedules, besides control of strain rates and strain per rolling pass, the main issue was shown to be temperature control s semi at in et al., 1991; Semiatin and Seetharaman, 1995: Clemens et al., 1995). Temperature control within an extremely narrow regime is important, first to optimize flow characteristics of the y(TiA1) alloy (within the whole set of process variables) and to avoid microstructural dainage such as wedge cracks and cavitation, but secondly it is also i ~ p o r t a n tin achieving o p ~ i ~ u ~ microstructural uniformity. Temperature coiitrol is needed not only for the rolling temperature itself but also for preheat temperatures (Semiatin and Se man, 1995). Temperature transients during arise from chilling effects due to contact of WO and rollers during transfer operations (~emiatinet al., 1991), and pronounced te~perature~radientsalong the rolling direction of the work-piece evolve due to finite processing times (low rolling speed) (Semiatin et aE., 1991; Clemens et al., 1993). From these considerations, is0 thermal rolling of y(TiAl) sheet as developed by Fujitsuna et al. (1993) and M seems to be the process of choice.
632
Processes and Phen ommu
considerations, cost issues and lack of availability of large, industrial-scale isothermal rolling equipment are making more conventional hot pack-rolling techniques as developed in the USA and Europe more attractive. Again, the most challengi~gtask with hot pack-rolling is temperature control. Therefore, techniques have been successfully developed to provide accurate heat control throughout tlie rolling process by canning the workpiece (Semiatin et al., 1991; Clemens et al., 1993). In order to ensure the homogeneity and quality of the sheets it is impo~tantthat the as-rolled y(TiAl)/ a,(Ti3Al)-phase distribution, which very sensitively depends on the rolling temperature, should be uniform over the whole sheet area. Figure 14 shows the y(TiAlj/ a2(Ti,Al)-phasedistribL~tiomsat different positions of a l m long Ti-48Al-2Cr sheet rolled at temperatures in the middle of the (a+y)-phase field (Clemems et al., 1994). Note that there is no significant change in phase distribution between the ends. Froin this finding it is evident that, even when using conventional rolling equip~ent, it is feasible to minimize temperature gradients and, therefore, to provide quasi-isothermal conditions. This is further confirmed by the
absence of any edge cracks over the entire sheet length. In addition, Figure 14 shows the y(TiAl)/a2(Ti3Aljphase distribution of a 700 mm long Ti-48A1-2Cr sheet that was rolled under n o n ~ Q p t i ~ i rolling z e ~ conditions. If one compares the phase distri~utionof this sheet to that of the ASRP sheet, it is obvious that during rolling under non-is~thermalconditions substantial heat loss takes place, which in turn limits the maximum sheet length. Employing these methods, treme~idous s ~ c c ~ s s e s have been achieved in near-i~othermalhot ~ack-roll in^ of y(TiA1) alloys, allowing upscaling of sheet dimensions as well as providing excellent microst~uctural homogeneity and m ~ c h a n i c ~properties l (Clemens et al., 1999a,b). Currently, y(TiA1) alloy sheets in dimensions of about 800 x 400 x 1.Ornm can be routinely produced, The dimensions of the largest y(TiAl) sheets that have been rolled so far are approximately 1900 x 500 x l.Oimn (Clemens and Kestler, 2000). Further upscaling of sheet dimensions seems to be feasible as far as the rolling process is concerned. However, factors that limit both sheet dimensions and microstructural h o ~ o g ~ n e i may t y depend on the
y ( ~ i A ~ ) ~ ~ ~ ( phase T i ~ A distribut~on l) within Ti-48A1-2Cr sheet after processing under q u a s ~ - ~ s o t h erolling ~a~ conditions and non-isothermal rolling conditions (as-rolled microstructures). Note: the a,(Ti,Al) appears brighter than the y(TiA1) phase. (SEM-Back-s~~tter~d-electroii mode (BSE)). Reproduced with permission of The Minerals, Metals and Materials Society
633
Forming particular prematerial used, as is generally true for other forming processes. The availability of prematerial for rolling in adequate dimensions and also homogeneous in chemical composition is manda~ory for upscaling of sheet dimensions. Although wroughtprocessed y(TiAl) prernaterial is available in the forin of forged pancakes or extruded bar material (see section 2.4), the use of these materials is considered to be relatively costly because of the expense of ingot break-down and the low yield in preparing rectangular rolling slabs from circular forged pancakes. For extruded bar material, coiiip.plicatedcanning-decanning operations have in the past made this p r e ~ ~ t e r iroute al economically unattractive for further sheet rolling. A factor limiting sheet quality is undoubtedly an improper, or at least not homogeneous, starting micrQstructure. The importance of close control of the AI-content must not be underestimated and can easily be figured out when the rolling temperature relative to the a-transus temp~rature(and thus the width of a potential processing window) of a specific y(TiA1) alloy is considered. As stated above, the processing temperature relative to TE and its interaction with other ~rocessingparameters govern the formability and the resulting microstructure o f rolled y(TiA1) sheet. Translatin~the Al-content variation into a relative temperature shift, fluctuations of 1 at % in A1 content would lead to an uncertainty in TEof about 25°C (as estimated from the binary Ti-A1 phase diagram) and therefore cannot be tol~rated. Even these small AI fluctuations (which normally occur in larger ingots) would lead to a scatter in the mechanical properties from sheet to sheet and also to varying, and thus unpredictable, sheet forming behavior (see also section 3.4). In this context, there is a certain advantage of the powder~metallurgy (PM) route over the ingot route (IM). However, recent achievements in ingot production have shown that a close control of the Al-content and other alloying elements is both possible and reproducible (Clernens et al., 1999a; Clemens and For a d ~ ~ ~ cthermal ed protection systems and honeycomb structures, thin sheets and foils with a thickness of -Sopin are required (LeHolm et al., 1999). Currently, thin foils with a thickness down to 150pm have been rolled (Glemens and Kestler, 2000). However, for processing of y(TiA1)-base alloy foils to industrially relevant dimensio~s,the requirements for rolling are more challen~ingthan for sheet processing. A critical point i s the surface quality that is developed during the rolling process. In the case of foil
processing, surface quality must be good enough that any final treatments become before industrial production o can be considered, additional research and development is required. All processing parameters must be defined and accompanying quality-control regulations established with w~ll-definedtest standards for foils. Because of the problem in achieving fully lamellar microstructures with colony sizes well below 50 pia, it is anticipated that thin sheets and foils will have to be used in the fine-grained condition. Thus, the choice of 1 a suitable alloy is a prerequisite in order to achieve sufficient mechanical pro~erties,e.g. creep strength.
3.2 In this section, the mechanical properties of ysheet material in hot forniin mechanical properties will be f resented mainly in terms of tensile data and data obtained from strainrate step tests, which give an indication of flow-stress/ strain-rate sensitivity and ductility at elevated ternperatures. The influence of microstr~c~ureon hot forming is noted with emphasis on fine-grained globular microstructures. (Transformed lamellar microstructures have been shown to exhibit a relatively poor formability due to their relatively large grain sizcs and anisotropic flow behavior.) prematerial routes (starting fro on damage incurred during forming as influence of texture will also be described. Figure 15 summarizes the tensile rolling direction of PM and 4(Cr,Nb,Ta,B) sheet material with mary-annealed, microstructure. The dependence of yield strength and ultimate tensile strength (U temperature generally exhibits characteristics to those of bulk material with similar microstr~c~ure. It is important to note that the flow stress is relatively low at temperatures above the stresses, in combination with high elongations to fracture, imply good hot-forming capabilities of yTiAl sheet material (Glemens et al., 1999b). Tensile ductility is also characterized by the which depends on alloy c o ~ p o s i t i o n~icrostructure ~ and strain rate. The BDTT may be rationalized in t e r m of the mechanisms described in section 2.2, that means by the occurrence of di~usion~assisted dislocation processes and enhanced mechanical twinning. Generally, the BDTT decreases with size (Koeppe et al., 199s). Above the
634
Processes and Phenomena
T, "C operties in the rolling direction of (1000 *C/2h/FC) PM Ti-46.5Al-4(Cr, ashed line: high-temperature plastic fracture eionga~ionof an IM sheet with identical composition (see text). engineer in^ strain rate: 8 x 10-5 s-'
rolled at a te~peraturewithin the (a+?)-phase field and subsequently annealed at 1000 "C for 2 hours shows the following texture components: (01 1) (211) (brass-like), (1 l2>,(111)(copper-like), (123>(632) (8, and ~OlO>(lOO) (cube). A special feature of the cube component is that the c-axis of the y(TiA1) cell is aligned in the sheet plane, perpendicular to the rolling direction. The influence of texture on tensile and creep properties in both the rolling and transverse directions is reported by Chatterjee et al. (1999), Kestler et al. (1999) and Bartels et al. (1997, 2000). The influence of texture on the tensile properties of a PM Ti-47-Al) sheet is shown by Figure 16 where the yield strength in the rolling direction (RD) and transverse to the rolling direction (TD) is plotted as a function of temperature (Kestler et al., 1999). I ,..
.
. .
i n c r ~ ~ s due e s to m e c ~ ~ n i ctwinniiig al and dislocation T'T of primary-annealed PM
sheet inaterial is around
15). At 1000 "C, wliich corre1 hot-forming temperature for y(TiA1) sheets, the maximum elongation is approx. 70%. This behavior has been found for all y(TiA1) sheets that were produced from gas-atomized powders lernens et al., 1997; Clemens et al., 1999a; Uolton et , 1997). This behavior results from the fact that the material develops internal cavitation at relatively low strains, which i s speculated to be thermally induced. 'The cavities grow rapidly and interlink perpendicular axis, producing early failure. TEM i~vestigatio~s have shown evidence for micropore segregation at gram boundaries (Appel et al., 1997b). lt remains d e b a t a ~ ~however, e, whether other factors such as stress concentrations at grain boundaries or inclusions could lead to similar effects. In comparison, the high-teinperature fracture elongations of an IMsheet with identical com~ositionare included in Figure 15. For sheet rolled from a forged ingot, the onset of cavitation is shifted to higher elongations. For e x a ~ p l e ,at 1000°C the fracture elongation of the sheet is twice as high as that of the PM sheet prox. 15O0/;,). The texture of y(TiA1) sheets in the as-rolled condition and after subsequent heat t r e a t ~ e n t shas iivestiga~e~ thoroughly by Koeppe et al. (1997) artels et al. (1997). For example, sheet material
c
I
Figure 16 (a) UTS (open symbols) and 0.2Yo-yield strength (filled symbols) vs. t~mperatureplots for PM T i - 4 7 A l - 4 ( ~ , Wln, Cr, Si, B) in primary annealed condition. Squares: tensile axis parallel to RD; Triangles: tensile axis parallel to the transverse direction. Note the anisotropy in UTS and yield strength at temperatures between 600°C and 900°C; (b) Elo~gation-to-fracturevs. tempe~aturefor PM T i - 4 7 A l - ~ ( ~ , Mn, Cr, Si, B) in primary annealed condition. Squares: tensile axis parallel to RD; Triangles: tensile axis parallel to the transverse direction
Forming Generally, the yield strength vs. temperature curves follow a trend similar to that shown in Figure 15 for PN Ti-46.5-Al-4(Cr7Nb,Ta,B) sheet material. However, pronounced anisotropy an yield strength is observed for temperatures between 600 "C and 900°C. Specimens with the tensile axis parallel to TD are stronger than the RD-oriented specimens. At temperatures =. 800 "C the anisotropy decreases and vanishes at T > 1000 "C. From texture analysis (Kestler et al., 1999) it is assumed that this temperature-dependent anisotropy is related to the modified cube texture mentioned above. Bartels et al. (1997) concluded that, due to the prevailing texture, plastic deformatioii is facilitated for tensile specimens loaded parallel to RD because the Schmid f s t o r for ordinary (1101-dislocations in y-grains is close to 0.5. In specimens loaded parallel to TD, however, the y(TiAl) unit cell in the majority of grains is oriented with the c-axes parallel to the loading direction; thus, the Schmid factor becomes close to zero for ordinary dislocations. Although, the Schmid factor for (0I 11-superdislocations is nearly 0.5, they are rarely activated due to their higli critical resolved shear stress (CRSS) compared to that of ordinary dislocations. The observed anisotropy is increased by mechanical twinning which, under tensile stress, can be activated in (1001 direction (i.e. RD), but not in (0011 direction (i.e. TD) (Bartels et al., 2000). The vanishing anisotropy in yield stress at temperatures >90O"C might be caused by increased grain boundary sliding and enhanced plasticity due to dynamic recrystallization, where texture-related effects play a less dominant role (Clemens et al., 1995). More detailed information concerning texture evolution and its complex influence on the deformation behavior of y(TiA1) sheet material is given by Bartels et al. (1997, 2000) and Schillinger et al. (2000~. In summary, secondary y(TiA1) sheet-forming processes should be carried out well above 900°C to avoid anisotropic flow that could lead to local thinning of sheet and, therefore, to early failure during forming. However, in PM y(TiA1) sheet, tliermally induced porosity might become a problem at elevated forming tem~eratures.Therefore, careful design of the hot sheet-forming process is mandatory in order to avoid undesirable geometric eAFects as well as damage on both macroscopic and microscopic scales. 3.3 As pointed out in section 2, significant understanding of the flow behavior and its interrelationship to
635
microstructure relevant to hot for ing of y(TiAl)-base alloys was gained during the last decade. Most of the published work was carried out on ingot breakdown routes and on the associated kinetics of microstructural refinement and chemical homogenization of the cast microstructure. Regarding practical forging of y(T~Al)-basealloys, very little is found in the open literature. However, activity in the hot forging of y(TiA1) increased s~bstantiallyduring the nineties and concentrated on components for aero-engines (Brooks et al., 1998; Millet et al., 3999; Appel et al., 2000b; Tetsui et al., 1997) and high~performanceautomotive valves (Kirn, 1994; Knippsclieer et al., 1999). As is clear from section 2, conventional forging is not appropriate for y(TiA1)-base alloys due to cold tools and high strain rates; therefore most of the components made thus far were forged isothermally. An example of the combination of different hotforming processes is the work of ~ ~ i p p s c h e eetr al. (I 999), where non-isothermal multi-step extrusion, hot bulging and isothermal near-net-shape forging were employed to manufacture y(TiA1) valves. The individual processing steps are depicted schematically in Figure 17. These valves were entirely produced using industrial production equipment. The valves had homogeneous, fine-grained microstructures and have been successfully tested in engines. However, as in primary wrought processing, microstructural refinement during secondary hot~working operations is one of the key concepts to improving processing reliability and optimizing mechanical properties. Isothermal closed-die forging of high-pressure compressor blades was reported by Appel and coworkers (Appel et al., 2000b). The use of extruded y(TiA1) bar with a refined ~ i c r o s t r u ~ t u as r e starting material reduced the susceptibility to cracking during hot working. More than 200 blades were forged in this study using the optimum processing parameters obtained from laboratory compression tests. Insight into the microstructural processes and the correspond= ing flow behavior during forging of y(TiA1)-base alloys and Brooks et has been reported by ~ i ~et ~al. e~1999) t al. (1998), wha studied isothermal forging of aeroengine airfoils. These authors developed a flow-stress model which was incorporated into finite element code, allowing specification of the forging process itself, the microstructural evolution, and the ~ow-softening behavior of y(TiA1) during hot working. The microstructural evolution is represented via an internal state variable in a coiistitutive equation. This model makes use of the Zener-Hollomon parameter, the peak
636
Processes and P ~ e n o ~ e ~ a
F i 17:~ Basic ~ steps ~ of ~ the fabrication process of T i A l ( ~ o , ~ automotive i) valves based on ingot metallu~gy and ther~nomechan~calprocessing - hot extrusion, hot bulging and quasi-i~o~hermaldie forging (Knippscheer et al., 1999). ~ e p ~ o d u c with e d permission of Wiley-VCH
stresses, and the steady-s~~te stresses of individu~lflow curves, which can be readily measured in laboratory compression tests. The model allowed prediction of the resulting microstructures in different sections of the airfoil in terms of the volume fraction of the recrystallized y-phase, in quite good agreement with experimental results obtained from quantitative metallography.
Superplastic forming (SPF) is widely used in the aerospace industry to manufacture co~plex-slia~ed parts from Ti-base alloys. The substitution of y(TiA1)base alloys for Ti-base alloys (and also for Ni-base alloys) in those compoiients can potentially provide better hi~h-temperatureca~abilities,improved stiffness,
Forming
637
and weight reduction ( eppe et al., 1995; Clemens et al., 1997; KestXer et , 1999; Das and Clemens, I00 1999). Fundamental studies of hot-deformation behavior showed that fine-grained y(TiA1) sheet materials have some characteristics of superplasticity, i.e. in terms of strain-rate sensitivity (m> 0.3) and high elongations (up to 800% (Lombard et. al., 1995)). For example, Figure 18 shows the strain-rate sensitivity exponent and the corresponding peak stress as a function of strain rate and temperature for PM Ti-47Al1 sheet material as determined from tensile strain-rate step tests (Kestler et al., 1999). Microstructural investigations on specimens which have been tested under different conditions have revealed that conventional y(T~Al)-basealloys do not obey the classical theory of supe~lasticity(Koeppe et al., 1995). Grain boundary sliding accommodated by di~usion-coiitrolleddislocation motion is the essential deformation mode, but the grain-size dependence of superplastic properties holds only at the beginning of deformation. After an incubation strain, a steady-state grain size is produced that is related to the defor~ation conditions. The steady-state grain size can be related to deformation strain rate and temperature through the eter (Koeppe et al., 1995). ize is smaller than the steadysening is observed; whereas grain refinement due to dynamic recrystallization takes ~ i ~ u 18 r e (a) Strain-rate dependence of the true peak stress place, if the initial grain size is larger than the steadyof PM Ti-47A1-4(Nb, Cr, Mn, Si, B) sheet material in the primary annealed condition, obtained froni tensile strain-rate state one. step tests between 950 "C and 1050 "C. Tensile axis parallel to From Figure 18 it might be speculated that y(TiA1) RD, (b) Strain-rate sensitivity (m)of PM ~ i - 4 7 A 1 - 4 ( Cr, ~, sheet materials potentially exhibit the properties Mn, Si, B) sheet material in the pr~maryannealed condition, required for industrial superplastic forming processes obtained froni tensile strain-rate step tests between 950 "C and (SPF), i.e es >0.3 associated with low stresses 1050°C. Tensile axis parallel to RD. Reproduced with permission of The Minerals, Metals and ~ a t e r i a l sSociety even at ternperatures below 1000"C. The latter is important, because SPF-facilities designed for SPF of Ti-alloys could then also be used for SPF of y(TiA1)-base alloys. failure takes place (Lombard et al., 1995; Clernens et From the Woodford correlation between the m value and the tensile ductility, fracture elongations of al., 1999a). Metallographic studies have revealed that > 200% might be expected for y(TiA1) alloys showing grain boundary separation takes place in the early stra~n-rate s ~ n s i t i v ~ tini ~ ~the range of 0.4-0.7 stages of deformation and leads to the formation of (Lombard et al., 1995). However, at 1000"C maximum isolated cavities during further deformation. Cavitaelongations of only 180% have been measured for IM tion starts at relatively low strains in PM sheet material Ti~47A1~2Cr-O.~Si sheet material; for PM sheets, the (Clemens et aE., 1999a). Failure is caused by subseelongations reported are even more limited to about quent growth of these cavities. At present, no complete explanation for the degraded SPF behavior of PM 100% (Clemens et al., 1999a). Tlvs difference in fracture elongation between PM and IM sheets was also found sheet material can be given. However, it can be assumed that in superplastic deformation the presence in high-temperature tensile tests. The reason for these low ductilities lies in the development of moderate-toof thermally induced microporosity eases grain boumdand consequently enhances the extensive cavitation., and thus f r a c t ~ ~ ~ ~ - ~ ~ n t rary ~ l l eseparation d
t
I
638
Processes and P h e n o ~ ~ ? ~ a
nucleation rate for microvoids (Clemens et al., 1999a,b). SPF experim~ntson a laboratory scale have shown the for~dbilityof y(TiAi)~basedalloy sheet materials. For example, gas-loaded, biaxial-forming tests have been conducted on 1M sheet material and a maxiinum true strain of 6 0 0 * was ~ realized (Clemens et al., 199’7). Generally, the best results are obtained for y(TiA1) alloys when those SPF techniques, used for other superplastic materials showing deformation-enhanced cavitation, are employed. For example, in utilizing driver-sheet techniques or by applying a ‘backSPF, the onset of cavitation is shifted to higher elongations (Clemens et al., 1999a), a fact that is highly beneficial in the case of PM sheets. At present, the main challenge is to establish a large-scale production of optimized y(TiA1) powders which have little entrapped atomization gas and non-metallic inclusions (Clemens and Kestler, 2000). (See also the chapter by Seetharaman and Semiatin in this volume.) In several technology demonstration programs, large sheet-based parts were successfully manufactured via SPF and/or more conventional ho~-forming techniques. For example, in the High Speed Research program, y(TiA1) sheet was used to manufacture for the so-called divergent flap (~artolotta e, 1999). The divergent flap is composed of loy box beams supporting a series of subelements made of y(TiA1) sheet as shown in Figure 19. The subelements were manufactured by F Goodrich Aerostructwes Group (USA) using Ti-46.5Al-~(Cr,~b,Ta,B) sheets with a thickness of 0.635mm. The struct~resshown in Figure 19 were fabricated using production equipment and at production fabrication rates (Bartolotta and Krause, 1999; Das and Clemens, 1999). The forming and joining processes were developed at NASA Glenn Research Center and Pratt $t Whitney. The corrugations were hot formed at relatively low temperatures in an argon environinent using standard tooling. The parts were assembled by vacuum brazing with a TiCuNi filler alloy (~artolottaand Krause, 1999; Das and Clemens, 1999). The overall dimensions of the part are approximately 66 mm (height) x 146min (width) x 610 mm (length). In the mid~nineties,within the German Hypersonic Technology Program, the feasibility of manufacturing y(TiA1) hot-structure components via SPF has been inve~tigated.At the end of 1995, a panel structure was fabricated out of wrought y(TiA1) sheet. ~ a nuf act ur e and asse~~ibly ofthc panel components as well as the results of a structural stability test conducted on the
~ i ~ u 19 r e y(TiA1) subelements of the divergent flap concept developed during the High Speed Rese~rchprogram with salient features of the full-scale flap (Bartolotta and Krause, 1999). The parts were hot formed frain Ti-46.581-4(Cr, Wb, Tay B) sheet. Overall dimension of the structures: 66mm (height) x X46rnm (width) x 6 1 ~ (lengt~). m ~ Sheet thickness: 0.635 mm. Courtesy of NASAjNASA Glenn Research Centre
y(TiA1) panel shown are summarized in Clernens et al. (1996). Recently, a German ma terials technology program (MaTech) has been established to demonstrate the feasibility of manu~dcture of hollow, low-pressure turbine (LPT) blades out of y(TiA1) sheet material by using SPF and diffLision~bondingtechnologies (Kestler et al., 1999). More ~nform~tion about tlie successful application of SPF and other hot-fori~ningtechniques on y(TiA1)-bascd sheet material is sumnnarized in Clemens and Kestler (2000).
Ordered intermetallic phases exist with a large variety of lattice structures, often involving complex unit cells.
Forming However, most compounds are too brittle to deform significantly, which makes ingot conversion and homogenization and refinement of microstructLires d i ~ c ~ In l t .many cases there are also serious problems in the large-scale production of ingots. For such materials, powder-processing routes thus play an elative to conventional ferrous and non-ferrous alloys, successful forming of y(TiA1) requires more precise control of forming variables and careful correlation of microstructural and chemical homogeneity. Some successes with other interinetallic conipounds using this approach are summarized below and in the chapter by Lipsitt et al. in this volume.
2s t ~ ~ c t ~ ~ e ) Hot extrusion of canned FeAl powders at 900°C and reduction ratios of 8:1 and 12:1 seems to be the niost effective method for obtaining fine-grained material. Canned extrusion was also demonstrated on small ingots (Gaydosh and Crimp, 1985). FeAl sheets were successfully produced from water-atomized powders (Deevi et al., 1999). The powders were roll compacted with a polymeric binder, then de-biiidered and sintered in vacuum. The sintered sheets of 0.66 mm thickness were rolled in several passes down to a final thickness of 0.2mxn.
above 550 "C)
Powder compacts and ingot castings were extruded and forged at temperatures between 900°C and 1200°C. These wrought products could be subsequently rolled to sheet at significantly lower temperatures of 500 to 6 0 0 " ~(Sikka et al., 1991; Sikka, 1993; Sun et al., 1993).
Cast Ni,AI alloys exhibit appreciable ductility at ambient and i n t e ~ e d i a t etemperatures. Hot working and recrystallization are sensitive to alloy composition and the cast microstructure. Forging and canned e x t r ~ s ~ ohave n been performed between 1050 and 1200°C on alloys containing less than 0.3 at,% Zr, which is added to improve the high-temperature strength (Sikka, 1992). Several alloys (e.g. IC-50) are even cold-workable in the as-cast condition. The development of recrystallization and texture during cold rolling and annealing has been investigated in a
639
series of detailed studies (Ponge and Gottstein, 1998; Escher et al., 1998; Escher and Gottstein, 1998). The fine microstructure established by wrought processing leads to significantly improved strength properties and allows superplastic forming.
Multi-step forging at teinperatures between 650 and 1050 "C was successfully performed on a single-phase a2(Ti3Al) alloy with the composition Ti-25Al and a lamellar coloiiy size of 200 to 300pm. This processing resulted in a nearly fully-recrystallized microstructure with an average grain size of 0.3pm. This grain refinement leads to an appreciable increase of the room temperature yield strength and ductility and enabled superplastic d e f o ~ a t i o n(Salishchev et aE., 2000). Ti,Al-derivative alloys can easily be processed by forging operations, provided A1 contents do not exceed 25 at. % . Alloy compositions of technical sign~ficance are, e.g., Ti-24A1-11Wb and Ti-25Al-IO~b-3V-1~0. Processing windows for forging, rolling and superplastic forming were clearly identified in terms of strain rates and temperatures, which allow the fabrication of complex semi-finished products or components (Banerjee et al., 1993).
The processing technologies and the understanding of composition-microstructure-properties relationships of today's y(TiA1)-base alloys have been developed in the last decade to an extent that semi-finished products and components can be man~facturedwith proi~ising engineering properties. Therefore, y(TiA1)-base alloys can be considered as the most important and advanced candidates amongst other intermetallics for applications in aerospace, automotive and related industries. Industry appears to be on the threshold of significant use of this new class of structural materials. In particular, all major aircraft and automotive engine manufacturers are advancing the qualification and introduction of y(TiA1) components. y(TiA1)-base alloys can be processed using conventional metallurgical methods - a factor, which is necessary for these specific materials to be economically competitive with other state-of-the-art materials. The processing of y(TiA1) alloys via ingotand powder nietallurgical-routes on an industrial scale has been successfully demonstrated. Also the feasibility
640
Processes and P h e ~ o ~ e n u
of sheet form~ngby means of superplastic forming and other forming techniques has been shown on industrial facilities as well as on a laboratory scale. However, for widespread application of y(TiA1) alloys it must be shown that semi-finished products as well as components with specified mechanical properties can be manufactured in large quantities at reasonable cost. Further, for structural applications, appropriate joining and repairing methods must be made available that guarantee achieve men^ of reliable joints exhibiting good mechanical properties, especially at temperatures below the ductile-to-brittle transition temperature. Further engineering of y(TiA1)-base alloys should achieve an i ~ p r o v e d balance between roomtemperature ductility, fracture toughness, highth, creep- and oxid~tion-resis~ance. Nex~-g~neration y(TiA1)-base alloys are under development aiming to provide these improved properties, In this connection, y(TiA1)-base alloys with. increased Nb promise improved high-temperature mechanical properties and oxidation resistance, thus rendering y(TiA1) as a true weight-saving structural material for replacing Ti-base alloys and Ni-based superalloys.
rence Appel, F. (1999). In Advances in T~jnning(eds S. Ankeni, C. S. Pande). TMS, Warrendale, PA, p. 171. Appel, F-,Beaven, P. A. and Wagner, R. (1993). Acta MetoEl.
, U., and Wagner, R. (1995). Philos. Appel, F., Loreiiz, U., and Qehring, M. (1997a). German Patent, DE, 1974257 Al, FR. Germany. Appel, F., Clernens, H., Glatz, W., and Wagner, R. (1997b). In ~ i g h - T e ~ ~ p e r a Ordered ~ u r e r~ite~metallic Alloys VII (eds C. C. Koch, C. T. Liu, N. S. Stoloff, A. Wanner). Mat. Res. Soc. S-ymp. Proc., Vol. 460, Pittsburgh, PA, p. 195. Appel, F., Lorenz, IJ., Paul, J. D. H., and Oehring, M. (X999a). In Gamma Titanium Alurni~~de~s 1999 (eds Y.-W. Kirn, D. M. Dimiduk and M. H. Loretto). TMS, Warrendale, PAYp. 38 1. Appel, F., Sparka, U., and Wagner, R. (1999b). Intermetallics, 7, 325. Appel, F,, Oehring, M., and Ennis, P. J. (1999~).In Gamma T ~ ~ a n i Alum~nides ~m 1999 (eds Y -W. Kirn, D. M. Dimidlak, M. H, Loretto). TMS, Warrendale, PA, p. 603. g, M., and Wagner, R. (2000a). Novel ts for Gamma-Base Titaniurn Alurninide Alloys. Interme~a~lics, 8, 1283. Appel, F., Brossmann, U,, Christoph, U., Eggert, S., Janschek, P., Lorenz, U,, Mullauer, J., Oehring, M., and
Paul, J. D. H. (2000b). Advanced ~ ~ g i ~ e~ea rt ~ r~~ ga l s , 699. Appel, F. and Wagner, R. (1998). t at er^ Sci.Eng., Banerjee, D., Gogia, A. K., Nandy, T. K., Mural K., and Mishra, R. S. (1993). In Structural Intermetallics (eds R. Darolia, J. J. Lewandowski3 C. T. Liu, P. L. Martin, D. B. Miracle, M. V. Nathal). TMS, Warrendale, PA, p. 19. Bartels, A., Clemens, H., Wartig, H., and Mecking, H. (1997). In High- Temperature Ordered r~termetallicAlloys VII (eds C. C. Koch, C. T. Liu, N. 5. Stoloff, A. ~ a n n e ~Mat. ). Res. Soc. Symp. Proc., Vol. 60, Pittsburgh, PA, p. 141. Bartels, A., Kestler, H. and Clemens, W. (2000). Mater. Sci. Eng. A (Proc. of 5th Int. Conf on Structural & Functional in press. Intermetal~i~s), Bartolotta, P and Krause, D. L. (1999). In Gamma Titanium A l u ~ i n i d e1999 ~ (eds Y.-W. Kim, D. M. Dimiduk, M. €€. Loretto). TMS, Warrendale, PA, p. 3, Blackburn, M. J. (1970). In The Science, ~ e c h n o ~ and ~)g~ Applications of Titanium (eds R. Jaffee and N. E. Promisel). Pergamon Press, London, p. 633. Brooks, J. W., Dean, T. A., Hu, Z. M., and Wey, E (1998). Materials Processing Teclznology, (88-81, 149, Bryant, J. D., Christodoulou, J. A., and Maisano, J, R. (1990). S c r i ~ t aMetall, mat^^,, 24, 33, Cahn, R. W., Takeyama, M., Horton, J. A., and Liu, C. T. (1991). J . Mater. Res., 4,57. Chatterjee, A., Clernens, H., and Kestler, H. (1999). In G ~ n ~ r n atitan^^^ A l ~ ~ ~ n (eds i d ~Y.-W. s Kim, D. M. Dimiduk, M. H. Loretto). TMS, Warrendale, PA, p. 401. Chen, G., Zhang, W., Wang, Y,, Wang, J., and Sun, Z. (1993). In Structural Inter~etallics(eds R. Darolia, J. J. Lewandowski, C. T. Liu, P. L. Martin, D. B. Miracle, M. V, Nathal). TMS, Warrendale, PA, y . 3 19. ., Kack, W., and Knerxnger, 6. Clemens, H., Glatz, W., Schretter, P., Koeppe, C., Bartels, A., Behr, R., and Wanner, A. (1995). In ~ ~ r n mTitonium a A l u m ~ ~ ~ d(eds e . ~ Y.-W Kirn, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, p. 717 Clemens, H., Glatz, W., Schretter, P., Klassen, M., Scliubert, E., Sepold, G., Fleischer, T., Schrock, W.W., and Franke, R. (1996). In S~acecraft Structures, Materials and ~ e c h ~ ~ nTesting. i ~ a l ESA, Noordwijk, p. 1297. Clemens, H., Glatz, W., Eberhardt, N., and Kabl, W. (1997). In ~ i g ~ - T e r n p e r ~Otr~dreer e Interm~tallic ~ Alloys VII (eds C. C. Koch, C, T. Liu, N. S. Stoloff, A. Wanner). Mat. Res. Soc. Symp. Proc., Vol. 460, Pittsburgh, PA, p.29. Clemens, H., Kestler, H., Eberhardt, N., and Knabl, W. e ~Y ~ -W. Kim, (1999a). In G a m r n ~~ i t a n i u mA l u ~ i n ~ d(eds B. M. Dimiduk, M. H. Lorctto). TMS, Warrendale, PA, p, 209. Clernens, H., Lorich, A., Eberhardt, N., Glatz, W., Knabl, W., and Kestler, H. (1999b). Z. ~ e t a Z l ~ d90(8), ., p, 569. Clemens, H. aiid Kestler, H. (2000). Adv. ER^. Mater., 9, p. 551.
Forming
641
and Froes, F. A. (1999). In G a ~ m Ta i ~ a n i ~Alumin~des m 1999 (eds Y.-W. ISKim, D.M. Dirniduk and M.H. Loretto). TMS, Warrendale PA, p. 565. Kestler, H., Clernens, H., Baur, H., Joos, R., Gerling, R., Cam, G., Bartels, A., Schleinzer, C., and Srnarsly, W. (1999). In Gamma Ti~a~~iuFn A l u ~ i n ~ (eds ~ e sY.-W. Kiln, D.M. Dimiduk, M. H. Loretto). TMS, ~ ~ r r e n d a iPA, e, p. 423. Kim, Y.-W. (1994). JOM, '7, 30. 1Kirn, Y.-W. (1995). In G a m ~ aT i ~ a n ~~ lu ~~ m i ~(eds i ~Y.-W. es Kirn, R. Wagner, M. Yama~~chi). TMS, Wa~endale, p. 637. Kiin, Y.-W. and Dirniduk, D. M. (1997). In Structural Intermetallies (eds M. V. Nathal, R. Darolia, C. T. Liu, P L. Martin, D. €3. Miracle, R. Wagner). TMS, Warrend~~e, PA, Dimiduk, D. M. and Vasudevan, V. K. (1999). Gamma p. 531. Titan~um Alumin~des 1999 (eds Y -W. Kirn, D. M. Knippscheer, S., Fromrneyer, G., Baur, H., Joos, R., Dkniduk, M. H. Loretto). TMS, Warrendale FA, p. 239. Lohmann, M., Berg, O., Kestler, H., Eberhardt, N., Ernst, F., Finnis, M. W., Hofmann, D., Muschik, T., Guther, V., and Otto, A. (1999). In Proc. of ~ ~ R O M A ~ Schonberger, U., and Wolf, U. (1992). Phys. Rev. Lett., 99, Symp. BI, Muterials for Tran~sport~~tion Technol~gy. 69, 620. Koeppe, C., Bartels, A., Clemen Escher, C., Neves, S., and Gottstem, 6. (1998). Acta M ~ ~ e r . , W (1995). Mater. Sci. Eng., 6, 441. LeHolrn, R., Clemens, H., and Kestler, H. (1999). In Gamma Escher, C. and Gottstein, G. (1998). Acta Mater., Tiianium Alumin~des 1999 (eds Y.-W. Kirn, 0.M. Fujitsuna, N., Miyamoto, Y., and Ashida, Y. Dirniduk and M. H. Loretto), TMS, Warrendale PA, S~ruc~urulInter~etallics (eds R. Darolia, J. J. p. 25. Lewaiidowski, C. T. Liu, P. L. Martin, D. 13. Miracle, Lombard, C. M., Gosh, A. K., and S. L., Serniatin (1995). In M. V. Nathal). TMS, Warrendale, PA, p. 187. (eds Y.-W. Kim, R. Wagner Gamma Titanium Aluminide~~ Fujiwara, T., Nakamura, A., Hosomi, M., Nishitani, S. R., and M. ~amaguchi).TMS, Warrendale, PA, p. 5'19. Shirai, Y., and Yarnaguchi, M. (1990). Philos. Mug., A61, Martin, P. L., Rhodes, C. G., aiid McQuay, P. A. (1993). In 591. Structural Intern~~tallics (eds R. Darolia, J. J. Gaydosh, D. J. and Crimp, M. A. (1985). In HighLewandowski, C. T. Liu, P. L. Martin, D. B. Miracle, T e m p e ~ a t ~ Interme~all~c re Alloys (eds C. C. Koch, C. T. M. V. Nathal). TMS, Warrendale, PA, p. 177. Liu, N. S. Stolaff). M m r . Res. Sym. Proc,, Vol. 351, Martin, P L., Jian, S. K., and Stucke, M. A. (1995). In Pittsburg, PA, p. 429. Gamma T ~ ~ a n ~ ul umm ~ n i(eds ~ e sY.-W, Kim, R. Wa~ner, Greenberg, B. A. and Gornostirev, Y. N. (1988). S c ~ ~ t a M. Yamaguchi). TMS, Warrendale, PA, p. 727. McCullough, C., Valencia, J. J., Levi, C . G., and Mehrabian, Werzig, Ch., Przeorski, T., and Misliin, U. (1999). R. (1989). Acta Metall., 37, 1321. ~ntermetallics,'7, 389. McQuay, P. A., Simkins, R., Seo, D. Y., and Bieler, T. T. H~iang,S.-C. (1993). In Structural Intermetallies (eds R. (1999). In Gamma ~ i ~ a n ~ Aluminides uin 1999 (eds Y.-W. Darolia, J. J. Lewandowski, C. T. Liu, P. L. Martin, D. B. Kim, D. M. Dimiduk, M. H. Lsretto). TMS, Warrendale Miracle, M. V. Nathal). TMS, Warrendale, PA, p. 299. PA, p. 197. Huang, S. C. and Chesnutt, J. C. (1995). In Interme~allic McQuay, P. and Sikka, V. K. (2002). In Inter~etallic C o ~ ~ o u n d P~ rs~: ~ c i p l eand s Practice, Vol. 2 (eds J. H. Compounds, Vol. 3, Prugress (eds J. H. Westbrook and Westbrook aiid R. L. Fleischer). John Wiley, Chichester, R. L. Fleischer). Wiley, Chicbester, UM. UK, p.73. Millett, J. 6 .F., Brooks, J. W., and Jones, I. P. (1999). Mater. Hu~phreys, F. J. and Hatherly, M. (1995). In Sci. TechnoE., 15, 697. R~cyystallizat~onand Related Annealing Fhenon~e~a, Morita, A., Fujitsuna, N., and Shigeo, H. (1993). In Symp. Pergamon, Oxford, p. 364. H ~ m a nM. , E., NIc~u~lough, C., Levi, C. G., and N I e h ~ ~ ~ i a n , Proc. for Basic Tec~nologiesfor Future Industr~esHighPerformance Materials .for Severe ~ ~ v i r o n m e n t4th s R. (1991). Metall. Trans., 2 2 ~ 1647. , Meeting, Japan Industrial Technology Association, Imayev, R., Salishcliev, G., Imayev, V., Shagiev, M., and Kuznetsov, A. (1995). In Gamma T i t a n i u ~Aluminides (eds Tokyo, p. 215. Nobuki, M., Hashimoto, K., Takahashi, J., and Tsujirnoto, U.-W ISirn, R, Wagner, M. Yamaguchi). TMS, Warrendale, PA, p. 665. T. (1990). Mater. Trans. Japan Inst. Met., 31, 814. Imayev, R. M., Salishchev,G. A., Imayev, V. M., Shagiev,M. R., Oehring, M., Appel. F., Ennis, P. J., and Wagner, R. (1999). Kuznetsov, A. V., Appel, F., Oehring, M., Senkov, 0. N., Interm~tallics,7, 335.
Das, 6. and Clemens, H. (1999). In Gamma Titan~um A1u~n~n~de.s (eds Y.-W. Kim, D. M. Dimiduk, M. H. Loretto). TMS, Warrendale, PA, p. 579. Davey, S., Loretto, NI. H., Evans, R. W., Dean, T. A., Huang, Z. W., Blenkinsop, P., and Jones, A. (1995). In Gamma Titanium A l u m i n i ~ ~(eds s Y.-W. Kim, R. Wagiier and M. Yaniaguchi). TMS, Warrendale, PA, p. 539. Deevi, S. C., Wajaligol, M. R., Sikka,V. K. McKernon, J., and Scorey, C. R. (1 999). In High-Temperature Ordered Intermetallies VIIl (eds E.-P. George, M. J, Mills, M. ~arnaguchi).Mater. Res. Soc. Symp. Proc., Vol. 5 Pittburgh, PA, p. KK 4.6.1. Dimiduk, D. M. (1999). Mater. Scr. Eng., Dimiduk, D. M., Martin, P. L., and Kirn, Y.-W. (1998).
642
Procrsscs and Phenomena
Paul, J. D. H., Appel, F., and Wagner, R. (1998). Acta Mater. 46, 1075. Ponge, D. and Gottstein, G. (1998). Acta Mater., 46, 69. Reed, S. (1995). In Gamma ~ i t u n ~ uAml i ~ ~ ~ n i(eds ~ ~ eY.-W. s Kirn, R. Wagner, M. Yamaguchi). TMS, Warrendale, PA, p. 475. Salishchev,G. S., Iniayev, R. M., Senkov, 0.N., Imayev, V. M., Cabdullin, N, K., Shagrer, M. R., Kwetsov, A. V., and Froes, F. €3. (2000). Mater. SCE.Eng., A%%,236. Sauthoff, G., (1995). In Intermetallic Compoundi;, Vol. I , Principles (eds J. H. Westbrook and R. L. Fleischer). Wiley, Chichester, UK, p. 91 1 . Schillinger, W., Lorenzen, B., and Bartels, A. (2000). Mater. Sci. Eng. A (Proc. of 5th hit. Conf. on Structural & Functional ~ntermetallics),in press. Seetharaman, V., Goetz, R. L., and Semiatin, S. L. (1991). In H i g ~ ~ T e ~ p e r ~ iOrdered t u r e ~ntermetallicAlloys IV (eds L. A. Johnson, D. P. Pope, J. 0. Stiegler).Mater. Res. Soc. Symp. Proe., Vol. 213, Pittsb~rgh,PA, p. 895. Seethara~an,V. and Semiatin, S. L. (1996). Metall. Trans., 27A, 1987. Segal, V. M., Reznikow, V. I., Drobyshevsky, A. E., and Kopylov, V. 1. (1981). Ru~ssianMetall~rgy,1, 99. Semiatin, S . L. (1995). Gamma Titanium Aluminides (eds Y -W. Kirn, R. Wagner, M. Yamaguchi). TMS, Warrendale, PA, p. 509. Semiatin, S. L., Volrner, D. C., El-Soudani, S. M., and Su, C. (1990). Scripta Metall. Mater., 24, 1409. Se~iatin,S. L., Ohls, M., and Kerr, W. R (1991). Scripta ivpuri, R. (1994). Metall. and Weiss, J. (1998). Mater. an, V. (1994). Metall. and
an, V, (1995). Metall. and Semiatin, S . L. and Seetharaman, V. (1997). Scripta Mater., 36, 291. A l u ~ i n i ~ eand s Sikka, V. K. (1992). In ~i~h-Temperature Int~rm~tullie,s (eds S. H. Whang, D. P Pope, C. T. Liu). Elsevier Science Pub~ishers,New York, p. 712.
Sikka, V. K. (1993). In ~ i g ~ - T e m p ~ r a t uOrdere~ ~e Interm~tallic Alloys VI (eds J. Horton, J. Baker, S. Hanada, R. D. NoebC, D. S. Schwartz). Mater. Res. , Pittsburgh, PA, p. 873. Soc. Symp. Proc., Vol. and Baldwin, R. H. (1991). In Sikka, V. K., Gieseke, B. ~eat"~e.si.stunt ~ateria~.i; (eds K. Natesan, D. J. Tillack). ASM International, Materials Park, OH, p. 363. Singh, J. P., Tuval, E., Weiss, I., and Srinivasan, R. (1995). In ~ a m m a~ i t a n ~A luu~~ i~n i d ~(eds s Y.-W. Kirn, D. M. Dimiduk, M. W. Loretto). TMS, Warrendale PA, p. 547, Sun, 2;. Q.. Huang, Y D., Yang, W. Y., and Chen, G. L. Ordered Inter~etallicAlloys (1993). In ~igh-T~mperature VI(eds J. Horton, J. Baker, S. Hanada, R. D. Noebe, D. S. Schwartz). Mater. Rcs. Soc. Symp. Proc., Vol. 364, Pittburgh, PA, p. 885, Tetsui, T., Higuchi, W., and Tacita, IS.(1997). Development of TiAl Forged Alloy for High-Te~~erature High-speed Rotating Components. Mitsubishi Heavy lndustries, Ltd., Technical Review 34. Umakoshi, Y., Nakano, T., and Yarnane, T. (1992). Mater. Sci. Eng., A152, 81. Veyssikre, P. and Dounin, J. (1995). in ~ v k ~ ~ r ~ e t ~ C o ~ p o ~ n dVol. s I , Principle~s(eds J.H. Westbrook and R. L. Fleischer). Wiley, Chichester, UK, p. 559. Westbrook, J. H. and Fleischer, R. L. eds. (1995). In Inter~etallic Compoi4ndi;: P r i n c ~ l e . ~and Practice, Volumes 1 and 2. Wiley, Chichester, UK. g, S . H. and Hahn, Y. D. (1990). Scrlptu Metall. Muder., 1679. Wiezorek, J. M. K., DeLuca, P. M. Mills, M. J., and Fraser, H. L. (1997). Philos. Mag. Left., 75, 271. Yamaguchi, M. and Umakoshi, Y. (1990). Progress in ~ a t e r i a l sScience, 34, 1 . Yolton, C. F., Habel, U. and Clernens, €3. (1997). In ~ d v a n c e d Particulnte Materials avkd Processes (ed F. H. Froes). Metal Powder Industries Federation, Princeton, New Jersey, p. 161. Yoo, M. H., Fu, C, L., and Lee, J. K. (1994). T~inningin Advazced Materials (eds M. H. Yoo, M. Wuttig). TMS, Warrendaie, PA, p. 97. Yoo., M. H. and Fu, C. L. (1998). Metall. Trans. A, Yoo, M. H. (2002). In ~nderme~allic ~ompounds,Vol. 3, Progress (eds J. H. Westbrook and R. L. Fleischer). Wiley, Chichester, UK.
Powder metallurgy (P/M) is one of the most diverse and comprehensive approaches for inanufact uriiig metallic and ceramic parts. It must be recognized that P/M is an ancient technology: almost every metal or ceramic material was initially made via the powder route (Exner and Arzt, 1996). odern applications of P/M in materials technology are widespread: connecting rods and gears in automobiles, self-lubricating bearings, porous metallic filters, tungsten wires for lamp filaments, soft and hard magnetic materials, electrical contacts, cemented carbides for cutting tools, damage-tolerant superalloys for gas turbine engine discs, amalgams for dental applications~and composite packages for microelectronic devices (see Volume 2 for applications of PjM internietallics).The main attraction of P/M is the ability to fabricate high-quality, complete parts to close tolerances in an economical manner. Indeed, the P/M approach plays a crucial role in the development of the near-net-shape fabrication technology for the aerospace and automotive iiidustries. Despite these advantages, the growth of P/M technology has suRered from some major lhitations imposed by the processing equipment, and furnaces, and by the reactivity of ates. Large-sized parts are more easily fabricated using the ingot metallurgy route. Tlie presence of oxides and other nonmetallic inclusions at the prior-particle boundaries has caused wide scatter in "See List of Contributors for current address.
the fatigue and fracture properties of ~igh-streii~th aluniinum and titanium alloys. The associated concerns regarding the reliability and safety of the c o ~ p o ~ e n t s have restricted the use of lion-fracture-critical appl partly overcome by comb extensive hot working. A review of the use of the processing of intcr~etallics and their co~nposites shows some interesting trends. Initi~lly,monolithic intermetallics based on binary, stoicliiometric compositions and prepared by ingot metallurgy (I/M) were invariably selected for research and development studies. In general, these materials exhibited very low ductility and toughness, even at fdirly Iiigli temperatures. Moreover, even the relatively small size in obtained by melting and casting of these ~ a t e r i a l s contained coarse microstructures, revealed high levels
ventional i~etal-working methods. Under these circumstances, the P/M approach often proved to be a viable alternative for processing relatively small quantities of the intermet~llicsinto sound products that were suitable for evaluation of the meclianical and other properties. $imultaneous efforts in the understanding of the phase equilibria and phase transforinatioiis in these materials led to the development of new, multicompone~t alloys that were
I n t i ~ ~ ~ Compounds: ~ ~ ~ ~ l i cVol. 3 , Principles m d Practice. Edited by J. W, Westbrook and R. L. Fleischer. 0 2 0 0 2 John Wiley & Sons, Ltd.
Processes and Phenomena
644
amenable to processi~gvia the conventional, ingot metallurgy route, At this stage, it became feasible to optimize the microstructure and tailor the properties for a specific application by selecting the chemical composition of the inter~etallicalloy and the appropriate processing method. ell-known examples for this type of evolution include alloys based on the Ni,Al (Dimiduk, 1999). ress in the development ch for a variety of intermetallics. In ctive to summarize the basic features of the different methods of syiithesizing powders of the intermetallic alloys. A brief discussion ethods of powder consolidation is then working of the consolidated billets is reviewed with emphasis on the use of process models the densification behavior. The application approach to the processing of intermetallics Al, Ni,Al, NiAl, Fe,Al, FeA1, Nb,Si, and iz is described in detail.
esis size, shape, microstructure, and chemistry of ameters in d e t ~ ~ i n i nthe g quality of the final product. Accordingly, a variety of techniques have been developed to synthesize metal powders with the required properties. These techniques can be broadly classified into four groups: i~echanica~ comminution, chemical reactions, electrolytic deposition, aiid liquid-metal atomization. The choice of the technique is determined by the properties of the material, the intended application, the desired purity, aiid the process economics, Detailed descriptions of the different synthesis te~hniquescan be found treatments by German (1994), Arunachandarcsan (199l), Klar (1984), and Lenel (1980). ~ m o n gthe wide range of methods available, only a few are suitable for producing intermetallic powders because of their high melting points and extreme chemical reactivity. A short description of these selected t e c h i ~ i ~ is u ~given s below.
tomization is a process of breaking up a liquid metal roplets and allowing them to solidify use of its ability to control the composition and shape characteristics of the powders, atomization has become increasingly popular and accounts for most metal powder syntheses. Both
elemental and pre-alloyed powders can be produced by atomization. Two common methods of atomization are inert gas atomization and centrif~galato~ization. In gas atomization, a liquid metal stream is produced by pouring molten metal through a tundish; the metal stream is then broken up into droplets by the impingement of high-pressure jets of inert gas. During flight through the collection chamber, the droplets lose heat and solidify into spherical powders. The mean particle size of the powders depends both on the therm~physicalproperties of the alloy and the operating conditions of the ~ t o m i ~ a t i oprocess. n In general, fine ( 4 100 pm) powders can be obtained for metals of high density and low surface energy and by operating at high gas velocities. Gas-atomized powders exhibit good packing and flow properties with tap densities approaching 65% of the theoretical values. Centrifugal atomization involves a rapidly rotating consumable spindle, which is melted by an arc using a tungsten cathode or a plasma torch. The centrifugal force throws off the molten metal as a fine spray, which solidifies into fine, spherical powders. The mean particle size is inversely proportiona~to the angular velocity of the anode and ~xhibitsa relatively weak dependence on the properties of the liquid and the radius of the electrode. The plasma rotating electrode process R RE^) developed by Nuclear capable of producing spherical powders in the size range 5-8Opm with a relatively arrow size distribution. Furthermore, by avoidin interactions of the liquid metal with gas jets and atomization nozzles, the centrifugal atomization process can yield powders of relatively high purity and cleanliness. Typical cooling rates of droplets during the atoms, a ization process are in the range 10, to 1 0 ~ ~ C /As result, the solidification processes occur under highly non-equilibrium conditions, leading to very fine dendritic microstructures with miiiirnum segregation ailcl with enhanced solid solubility for solutes. As the cooling rate increases, transitions from dendritic to equiaxed to microcrystalline structures are observed.
Ball mills or attritors use the high-energy impact of balls to produce powders from brittle materials. Even ductile materials can be powdered by using lowtemperature milling (cryoniilling). Alternatively, some of the titanium and zirconium alloys can be converted into powders by using the hydriding-n~illing-deliydriding approach. The fineness of the powders obtained by milling spans a wide range (1-500prn) depending on
the ball size, friability of the charge, and the comminution time. The powders are usually of irregular shape and exhibit poor packing and flow characteristics. Contamination of powders with the materials of the balls and the pickup of nitrogen in cryomilling are also important problems. Milling is also used for an effective mixing of the components in a powder blend. A novel extension of milling wa;s developed by ~enjamin(19%) for producing alloyed and dispersion~strengthenedmaterials from powder blends. This process, known as mechanical alloying (MA), is carried out in attritors or high-energy ball mills. The powders are welded, fractured and rewelded repeatedly during milling to produce a uniform dispersion of the constituents. Using this method, fine uniform dispersions of oxides (A1,O3, Y,O, or Th0,) in nickel-base superalloys were obtained. In addition, one can produce ‘true alloying’ by milling blended elemental powders. Micromechanisms involved in this type of mechanical alloying include microforging, cold welding, forniation of micro-laminates, and accelerated diffusion (~enjaminand Volin, 1974). Koch (199 1, 1998) and Froes et al. (1995) have described detailed models of the mechanics and kinetics of mechanical ecause of the non-equilibrium nature of the MA process, the powders synthesized by this method exhibit several novel characteristics including extended solid solubility, new crystalline phases, amorphous structure, or micro/n~nocrystalline structures (Koch, 1991; Froes et al., 1995; Counihan et nl., 1999; Chung et al., 1995). Powders synthesized by mechanical alloying are already work-hardened and are not easy to consolidate by cold compaction. However, these powders have been coiisolidated successfully using hot pressing^ hot extr~sion/rolling,and hot isostatic pressing. See also the chapter by Thadhani in this volume for discussion of niechaiiical alloying and reaction synthesis.
Solid-state reaction synthesis, also known as combustion synthesis or self-propagating high-temperature synthesis (SHS) provides an attractive, practical alternative to the conventional methods of producing intermetallics that exhibit higher thermodyna~ic stability than tlieir constituent elements. In this process, a porous compact made from a mixture of the elemental powders reactant^) is ignited using a hot tungsten coil or by heating in a furnace. An exotherniic reaction is initiated and continues as a self-sustaining
combustion wave propagating through the porous inass, If the peak temperature resulting from the adiabatic heating of the preform can be controlled to remain below the melting point of the reaction product, the process will be suitable for the production of alloy powders. Developed originally in the former reaction synthesis process has been ing a variety of materials, includi elements, TiNi shape memory alloys, funct~onallygraded materials, composites and i nt e~et al l i calloys. Moore and Feng (1995a,b) have reviewed the reaction parameters, applications, and modeling of this process. Major advantages of this method include: (i) removal of volatile impurities, (ii) short simple exothermic nature of the alleviates the need for expensive facilities and equipment) and (iv) rapid cooling rates and high thermal gradients that allow the development of non-equilibrium microstructures or phases. The r ~ a c t ~ o n synthesis method provides o~~portunities for producing and consolidating powde cornpounds such as Ni,Al, TiAl, Fe&, and FeAl (Counihan et al., 1999; Joslin et al., 1995; Deevi and Sikka, 1995; Mumgesh et al., 1994; Gialanella et al., 2000; Gauthi~ret al., 1999).
3.
S
In conventional P/M processing, the powder is generally consolidated into a compact by the application of pressure at the a ~ b i e ~t te ~ p ~ r a t u r e . Compaction is followed by sintering at high teniperae type of cold tures to obtain the final part. ~ h i l this compaction + pressureless sintering has been very successful in processing a lar number of alloys, structural intermetallics usually to consolidation by such an approa high strength and resistance to plastic deformation at rooni temperature, hot-consolidation techniques are required to obtain high-density products. )
#
HIP is now widely adopted for powder consolidation to complex shapes. The powder is tap-filled in a shaped container, degassed under vacuum at an elevated temperature and sealed erm me tic ally . The canned preform is loaded into the HIP vessel and processed through a prescribed pressui-e-temperature-time cycle. Argon is the most common pressurizing gas. While the
646
Processes and Phenomena
vessels are rated for a maximum pressure of 400 MPa, typical pressures used in the HIP process range from 100 to 200MFa. After HIP, the can material can be removed by machining or chemical usually results in complete densification of the part. Price and Kohler (1984) have provided a detailed account of the process equipment, design of the cans, process cycles and applications of HIP to the near-net-shape forming o f superalloys and titanium alloys. These methods and equipment are generally applicable for the HIP consolidation of intermetallic powders as well. N
3.2
essiRg
ot pressing of metal powders in closed dies is a relatively mature technology, developed originally for the hot consolidation of refractory metals and cemented carbides. In this technique, pressure is applied on the powder in a die, either in a singleacting or in a double-acting press. Hydraulic or pneumatic presses with proper provisions for ejection of the finished parts are usually employed. Dies and punches used for hot pressing are made of heatIs, graphite, cobalt/nickel-base molybdenum alloy. Heating of the dies is accomplished via induction heating, resistive heating or through electric-spark discharges between the powder particles. In general, the metallic dies and punches are water cooled to prevent their plastic deformation and distortion. A controlled environment, such as vacuum or inert gas, is usually necessary. The quality of the hot-pressed compact is determined by the interrelat~onship of three process variables: pressure, tennperature, and time. For practical reasons, hot-pressing ~emperaturesshould be su~cientlyhigh to achieve complete desensification at moderate pressure in a reasonably short time.
Hot extrusion of powder preforms combines hot cornpaction and mechanical working, yielding a fully dense product in a single step. The process consists of filling the powder inside a metallic can or capsule, e v ~ ~ u a t i oand n sealing, followed by extrusion at fairly high temperatures. Aside from its obvious role as container, the can also serves to protect the powder from contamination from the atmosphere, extrusion lubr~cants and the tooling. Can materials include copper, mild steel, austenitic stainless steel, Ti-6Al4V, and molybdenum, and are selected on the basis of
their hot workability, physical and chemical compatibility with the powders, and cost. Powders of superalloys and intermetallics are usually extruded using conical dies in order to improv~the metal flow characteristics and minimize the redundant work. Often, the extrusion step is preceded by blind compaclion of the canned assembly. Extrusions with simple geornetries such as circular, elliptical, or rectangular cross-sections are produced quite readily. An innovative method called the ‘filled billet’ technique has enabled the fabrication of complex structural shapes by extrusion. As described by Bufferd (1972), this technique consists of producing a low cost ‘filler’ billet conta~ninga fairly large cavity with the desired shape, pouring and tapping the powder inside the cavity, sealing of the billet inside a can and extrusion of the canned assembly through a conical die with a round orifice, After extrusion, the filler material is dissolved leaving the desired shape of the fully deiise P/M product.
Powder forging involves pre~arationof a pressed and sintered preform, followed by forging of the porous preform into a highly densified final product. Forging is generally performed in one operation in confined dies to eliminate flash formation and achieve a nearnet-shape. The main advantages of the P/M forging method include flexibility in alloy selection, less mechanical anisotropy in the forged P/M products, and fine and uniform ~icrostructureof the forgings. Currently, P/M forging is used widely in the manufacture of automobile components. Powder rolling refers to the continuous compaction of powders in a rolling mill. In this process, the metal powders are fed from a hopper to a set of compacting rolls that produce a continuous green sheetlstrip. These materials generally require further processing by sintering or re-rolling to produce final products with desired mechanical properties. Powder rolling processes have been used to produce nickel-iron alloy strips having controlled e~pansioncharacteristics and sandwich (bimetallic) strips used in composite bearings.
illets:
Hot deformation processes such as hot pressing, powder extrusion, aiid powder forging serve to produce fully dense P/M parts with desired micro-
Powder Metallurgy structures and mechanical properties. The kinetics of densification during the hot consolidation/hot working steps depend critically on process parameters such as temperature, pressure, strain path, strain rate, and time. The selection of process parameters for the manufacture of conventional alloys is usually based on trial and error methods or semi-empirical approaclies. Challenges associated with the processing of advanced materials require the use of science-based models for the prediction of the densification behavior' of powder aggregates during hot consolidation. Sintering mechanism maps developed by Ashby (1974), densification mechanism maps for pressure sintering evolved by Arzt et al. (1983) and HIP consolidation diagrams proposed by Helle et al. (1985) represent important contributions toward a mechanistic understanding of these processes. Figure 1 shows a typical densi~cation mechanism imp for the HIP consolidation of the yTiAl powder compacts (Ashby, 1990; Martin and ~ a r ~ w i c 1995). k, The map displays different regions
647
representing dominant densification mechanisms. The dotted lines represent the time contours for achieving the desired levels of densi~cation. A crucial aspect of ~echanistica~proachesis the need for accurate data for thermophysical properties and transport properties. This requirement limits the use of the mechanistic models to we11-characterized alloy systems with known material data. continuum models such as those advanced by Lee and Kim (19921, Dutton et aE. (1995), and Park et al. (1999) deal with macroscopic parameters, such as Poisson's ratio and the stress-intensification factor, and their dependence on the ins~antan~ous relative density of the compact. These parameters can be derive experiments, rnsually uniaxial upset tests. ( 1995) have successfully incorporated microstructural effects such as grain growth, pore size and pore geometry into the continuum yield functions and the associated flow rule. Figure 2(a) shows the relationship between the stress intensity factor and the relative
.*. ... * . :
f
.7
re I Densification mechanism map for HIP consolidation of the y-TiAl powder compacts. The HIP temperature of 1100 "C corresponds to a homologous temperature, T, = T/T,z0.8 where I;, represents the peritectic tratsformatisn (L + a-q~) temperature. Different regions separated by solid lines correspond to dominant densification mechanisms (Martin and Wardwlck, 1995)
648
Processes and ~ h e ~ ~ ~ e ~ ~ a have described the application of a numerical model called PreCAD for HIP consolidation of aerospace components in France.
U) In
E 3;
(a)
s 1
0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95
1
Relative Density 10 Experiment
8
--
Shlma-Oyme(19763
Dutton et al. (1995)
2 0 Stroke (mm)
(a) Comparison of the measured stress intensity factor, rp, for the alpha-2 titanium aluminide (Ti-24Al-llNb) with model predictions of v, using various idealized powder geometries and the empirical model of Shima and Oyane (1976), and (b) comparison of the finite element model predictions and experimeiital measurements of the loadstroke curves for die pressing of alpha-2 titaiiium aluminide powder compacts with an initial relative density of 0.63 (Dutton et al., 1995)
density for alpha-2 titanium aluminide, while Figure 2(b) provides the validation for the use of the continuum modeling approach. In related work, Semiatin et al. (1996) and Dutton et al. (1996) have med finite element method (FEM) calculations to construct HIP consolidation diagrams for the processing of intermetallic composites based on alpha-:! Ti3Al.The results shown in Figure 3(a) and (b) are analogous to the HIP diagrams proposed by Helle et al. (1985). ~xperimentalresults shown in Figures 3(c) and (d) indicate that the relative densities of the matrix are approximately 0.80 and 0.97 after consolidation at 982 "C and HIP pressures of 7 MPa and 35 MPa, respectively. These data are in reasonable agreement with the model predictions. Recently, oret (2000) and ~ a c c ~ netoal. (2000)
Near-gamma titanium aluminide alloys based on the intermetallic compound LI (tP4) TiAl are rapidly emerging as engine materials with potential applications in aero , a ~ i t o ~ o t i v eand , powergeneration industries. Excellent reviews by ~ainagLi~hi et al. (2000), Dimiduk (1995, 1999), Mim and Dimiduk (1997), Kim (1994), and Austin and summarized the engineering po knowledge, technological matu~ation~ and cost competitiveness of these alloys. ~omprehensiveaccounts of the physical metallurgy of gamma titanium aluminides can also be found in papers written by Titkeyama et al. (1993), Huang and Chesnutt (1995) and ~ i s h u r d aand Perepezko (1991). Other chapters in this volume review high temperature applications (Lipsitt et al.), casting ( ~ c ~ u and a y Sikh), and f ~ ~(Appel i ~ et gal.). In the early stages of the development of gamma titanium aluminides, attention wits focused mainly on alloys with nearly stoichiometric composition^ These alloys were invariably processed via the powder metallurgy route (Shechtman P I al., 1974; Lipsitt et al., 1975; Blackburn and Smith, 1980). ~ r ~ a l l o y e d powders were produced by both ar on gas ato~ization and by the plasma rotating electrode processes. Powders were canned in titanium cans, blind compacted, and extruded with fairly high ~ x t r ~ s i oratios n (8-24) at tem~eratures higher than 1300°C. The extrusions were invariably 100% dense and possessed moderate levels of tensile strength but very low ductility at room t e ~ p e r a t ~ r e , Studies by Valeiicia et al. (19$7) and McCullough et al. (1989) on conventionally solidified ingots and rapidly solidified powders of gamma titanium aluminide alloys led to early understanding of the hightemperature phase equilibria in the Ti-A1 system and the effect of compo§ition and under cool in^ on microstructural evolution in these alloys. The occurrence of double-cascading, peritectic reactions in the composition interval 45 < A1 55 at.% is re$~onsiblefor the microse~regationpatterns observed in the solidification structures. These findings enabled the selection of alloy co~positionswith i ~ p r ~ v i~trinsic ed wor~abili~y
Powder ~ e t u l l ~ r g ~
649
ure 3 Predictions of relative density versus time for the HIP consolidation of alpha-two tita~iL~m alum in id^ powder/siliconcarbide fiber tapecast monotapes at 982 "C under applied pressures of (a) 7 MPa or (b) 35 MPa. ~~crostructures of the composites consolidated by HIP fur 2 hours at 982 "C and pressures of (c) 7 MPa and (d) 35 MPa are also shown (Dalton et al., 1996)
and allowed the successfbl design of thermomechanical processes. Current alloys of near-gamma titanium alu~inidestypically contain 4548% A1 along with small additions of niobium, chromium, vanadium, manganese, tantalum, and boron. A great deal of research has been performed on defo~mation-processing of ingots of near-gamma titanium aluminides. This work Iias ranged from basic studies on the hot-compr~ssionbehavior and microstructu~alevolution in cast and wrought alloys to the development of various extrusion, forging, and rolling techniques. Excellent summaries of this research can be found in exhaustive reviews by iatin (1995) and and his coworkers 7a,b, 1998, 1999). ast, much less effort has been expended on the processing of gamma titanium aluminides synthesized from powder. Prealloyed powders of near-gamma alloys such as Ti-48Al-2Cr-2~bhave been synthesized by argon gas
atomization (GA), and the plasma rotating electrode process (PREP). Moll et al. (1990), Yolton et al. (1994) and Habel et al. (1999) have analyzed the characteristics of these powders in detail. Gas-atomized powder generally contains significantly higher levels of oxygen than the powder. Because of the interaction of the melt wi the graphite nozzle assembly, tains high levels of carbon, in the depending on the size fraction ( Recently, Gerling et al. (1998) systematically investigated the porosity, pore size distribution and the argon concentration of the Ti-47Al-4 (Nb, powders produced by a novel proce melted, induction-g~idin~, gas atomization (PIGA). They demonstrated that both the volume fraction of the pores and the argon concentration of the powders increase significantly with the powder partic Distinct advantages of the @A. process over
650
Processes and Phenomena
processes (which employ drip melting of relatively small amounts of the feedstock) are the mity and structural homogeneity el et al., 1999; Moll et al., 1990). owder consolidation has been performed primarily by IP and/or hot extrusion. Microstructures and echanical properties of the consolidated material stro~glydepend on the temperatures at which HIP and thc subsequent hot-working operations are performed meiis et al., 1995; Fuchs, 1993). (Eylon et al., 1993; Figure 4(a) and (b) w the typical ~icrostructuresof a near-gamma titanium almiinide alloy after HIP consolidation at 1010 "C/200MPa/4 hours and HIP + i s o t ~ e r ~ hot a l co~pressionat 1260 "C, respectively emiatin et al., 1994). Fuchs (1993, 1995) conducted a series of consolidation experinients with diEerent combinations of the powder type, parameters, and post-HIP ex~rusio~/isotherma~-forging operations. The tensile
Polarized light optical micrographs of a HIPconsolidated Ti-48A1-2Nb-2Cr alloy: (a) as-HIP'ped material (b) HIP'ped+upset at 1260 "C and 0.1 s-' (Semiatin et al., 1994)
properties of the consolidated billets evaluated at different test temperatures showed some broad trends. The strength and ductility of the GA billets at test tei~peraturesless than 500°C increased with the HIP temperature up to 1230"C, followed by gradual decreases up to 1300 "C. Tensile properties nieasured at test temperature^ above 500°C were by and large independent of the H P temperature. For a fixed HIP cycle of 1230"C/103 ~ ~ ahours, / 4 the CA and PREP materials exhibited similar properties. This work also reported that the properties of the PjM materials were equal to or better than those of the ingot-metallurgy materials. Fuchs (1998) also evaluated the effect of supertrans~s processing on the tensile and creep properties of the same alloy. It was observed that the resulting refined lamellar structures confer high tensile strength and i ~ p r o v e ~ e nint sthe creep resistance. The hot workability of the P/M-consolidated Ti48Al-2Cr-2Nb alloy has been evaluated by Semiatin et al. (1994) and Fuchs (1997). Figure 5 c o ~ p a r e sthe flow curves of the PjM consolidated materials with those of the cast+HIP'ped ingot material and the wrought material. It i s clear that the P/M materials exhibit low flow stresses as compared to those of the ingot metallurgy (I/M) materials. The strain-rate sensitivity data for the P/M material are slightly higher than those measured for I/M materials. Figure 6 shows the temperature dependence of the steady-state flow stress data for both these materials. An apparent
F i ~ ~ 5r e Flow curves obtained from isothermal compression tests performed on P/M and I/M materia~sof the Ti-48A12Nb-2Cr alloy. TA22A, TA22C and TA22X correspond to I/M materials subjected to annealing at 1200 "C for 96 11, annealing at 1440°C for 20 minutes, and isothermal forging followed by annealing at 1200"C for 48 hours, respectively. TA47 and TA50 are PjM materials obtained by HIP consolidation at 1230"(7172 MPa/4 hours and hot extrusion at 1300"C, respectively (Fuchs, 1997)
Powder ~ e t a l l ~ r g ~
"I 300
W
I
30t 2
1000/T(K)
Figure 6 Semi~ogarithi~ic plots of the steady-state flow stress vs. reciprocal te~peraturefor the ~ i - 4 8 A l - 2 ~ - 2 Calloy. r Data for the HIP-consolidated P/M material (Semiatin et al., 1994) are compared with those for the cast+HIP'ped ingot material (Shih and Scarr, 1991) and wrought I/M material ( S e ~ i a t i net al., 1992)
activation energy of Q = 385 kJ/mol obtained from the P/M material was in good agreement with the activation energy values of the I/M material. Enhanced workability of the P/~-consolidated gamma titanium aluminide alloys permits extensive deformation processing of these materials by isothermal forging/upsetting at relatively low temperatures ( 8 0 ~ 1 0 0 0"C) and at low strain rates (- 10-3 s-l). Imayev and Imayev (1991), Imayev et al. (1992), and ~alishchevet al. (1995) have investigated the dynamic recrystallization of the gamma grains and the attendant grain refinement encountered during upsetting of the powder compacts of the Ti-5OAl and Ti-46A1 alloys. Similarly, Cheng et al. (1992) succeeded in producing a recrystallized two-phase microstructure with an average grain size of 5 prn in a Ti-43A1 alloy by using a thermomechanical processing sequence involving blending of prealloyed powders, blind compaction + canned extrusion at temperatures above
65 1
1200"C, and hot upsetting at 1050°C. Both sets of investigations reported superplasticity in the hotworked materials at temperatures close to 1000 "C. Hot rolling of the near-gamma tita~iumal~rninide alloys, using both the I/M and P/M preforms, has attracted intense research and development efforts over the last decade. Potential h~gh-temperature applications of the near-gamma titanium aluininide sheets in the aerospace industry include engine components, exhaust nozzles and plugs, and thermal protection system panels for space vehicles (LeHolm et al., 1999). Plansee AG, Austria has spearheaded the development efforts for the rolling of large sheets on an industrial scale using P/M preforms (see chapter by Appel et al. in this volume). The P/N route for sheet rolling (based on early work at Battelle Colurnbus Laboratories in the 1980s) consists of the following steps: synthesis of fine, spherical, prealloyed powders by argon-gas atomization, sieving and blending of the powders to obtain proper particle-size distributi~n,canning and sealing in argon, and hot isostatic pressing of the cans at 1300 "C and 200 MPa for 2 hours to produce a fully dense material having a fine, globular structure with an average g a i m a grain size of 10 ,urn. The preforms are pack-rolled using conventional hot-rolling mills at low speeds (< 10m/rnin) and a narrow range of temperatures within the a+y phase field. Using this process, sheets with dimensions of 750 x 350 x l.Omm have been produced quite readily. Clemens et al. (1995, 1997, 1999a,b), Kestler et al. (1999), LeHolm et al. (1999), Chatterjee et al. (1999) and Inksen and Clemens (1999) have documented different aspects of the sheet rolling technology: processing, rnicrostructures, mechanical properties, and applications. The main advantages of the P/M route to sheet rolling include (a) elimination of the homogenization and forging steps; (b) the much higher yield in the PjM route than that in the IjM route, because of the nearnet-shape HIPing capability; and (c) large-size sheets can be produced because of the absence of any constraints 011 the size of the HIP'd preforms. The microstructures and the mechanical properties of the sheets praduced by the I/M and P/M routes are comparable (Clemens et al., 1995, 1999a). Figure 7 shows the tensile properties of the P/M Ti-46.5A1-4 (Cr, Nb, Ta, B) alloy sheets as a function of the test temperature. The fracture elongation of the sheets (after primary annealing at 1000"C for 2 hours) reveals a broad maximum at -800°C. The decrease in ductility in the temperature range 8 0 ~ 1 0 0 0 ° Cis cominon to all gamma titanium aluminide sheets produced from gas atomized powders (Yolton et al.,
652
Processes and Phenomena
1000
160 140
120
-
100
gi
c
80
0
60 9 40
20
ure 7 Tensile properties of the P/M Ti-46.5Ai-4 (Gr, Nb, Ta, B) alloy sheet evaluated in the longitudinal orientation. The sheets were in the primary annealed condition (1000 "C/2 hours). The dotted line represents the tensile d~ctiiityof an l/M sheet with an identical composition (Clemens et al., 1999a)
estler et al., 1999). It is interesting to note that the fracture elongation of the I/M sheets increases 1000 "C, it is almost sheet. The inferior attributed to the development of mod~rate-to-extensive cavitation at fairly low strains, possibly because of thermallyinduced, intergranLilar nucleation of micropores (Clemens et al., 1499a,b). Microstructural heterogeneities may also be responsible for the reduction in uctility at high temperatures. Despite this limitation, 0th the P/M aiid I/M sheets are considered to be a ~ e ~ a bto l esuperplastic forming. efined, fully lamellar microstructures provide the best nieans of obtaining a balance of mechanical ~roperties in the near-~amma titanium aluminide (Kim, alloys at both low and high tempe~~tures 1994). The scope i s somewhat limited for refining the lamellar grain size in cast or wrought IIM materials. Therefore, Liu et al. (1995), and Wang et al. (1995) have studied the role of the PJM approach to achieve microstructural refinement and improvei~entsin the mechanical ~ropertie$of a Tia47AI-2Nb-2Cr alloy. They extruded prealloyed powders obtained by the rapid-solidificat~o~~rate process at temperatures above the alpha transus (1320 "C) and then heat treated the extruded material at temperatures in the range 900 to 1350°C for 2 hours, followed by cooling at less than 0.1 "C/s. Under o p t i m u ~extrusion and heat treatment
conditions, a refined lamellar structure with a colony size of -65pm and an interlamellar spacing of -0.1 pm was obtained. This structure led to s i ~ n i ~ c a n t i~provements in tensile strength, ductility, fracture toughness, and creep resistance. Subsequently, Maziasz et al. (1997) and Maziasz and Liu (1998) investigated the stability of the ultrafine lamellar structures at elevated temperatures and the associated changes in mechanical properties. They found that lamellar structures in the P/M Ti are quite resistant to coarsening d for at least 5000 hours. The yield strength of the alloy at 20°C and 800°C 11-Petch type relationship with the lame1 over a wide range of spacings. Figure 8 he room temperature tensile ductility values of the P/M and I/M materials after di~erentheat treatn~e~ts, The large increase in ductility caused by a change in the heat treatment temperature from 900 "C to 1320 "C is noteworthy. As an alternative to the use of c re alloyed powders, blended elemental powders can be converted into alloy powders via mechanical alloying (MA) or by reaction esis (RS). ~ e h r i et ~ gal. (1993) have synthesized A1 powders by milling blended elemental powders in a planetary mill for 10 hours in a high-purity argon atmosphere. X-ray diffraction analysis of M A powders showed the formation of a metastable, disordered hcp phase with an average crystallite size of 15nm. Upon hot isostatic pressing above 50O"C, the hcp phase 5
F ~ ~ 8~ Plot r e of the room temperature ductility of P/M and I/M near-gamma titanium aluminide sheets processed by extrusion at supertransus tempe~atures(T2 = 1400 "C and T3 = 1350"C) and then. heat treated to either preserve (900 "C) or coarsen (1320 "C) the as-extruded, ultrafine lamellar structure without altering the refined coloiiy size (Maziasz and Liu, 1998)
653
Powder ~ e t a l l ~ ~ ~ y
decomposes into a mixture of y and a2 phases. However, a HIP cycle of ~OO"C/20OMPa/2hours is needed to achieve a fully dense product with porosity below 1%. It should be noted that prealloyed powders of the same composition require HIPing at temperatures above 1000°C to obtain a fully dense product. Apparently, the small crystallite size of the MA powders enables superplastic deformation at 800 "C, thereby causing complete densification at ly, Calderon et al. such a low temperat~re. (1999) investigated the use and plasma-assisted sintering to produce nanocrystalline titanium aluminide alloys. Prolonged milling up to 1000 hours results in powder particles comprising an amorphous matrix and smdl crystallite domains. Plasma-assisted sintering of this powder at a pressure of 50MPa causes ~ i c r o s ~ o p electric ic discharges between the particles, which in turn allow full densification to occur in relatively short sintering cycles ( 15 minutes). Two limiting factors of the MA process for synthesis of the titanium aluniinide alloy powders are (a) long milling times and (b) poor yield. Takasaki and Furuya (1999) and Hashimoto et al. (2000) have shown that introducing small amounts of n into the argon atmosphere can se the kinetics of mechanical alloying and improve the powder yield substantially. However, these approaches are fraught with the problems of hydro~enpickup and the formatio~of TiN and Ti2AIN in the powders. Reaction synthesis or reactive powder processing of near-gamma titanium aluminide alloys has been i n v ~ s t i ~ ~ tby e d Wang and Dahms (1992, 1993), Taguchi et al. (1995), Schneider et al. (1997), and Kin1 et al. (1999). typical se~uenceof the opera~ionsinvolved in the reactive powder processing of rrear~~amma titanium aluininide alloys. Elemental powders are mixed in the desired ratio, compacted by die pressing or cold isostatic pressing and then extruded at room temperature. The reaction sintering of the extruded fibrouscomposite-like material is carried out in a vacuum furnace (pressureless s~ntering)or in a HIP vessel. The Kirkendall porosity formed due to the differences in the diffusivities of A1 and Ti atoms can, of course, be eliminated by the HIP process. The porosity formed in the pressu~eless sintering process decreases with increases in the sinteriiig temperature and the extrusion ratio used in the previous step, ~ a n and g Dafnms (1993) have shown that the tensile strength and ductility of the reactive-powder-processed materials are somewhat inferior to those of the I/M materials or
-
I
1
-
~ i ~ u r9 e A typical sequence of ope~ations used in the a ~ ~ ~ reactive powder processing of n e a ~ - ~titanium aluminide alloys (Wang and Dahnis, 1993)
the prealloyed P/M inaterials. In an attempt to eliminate the expensive HIP operation, Schneider et al. (1997) have modified the processing sequence by introducing additional forging or extrusion steps before the reaction-sirrtering step. In this process, designed to produce n e a r ~ n e t - s h avalves ~~ for the automotive industry, total extrusion ratios of the order of 400 are specified. Two-step sintering treatments at 1400°C for 0.3h and 1000°C for 1h are employed to reduce the amount of porosity and minimize the pore size. In related work, Kim et al. (1999) devised a reactivepowde~"~rocessingmethod in which the blended elemental powders are compacted in stainless-steel cans and then hot extruded at 1250°C. They also systematically investigated the effect of the heating rate to the extrusion temperature on the distributions of the different phases. On the basis of extensive evaluation of tensile and creep properties, they have sought to establish the viability of the react~ve-powderprocessing route for the fabrication of the near-gamma titanium aluminide alloys.
654
Processes and P ~ e ~ o ~ e n ~
Iron aluminides based on DO, (cF16) Fe,A1 offer a corn bination of attractive properties, such as excellent resistance to oxidation and sulphidation at elevated temperatures and a high strength-to-weight ratio (Vedula, 1995; Deevi et al., 1997). Fe,Al has an 0, structure below 500 "C and is stable over a wide range of aluminium contents (23 to 36 at.%). Above 550 "G,it transforms to an imperfectly ordered B2 structure, which ultimately changes to a disordered solid solution. The Fe,Al-base alloys generally contain small amounts of chromium, zirconium, boron, and carbon to reduce environmental embrittlement and refine the microstructure (Sikka et al., 1993; Stoloff, 1998). Fe~Al-basealloys are used in a wide variety of applications including metalworking dies, heat shields, furnace fixtures and heating elements, and automotive components. However, limited duc~~lity at room temperature and a sharp drop in strength at temperatures above 600 "C have been the major deterrents to their acceptance as high-temperature structural materials. The room~temperature ductility has been improved by alloying with chromium and by introducing a highly eloiigated (unrecrystallized) grain structure^ resistant to hydrogen diffusion. High-temperature strength has also been improved by alloying via solid-solution strengthening and precipitation hardening (McKamey et al., 1991; Mc~aniey,1996; Sikka et al., 1993). Fe,Al-based alloys are probably the structural intermetallic materials that have been produced in the largest quantity to date. Processing methods based primarily on melting, ingot casting, and hot working have been well established and commercialized (Sikka et al., 1993; Sundar et al., 1998). The wrought material can be also be warm-rolled between 500 and 600 "C to manufacture products with room-temperature tensile ductility of 15 to 20% (Sikka, 1995). Processing of Fe,Al alloys by powder metallurgy methods has been studied to a limited extent. Prealloyed powders have typically been synthesized by gas atomization, using nitrogen, argon, or helium. Wright and Wright (1994) demonstrated that spherical powders, with oxygen contents approximately the same as those in the melt, can be o b ~ ~ i if n ~suitab~e d melt-pur~ingsystems with pure carrier gas are used. The powders can be used to form near-net-shape parts by hot isostatic pressing (Stoloff, 1998). Sikka et al. (1991) produced ~e~Al-a l l oy powders containing 2 to 5% chromium using gas atomization and then consolidated the powders by hot extrusion at 1000 "C. Hot
forging and rolling of the consolidated material at 1000 "C,followed by warm rolling at 650 "C were also successful. Considerable work has been devoted to the processing of Fe,Al-based alloys using reaction synthesis. Rabin and Wright (1991, 1992) and Rabin et al. (1992) have studied the reaction mechanisms and the densification behavior of Fe& and Fe3AI+ Cr alloys. The exotherniic reaction between the constituent powders was accompanied by rapid f o r ~ a t i o nand outward spreading of an a1uminum"rich liquid from prior aluminram-particle sites. This phenomenon led to an increase in porosity and swelling during pressureless sintering. While careful control of the relative sizes of the iron and nluiniiiurn powders and of the green density can ~i ni m i zethe extent of swelling, it is usually necessary to apply pressure during or subse~uentto the exothermic reaction to achieve complete densification. The typical procedures used for the reaction s y n t h e s i s / ~ cycle I ~ include preheating of the furnace 140 MPa and to 400 "C, ramping the pressure to then heating o f the furiiace to 1000°C at a rate of 20 "C/min while maintaining a constant pressure. The reaction is initiated at 600 "C and goes to completion within a few seconds. As-synthesized materials exhibit a fine, equiaxed grain structure (grain size < 1 0 ~ m ) . Since chromium does not dissolve within the short reaction time, prolonged solution treatments (1 200 "C/ 8 hours) and secondary HIP operations are necessary to achieve complete homogeniz~tionand full densification in Fe,A1 -t Cr alloys. Mechanical alloying has been used to produce nanocrystalline powders of Fe-25A1 and Fe-25Al-10% alloys. Zhu et al. (2000) have shown that the MA powders of these alloys contain metastable Fe (Ti, Al) solid solutions, By hot isostatic pressing at very high pressures (- 1 GPa) and low temperatures ( 800 "C), they produced fully dense products containing 200 nm size graiiis composed of the DO, and €32 phases. ~anocrystallineFe-Al-Ti alloys are being developed for tribological applications (Zhu and Iwasaki, 1999). FeAl-based alloys have an ordered cubic, €32 (cP2), structure for aluminum contents between 36 and 50 at.%. These alloys exhibit better oxidation and corrosion resistance than Fe,Al-based alloys and offer 3040% weight savings c o ~ p a r eto~ at-resista~t steels and superalloys (Deevi and Sikka, 1996; Deevi et al., 1997; Veduia, 1995; Liu et al., 1997). Because of their extreme sensitivity to environmental embrittlement in the presence of water vapor, these alloys suffer from poor room-temperature ductility in air. Some improvements in strength, ductility, and resistance to ~..*i
-
N
Powder ~ e t a l l ~ r g y
environmental embritt~ementhave been achieved via alloying with boron, zirconium, and molybdenum, aiid by maintaiiiing aluminuin contents in the range 3640% (Alexander et al., 1998). Because the B2 phase remains ordered and stable up to the melting point, FeAl-based alloys are much more difficult to process and are quite far from becoming commercially viable ma terials. Hot extrusion of small castings of €32 FeAl has been shown to be ef3Fective in breaking up the cast structure and refining the grain size through dynamic recrystallization (Gaydosli aiid Crimp, 1985). However, hot rolling of the X/M material has been unsuccessful. Powder processing by canned, hot extrusion of FeAl powders at 900 "C with extrusion ratios of 8: 1 to 12:1 is an effective method to obtain fully dense material with fine, equiaxed recrystallized grains (Vedula, 1995). The P/M-extruded material developed a (1 11) fiber texture, which led to abnormal grain growth during subsequent annealing (Stout and Crimp, 1992). Vacuum hot pressing and hot isostatic pressing have also been used to consolidate FeAl powders. However, only hot extrusion was found to be effective in breaking up the prior particle boundaries containing oxide inclusions. FeAl powders liave been produced by gas atomization, water atomization, and polymer atomization techniques. While the gas atomization process yields spherical powders, water/polymer atomization processes produce powders with irregular shapes (Hajaligol
655
et al., 1998; Strauss et al., 1998). Varin et al. (1999) synthesized nanocrystalline powders o f the Fe-4541 alloy by controlled ball milling. The as-milled powder is a disordered bcc solid solution, but it can be transformed to the ordered €32 structure by annealing at or above 600 "C. Schneibel et al. (1992) investigated the synthesis o f FeAl-2% Y203particulate composites using gas atomization 1- mechanical alloying + They demonstrated that the creep strength of the FeAlU20, composite is approximately four times that of the matrix. Recently, consolidation of FeAl alloy powders by tape casting (Mistler et al., 1998) and roll co~paction (Hajaligol et al., 1998; Deevi, 2000) was demonstrated. A complete thermomechanicdl processing sequence (Figure 10) involving roll compaction, binder burnout/ initial sintering, multistage cold rolling with intermediate annealing treatments, and final heat treatments was developed to produce very thin (0.2mm) sheets. The sheets are fully dense and exhibit a fine-grained matrix of €32 FeA1 with a uniform dispersion of alurnina particles. The room-tempe~ature mechanical properties of the P/M sheets are superior to those of I/M products.
The nickel aluminide Ni,Al has a Llz (cP4) crystal structure and remains ordered up to its melting point
Figure 10 A ~ l ~ e r ~ o ~ e c h aprocessing n i c a ~ sequence for obtaining fully dense, thin sheets of FeAl using roll-compacted material (Hajaligol et al., 1998)
656
Processes and P h e ~ o ~ ~ ~ a
( 1395 "C). Ni,Al (y') serves as the strengthening constituent in many commercial nickel-based superalloys and in this use exhibits excellent hightemperature strength and creep resistance. Unalloyed Ni3Al has negligible tensile ductility at room temperature and fails by brittle, intergranular fracture, mainly because of moisture-induced hydrogen embrittlement. This problem has largely been overcome by intense research efforts at Oak Ridge National Laboratory. The current generation of hypostoichiometric Ni3Al alloys doped with boron ( < 0.5 at.%) and alloyed with elements such as chromium, zirconium, and molybdenum exhibits excellent strength and ductility at ambient and elevated teI~peratures(Liu and Pope, 1995; Deevi and Sikka, 1996; Sikka, 1996; Deevi et al., 1997). Because of their excellent oxidation, carburization, and wear resistance at temperatures up to 1000 "C and superior mechanical properties, the Ni,Al-based alloys have found widespread applications including tur~ochargersfor heavy-duty diesel engines, automotive valves, valve seats, and pistons, dies and molds, cutting tools, and directionally-solidified blades and vanes for jet engines (Liu, 1993a,b; Sikka et al., 1992). Processing of Ni,Al has generally been accomplished via ingot metallurgy and only limited work lias been devoted to P/M processing. Because of their excellent hot and cold workability. Ni,Al-based alloys have been fabricated by several methods including hot extrusion, hot and cold rolling, isothermal forging and superplastic forming (Sikka et al., 1992). Prealloyed Ni,Al powders have been produced by atomization in nitrogen or argon, The powders are encapsulated in mild-steel cans and extruded at 1100 to 1200 "C, Extrusion ratios higher than 8:l are required to achieve full densification and to break up oxides at prior-particle boundar~es(Sikka, 1989; Wright et al., 1989). The hot-extruded material has a fine-grained structure (10 to ~ 0 ~ with m ) few micropores. Because of the su~erplastic properties, the fine"grained extruded material is suitable for near-net-shape forming by isothermal forging. Reaction synthesis of elemental powders has been N
These investigators studied the influence of parameters such as the particle size, heating rate, atmosphere, preheating temperature and extrusion temperature on the extent of chemical reaction(s) and the density of the c~n~olidated products. Nishimura and Liu (1993) showed that relative density in excess of 99% can be obtained by combining reaction synthesis and HIP operations. The use of injection molding to consolidate
~eaction-synthesized powders has also been explored (Cooper, 1993). The intermetallic compound 2 (cP2) NiAl has several a~tractiveattributes, including excellent oxidation resistance, high thermal conductivity, low density, and high melting point. The two principal limitations of NiAl are poor toughness and damage tolerance at room temperature, and inadequate strength and creep resistance at elevated temperat~res.These problems have been overcome to some extent through microand macro-alloying with elements such as zirconium, iron, molybdenum, and chromium, dispersion strengthening, and fiber reinfo Walston, 1997; Miracle and principal intended use of NiA turbine blades and vanes in jet engines, In this application, single-crystal iAl, produced by directional solidification, is hoped to replace n i c ~ e l ~ b a s e ~ superalloys. Other potential applications include hight~mperature heater exchan~ers, injection molds, automotive components and substrate/thin-film components in electronic and magnetic devices (Noebe et al., 1993). Processing NiAl alloys is focused on two approaches (a) directional solidification of single crystals and eutectic alloys and (b) P ~ ~ - ~ r o c e sofs i~ne-gr~i ~g ned, polycrystalline alloys. The P/M processing of solidsolution-strengthened and ~~spersion~st~engthened NiAl alloys consists of the following steps: (a) powder synthesis by inert gas atomization, mechanical alloying or reaction synthesis and (b) powder consolidation via HIP, vacuum hot pressing, hot extrusion, or swaging. For example, Bowman et al. (1992) used hot extrusion ized at 900 "C to 1175 "C of a r g o ~ - ~ ~ s ~ a t o mpowders (encapsulated in mild-steel cans) t bars of NiAl and NiAl-Zr alloys. al. (1999) have consolidated the of injection molding NiAl-27Cr using a co~~bination and HLP. The microstructure of the alloy contained a mixture of very fine NiAl grains and uniformly dispersed chromium particles ( 1.7pm), and was found to be resistant to grain rowth at tem~e~atures up to 1350°C. NiAl alloys containing fine dispersions of Al,Q, or AlN particles have been developed at the Max Planck Znstitut fiir ~etallforschung~ Germany and the NASA Clenn Research Center, USA, respectively. Arzt et al. (1993) and Crahle and Arzt (1997) describe the develop~entof o~ide-dispersioI~-strengthened(ODS) NiAl alloys by a conventional m~chanical alloying approach, followed by c~nsolidationand coarse-grain recrystallization. ODS-NiA1 alloys exhibit excellent N
Powder ~ e t a l l ~ r g ~ creep resistance above 1200°C and are envisioned for applications in which ODS-Ni-based or Fe-based superalloys are currently used. Cryomilling of prealloyed NiAl at liquid nitrogen atmosphere is used to produce NiAl powders containing a dispersion of AlN (Aikin et al., 1994; Noebe and Walston, 1997). The creep strength of hot extruded + heat treated NiA1-AlN alloys increases dramatically with the volume fraction of ALN particles and approaches that of a firstystal superalloy (NASAIR 100). ned NiAl alloys also suffer from 400 "C. Currently, low fracture toughness below concerted efforts to improve the fracture toughness of Y~O,/AlN-dispersed NiAl alloys via ductile-phase toughening are underway (Noebe and Walston, 1997). Hence, NiAl alloys strengthened by dispersions o f Y20,/A1N particles and toughened by ductile reinforcements such as Cr or MO particles hold promise for a balance of creep and fracture pr Reaction synthesis of NiAl alloys has been inv by Deevi and Sikka (1995). They used blended elemental powders sealed in aluminuni cans. The preforms were preheated to 425 "C, and extruded with an extrusion ratio of 20:1 to produce fully reacted and consolidated NiAl rods. Recently, Morsi et al. (2000) lxwe re-examined this combined process (hot extrusion+ reaction synthesis) and identified optimum conditions to avoid both macroscopic fdure and microscopic defects. The RS process has also been extended to the preparation of NiBl-AL,O, co~posite$via vacuum hot p~essingof a mixture of A1 and NiO powders at 1200°C for 2 hours (Zhu and Abbaschian, 2000). N
Refractory metal silicides have been developed for a wide range of applications including heating elements in high-temperature furnaces, structural materials operating at temperatures in the range 1200 to 1600 "C, and protective coatings for refractory metals (Vasudevan and Petrovic, 1992; Kumar, 1995; Petrovic and Vasudevan' 1999). Molyb~enumdisilicide, MO C11, (tl6) i s the most well-known refractory metal silicide. It has a high melting point (2030"C), low density (6.24 Mg/m3), and excellent oxidation resistance at high temperatures. Its major drawbacks are: (a) poor tensile ductility and fracture toughness at temperatures below 1000 "C, (b) inadequate strength and creep resistance at temperatures above 1200 "C, and (c) susceptibility to 'pest' oxidation at intermediate temperatures ( w 500 "C). The low-temperature fracture resistance of MoSi2 has been improved through N
657
ductile-phase toughening, A ~ p r o ~ c h eto s improve elevated-temperature strength have included alloying with tungsten and introduction of a variety o f reinforce~entssuch as ~hiskersand fibers of Sic Si,N, and particulates of Sic, Tic, (Hardwick et al., 1993; Jayashankar et al., 1997). owder metallurgy has played a dominant role in the synthesis and fabrication of MoSi, and its coinposites (Hardwick et al., 1993; Patrick and Van Aken, 1994; Scholl et al., 1999)" MoSi,! powder is produced commercially by reacting elemental powders of MO and Si. Unfortunately, such powders contain undesir~blyhigh levels of oxygen ( 6000 ppm) which are manifested as silica particles or films on grain boundaries of the consolidated products. The presence of intergranular SiO, particles causes crack nucleation at low temperatures and promotes viscous flow at high temperatures. Powders produced by mechanical alloying of elemental powders (Schwartz et al., 1992) generally contain much higher levels of amorphous silica than are found in powders produced by reaction. silica content include additions of elements powders have been CO hot pressing. Hardw reaction synthesis of 200 MPa and 1400 "C 600 wppm oxygen and an average grain size of 40pm. Vas~devanand Petrovic (1992) and Jayashankar et al. (1997) have documented the mechanical properties of MoSi, and its alloys/composites. Because of interfacial reactions between the ductile reiiiforcements and MoSi,, the interest in molybdenum silicides has recently shifted from MoSi, to Mo5Si3, D8, (t132). While MOSS3 exhibits superior creep resistance at high temperatures, its oxidation resistance at temperatures higher than 1000 "C is However, the oxidation properties of improved substantially by the addition effect is attributed to the formation of a protective borosilicate glass. ment of oxidation1.0 to 4.5 wt.% Si Vasudevan, 1999; processing history, these alloys may contain multiple phases including MO&, MosSi3 (Tl), Mo,SiB2 (T2), MOB and molybdenum solid solution (Liu et al., 1999; Nunes et al., 1997; Sakidja et al., 1999). The threephase field Mo-Mo3 i-T2 i s of particular interest for ductile-phase toughening of Mo3Si and T2 phases by MO particles. These alloys are currently processed by a N
658
Processes and Phenomena
P/M approach involving HIP consolidation of prealloyed powders and hot extrusion of the P/M billets. Much work has been performed on niobium-based intermetallic alloys over the last decade at the Air Force Research Laboratory and at the General Electric Company. Dimiduk et al. (1993) have summarized the syste~aticefliorts to develop in-situ composites containing a solid solution of niobium and Nb,Si,, D8, (tI32). Several alloys containing a wide range of silicoii contents (0.6 to 37.5at.94 Si) have been examined. However, the best blend of properties is o b t ~ n e dfor compositions between 6 and 18.7at.Yo Si. Fracture toughness of the in- sit^ composites increases with the amount of the niobium solid solution providiiig the ductile-phasetoughening mechanism by crack bridging. Processing of the Nb/NbSSi, alloys was accomplished mainly by ingot metallurgy using vacuum arc melting, canned hot extrusion of the ingots between 1485°C and 16SO"C, and then heat treatments at 1500 "C for 100 hours to allow the decomposit~onof Nb3Si to a mixture of Nb+Nb,Si,. Limited studies have been conducted on P/M processing of these alloys by milling and blending of NbjSi3 and Nb powders, and vacuum hot~pressingat 1650"C under a pressure of 5 ksi for 2 to 4 hours. Hot-pressed compacts were at least 98% dense and had an average oxygen content of 1700ppm (Nekkanti and ~ i m i d u k ,1990). While the I/M route has been successful only for the preparation of low-silicon alloys, P/M processing could produce dense compacts of both low- and highsilicon-containing alloys, including 1OO0/o Nb jSi3. The catastrophic oxidation behavior of Nb-Nb,Si, alloys makes binary Nb-Si alloys essentially unsuitable for high-temperature applications. In order to seek improvements in oxidation resistance and creep resistance at high temperatures, Nb-Si alloys containing different amounts of Al, Ti, Hf and Cr have been explored (Subramanian et d.,1997; Jacksoir et al., 1996; Bewlay et aE., 1999 and the chapter by Bewlay et al. in this volume). These alloys generally contain multiple phases including P-N b, (Nb,Ti)jSi3, and Cr2Nb. With an increase in the amount of the C1S (cF24) Laves phase, Cr2Nb, the hot workability of the alloys deteriorates rapidly. Under these circumstances, P/M processing may provide a viable alternative to ingot casting and wrought processing.
P/M processing of internietallic niaterials has witnessed tremendous advances in the last decade.
Synthesis of pre-alloyed intermetallic powders by atomization techniques remains the most developed and reliable approach for producing high-quality P/M parts. Novel powder s y n t ~ ~ s te~hniques, is such as cryomilling, mechanical alloying, and reaction synthesis, have provided opportunities to produce intemetallic alloys and composites with nanocrystalline or amorphous structures, often with unusual properties. Standard consolidation methods, viz., HIP, hot pressing, and hot extrusion have been applied very successfully to intemetallics to manufacture nearnet-slmpes of coniplex geometries and controlled microstructures. At the same time, new techniques, such as shock consolidation, plasma-assisted sintering, tape casting, and reactive sintering, arc being used increasingly to fabricate small ~ u ~ n t i ~ of i e sintermetallic materials. Advances in the synthesis, consolidation, and the~omechanicalprocessing of P/M intermetallics have benefited immense~yfrom f u n d a ~ e n t studies a~ on the phase equilibria, kinetics and mechanisms of phase transformations, microstructure evolution, and workability, as well as by the design and imple~entationof novel processes. It is also i ~ t e r ~ s t i nthat g processing of intermetallics can be performed on existing maiiufactu~ingequipment with minimal modifications. F u r t h e ~ o r e ,development and application of process models and material models is rapidly transforming powder metallurgical processing from a trial-and-error approach to a science-based rnanufacturing practice, Future activities that will enhance the transition of intermetallic materials from scientific curiosities to engineering materials include: (a) definition of processing windows and demonstration of process robustness in real manufacturing environments; (b) further development of models of material behavior and process models for specific inter~etallics; (c) changes in design methodologies for components made of intermetallics with specific allowanc~sboth for their attractive as well as for their limiting properties; and (d) assessment of production and lifecycle costs for intermetallic materials and ~omponents manufactured through diflierent routes ~~~
Aikin, B. J. M., Dickerson, R. M., Jayne, D. T., Farmer, S., l ~ . and ~ h i t ~ e n b e r gJ.D. ~ r , (1994). S ~ r i ~ E~a ~ t aMater., 30, 119. Alexander, D. J., Maziasz, P. J., and Wnght, J. L. (1998). Mater Sci. Eng., ~ ~276.~ g ,
I
Powder ~ e t ~ l l ~ ~ g y
659
Arunachalam, V. S. and SundareSan"R. (1991). In Materials Science and Technology: Processing of Metals and Alloys (eds R.W. Cahn, P. Haasen, and E.J. ICramer), vol. 15, VCH Publishers, New York, 137 Arzt, E., Ashby, M. F., and Easterling, K. E. (1983). Metall.
Ashby, M. F. (1990). Operating ~ a n ~ a l HIP ~ o 487, r Version HIP 6.0. Cambridge University, Cambridge, UK. Austin, C. M. and Kelly, T. J. (1995). In G a ~ m aTitan~um Aluminides (eds Y-W. IGm. R. Wagiier, and M. Yamaguchi). TMS, Warrendale, PA, 21. Baccino, R. and Moret, F (2000). Materials aid Design, 359. Bacciiio, R., Moret, F., Pellerin, F., Guichard, P.. and Raisson, G. (2000). ~ a t e r ~ and a ~ Design, s 21, 345. Benja~in,J. S. (1976). Scientific American, 234(5), 40. Benjamin, J. S. and Volin, T. E. (1974). Metall. Tram., 1929. Berczik, D. M. (1997). U.S. Patent No. 5,595,616 issued on Jaiiuary 21, 1997, United Technologies Corporation, East Hartford, CT. Bewlay, B. P., Zhao, J-C., Jackson, M. R. (1999). Mater. Res. Soc. Symp. Proc., Blackburn, M. J. and Smith, M. P. (1980 Aircraft, United Technologies Corporation, West Palm Beach, FL, unpublished research. Bose, A., Rabin, B. H., and German, R. M . (1988). Powder ~ e t a l l u Ir d~. ~, 20, 25. S. V., and Locci, 1. E. Bowman, R. R., Noebe, (1992). ~ e t ~ lTruns. l. Buffered, A. S. (1972). Metallurgy for High Performance Applica~ion,s(eds J .J. Burke and V Weiss). Syracuse, New York, 303. Calderon, H. A., Garibay-Febles, V., Cabrera, A., Umemoto, M., Cabanas-Moreno, J. C., Matcr. Res. Soc. Symp. Proc., Cbatterjee, A., Baur, M.,Appel, F., Bartels, A., Knabl, W., Kestler, H., and Clemens, €4. (1999). In Gumma Titurizicm A l u m i ~ i ~ e1999 s (eds. Y-W. Kirn, D.M. Dimiduk, and M.H. Loretto). TIMS, Warrenda~e,PA, 401. Cheng, S. C., Wolfenstine, J., and Sherby, 0. D. (1992). Metall. Tram., B A , 1509, Chung, W.H., Hwang, S. J., and Kini, N. J. (1995). In Gamma ~ ~ ~A al ~ n~ ~i(eds. z ~ ~Y-W. ~i ~ Kim, ~ s R. ~ ~ g n eand r , M. Yarnaguchi). TMS, Warrendale, PA, p. 8 1 1. Clemens, H., Glatz, W., Schratter, P., Yolton, C. F., Jones, P. E., and Eylon, D. (1995). In G u ~ m uTitani~mAluminides (eds Y-W. Kirn, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, p. 555. Ckmens, H., Eberhardt, N., Glatz, W., Martiz, H-P., Knabl, W., and Rehels, N. (1997). In S t r ~ c t ~ r aIntermet~llics l 1997 (eds. M. V. Nathal, R. Darolia, C. T. Liu, P L. Martin, D. B. Miracle, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, 277. I
Couniban, P J., Cr Mater Sci. Efig., Deevi, S. C.(2000).
A., and Thadhani, N. N. (1999).
Proc., 364, 917. ~ ~357. ~.s~ Deevi, S. C. and Sikka, V. IC.(1996). I n t e r ~ i e t a ~ 4, ,V. K., and Liu, C. T. (1997). Prag. at er (1995). In Ganzmcr Titamiim Alunar (eds U-W. Kim, R. Wagner, and M. Yai~a~uchi). T Warrendale, PA, 3. Dimiduk, D. M. (1999). ~ a t Sri. ~ r Dimiduk, D. M., Mendiratta, M. G., (1993). In S t r u c t ~ r a~~n t ~ r m ~ t a ~ l i Lewandowski, C. T. Liu, P. L. M M. V Nathal). TMS, Warrendal Dutton, R. E., S h a ~ a s u n d ~S., r , and S ~ ~ i a t iS. n ,L. (1995).
Amsterdam, The Netherlands, p. 2627. Eylon, D., Cooke, C. M., Yoltoii, C. F., Nachtrab, W. T., and Furrer, B. U. (1993). In ~ 1 ~ n . s eSe~zinar e '93 (eds. Bildsteiii and R. Eck). Plarisee, Reutte, Austria, 552. Froes, F. H., Suryaiiarayan sell, K., and LI, C-G. (1995). Mater. Sci. Eng., Fuchs, G. E. (1993). Mater.
Gaydosh, D. 9. and Crimp, M. A. (1985). Mater. Re,r. I S I C f ~ . Symp. Proc., 39, 429. Ceding, R., Leitgeb, R. and S c ~ i ~ aF-~ s ~ ~ , Scz. Eng., ~ 2239 ~ ~ , German, R. M. (1994). Powder M e t ~ ~ Science. ~ ~ ~ rMetal ~ y Powder lndustries Federation, Princeton, N 9. Gialanella, S., Lutterotti, L., Molinari, A., Kazior, J., and Pieczoiika, T. (2000). ~ ~ ~ t e r i ~ i 8, e t279. ~ll~~s~ ., and Loretto, M. H. (1998). Mater. 1997). Arta Metall. Mater., Habel, U., Yolton, C. F,, and Moll, J. H. (1999). In G a ~ ~ a T i t u n i u ~ A l ~ n ~ i n ~ d1999 e s (eds Y-W Kim, D. M.
660
Processes and Phenomenu
Dirniduk, and M. H. Loretto). TMS, Warrendale, PA, 301. ajaligol, M. R., Deevr, S. C., Sikka, V. K., and Scorey, C. R. (1998). Mater. Sci. Eng., ~ 2 5 $ 249. , Hardwick, D. A., Martin, P. L., Patankar, S. N., and Lewandowski, J. J. (1993). In Structural Intermetallics Lewandowski, C. T. Liu, P. L. le, and M. V. Nathal). TMS,
Lebrat, J-P., Varrna, A., and Miller, A. E. (1992). Metall. Trans., 23A, 69. Lee, D. N. and Kim, H. S. (1992). Powder ~ e ~ a l l35, . , 275, Lee, T. K SKim, J. H., and Hwang, S. K. (1997). Metall.
LeHolm, R., Clemens, H., and Kestler, H. (1999). In Gamma Titani~~ A l u ~ i n ~ d e1999 ~ s (eds U-W Kim, D. M. Dirniduk, and M. H. Loretto). TMS, Warrendale, PA, 25, Lenel, F. V. (1980). Powder Metallurgy: Principles and Ap~lications. Metal Powder Industries Federation, ashimoto, H., Abe, T., and Sun, Z. (2000). Irzter~etall~cs, 8, Princeton, NJ. Lipsitt, H. A., Shechtman, D., and Schafrik, R. E. (1975). Helle, A. S., Easterling, IS.E., and Ashby, M. F. (1985). Acta Metall. Tram., tiA, 1991. Metall., 33, 2163. Lm, C. T. (1993a). Mater. Res. Soc. Symp. Proc., 2 U, W-C. (1995). Mater Res. Soc. Symp. R. Liu, C. T. (1993b). In S t r ~ c t ~ r aI l~ t e r ~ e t a l l i c(eds. s Darolia, J. J. Lewandowski, C. T. Liu, P. L. Martin, D. B. Chesnutt, J. C. (1995). In ~ n t e r ~ n e t a l l i ~ Miracle, and M. V. Nathal). TMS, Warrendale, PA, 365. CoFnpounds: P r ~ n c ~ ~and l e ~ ~Practice (eds. J. H. Liu, C. T. and Pope, D. P. (1995). In Inter~etallic Westbrook and R. L. Fleischer), vol. 2. Wiley, C o m p o u n ~ Principles : and Practice (eds J.H. Westbrook Chichester, UK, 73. and R.L. Fleischer), vol. 2. Wiley, Chichester, UK, 17 Irnayev, R. M. and Imayev, V. M. (1991). Scriptm Metall., 25, Liu, C. T., Maziasz, P. J., Clemens, D. R.,Schneibel, J. H., 2041. Sikka, V. K., Nieh, T. G., Wright, J., and Walker, L. R. M., and ~alish~hev, G. A. (1992). (1995). In G a ~ ? n T~tan~um a Alu~inides(eds U-W. Kim, R, Wagner, and M. Yamaguchi). TMS, Warrendale, PA, Inkson, B. 5. and Clernens, H. (1999). Mater. Res. Soc. Syinp. 679. C. T., Stringer, J,, Mundy, J. N., Horton, L. H., and . P., Rowe, R. G., Skelly, D. W., Liu,Angelini, P. (1997). I ~ ~ e r ~ e t a l l5,~ c579. s, and Lipsitt, H. A. (1996). JOM, 48(1), 38. Liu, C. T., Schneibel, J. H., and HeatherXy, L. (1999). Mater. Jayashan~ar,J. S , , Ross, E. N., Ea Res. Soc. Symp. Proc., 552, KK 6.2.1. Martin, P. L, and Hardwick, D. A. (1995). In I n t e ~ m e ~ a ~ ~ i c Compoundr: Principles and Practice (eds. J. H. Westbsook Frommeyer, G. (1999). Acta Mater., 47, 3655. and R. L. Fleischer), vol. 1. Wiley, Chichester, UK, 637. Joslin, D. L., Easton, D. S., Liu, C. T., Maziasz, P. J., Liu, C. T., and Wnght, J . L. (1997). In (1995). Mater Res. Soc. Sjmp. Proc., Structural ~nter~netalli~~s 1997 (eds M. V Nathal, R. Kestler, H., Clemens, H., Baur, H., Joo Darolia, C. T. Liu, P. L. Martm, D. B. Miracle, R. Cam, G., Bartels, A., Schleinzer, C., and Smarsly, W. Wagner, and M. Yarnaguchi). TMS, Warrendale, PA, (1999). In G a ~ T~i t ~a n i u tAl~minides ~ 1999 (eds Y-M. 245. ICim, D. M. Dirniduk, and M. H. Loretto). TMS, Maziasz, P. J. and Liu, C. T. (1998). Metall. Mater. Trans., Warrendale, PA, 423. 2 9 ~ 105. , Kirn, U-W. (1994). JOM, McCullough, C., Valencia, J. J., Levi, C. G., and ~ e h r a b i a n , ICirn, J. K., Kim, J. H., Le wang, S. K., Nam, S. W., R, (1989). Acta Metall., 3'9, 1321. mid ICini, N. J. (1999). 111 Gamma Titcimim Aluminides McKamey, C. G. (1996). In Physic~l Metallurg~ dnd 1999 (eds U-W. Kim, D. M. Dimiduk, and M. H. Proces.sing of ~ n t e ~ ~ ~ e t a~ l loi c~ p o (eds, u n N. ~ S, StoloR Loretto). TMS, Warrendale, PA, 23 I. and V K. Sikka). C ~ a p ~ and a n Hall, New York, 351, Kini, Y-W. and Dimiduk, D. M. (1997). In Structural McKamey, C. G., DeVan, J. H., Iatermetallics 1997 (eds M. V. Nathal, R. Daroilia, C. T. V K. (1991). J. Mater. Res., Liu, P. L. Martin, D, B. Miracle, R. Wagiier, and M. Miracle, 2). B. and Darolia, Yarnaguchi). TMS, Warrendale, PA, 531. C o ~ p o ~Principles ~ n ~ ~ :and Practice (eds. J. H. Westbrook lar, E. ed, (1984). A S M Metals Hmndbook, 7, 21. and R. L. Fleischer), vol. 2. Wiley, Chichester, UIC, 53. och, C. C. (1991). In Materials Science and Technology: Mishurda, J. C. and Perepezko, J. H. (1991). In Processing of Metals and Alloys (eds. R. W. Cahn, P. ~ i ~ r o s t r ~ c t u rProperty e R e l a t i o n s ~ ~ sin ~ i t a n ~ u ~ Haasen, and E. J. Kramer). VCH Publishers, New York, A~urnini~~es aad Alloys (eds. Y-W. Kim and R. R. vol. 15, 193. Boyer). TMS, Warrendale, PA, 3. Koch, C. C. (1998). Mater. Sci. Eng., A2 Mistler, R. E., Sikka, V. K., Scorey, C. R., McKe Kuniar, K. S. (1995). In ~ n ~ e ~ ~C eo ~ ~~ ol ulPrincples ni ~~ ~ : and Elajaligol, El. R. (1998). Mater. Sci. Eng., and Practice (eds. J. H. Westbrook and R. L. Fleischer), Moll, J. H., Yolton, C. F., and McTiernan, B. 3. (1990). Inter. J. Powder Metallurgy, 26, 149. vol. 2. Wiley, Chichester, UK 21 I.
Powder ~ e t ~ l l ~ r g ~ Moore, J. J. and Fen , H. J. (1995a). Prog. Mater. Sci., 39, 243. Moore, J. J. and Feng, H, J. (1995b). Prog. Mater. Sci., 39, 275. Morsi, K., McShane, H. B., and McLean, M. (2000). Mater. Sci. Eug., A290, 39. Murugesh, L., Venkateshwara Rao, K. T., and Richie, R. 0. (1994). M a t ~ r Sci. . Eng., . (1990). Mater. Res. 993). Acta Metall. Mcater~, 113. Noebe, R. D., Bowman, R. R., and Natthdl, M. V (1993). Noebe, R. D. and Walston, W. S . (1997). In Structural I n t c ~ ~ e ~ u i1997 ~ i c s(eds. M. V. Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. ~arnaguchi),TMS, ~arrendale,PA, 573. Nunes, C. A., Sakidja, R., and Perepezko, J. H. (1997). In Structural 6itermerallics 1997 (eds. M.-V. Natlial, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. ~ a g n e r ,and M. Yamaguchi). TMS, Warrendale, PA, 831. mann, R. (1993). Mater.
66 1
Semiatin, S. L. (1995). In Garn~caT i t a n i u ~A~l u m i ~ i ~ (eds. es Y-W. Kiln, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, 509. Serniatin, S. L., Frey, N., ~l-Soudani,S. M., and Bryant, J. D. (1992). Metall. Trans., 2 Semiatin, S. L., Cornish, G. ylon, D. (1994). Mater. Sci. Eng., 19186, 45. Semiatin, S. L., Dutton, R. E., and Goetz, Scripta Metall., 35, 85.5. Semiatin, S. L., Seethararnan, V., and Weiss, I. (1997a). In Advances in the Science and Technolog~of Titaniu~Alloy Processing (eds I. Weiss, R. Ssinivasan, P. J. Bania, D, Eylon, and S. L. Semiatin). TMS, Warrendde, PA, 3. Serniatin, S, L., Chesnutt, J. C., Austin, C., and S e e t ~ a r a ~ a n ? V. (1997b). In Structur~lI n ~ ~ r ~ e ~ 1997 ~ l l i(eds. ~ , sM. V. Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, 263. rainan, V., and Weiss, 1. (1998). mate^.
Semiatin, S. L., Seethara~dn,V., and Chosh, A. K. (1999). Phil. Trans. R. Soc. London A, 3 Shechtman, D., Blackburn, M. J., and Lipsitt, H. A. (1974). Me~call.Trans., 5, 1373. Shih, D. S. and Scarr, G. K. (1991). c at er. Res. Soc. S ~ ) m ~ . Proc., 213, 727. ., aiid Lee, D. N. (1999). Shimnit, S. and Oyane, M. (1976). Inr. J . Mech. Sci., 18, 285. (1994). Mater. Sci. Eng., Sikka, V. K. (1989). Mater. M a n u ~Processe~~? Sikka, V. K. (1995). Mater. Res. Soc. Symp. P Petrovic, J. J. and Vasudevan, A. K. (1999). Mafer. Scz. Eng., Sikka, V. K. (1996). In P~~ysi~al Me~cal1urg.yand Processing I~ter~~etallic Compounds (eds. N. S. Stolof? and V. ohler, S. P, (1984). M ~ ~ a ~l sa n ~ ~ o o k : Sikka). Chaprnan and Hall, New York, 561. gy, vol. 7. ASM International, Metals Sikka, V. I<., Gieseke, B. G., and Baldwin, R. H. (1991). In Park, OH, 419. Hecat Resistant ~ c a t e ~ ~ (eds. ~ a l ~ K. s Natesari and 0. J. Rdbin, B. H. and Wright, R. N. (1991). Metall. Trans.,22A, Tillack), ASM International, ~ a t e r i a l Park, s OH, 363. 277, Sikka, V. K., Mavity, J. T., and Anderson, . H. and Wright, R. N.(1992). Metall. Trans., 23.4, Sci. Eng., A153, 712. 35. Sikka, V. K., Viswaiiathan, S., and McMamey, C. 6. (1993). Rabin, B. H., Wright, R. N,, Knibloe, J. R., Raman, R. V., In Structu~a~I n ~ ~ r ~ e t a l l(eds. i c ~ ~R, Daolia, J. J. and Rale, S. V. (1992). Mcater. Sci. Eng., A153, 706. Lewandowski, C. T. Liu, P. L. Martin, D. B. Miracle, Sakidja, R,, Wilde, G., Sieber, H., and and M. V. Nathal). TMS, Warrendale, PA, 483. (1999). Mater. Rcs. Soc. Syrnp. Proc., StoloR, N.(1998). &Ifiter. Sci. Eng., Sali~hchev,G., Imayev, R., Imayev, V., and Shagiev, M. Stoloff, N. (1999). Mater. Sci. Erzg., (1995). In G a m ~ aT ~ ~ a nAi luu~~ ~ ~ ~(eds. ~ d Y-W. e s Kirn, Stoutt, J. J. and Crimp, M. A. (1992). Mufer. Scr. Eng., R. Wagner, aiid M. Yamagtichi). TMS Warrendale, PA, 33s. 595. Strauss, J. T., Scorety, C , R., McK le, P,, and Rosler, J. (1992). Mater. Sci. M. R. (1998). Mater. Sci. Eng. Submmanian, P. R., Mendiratta, T., Gente, T., Segtrop, K., and Dahms, and Stucke, M. A. (1997). Mater, Set. EBg., M. (1997). In S t ~ ~ c t Iun ~ e~r l~ e ~ a Z1997 l ~ c s(eds. M. V. Sundar, R. S., Baligidad, R. G., Nathal, R. Darolia, C . T. Lm, P L. Margin, D. B. Sastry, D. H. (1998). Mafer. S Miracle, R. Wagner, and M. Yarnaguchi). TMS, Taguchl, K., Ayada, M., Ishihara, K. N., and Shingu, F. W. Warrenda~e,PA, 453. (1995). In G a ~ Tit~nium ~ ~ a A l u ~ i ~ i (eds. d ~ s Y-W. Kim, ., and Kieback, R. (1999). Mater. Sci. R. Wagner, and M. ~ a i ~ a g u c h iTMS, ). ~arsendale,PA, p. 619. Schwarz, R. B., Srinivasan, S . R., Petrovic, J. J., and Takasaki, A. and Furuya, Y (1999). Nan{)~~tr~ctura~ Maggiore, C. J. (1992). Mater. Sci. Eizg., G155, 75. Ma~erial~s? 11, 1205.
662
Processes and Phenomerm
umagai, T., Nakamura, M., and Kikuchi, Structural Intervnetallics (eds. R. Darolia, J. J. Lewandowski, C. T. Liu, P. L. Martin, D. B. Miracle, and M. V. Nathai). TNS, Warrendale, PA, 167. Valencia, J. J. McCullough, C., Levi, C. G., and Mehrab~an,
Alun~inides (eds. Y-W. Kim, R. Wagner, and M. Yamaguchi). TMS, Warrendale, PA, 949. Wnght, J. K. and Wnght, R. N. (1994). In Processing, Properties and A p ~ l i c a t i o nof~ ~Iron Alu~inides(eds. J. H. Schneibel and M. H. Crimp). TMS, Warrendale, PA, 47. Wright, R. N., Rabin, B, N., and Knibloc, J . R. (1989). Mater. Manuf. Processes, Calka, A. (1999). ~ a ~ a g u c hM., i , Inui, H., and Ito, K. (2000). Acta Mater., 301. Vasudevan, A. K. and Petrovic, J. J. (1992). Muter Sci. Eng., Yolton, C. F (1990). In P / M in AcJrospace Q F I Dejknse ~ ~ e c ~ n o l o g i(ed. e ~ F. €€. Fres), vol. 1. Metal Powder Vedula, K. (1995). In In~ervnetullicCo~pounds:P r i n ~ ~ l and es Industries Federation, Princeton, NN, 123. Practice (eds. J. If.Westbrook and R. L. Fleischer), vol. 2. Yolton, C . F., Habel, U., and Clemens, H, (1997). In Wiley, Chichestes, UK, 199. Advanced I ~ a r t i c ~ l u~ t~~ t e r i uand l . ~Processes (eds. F. €3. Wang, G, X. and Dahms, M. (1992). Powder Metall. Int., 2 Froes and J. C. Hebe~sen). Metal Powder Industries 219. Federation, Princeton, NH, 161. Wang, G. X. and Dahms, M. (1993). In Strticttirnl Zhu, H. X. and Abbaschian, R. (2000). Mater. Sci. E q . , I n t ~ r v n e t ~ l l (eds. i c ~ R. Darolia, J. J. Lewandowski, C. T. 2 8 ~ ,1. Liu, P. L. Martin, D. B. Miracle, and M, V. Nathal). Zhu, S. M. and Iwasaki, K. (1999). Mater Scz. Eng., TMS, Warrendale, PA, 215. 170. Wang, J. N., Schwarz, A. J., Nieb, T. G., Liu, C. T., Sikka, Zhu, S. M., Tamura, M,, Sakamoto, K., and lwasaki, K. V. K., and Clemens, I). (1995). In ~ u ~ v n~ ui t a n i u ~ (2000). Mater. Scz. Eng.,
Troy, N Y , USA
Over the last several decades, there have been rapid advances in the field of thin-film metallurgy, which has become an increasingly important area of materials science and engineering. While there is no rigorous definition of how thin t ~ films i ~ really are, typically films thinner than 1OOyin can be characterized as thin films. The advantages of thin films can be harnessed for engineering applications in two general ways. Firstly, thin films can be used to tailor and/or modify surface and interface properties of bulk materials. De~ositinga corrosion-resist an^ film onto a turbine blade made from a high-temperature material, using a coating of TiCN (Donnelly et al., 2000; Karlsson et al., 2000) on WC-based cutting tools to improve the wear resista~ce, and ~epositing a Tin coating on steel (Randhawa, 1986; Hedenqvist et al., 1990) to improve surface finish and obtain a golden color are but a few examples of thin films used to modify the properties of bulk materials. A second important way in which thin films can be used i s where the film itself plays the primary functional role and the substrate merely provides mechanical support to the thin film. Examples of this mode of use include Cu-based interrnetallics in soldering and chip packaging applications (Ray et al., 1996; Zribi et al., 1999), CuInSe, films in solar-cell applications (Basal, 1993; Rockett et al., 1994.1, FePd alloys in ~agnetic-storageapplications (Durr et al., 1999>, and metal-silicides (Wittner, 1983; Murarka,
1993) used as electrical contacts in integrated circuits. The exacting requircmciits of microstructural and compositional control of thin-film alloys and compounds placed by this class of ~ ~ p l i c a t i o nhave s contributed tremendously towards pushing the frontiers of deposition technologies to their limits and enabled a deeper understanding of i n t e r d i ~ u s i o ~ and reaction mechanisms in multilayer thin-film stacks. An intermetallic c o ~ p o ~ nis dtrad~tionallydefined as a compound between two metals, often with components of covalent and ioiiic bonding. In this chapter, however, we will also ~ o n ~ i d ~e r ~ ~~-~~ ~ - V semiconductors and their related ternaries, metal- and semiconductor-nitrides, carbides, and silicides as intermetallics. The justification for this is the fact that these c~mpoundsare in many ways s i ~ i l a rto traditiona~ intermetallic compounds. Some common features are: long-range order, definite compound stoichioi~etry, well-defined melting points, high strength and low ductility (e.g. MoSi,) (McKamey et al., 1992). Another reason for our expanding the scope of the discussion is the importance of these materials in thin-~lmform in real applications. For instance, 111-V compounds such as GaAs, GaN aiid AlN offer unique properties for several applications including optoelectronic and surface-acoustic-wave devices, and sensors ( 1991). Our effort to integrate discussions on thin films on the one hand, and on bulk intermetallics on the other,
~ n ~ e r m e t ~Cornpounds: ~lic Vol. 3, Principles aizd Practice. Edited by J. H. Westbrook and R.L. Fleischer. @ZOO2 John Wiley & Sons, Ltd.
is aimed at p r o ~o tin ga synergy between the traditional intermetallics community and the thin-film c o m u n ity . y of the fundame~italissues regarding structure and se transfor~ationsare well understood in several intermetallic systems in the bulk form. This knowledge can serve as a very useful foundation to understand the fundamental issues in their thin-film counterparts. The phenomenal flexibility of the inherently non-equilibrium nature of thin-film processing allows a greater variety of microstructures and metastable phases to be synthesized as compared to bulk intermetallics. Revealing the relationships between these microstructures and processing techniques can give insights into new methods of synthesis of new bulk phases. Broadly, there are two challenges in the area of thinfilm intermetal~ics.One is to create desired phases and rations (alloys and n~ultilayers), specified ructures (e.g. grain size and orientation), and desired compositions. The other main challenge is to correlate - at the atomic level - the relationships
is critical from the viewpoint of understanding the influence of p~ocessin parameters on properties, and developing new and alable routes for synthesizing thin-film interinetallics for real applications. In this chapter, we focus our discussion on the tec~niquesfor forming thin-film inte~etallics,in the context of the first challenge. Depositing a thin film of an intermetallic compound with a specific s~oichio~etry can be very difficult since boiling points, sputter yields, light absorption, chemical reactivity and electrochemical potentials of the elemental co ~p o n en tsconstituting the compound are generally quite different. Controlling the microstructure can be equally challenging because thin films have a higher sur~ace/volumeratio and higher defect concentrations when compared with their bulk counterparts. Consequently, a large fraction of the atoms of cinity of point defects, surfaces, oundaries, causing many struts of the film to be very different froin those of bulk materials (see the chapter by oken). These actors also result in niult~pledriving rces for compositional changes and fast diffusion kinetics, making it common for as-deposited films, or even annealed thin-film stacks, not to have the desired composition, phase (Baeri et al., 1988; Chen et al., 4), or niicrostructure (Vang and Rockett, 1994; anath er: al., 1995). As a result, post-deposition or
multiple thermal processing steps are often required in thin-film processing. For example, annealing Ti/Si thin-film bilayers, does not result in the formation of either C54 (oF24) TiSi, or bP16 Ti,Si3 both of which should form in the 33-63 at.% Ti composition range according to the equilibriuni Ti-Si phase diagram (Yu er al., 1998). Instead, the C49 (oC12) TiSizphase forms first because of its much lower nucleation energy, and an additional thermal treatment is required to transform it into the low-resistiv~tyC54 phase. In addition to the above factors, submicron lateral dimensions and non-planar substrates - both frequently encountered in device applications - pose further challenges to control over film continui~y and uniformity, For example, the deposition rate on the sidewall and the bottom of high width~to-depthaspect-ratio holes can be very different from that rate outside the holes, resulting in n o n - ~ i n ~ ffilins o r ~ with varying properties. In many cases, thermal t r e a t ~ e i ~alone t is not sufficient to achieve the required film com~osition and i~icrostructure,necessitating the use of different variants of physical and chemical vapor deposition techniques that incorporate non-thermal means such as low-energy ions and lasers. ~ost-d~position processing techniques such as laser annealing, rapid thermal annealing (Roozeboom, 1993) and ion beam mixing (Wolf, 1992) are also becoming widespread for modifying the microstructure and composition. While the inclusion of such additional processes and process variables has expanded the flexibility necessary to control film microstructure, the identification of process windows has become more coniplex and time r and consuming. The use of c ~ m p u ~ esimulati~n modeling for process optinii~ationand control has alleviated this difficulty to some extent (Cale et al., 1991; McJnerney, 1996). However, understanding the atomistic mechanisms of thin-film phase f o r ~ a t i o n and microstructural evolution paths continues to be the predominant challenge limiting the full exploitatation of the various deposition technologies for routine fabrication of thin-film interrnetallics. The rapid development and increased use of high-spatial-resolution rnicroscopy and spectroscopy techniques in the last decade and a half has facilitated substa~tial progress on this front in several intermetallic ~ ~ ~ i l i e s such as silicides (Setton and Van der d'Heurle, 1998), nitrides (Greene et al., 1995; Ronning et at., 1998; ~ u h and l Mendez, borides (Ikushima et aE., 1994; Zhang et al., 1995). In this chapter, we wifl first discuss physical vapor deposition (PVD) and chemical vapor deposition
( C V ~ techniques. ) This will be followed by discussion of electroplating aiid hybrid deposition methods. Finally we will describe post-deposition annealing as a synthesis route for th~n-filmintermetallics. Important features of the difl6erent synthesis methods are highlighted in the context of example applications, where appropriate. The reader is also referred to previous volumes in this series: For synthesis, Martin and Hardwick, Chapter 27 of Volume 1 and Vijh, Chapter 23 of Volume 2; for thin-~lmapplications, ~ a s u m o t o et al,, Chapter 15 of Volume 2, Greenough and Schulze, Chapter 17 of Volume 2, England and Arakawa, Chapter 18 of Volume 2, ~ c G a h a n , Chapter 19 of Volume 2.
Sputtering is the most common method of depositing metal thin films and is particularly advantageous for high-rate deposition of large areas (up to several m2) of uniform, smooth films of high purity. The sputtering process requires a vacuum system wherein a glow ~ i s c h a r plasma ~e is used to generate positively charged as ions. These ions are accelerated by a large negative oltage, of the order of 100s or 1000s of volts, applied to a target with a composition desired for the thin-film deposit. Bornbardmcnt of the target by the accelerated ions and high-en~rgyneutrals results in ~ o m e n t u m tr~nsferto the target atoms which are thus ejected or sputtered into the vapor phase from which they condense on the substrate and walls of the vacuum chamber. A schematic of a typical sputter deposition apparatus is shown in Figure 1. There are many different types of sputter-deposition processes. The basic process described above is referred to as d i o ~ e ~ ~ ~ t t e (Westwood, ri~g 1976) where the target and chamber walls form the two electrodes. The two main variants of diode sputtering g RF s ~ ~ t t e ~ i In n g the . former are DC s ~ ~ t t e r i nand case, a direct-current power supply provides the target voltage, while a high-frequency alternating-current (commonly 13.56MHz) is used in the latter case (Wehner, 1955). Figure 1 shows a schematic sketch of a sputter~depositionsystem with two targets. Often, a system of magnets (fixed or rotating) is placed near the target to increase the residence time of the electrons and thereby enable a larger number of ionizing col~~sions in the tar et zone. This technique - refarred to as ~ ~ ~~ n ~ ~ ~(Thornton, ~ ~ r t 19’98) o e ~-~is useful i ~ for depositing films at low pressures (x10mtorr) and target voltages ( x 500 V). In r~active s ~ ~ t t e r i n g
ure 1 Schematic represe~tation of a typical glowdischarge sputter-deposition process used to fabricate a thin layer of an intermetallic compound in a chamber equipped with two targets
(Westwood, 1989) a reactive gas such as nitrQgen is used (either alone, or mixed with an inert gas) to sputter from elemental targets to deposit filnis of compound materials. For example, transition metal nitrides such as TiN ( ~ u ~ t m aetn al., 1988), (Manaila et al., 1998), CrN (He et al., 2000), and HfN (Sproul, 1985) are most commonly deposited by this technique. sputter-deposition In addition to the glow-~i~schar~e techniques discussed above, there is another class of sputtering techniques that uses focused ion beams created from remote ion guns such as Kaufman sources (Kaufman aiid Robinson, 1987). The focused ions can be deposited directly onto the substrate @rimary ion beam ~ e ~ o s i t ~ o(Kim n ) et al., 1995) or directed to sputter target materials onto the substrate ( s ~ ~ o n ion ~ ~ rbeam y d e ~ o s ~ t i o(Itoh, ~) 1989). Arc ~ v a ~ iso a ~variant a ~ of~the~ ion ~ beam deposition method where the ions are produced from an electricalarc discharge (Sanders, 1994). This technique is characterized by high deposition rates, and offers excellent control over ion energy and trajectory, enabling a high degree of control over microstructure and cornposition (e.g. TiN, TiCN) (Karlsson et al., ~2001). However, ion beam deposition, in general, has found only limited use in industrial applications due to scaling and throughput issues.
666
Processes and P h e n o ~ ~ n ~
putter-deposited films typically have a columnar rostructure in which the grain sizes increase with substrate tem~erature, but decrease with increased pressure (T~oriiton,1978). Contamination can also strongly limit grain sizes by limiting adatom mobilities on tlie surface of the growing film (Leanly and Dirks, 1978). ~nderdensefilm microstructures due to the presence of intercolumnar voids is a comiaon problem of sputtcr-de~)o~itediiitermetallic films that have
ing a negative substrate bias to enable enhanced ion the usual solutions to minimize void rature increases, however, are often limited by thermal damage to the substrate and chemical reaction o f the thin film with the substrate. can also have detrimental impact - such as increasing film stress, altering the grain size and orientati~n, and the implantation o f sputtering-gas atoms and surface contaminants into the film Increasing the ion-toneutral ratio of particles boi~bardingthe substrate at low ion energ~esoffers superior control of microstructure at low tenipterature t inducing substrate n et al., 1988; Petrov
r, there are some ~ n u s u a lcases where the and deposition ratios are not equal, such as when the components have very different angular flux distributions (Greene et al., 1978) or when there are film-incorporation coefficients o f less than one (Bergputter-deposited alloys are commonly used in the microelectronics industry. For example, A1 films used ted circuits are typically doped with in order to reduce el~ctro~igration(d’Heurle and Ho, 1978) and Si dissolution into A1 leading to pyramidal pit formation in Si devices (Pai et al., 1985). A1 alloys are almost universally deposited by sputter de~ositionfrom Cuor Si-doped Al targets creating films of the same composition. A common example of a sputter-deposited ~ntermetal~ic is TixWI-, (x = 0.1-0.2), which is commonly used as a diffusion barrier to further limit spiking of Si by AI and improve adhesion. Controlling the co~positiono f TixWl-x layers is ~ i ~ c ubecause lt the backscattered Ar atoms preferentially resputter the Ti from the film, resulting in a lower Ti co~~~entration than the target. This eEect can be quite large since the Ti incorporation probability can vary by an order of magnitude for different deposition conditions ergstrom et al., 1995).
ticularly well suited for films because multicomdeposited from targets of a binary alloy target with CO the ~ ~ o s esputte~ing n coiiditions the ejection flux of c o ~ p o n e nA t can be written as SA@,where SA is the sputter rate o f A and 0,4is the surfidcc coverage of A on the target. If we assume the sputter yield of A is less , tlie r e ~ ~ a i n i nconcentration g of A will steadily layer within about 10nm o f increasc to form a11a1 the target surface as preferentially sputtered. At steady state, it can hown that (Eltoukhy and reene, 1980)
wbcre CAand CBare the on cent rations of A and B in et. Thus at steady state, the ejection flux of each coi~ponentis equal to its concentration in the target. In most cases the ejection flux ratio is also equal to tlie deposition flux ratio, resulting in films with ~ o ~ ~ o s i tii~o en~ t i to c ~the l target, provided that the target has been srdcquatcly ‘conditioned’ prior to deposition such that a steady state has been achieved.
Evaporation of a material in vacuum is another common thin-film deposition technique where the source material i s heated resistively, by induction, by electron beam, or by laser beam ( ~ e s h p a n ~ eand y Bunshah, 1991). In resistive heating, typically a refractory metal wire or foil is used to hold a fixed amount of the evaporant material, The evaporation chamber is evacuated and a current is passed through the refractory metal until the source material vaporizes, and condenses on the substrate. In the inductive heating technique, the metal source placed in a ceramic crucible surrounde~by RF coils is heated by passing a large current (typically several tens of amperes) through the coils to create a molten metal pool with a high vapor pressure. In the electron beam technique, an electron gun is focused on to the surface of the source material, which i s typically placed in a water-cooled container. The electron beam heats the surface of the source beyond the melting point to create a molten region with a high vapor pressure. Laser ablation is the term used for the technique where high-power laser pulses are focused on the source
Thin-film Deposition and Treatment
material, which create local heating and small-scale evaporation. Like sputter-deposited films, evaporated films typically have a columnar microstructure in which the grain size increases with substrate temperature, but because of the effect of ion bombardment, sputterdeposited films tend to be denser while evaporated films tend to have more intercolumnar porosity (Bunshah, 1982). In general, evaporation is difficult to use for the direct deposition of intermetallics since the vapor pressures o f iiidividual components will typically be very different functions of temperature, resulting in films with varying composition. One way around this problem i s to use sinall pellets of the source material and feed them one at a time onto a resistive heater kept at high temperature. The entire pellet evaporates instaiiteously and thus the vapor flux of each component is equivalent to its concentration in the source. This technique is known as flash e v a ~ o r a t i o ~ unshah, 1982). Laser ablation and electron-beam ablatioii can also be used iii a similar way to evaporate small amounts of the source material to obtain the desired composition (Inam et al., t 990). For instance, alloy deposition with good compositioiial control by electron-beam evaporation of an alloy source has been demonstrated (~immagadda et al., 1972). In this process, illustrated in Figure 2, a molten pool is maintained on the surface of an alloy rod that is fed
667
upward. Assuming the source has two components, A and B, of equal concentration where the vapor pressure of B is 10 times that of A, the steady-state composition of the molten pool becomes 10 parts A to one part B. The resulting vapor fluxes are thus equal, and a stoichiometric deposit is obtained. Another PVD technique for depositing alloy or iiitermetallic films is co-evapor~tio~from multiple sources. Precise calibration and control of the evaporation rates is essential for con troll in^ the film composition. Even when this is ensured, co~positional uniformity across the substrate inay be difficult to achieve. A variant of the co-evaporation method where the deposition rate is kept low enough for accurate control of the depositing fluxes is called ~ o ~ e ~ ~ ~ beam epitaxy (MBE) (see the chapter by Ng and Moustakas). The most common eva~orationsource, called a Knudsen cell (or K-cell) is a heated crucible full of molten material kept near its boiling point (Knudsen, 1909). The distance from each K-cell to the substrate is relatively large (30s of cm) to ensure a uniform deposition. The low deposition rate allows very precise control of the depositing flux, but r ~ q u i r ~ s ultra-high vacuum environments ( c 10e9 torr) to prevent film contamination. As the name implies, the e of control possible with M ique especially well qualified epitnxid films on single crystal substrates. The advent o f MBE has enabled the study of a wide variety of compound semiconductors such as CaAs (Yang et al., 19961, InSb (Okamoto et al., 1999), and CdTe (HuertaRuelas et al., 2000). Multilayers and superlattices also can be grown by using computer-contr~lled shutters (Kwo et al., 1985). However, conimercialization of' MBE is difficult due to low deposition rates.
CVD is a process in which deposition is efYected on a heated substrate via a chemical reaction or decomposition of volatile precursor molecules that contain the constituent(s) of the material to be deposited. In this process, chemical reactions such as pyrolysis, reduction, disproportionation, carburization, and nit can take place singly, sequentially, or in tandem. can be activated thermally, aiid assisted by non-th means such as RF or electron-cyclotron resonance plasmas plasma-e~hanced GVD: PECVD), or laser source using an electron. bean;
enhanced CVD: WECVD) to enhance the film quality
668
Processes and P ~ ~ ~ o ~ e n a
by minimizing the influence of plasma fields on the reaction paths and the damage caused by energetic particles (Park et al., 1996). The use of non-the~nal energy sources (e.g., photo-act~vat~on by laser beams) usually decreases the deposition temperature. CVD usually does not require ultra-high vacuum c h a ~ b e rsand , deposition is possible at a wide range of pressures including atmospheric pressure. Some typical CVD reactors and processes are described schematically in Figure 3. CVD reactors can be of two types, namely, hot-wall and cold-wall. In hot-wall reactors, all parts of the reaction chamber are at high temperature, resulting in uniform temperature zones that are very conductive for depositing u ~ i f o layers r~ on large areas. However, apart from causing high thermal budgets and increasing precursor consumption, hot-wall reactars require etching cycles to periodically remove the deposits from unwanted areas, leading to lower process efficiency. In cold-wall reactors only the substrate is heated, either ind~ctively or resistively. Although these reactors are more complex in teims of the variations in temperature and gas flow at different parts of the reactor, cold-wall reactors allow greater process control at lower costs.
~ ~ s ~ g ~ ~ Unlike PVD, which involves l ~ ~ e - o de~osition due to near-unity sticking coefficients, CVD exhibits low reactive sticking coefficients, typically ranging from 0.001 to 0.1. This salient characteristic of CVD is perhaps its major advantage, because it enables the conformal coating or filling of depth-to-width aspect ratio (e.g. 5: 1) holes and complex non-planar topo~~aphies frequently encountered in microelectronic device structures. Another advantage of CVD over PVD is that near-theoreticaldensity films can be easily deposited. CVD is also suited for selective growth of intermetallic thin filins on certain surfaces. For instance, TiSi, can be selectively grown on Si substrates from TiC1, and SiH4 ( ~ e n d i cino et al., 1993). In the case of h i g h " ~ e m p e ~ ~ ~ ~ ~ r intermetallics, CVD processes usually have a high ~epositio~ rate, ~ a k i nthis ~ route econo~ically attractive. For example, the rate of SIC deposition by CVD is several orders of rnagnitudes higher than that by sputter deposition (~~ and Lin, 1993). CVD also offers a high degree of control and flexibility over the process parameters, enabling the deposition of a wide range of film thic~nesses from 1Onm to centimeters. The primary isa advantage o f CV
Figure 3 Schematic sketches of CVD chambers in (a) hot-wall and (b) cold-wall configurations; and FECVD ChdmherS with (c) conventional and (d) remote plasma sources
~ h i n - Deposition ~ l ~ and T r e ~ ~ t ~ e n t high deposition temperature, which can result in the formation of undesirable phases due to interdiffusion and interfacial reactions, either between adjacent layers of the substrate or between the precursor and the substrate (Ramanath et aE., 1999). High temperatures combined with large mismatches in thermal expansion coe~cientsof the substrate and the film, can result in high stresses and film delamination. Also, many reactions leave solid byproducts that not only contaminate the film, but also are often toxic and corrosive, requiring expensive disposal procedures (Pierson, 1992). Such problems and the high thermal bud~etscan make CVD unfeasible, in the absence of significant offsetting a ~ v ~ n t a g e s . CVD has been extensively used for depositing thinfilm intermetallics for applications ranging from microand opto-electronics to wear- arid corrosio~~resistarit coatings for cutting tools (Pierson, 1992). Examples include III-V compounds (Jones and O'Brien, 1997), silicides (Madar and Bernard, 1989), metal- and semiconductor-nitrides (Ghatterjee et al., 1992), car-
669
bides and borides (Stinton et al., 1988), and related ternary compounds such as TiC,vNI- (Pierson, 1994), to name a few. Table 1 shows examples of the overall chemical reaction(s) involved in the CVD of several thin-film intemetallics. For CVD of intermetallics, the precursors should contain the constituent elements of the compound and be s u ~ c i e ~ treactive. ly ~igh"vapor~ pressure inorganic or metal-organic precursors in the gaseous form are preferred because they enable precise d or liquid sources (Matsuno control of flow rates, et al., 1995) can also sed by evaporating them into the gas phase before introducing them into the chamber. Condensation in the delivery systems (Weber and Klages, 1995) due to gas-phase nucleation of particulates (e.g. TiCI,(NH,), (x = 2-8) during TiN deposition from TiCl, 3.NH, ( U et al., 1999)) resulting ~ ~ nthat in high precursor reactivity is a ~ o m problem has a strong influence on film quality and process viability (Mochizuki et al., 1995). As seen in Table 1, in many cases intermetallic compounds can be deposited from different precursor
able 1 Examples of thm-film intermetallics deposited by CVD. The overall reaction path and deposition temperature are also shown
Co~~ound TiSi, MoSi, TaSi, WSi, Sic B4C Tic TiB, Si,N4 BM GaN
AlN TIN
InP (In,Ga)P GaAs ZnSe
Main reaction
TiCI, + 3SiH,+TiSi, + SiCIH, + 3HC1+ 3H, MoF, + 2SiH4+MoSi, + 6HF + H, TaCl, + 2SiH,-+TaSi, + 5HCI + 1.5H, TaCI, + 2SiH,C1, + 2.5H2-+TaSi,+ 9HCl WF, + 2SiH,+WSi, + 6HF + H, SiCl, (g) + CH, @-+Sic (s) + 4HC1 (g) 4BC1, + CH, 3- 4H,+B4C + 12HC1 2B,H, + CH,-+B,C + 8H, TiC1, (g) + CH, (g)-+TiC(s) + 4HC1 (g) TiCl, + 2BC4 + 5H,-+TiB, + lOHCl TiCl, + B,H,-+TiB, + 4HC1+ H, 3SiCl,H, (g) + 4NH, (g)+Si,N, (s) + 6H2 (g) + 6NCl (g, BF3 (g) + NH3 (g)-+BN (s) + 2-w(8) Ga(CH3),+ HCI + NH,-+GaN + C1+ €3, + CM, (incorrectly balanced) AlC1, (g) + 2NH, (g)-+AlN (s) + 2HCl (g) + NH,CI (g) TiCl, (g) +1/2 N, (g) + 2H,-+TiN + 4HC1 6 TiCl, + 8NN3-+6TiN+ N, + 24HC1 T i ~ ~ ( C+N,--+TiN ~ , ) ~ -t-~HxN(C€i3)3.-x +Ti(HCN) (KN = CH,) + HCN + NH, + H Ti[N(CH,),],# + 2NH,+TiN + 4HN(CH3),+ H, + -N2 T i ~ ~ ( G , H4-~NH,-+TiN ) , ~ ~ ~ + 3 ~ H ( C , H ~+) ,*N(C,H~), (CH~)~InP(C,H~)3 + PH,-+InP + 3CH, + P(C,H& 0.5 1 R,Ga + 0.49W'Jn + EH,-+In, ,,Ga, ,,P + nC,,H,, R, R & E alkyl or hydride GaCl + 114 As, + 1/2 H,+GaAs, 4- HCl (CH,),Zn + W,Se-+ZnSe + 2CH,
ffCommonlyknown as TDMAT *Commonly known as TDEAT
Comments
Reference ___
ifdep= 350-800 "C Tdep= 520-600 Tdep= 500-650 "C Tdep= 540-580 "C rdep = 230 TdeD = 1400 "G Tdep= 1200-1400 " c TdCp = 400 "C Tdep = 1000"C Tdep= 800-1 100"C Tdep = 600-1000 "C
"c
"c
ifdep= 1100 "C
Tcfcp = 550-800 "C Tdep= 750 "c T,, = 400-700 "C Plasma CVD
Madar and Bernard, 1989 Gaczi, 1986 Wieczorek, 1985 Wieczorek, 1985 Madar and Bernard, 1989 Ohring, 1992 Mullendore, 1985 Pierson, 1994 Ohring, 1992 Pierson, 1994 Pierson, 1994 Ohring, 1992 Ohring, 1992 Kryliouk et al., 1999 Roman and Adriaansen, 1989 Chatterjee et al., 1992 Hu et al., 1999 Weber and Klages, 1995 Intermann et al., 1993 Cale et al., 1993 Jones and O'Brien, 1997 Razeghi et al., 1986
Tdep= 250-350 "C
Ohring, 1992 Jones and O'Brien, 1997
670
Processes and Phenomena
combinations. Since film quality, purity and microstructure (and hence properties) are strongly influenced by factors such as deposition temperature, surfacereaction mechanisms and non-thermal activation sources, the choice of the precursor chemistry usually depends on the application. For instance, Si,N4 can be deposited either t h e ~ a l l yfrom SiC12H2and NH, at >750°C or from SiH, aiid NH, in a hydrogen plasma at 300°C While the latter process is desirable for deposition on low-temp~rature substrates, the high amounts of hydrogen incorporated in the film may be unacceptable for certain applications. Process viability and ease are also often important factors that determine the choice of precursors. as illustrated by the following example. GaAs can be deposited either from a liquid Ga source and ASH,, or by reacting a metal-organic precursor such as (CH,),Ga (trimethyl gallium (TMG)) with ASH,. Both these methods yield highquality films and high deposition rates. However, metal-organic CVD (MOCVD) from TMG is preferred because the easy delivery and pyrolysis of TMG obviates cumbersome delivery systems required for liquid Ga precursors (Field and Gandhi, 1984; Cruter et al., 1989). Optimal process conditions for a given precursor co~binationis usually achieved with the aid of computer odel ling (Kleijn, 1991) since the relationship between film properties and key variables such as tem~erature,pressure, reactant gas Compositions and flow rate can be complex (Pierson, 1992). Most microstructures obtainable by sputter deposition can be obtained by CVD as well, the exception being amorphous films at high temperatures, where crystallization is favored. Even though a columnar microstructure is typical in crystalline CVD films, equiaxed grains can be promoted by crystallite growth in amorphous films during deposition, or by high impurity concentrations favorable to nucleation. The two most important parameters that influence the niicrostructure of films deposited by thermal-CVD are reactant supersaturation on the substrate surface, and deposition tex~pe~ature. These, along with other secondary factors such as plasmas, laser activation etc., determine the nucleation rate and growth modes. Specifically, surface reactant concentration strongly influences the nucleation rate while substrate temperature affects the growth rate. Epitaxial growth on single-crystal substrates is favored at high temperatures (high adatom ~ o b i l i t y )and low reactant supersaturation (low nucleation probability), while amorphous films are Famed when these conditions are reversed.
Electroplating (a.k.a. electrochemical deposition) is an electrolyte process wherein metallic ions are reduced to neutral atoms in aqueous solutions or molten salts, aided by the application of an external electrical bias. The advantages of electroplating include low deposition temperatures, low e~uipmentcosts due to lack of vacuum, process simplicity, and the easy availability of reagents. Another advantage of electroplating is that - like CVD - it is not a line-of-sight method, and hence is attractive for filling high-aspect ratio topological features (Yung and Turlik, 1991). The primary disadvantage of this technique is that deposition can be eEected only on conducting substrates. This continues to be a niajor liniiting factor in exploring electrodeposition of thin-film inte~etallics for semiconductor applications. However, recent reports of electrodepositing Au and Pd/Ni multilayers on Sic with the assistance of excimer laser pulsing (Zahavi and Pehrsson, 1986) show that combining conventional plating with other energy sources that can influence chemical reactions, either locally or remotely, holds promise for obviating this difficulty. Electroplating of elemental materials and many metallic alloys has been extensively studied, well understood, and used in industrial applications for a long time (Srirnathi et al,, 1982; Choi et al., 1998; Ng et al., 1998; Liebscher, 1999). For i n t e ~ e t a l l i cmaterials, however, electrode~ositiondoes not belong among the mainstream methods for depositing thin films and is still a subject of active research (see Vijh in Chapter 23 of Volume 2). The main difficulty is the challenge of co-depositing elemental components with very large differences in electrochernical potentials. For instance, even though Cu forms intermetallic compounds with A1 (Massalski et al., 1986), the respective electrochemical potentials of Cu and AI are +0.34 and - 1.66V respectively, m a ~ i n git diffic~ltto deposit both the elements simultaneously without hydrogen evolution, which destroys adhesion and decreases the plating efficiency (Krishnamoorthy et al., 1999). Molten baths may be used to overcome some of the problems associated with aqueous chemistries; but this is not the preferred method since the deposition temperatures are high. Despite such limitations, a variety of intermetallic conipounds have been synthesized directly by electroplating. These include refractory metal carbides and borides such as TiB,, CrC (Domrachev et al., 1995), I-111-VI,, IT-VI, and IIIV compounds such as (In,Ga) (As,P) (Mulhoff and Muller, 1988), CuInSe,, CdS (Fatas et al., 1984), InP,
thin-^^^ Deposition and onigstein and Spallart, 1998), and conventional intermetdlics such as Ta,Ni, TaNi,, NbNi, NbNi, (Taxil and Mahenc, 1987), FeNi (Cheung et al., 1995), CoNiFe (Shinoura et al., 1994), and (Co,Ni,Fe)S (Takai et al., 1997). Typical deposition conditions for representative compounds are listed in Table 2. Even in cases where intermetallic thin films have been electrodeposited, the control over the phase selection and composition is often not sufficient because the r~lationsh~ps between bath chemistry and phase-selection mechanisms are not well understood. For example in the Ni-Sn system, Ni,Sn4 - despite being thermodynamical~ythe least stable and structurally the most complicated - forms first in preference over Ni,Sn, and Wi,Sn (Allen et al., 1985). As a result, the most common route in which electroplating is involved in the fabrication of thin-fil~inter~etallicsis reactive diffusion (discussed in detail in section 7) of electroplated component layers with either the substrate, or with each other, during annealing. For instance, thin layers of Cu3Sn and Cu,SnS can be formed by the reaction of electroplated Sn films with Cu substrates during thermal annealing for improving oxidation resistance and adhesion of subsequently applied Sn-Pb solders in packaging applications (Reynolds and Morris, 1995). Other examples of the interrnetallics fabricated by this route include Ti-Cu (Salelii a i d Hossei, 19961, AI-Mn (Li et al., 1998), NiCO (Pauiiovic et al., 1994), and Ni-Zn (Shibuya et al, 1985). A variant of electrodeposition is the electroless deposition method, where no external potential is applied to drive the reaction. Instead, film deposition proceeds by electrockemical oxidation of liquid-phase
67 1
Treatment
reducing agents that provide the electrons and the potential for reaction. This method has been used extensively in the past to deposit elemental layers, but it is only recently that there have been efforts to deposit binary alloys such as Ni-€3 (Zhang et al., 1993; Lee and Lm, 1997), Ni-P (Inaba et al., 1990) and ternary compounds such as CoWP (Lopatin et al., 1998), NiNbP and NiZrP (Osaka et al., 1989).
ositio The inherent limitations of the deposition techniques discussed above sometimes make it i~practicalto use them individually for depositing thin-film intermetallies. For example, it can be difficult to deposit intermetallic films by evaporation if the c o ~ p o n e n t elements have very different vapor pressures; by sputter deposition if the components have very digerent sputtering yields, or if one of the elements is a liquid at ambient temperatures. CVD has its limits as well, when suitable precursors of one of the components are not available or viable. Such limitations have led to the innovation of new processes con~istingof combinations of two or more variants of coiiventioiial deposition techniques in order to exploit the advantages o f the individual techniques and circumvent their disadvantages. This class of deposition inethods is called ~ y ~ ~ re pio ~~i t i o(HD). n Examples of the deposition of A,B,C, films from the sputtered fluxes of A and B, a i d the evaporated flux of C;or sputtered fluxes of A and B and a surface chemical reactions of a precursor c~ntainingC; the different steps occur either simultaneously (single-stage
Table 2 Examples of electrodeposited thin-film intermetallics shown along with the bath chemistry and plating conditions Intermetallic
Deposition conditions
p Cu,Al
60-40 rnol.% AlCl~-MeEtirnCl(l-~ethyl-3-ethylim1dazoliumchloride) 40 "C at 0.22 V Pretrniiiiig of copper with coatings of 75Sn-25Pb by electroplating followed by aging at 170°C for 2 hours in Ar Ni3Sn4,Ni,Sn, Sn electroplating on Ni coupons from a stannous fluoroborate bath (pH 0.2) at Ni,Sn room temperature followed by anneal in^ at 100-190 "C Ta,Ni, TaNi, Ni cathode in K,TaF,-LiF-NaF or K,NbF7-LiF-NaF baths at 850-1050 "C (X =: 1-3) NbNi & NbNi, TiCu, Ti2Cu, Cu electro~lat~ng oii Ti-6A1-4V substrates followed by thermal annealing Ti3Al,TiCu, TiB, NaF-NaBF, eutectic mixture + TiF, ---t fuse in carbon crucible Electroplating at 1.1 V (NaC/Na) ZnTe 0.15M ZnS04+0.005M K,SQ4-Te0,, pH=2.5-5.6, at -1 to -0.6V Cu,Sn, CuSn
CdS
0.002 M CdSO, + 0.1 M Na,S,O,, pH = 2.3. Potential = -0.75 V(SCE)
Reference Tierney et al., 1998 Reynolds and Morris, 1995 Allen et al., 1985 Taxil and Mahenc, 1987 Salehi and Hossei. 1996 Uamarnoto
et
al., 1996
Konigstein and Spallart, 1998 Fatas et al., 1984
Processes avld P h e v l o ~ e ~ a
672
I
I I e
c
igure 4 Schematic sketch of a hybrid deposition chamber with three sources, n a ~ e l y ,two sputter targets and one effusion cell, coilfigured for depositing a ternary intermetallic compound such as CuInSe,
processing) or sequentially (multiple-stage processing). Figure 4 shows an example of a hybrid deposition chamber with two sputter targets and an effusion cell for depositing thin layers of ternary intermetallics. has been used to deposit many intermetallics such as irkarimi et al., 1997), ternary chalcopyrite (I semiconductors like CuInSe, (Rockett et; al., 1989), inter~iietallic-co~poundoxides like 5), and multilayers of metal/ et al., 1397). Deposition of highInSe, films from sputtered fluxes of Cu. and In, and Se vapor is an excellent example of a D process that is commercially used to coat 3 10m2 glass substrates at inillion square meters per year for solar cell appl~cations(Rockett et al., 1990; Ashida et al., 1993; B a d , 1993). We use this as a model system to highlight the salient features of HD. Tables 3A ighlight the salient features of conventional synthesis routes, respectively, for Pdbricati e2 solar cells. The solar-cell efficiencies (Basol, 1993) are also shown. Sputter deposition is perhaps the most suited amon conventional deposition methods to deposit I CuInSe, (Rockett et al., 1989) on large-area substrates. owever, inherent limitations of this synthesis route espread use. For instance, the low causes unstable sputtering at high deposition rates, limiting process throughput. More-
over, three-target sputting results in the incorporation of a large number of ion-bonibar~~eiit-induced planar and point defects in the film, deleterious to device efficiency (Nakada et al., 1995). ~ l t h o u g hco-evaporated films yield high efficiency devices, scaling this process is not viable. The W D method provides solutions to such problems. The higher device efficiencies of WD-deposited CuInSe, films (see Table 3) testify to the effectiveness of this deposition strategy. For example, ~ i ~ ~ l ~ ~ v l sputter deposition from Cu and In targets, and Se evaporation from an effusion cell, allows precise stoichiometric control in films over a wide range of fn/Cu ratios between 0.8 and 1.4 due to a linear relationship between the composition and the target current ratio. The Se flux can be controlled independently by adjusting the egusion cell temperature (Rockett et al., 1989). HD lowers the deposition temperature, facilitating the CuInSe, film deposition on inexpensi~eglass substrates. Moreover, the absence of high-energy Se ions in this method results in better film crystallinity (Wakada et a/.>1495). This HD route is also amenable to growing epitaxial CuInSe, layers on substrates such as CaAs for infrared detector applications (Mullan et al., 1993). HD methods also come with their share of disadvantages that are usually related to the integration of the deposition methods. Typically, the shortcomings are eli~inatedby minor modifications in process variables, or choosing alternative component precursors, and/or modifying the sequence (e.g,, parallel vs. serial) of the different deposition processes, which is closely tied with deposition tool design, We note that the integration of differ~nt deposition methods can be quite coniplicated. However, the advent of sophisti~~ted cluster tool configuratio~san important develop~entin thin-film technology in recent years - should alleviate this concern. In the case of CuInSe,, the problems of the W D route were related to the source and reactivity of the Se flux (see Table 33). For instance, the Se vapor flux can contaminate the metal tar ets, and cause the formation of surface pits (Rockett et al., 1990). However, sputtering ~ # l ~ by o ~selenization e ~ clhinates the former problem (Nakada and Kunioka, 1998), while the latter can be sur~ount edby the use of ionized clusters (Sano et al., 1998). The successful implei~entation of HD in depositing CuInSe, has contributed to the extension of this technique to deposition of epitaxial C u b , __ .Ga,Se2 from magnetron-s~uttere~ Auxes of Cu, Ga and In and evaporated molecular Se (Rockett et al., 1994).
T
h
i ~ ~e ~~ o ~and ~~~i ~~ ri o e~
~
~
~
e
~
~
673
Table 3 Salient features and solar-cell efficiencresof CuInSe2 films deposited by (A) conventional and (B) hybrid processing routes A. Conventiona~processing
Processing approach
Solar-cell efficiency (%1 5
Spraying
5
Advantages
Disadvantages
Low cost Non-vacuum Low cost Non-vacuum
Low efficiency Poor control of stoichiometry Low efficiency Secondary phase formation Instabilit~esin starting solutions back of stoicbiometriccontrol Variations in target composition
~ ~ a p o r a t i o n / s from ~ng 6 CuInSe, targets Laser ablation of CuInSe, 8.5 targets 15-17 Co-evaporation of Cu, In and Se 6 ~o-sputteringof Cu, In and Se (Nakada and Kunioka, 1998)
Large volumes, simple process, high purity High efficiency
Screen printing of CuInSe, pastes
Low cost Non-vacuum
High efficiency Good quality Scalable
Scaling IS difficu~t Low efficiency Se-contaminat~onof targets High deposition rates due to low melting point of Se Film discontinuities cu-rich films
B. Hybrid processing Selenization of Cu-In layers using H,Se
12.4-12.6
High efficiency, high volumes
11.5 Selenization o f Cu-In-Se in Se vapors Sequential or simultaneous evaporation of IYr,Se, and Cu,Se (Ashida et al., 1993) Reaction of Cu/En/Se layers 10-10.5 (Gupta et al., 1993)
Good efficiency
Reaction of Cu with In,Se3-4-Se 10.8-13.5 or reaction of CulnSe,+ Cu,Se with In and Se (Tuttle et al., 1993)
Good quality Large grains Good-high efficiency
Large volumes Simple process Low cost Good efficiency Large grains
Many technologically important ~ntermetal~ic layers are created by solid~stated i ~ u s ~ oand n interfacial reactions - collectively referred to as reactive diffusion - between previously deposited elemental or alloy layers. Although reactive di~usioncan occur even during film deposition at high te~peratures, this synthesis method is more commonly employed as a post-deposition anne~lingstep in conventional furnaces or rapid thermal heating systems (~eygenson and Zemel, 1988). For instance, C54 TiSi, layers are formed by rapid thermal annealing of TijSi bilayers for
Toxicity of H,Se In depletion and H incorporation at tei~peratu~es > 400 'C Surface pitting, target conta~ination Compositional control i s a problem; Cu,Se splashes due to In,Se, melting Scaling is diffic~~t Poor adhesion and inferior film quality Heat treatment required to annihilate voids at film/substrate interface Involves multiple steps Scaling and compositional control may be difficult
making electrical contacts to Si-based transistors, as described previously. In many cases, however, intermetallic phase formation by reactive diffusion is ~ndesirab~e, For example, A1,Ti orm mat ion due to reactive diffusion between adjacent layers of Ti and A1 in interconnect structures of microelectronic devices results in large transfor~ationstresses, brittl~n~ss, and film delamination (Colgan, 1990). One major advantage of reactive diffusion is that it is perhaps the only way of synthesi~~ng inte~etallic phases at buried interfaces. This method is relatively inexpensive and can be integrated in series with almost any deposition technique without mu& difficulty, The
674
Processes and P ~ ~ e n o ~ e n a
main disadvantage of this method is the lack of sufficient control over phase selection and stoichiometry, which are related to a variety of factors such as microstructur~and relative thicknesses of the component films, lateral confinement, and annealing rate, to name a few (Rai~anatliet al., 1993; Svilan et al., 1997). Studying compound formation during reactive diffusion is a challenge because in situ surface analysis tools cannot be used to probe reactions at buried interfaces. In electrically conductive materials systems, however, measuring the changes in the sheet resistance of thin film stacks during annealing is routinely used to monitor interdiffusion and reactions (Allen et al., 1994; amanath et al., 1996). ~ ~ a ~ t ~ determination t ~ t i v e of netic parameters such as activation energy is possible by ~ombiningsuch in situ measurements with ex situ determination of the volume fraction transformed (Bergstrom et al., 1995). In this regard, cross-sectional transmission electron microscopy and compositional have played very important roles in revealing several key features of reactive diffusion processes. Reactive diffusion usually involves one dominant diffusing species (DDS), and occurs in three stages, namely, interdiffusion, phase nucleation, and growth. Based on the relative values of the grain boundary diffusion coefficient DGB and the bulk diffusion coefficient Dbulkof the DDS, thin-film diffusion can be classified into three types (Gupta and €30, 1980). Type A ~ ~ ~ uoccurs s ~ owhen n DCBis only slightly larger than Dbulk-the DDS diffuses both along grain boundaries and within grains, resulting in a nearly planar di~usionfront. When DGB> Dbulk(Type B ~ i ~ u ~substantial ~ i o ~ )flux of the DDS is injected into the grains from the grain boundaries in addition to the direct flux from the film interface with the diffusion source. In Type C ~ i ~ ~ DGB>>Dbulk s i o ~ and the diffusant is transported almost exclusively through grain boundaries. After sufficient amounts of the DDS have dissolved into the adjacent matrix layer, intermetallic phase nucleation can occur. Since the DDS concentration is highest at intersection points of grain bou~dariesand interfaces, especially for Type B or C cases, nucleation is favored at such high-energy locations. As a consequence, local microstructure can have a large effect on phase selection and nuclei density ( ~ e r g s t r o et i ~ nl., 1995). After nucleation, the reaction rate is limited either by the diffusion rate of the DDS through the new phase or the reaction rate at the parent-matrix and iiite~etallic-nucleiinterfaces. For T’ye A diflzisiun, the phase growth front is approximately planar so that
the volume of the intermetallic phase is given by (Turnbull, 1956):
where V(t)is the volume of the product phase at time t, Ea and C are the activation energy and pre-exponential constant describing the kinetics of the rate limiting step, k is Boltzmana’s constant, and T is the annealing temperature. The exponent i z = $ if the reaction is limited by diffusion, and n = 1 if the reaction at the interphase interface is the rate-limiting step. If grain boundary diffusion ~redoi~inates, the newly formed intermetallic grains grow in two or three dimensions. In these cases equation (2) can still apply with, YZ = 2 or 3 for interfacial-reaction-rate limited processes and n = 1 or 3/2 for 2-D or 3-D diffusion-limit~dgrain growth, respectively (Ham, 1958). Often interfacial ~eaction-rate-limited processes become limited by diffusion as the diffusion distance through the product phase increases. This general eqwation covering both regimes is called the linear-parabolic growth law, which for one-dimensional growth i s given by the Deal and Grove model (Deal and Grove, 1965): dx(t)
3
x
T= [XiD]
(3)
where x(t) is the product layer thick~essat time t; and K and D are the thermally-acti~dted interfacial reaction rate and diffusivity, respectively. ~nderstandingthe sequence and kinetics of interfacial reactions during reactive diffusion is extremely important for industrial applications for synthesi~ing new functional layers at buried interfaces, and preventing the formation of unwanted compounds that are deleterious to the functionality of the thin-film layers. An interesting case study of reactive diffusion involving two materials systems used for microelectronics device applications is found in AllSi contacts. Figure 5 shows a schematic sketch of a typical multilayer structure o f an AI contact in Sibased devices. A transition metal silicide layer (e.g., TiSi,) over the Si gate provides low contact resistance (Chittipeddi et al., 1992), while a diffusion barrier layer typically a refra~tory-metal or a transitian-metal nitride) is deposited on the silicide to prevent interdiffusion between A1 and Si. The challenge of silicide synthesis is to induce a low resistance inte~metallic phase at as low a temperature as possible, while the goal of the diffusion barrier is to mininiize aluminide formation at as high a temperature as possible.
675
~ h i ~Deposition - ~ l ~and ~ r e a t ~ e n t
A typical Al-contact structure invo~vingthin-film intermetallics in integrated circuits. The silicide layer provides electrical contact, and an elen~entalor alloy diffusion barrier layer prevents A1 d i ~ u s i into o ~ Si and altiminide formation
Transition-metal silicides are the most widely studied family of intermetallic thin film materials typically formed by reactive diffusion. Table 4 lists the transition metals and the corresponding silicide phases that f o m during anneali~gof metal/Si bilayers. The DDS can either be the metal or Si, depending upon the system, and in many cases multiple phases form during annealing. The silicide phase(s) usually nucleate at the m e t a ~ ~ Sinterface i and grow into columnar polycrystalline grains. Many epitaxial silicides, e.g., those of CO, Ni, Pd, and Pt have also been extensively studied and are being considered for future applications (Chen and Tu, 1991). In several cases such as TiSi,, which is the most commonly used silicide, an amorphous phase forms prior to nucleation of crystal-
lites. In addition the formation of ni~tastablephases (e.g., C49 TiSi, formation described earlier) - a common occurrence in thin-film reactions with nanometer-scale line widths and tliic~nesses- can also strongly influence phase selection and ~ i ~ r o s t r u c t ~ 6 r e evolution during annealing, For instance, transfori~in C49 TiSi, to the CS4 phase becomes more ~ i ~ ci ~ l t thinner and narrower lines due to a lower number of favorable nucleation sites (Svilan et al., 1997). Table 5 lists different aluminides formed during annealing along with their r e s ~ ~ c t i v eformation temperatures. The study of thin-film ~etal~alLimin~des is closely linked to di~usion-barrierfailure because alurninide formation correlates with the poorer barrier performance. (See de Reus in Chapter 29 ofVolu1ne 2.) Tungsten is among the best elemental barrier materials and has been widely used. tudies of interfacial
followed by the formation of a blanket that isolates the A1 and the underlyi and WAl, (Bergstrorn et al., 1995). 10--20°/0 Ti not only improves t
properties (Bergstrom et al., 1997). When multiple intermetallic phases are possible - as is the case in many interi~etallics ~ s t e i ~such s as silicides, aluminides, and nitrides - it is often difficult to predict which phases will form first, based on interfacial energetics, lattice m a t c ~ ~ ~ grams, or other t ~ e r ~ o d y n a ~considerations ic
Transition metal silicides and their approxima~eformation temperatures in "C. The order in which the silicides are listed for each element, in general, reflects the sequence of phase formation. Adapted with permissioii from Chcn and Tu (1991) and Nicolet and Lau (1 983)
Ti
V
TiSi (500) TiSi, (600)
VSi, (600)
zr ZrSi, (700)
fv
HBi (600) HfSi, (750)
Cr CrSi, (450) Cr,Si, (550) CrSi (600) Cr,Si (600)
ME MnSi (450) MnSi, (800)
Fe FeSi (500) FeSi, (550)
Nb NbSi, (650)
M(? MoSi, (525)
Tc
Ru Ru,Si (900) RuSi (900) Ru,Si (900)
RI2 RhSi (375) Rh,Si (400) Rh,Si, (825) Rh,Si, (9259
Ta TaSi, (650)
w
Re ReSi, (1100)
OS
WSi, (650)
Ir IrSi (450) IrSiI,7(1000) IrSi, (1000)
Os,Si, (1000)
CO
Co,Si (400) CoSi (400) CoSi, (550)
Ni i,Si (250) Nisi (400) Ni,Si, (400) Ni,Si (450) Nisi, (750) Pd Pd,Si (200) Pd,Si (350) Pd,Si (400) Pd,Si (650)
Pt Pt,Si (300) PtSi (300)
676
Processes and P ~ e n o ~ e n u
Transition metal aluminides along with their approxi~ateformation ~ e m p e r ~ t uinr ~"C. s Adapted with per~issionfrom Colgan's review article (Colgan, 1990) Ti TiAls (350)
Y VA1, (425) VAL,, (450)
zr ZrA1, (350)
Nh NbAl, (300) Nb,AI (750)
fff
Ta
HfAl, (350)
TaAl, (500) Ta,Al (550)
cr
MPZ
Cr,Al,, (375) CrA1, (385) Cr,Al,, (425) MO
Tc
RU
Rh
Re
OS
Ir
MoAl~,(475)
w WAl, (475) WAI,, (500)
strom et al., 1995). This difficulty arises because energy barriers for nucleation are low enough that phase selection is dominated by kinetics. For instance, if the diffbsivity of the DDS in a thermodynamically favored phase is low, the formation of a ighboring phase will be promoted (Philibert, 1991). hile there is a huge volume of literature involving both modeling and experimental investigations undertaken with the objective of predicting the first phase formed in several intermetal~ic thin film systems, a simple phei~o~neiiologicalmodel that Can explain the niation in a large class of systems et al., 1991; Miura et al., 1991). This gap is a testimony to the vast differences in phase formation mechanisms in thin films vs. bulk.
Interest in thin-film intermetallics has been steadily increasing over the last couple of decades due to the gro~iiig importance of thin functional layers of compound materials in several applications, the most nobble being micro- and opto-electronics device structures. We have discussed the major deposition and annealing techniques most commonly used for synthesizing layered intermetallic phases on surfaces and interfaces in the context of typical materials systems drawn from microelectronic applications. The flexibility of conventional deposition technologies and the emergence of new variants of these methods have enabled the synthesis of several iatermetallic compounds in the thin film form. The synthesis of metastable phases and new microstructures (those not observed in the bulk) have enhanced the scope of intermetallic compounds in general. While control over phase selection and niicrosti-ucture has been a ~ h i e v in e ~a few materials systems, the lack of a
Pd PdAl(200) PdAI, (200) Pd,Al, (250) PdAI, (250) Pt (225) PtAl, (250)
fundamental understanding of the atomistic inechanisms of phase formation and micros~ructureevolution paths continues to be the pr~dominantchallenge that limits the full exploitation of advanced deposition technologies for routine fabrication of thin-film interrnetallics. Revealing the atomic-level relationships between processing parameters and film characteristics will not only enable better control over film quality and properties in more materials systems, but will also shed light on new methods of synthesizing etas stable intermetallic phases in the bulk form. The ~icroelectroiiicsindustry will continue to be a major driving force for improved u~~erst andi ng and development of thin-film intermetallics due to the following factors. As traditional material systems reach their inherent p e r f o ~ a n c elimits, novel alloys and intermetallic compounds with diverse functionalities will be needed, and aggressively pursued. These layers can have a dramatic impact on the price and performance of devices. Future circuits will require: (1) a large number of vertically stacked metallization layers - increasing the number of interfaces where reactive diffusion must be controlled; and (2) decreased device dimensions which will require films to be deposited or formed in confined geometries. These trends will require advances in the ability to control atomic-scale features. Emerging deposition technologies such as atomic-layer CVD and analysis techniques such as scanning tunneling microscopy are beginning to address these needs, but much development will be needed to make them suitable for widespread use.
The authors would like to thank Ahila Krishnamoorthy for her input on the electroplating section, ~ a u s h i k Chanda for his help in searching the
T h i n - ~ l m~ e ~ ~ s iand t i Treatment ~ n
677
Devices and ~irc~4its - II, Materials Research Society Symposium, 260, 207. Clio~,C.-H., Lee, H.-J., Min, S.-H., Kirn, I<.-B., and Lee, D. I’J. (1998). J . Kor. fnst. Colgan, E. G. (1990). Mat Davis, R. F. (1991). Proc. Deal, B. E, and Grove, A 3770. Deshpandey, C. V. aiid Bunshah, R. 2;. (1991). In Thin Film Processes II, J. L. Vossen and W. Kern, Academic Press, New York, 79. d’Heurle, F. M. (1998). J. Electron. Res., 2”7(13), 1138. rences d’Heurle, F. M. aiid Hs, P. S. (1978). In Thin Fibs, J. M. Poate, R.N. Tu, and J. W. Mayer, eds. Wiley, Chichester, Adibi, F., Petrov, I., Greene, J. E., Hultman, L. and 243. Sundgren, J.-E. (1993). J. Appl. Phys., 73(12), 8580. Domrachev, G. A., Varuhin, V. A., and Zakurazhnov, A, A., Allen, C. W., Fulcher, M. R., Rai, A. S., Sargent, G. S., and (1995). 3rd Internutional S y m p o s ~ uon ~ Struct~raland Miller, A. E. (1985). Materials Research Society S-vmpo~sia Functional Gradient ~ a t e r i a l s , B. llschner and N. Proceedings: Electronic Packaging Muter~a~,s Science, Cherradi, eds. Presses Polytech. Univ. Roinandes, 139. Lausanne, Switzerland, 549. Allen, L. H., Ramanath, G., Lai, S. L., and Allman, D. J. Donnelly, N., McConnell, M., Dowling, (1994). Appl. Phys. Lett., O’Mahony, J. D. (2000). Mat. Sci. for^^ Ashidn, A., Hachiuma, Y., Yamamoto, N., Ito, T., and Cho, Durr, H. A., Dudzik, E., Dhesi, S. S., Goedko Y. (1993). Proceedings of the 9th International Conjererzce G. V., Belakhovsky, M., Mocuta, C., Mary, A., and on Ternary and Multinary Compo~nds,Japanese Journal Samson, Y. (1999). Science, 2 of Applied Physics, 32, Suppl. 32-3, 84. Eltoukhy, A. H. and Greene, J. E, (1980). J . A@. Phys., Baeri, P., Grimaldi, M. G., Pnolo, F., Rirnini, E., Cullis, 51(8), 4444. G., and Chew, N. G. (1988). J. Less-Co~~zon Met., 1 Fatas, E., Duo, R., Herrasti, P., Arjona, F., and Camare, E. 181. G. (1984). J . ~lectrochem.Soc., 131(10), 2243. Basal, B. M. (1993). ~ r o c ~ e d i nofg ~the ~ 9th International Feygenson, A. and Zemel, J. N. (1988). Thin Sol. Films, Confirence on Ternary arzd &Id 165(1), 109. Japanese Journal o f Applied Physics, Field, R. J. and Gandhi, S. K. (1984). J. Cryst. Gro~vth, Bergstrom, D. B., Petrov, 581. (1995). J . Appl. Phys., Gaczi, P. J. (1986). Proceedings of the S y ~ ~ p o s i c~mm Bergstrom, D. B,, Petrov, M~ltilevel ~etallization, I~terconnec~ion, and Contact Phys., 82, 2312. Technologies, L. B. Rothman and T. Herndan, eds. Bergstrom, D. B., Tian, F., Petrov, I., and Moser, J. (1995). Electrochem. SOC;Penniiigton, NJ, USA, 32. Appl. Phys. Lett., 67(21), 3102. Greene, J. E., Natarajan, B. R., and S~queda-Osorio,F Bunshah, R. F. (1 982). ~ e p o s i t i o nTech~ologyfor Films and (1978). J. Appl. Phys., oati in^^^ - ~ e v e l o ~ ~ e and n t s Applications, New Jersey, Greene, J. E.. Sundgren, J.-E., Hultrnan, L., , I., and Park Ridge, Noyes Publications. Bergstroin, U. B. (1995). Appl. Phys. Lett ), 2928. Cale, T. S., Cliaara, M. B., Raupp, G. B., and Raaijinakers, R., and Gruter, K., Deschler, M., Jurgenesen, I. J. (1993). Thin Sol. Films, 236, 294. Balk, P. (1989). J. Cryst. Growth, 9 Cale, T. S., Gandy, T. H., Raupp, G. B., and Ramaswami, M. Gupta, A., Ohno, T., Kariya, T., Shirakata, S., and Isomura, (1991). Thin Sol. Films, 206(1-2), 54. S . (1993). In 9th ~n~ernationul C o n f ~ r ~ n cofe Ternary and S., and Sudarshan, T. S. Multina~yC o m p o ~ nIS. ~ ,~ ~ s u m o tT. o ,Irie, S. lida, and N. Yamainoto, eds. Japanese Journal o f Applied Pbyslcs, Chen, L. J., Lur, W., Chen, J. F., Lee, T. L., aiid Liang, J. M. 32, 74. (1994). Rapid Thermal and Integrated Processing III Gupta, D. and Ho, P. S. (1980). T h ~ nSol, Films, 7 Materia~sResearch Society ~ymposium,99. Ham, F. S. (1958). J. Phys. Chem. Sol., Chen, L. J. and Tu, K. N. (1991). Mat. Sci. Eng.: Rei?orts,6, 53. He, X.-M., Baker, N., Kehler, B. A., Walter, M. C., Nastaw, M., and Nakamura, Y. (2000). J . Vac. Sci. Technol.. A, Cheung, C., Djuanda, F., Erb, U., and Palunibo, G. (1995). ~ a n o s t r $ ~ cMut., t . 5, 513. 18(1), 30. Chittipeddi, S., Dziuba, C. M., Kelly, M. J., Kannan, V. C., Hedenqvist, P., Olsson, M., Wallen, P., Kassman, A., Irwin, R. B., Kahora, P, M., and Cochran, W. T. (1992). Hogmark, S., and Jacobson, S , (1990). Surf: Coat. A~vanced~ e t a l l i ~ a t i oand n Processing for Sem~cond~ctor Technol., 41(6), 243.
ew Frederick for his help in uscript. We also thank MarcNicolet, King Tu, and E. G. Colgan for permitting the use of their data. G. Rarnanath and Harmeet Coindi gratefully acknowledge funding from the National Science Fou through the former’s CAREER 99844~8for making the ~ r i t i of n~ award grant this chapter possible.
678
Processes and Pheno~zena
M., Chen, F., Xi, M., Chang, M., Lando, is. P., ar, P., and Ellwanger, R. (1999). Confkrence Proceedings ULSI X I V , Materials Research Society, 593. Hu, U.Z., Li, M., Sunko, J., Andrews, J., and Irene, E. (1990). Proceedi?~~*~ of I 1 th International Conference of Electrochenzical Society, Electrocheniical Society, 166. Huer~~-Ruelas, J., Lopez-Lopez, M., and Zelaya-Angel, 0. (2000). Jpn. J. Appl. Phys. I , 39(4A), 1701. Hultinan, L., Barnett, S. A., Sundgren, J.-E., and Creene, J. E. (1988). J. Cryst. Growth, 92, 639. Ikushirna, K., Kikuchi, J., Yasuoka, H., Cava, R. J., Takagi, H., Krajewski, J. J., and Peck, W. W., Jr. (1994). J . Phys. Soc. @n., 63(8), 2878. Inaba, M., Yarnakawa, IS.,and Iwase, N. (1990). IEEE Tram. Comp. ~ y ~ Manuf. r ~ Techno~., ~ ~ s13, 119. Inam, A., Wu, X. D., Nazar, L., Hegde, N. S., Rogers, C. T., Venkatesan, T., Simon, R. W., Daly, K., Padamsee, H., ~irchgessner,J., Moffat, D., Rubin, D., Shu, Q. S., Kalokitis, D., Fathy, A., Pendrick, V., Brown, R., Brycki, B., Relohoubek, E., Drabeck, L., Gruner, G., Hammond, R., Gamble, F., Lairson, B. M., and Bravrnan, J. C. (1990). Appl. Phys. Lett., 56(12), 1178. Intermann, A., Koernerb, H., and Koch, F. (1993). J . Electrochrm. Soc., 140(11), 3215. Ito, W. (1995). J. Mat. Res., 10(9), 2216. Itoh, T. (1989). Ion Beam A ted Film G r o ~ t h Elsevier, , Amsterdam. Jones, A. C. and O’Bnen, P. (1997). CVD of Compound S e ~ ~ c o n d u c ~VCH, o r ~ ~New , York. Karlsson, L., H ~ ~ l t ~L., a nJohansson, , M. P., Sundgren, J.-E., and Ljungcrantz, H. (2000). Surf: Coat. Tmhnol., l26( I), I. Kaufman, H. R. and Robinson, R, S. (1987). Operation of B r ~ ~ ~ d - Sources, B ~ a ~ Commoiiwealth Scientific Corporation, Alexandria, Virginia. Kim, Y.-W., Petrov, I., Ito, H., and Greene, J. E. (1995). J . Vac. Sci. Technof.. A , 13(6), 2836. Kleijn, C. R. (1991). Thin Sol. Films,206, 47. nudsen, M. (1909). Ann. Phys., 29, 179 Spallart, M.N. (1998). J. Eleetroche~.
Lopatin, S., Kim, Y., and Diarnand, Y.-S. (1998). A d ~ a n ~ e d Irrterconnecis and Contact Materials and Processes f o r Future Integrated ~ircuit~~, Materials Research Society
Ma, E., Thompson, C. V., and Clevenger, L. A. (1991). J . AppI. Phys., 69(4), 221 1. Madar, R. and Bernard, C. (1989). J. P ~ ~ CoEE., , s ~C5,479. ~ ~ ~ ~ , Manaila, R., Biro, D., Devenyi, A., Fratiloiu, D., Popescu, R., and Totolici, J.E. (1998). App1. Surf. Sci., 134(1-4), 1. Massalski, T. B., Murray, J. L., Bennett, L. H., and Baker, H., Eds. (1986). B i ~ ~ a rAlloy y Phase D ~ ~ g r ~Am~se,r i c a n Society for Metals, Metals Park, Ohio. ,I?., and Ikeda, B. (1995).
McKamey, C. G., Tortorelli, P. Carmichael, C . A. (1992). J. Mat. Res., 9(10), 2747. Meiidicino, M. A., Southwell, R. P., and Seebauer, E. G. (1993). In Gas-Phase and Surface Chemistry in Electronic Mutericrls Processing Sywiposrum, T. J. Mouritziaris, G. R. Paz-Pujalt, F. T. 5. Smith, Materials Research Society Mirkanmi, P. B., Medlin, D. D. C., Clift, M. W., Knapp, J. A., and Barbour, J. C. (1997). J Appl. P h j ~ . , Mitterer, C., KTomenda-Stalliiiaier, J., Losbichler, P., W. S. M., and Stori, H. (1995). Thompson, C. V (1991). J. Appl. PhyLs.,70(8), 4287. Mochizuki, Y., Okamoto, Y., Ishitani, A,, and Hirose, K, (1995). Jpn. J . Appl. Phys., 3 Muhl, S. and Mendez, J. M. (1999). Diam. Related Mat., 8(10), 1809. rchiv f i r Electronik und
R. W., Yang, L. C., and Rockett, A. (1993). In Pr(~ceeding.sof the Institute of Phys~cs Electron and Analysis Group Conference, 10P Publishing, Liverpool, 305. Mullendore, A. W. (1985). InternationaI Coqjerence on the , Duquette, D. J., and Murarka, S. P. sics and C h e ~ ~ s t of r y Boron and ~ o r o ? ~ - Borides, ~~ch (1999). P r o ~ ~ e d i n gqfs the Syn~posi~ on ~ l e c t r o c h e ~ i c a l , APS, Albuquerque, 41. process in^ in ULSI ~ a ~ r i c a t i oand n Semiconductor f ~ e t a l Murarka, S. P. (1993). MetulIi~ation:Themy and Practicefor Deposifion ZI, Electrochemical Society. VLSI and ULSI, B~tterworth-Heinemann,Stoneham. Kryliouk, O., Reed, M., Dann demon, T., and Chai, B. Nakada, T. and Kunioka, A. (1998). Jpn. J. Appl. Phys., 37, (1999). Mat. Sci. Eng.. B, L1065. KWO,J., Cyorgy, E. M., McW B,, Hong, M., DiSalvo, ., and Kunioka, A. (1995). Nakada, T., Migita, K., F. J., Vettier, C., and Bower, J. E. (1985). Phys. Rev. Lett., Jpn. J. Appl. Phys., 3 5(13), 1402. Ng, W.-C., KO, T.-M., Chen, W., and Qi, G,-J, (1998). In Leainy, H. J. arid Dirks, A. G. (1978). J ; Appl. Phys., Proceedings of 2nd Electronics Paclcaging Teclzizology 3430. Conference, A. A. 0. Tay and L. Thiam Beng, eds. Lee, C.-Y. and Lin, K.-L. (1997). J. Mat. Scz., 8, 377 IEEE, Piscataway, NJ, 89. Li, J. C., Nan, S. NI,, and Jiang, Q. (1998). Surf. Coat. Nicolet, M.-A. and Lau, S. S. (1983). In VLSI ~ l e c ~ r o n i c ~ s ; Technol., ~ ~ 6 ( 2 - 3 )135. , ~icrostructure Science, N . G. Einspruch and G. B. Liebscher, H. (1999). Zeitschrijt jur Physikalische Chemie, Larrabee, eds. Academic Press, 6, 329. , 183.
T h i n - ~ Deposition l~ and Treatment Nimmagadda, R., Raghuram, A. C., and Bunshah, R. F. (1972). J. Vac. Sci. Tecktnol., 9(6), 1406. Ohring, M. (1992). The Materials Science of Thin Films, Academic Press. Okarnoto, A., Yoshida, T., M~iramatsu,S., and Shibasaki, 1. (1999). J. Crvst. Growth, 2 ~ ~ - 2 0 2765. , Osaka, T., Koiwa, I., Usuda, M., Arai, K., and Saito, I. (1989). 6. E~ec~rocktem. Soc., 136, 1124. Pal, C.S., Cabreros, E., Lau, S. S., Seidel, T. E., and Suni, I. (1985). Appl. Phys. Lett. Park, D. G., Tao, M., LI, rev, A. E., Fan, Z., Wang, Z., Moharnmad, S. N., Rockett, A., Abelson, J. R., Morkoc, H., Heyd, A. R., and Alterovitz, S. A. (1996). J. Vac. Sci. Tec~nol..8,1 Paunovic, M., Bailey, P. J., Schad, R. G., and Smith, D. A. (1994). J. Electrockem. Soc., Petrov, I., Hultman, L., Hehersson, U., Sundgren, J.-E., and Greene, J. E. (1989). Thin Sol. Films,169, 299 Philibert, J. (1991). Appl. Surf: Sci., 53, 74. o k Chemical Vapor Pierson, €3. 0. (1992). ~ I a i ~ d ~ o Qf ~eposition(CVD), Noyes Publications, New Jersey. Pierson, H. 0. (1994). CVD of Non-Se~conductor o ~ , International, Materials, ASM ~ a n d ~ oASM Ramanath, G., Carlssoii, J. R. A., Greene, J. E., Allen, L. H., and Allman, D. J. (1996). Appl. Phys.
679
Ronning, C., Felderm~inn,H., Merk, R., P., and Thiele, J.-U. (1998). Phys. R Roozeboom, F. (1993). Rapid ~ h e r ~ and ~ i l integrate^ Process~ng II, Materials Research Society Symposi~m, 303, 149. Salehi, M. aiid Hossei, R. (1996). Surj: Eng. Sanders, D. M. (1994). Arc Deposition, ASM ~ ~ ~ ~ d ~ o o l S ~ ~ a Engineering, ce T. M. Lucarelli, K. D. Wheaton, and M. S. Woods, ASM Int Sano, H., Ishibashi, T., and Sato, I(.(1998). Jpn. J. A p t . Phys., 37, L1070. Setton, M. and Van der Spiegel, J. (19911. J . Appl. Phys., 69(2),994. Shibuya, A., Kumnoto, T., and Kimoto, M. (1985). Sumiton~oSearch, 31, 75. Narumiya, Y. (1994). IEEE
Srimathi, S. N., Mayanna, S. M., Sur$ Technol., 56, 277. Stinton, D. P., Besmann, T. M., and Lowden, R. A. (1988). Ceram. Bull., 67(2), 350. Svilan, V., Rodbell, TC. P., Clevenger, L. A., Cabral, C. J., Roy, R. A., Lavoie, C., Jordan J. M. E. (1997). J. Electron. Res.,
e, J. E., Carlssoii, J. R. A., Hornback, V. C., Allman, D. J. and Allen, L. H. (1999). J Appl. Ramanath, G., Ma, Z., Allen, L. H., and Lee, S. (1993). In: Rapid Thermal and Integrated Processing II. J. C. Gelpey, J. K. Elliott, J. J. Wortman, and A. Ajrnera, eds. Materials Research Society Symposium, 303, 63. Ramanath. G . , Xiao, H. Z., Lai, S. L., Allen, L. El., and Alf'ord, T. L. (1996). J. Appl. Phys., 79, 3094. Ramanath, G., Xiao, H. Z., Rockett, A., and Allen, L. H. (1995). J. Appl. Phys., 7~(4),2435. Randhawa, H. (1986). J. Vac. Sci. Technol.: A , 4(6), 2755. Ray, U,, Artaki, I., Finley, D. W., Wenger, G. M., Pan, T,, Blair, H. D., ~ i c h o ~ s oJ.n ,M., and Vianco, P. T. (1996). Proceedings qf Surface ~ o u i ~Internationd. t Advanced Electronic ~ a n u f a c t u r i ~ gTechnologies, 826, Surface Mount Technol. Assoc, 656. Razeghi, M., Wagle, J,, Maurel, P., Qmnes, F., and Pocholle, J. P. (1986). Appl. Phys. Lett., 49(17), 1110. and Morris, J. W. (1995). J. Electron. Mat.,
F, L. (2000). J. Amer Ceram. Soc., 83(2), 245. Rockett, A., Agarwdl, A., Yang, L. C., Banda, E., and Kenshole, G. (1990). P~oceedings of the 21st IEEE ~ h o t u v o l t a ~special c is^ Conference, Rockett, A., Berry, G., Schroeder, D., Xiao, H. Z., and Yang, L. C. (1994). 1st WCPEC, IEEE, 172. Rockett, A., Lornmassoii, T. C., Campos, P., Yang, L. C., and Talieh, H. (1989). Thin Sol. Films, 371, 109 Adriaansen, A. P. M. (1989). Thin Sol.
261. Thornton, J. A. (1978). J, Vac. Sci. Tec Tierney, B. J., Pitner, W. R.,Mitchell, and Stafford, G. R. (1998). J. Elect 3110. Turnbull, D. (1956). Solid State Plizysics - Advances m Research and ~ p p l i c a t i o ~ R. ~ s , Turiibull and F. Seitz, New York, Academic Press. Tuttle, J. R., Contreras, M., Tennaiit, A., Albin, D., and Noufi, R. (1993). 231.6 IEEE Photovoltaic Specialisl Conference, 23, IEEE, 415. ~ a n g t s C. , H., Jackson, T. J., Somekh, R. E., Leake, J and Evetts, J. E. (1997). Measwem. Scz. Technol., 978. Weber, A. and Klages, (2.4'. (1995). J. E l e ~ t r o c ~ SOC., e~. Wehner, G. K. ( 1955). Adv. Electronics Electron P ~ J I s . , 239. Westwood, W D. (1976). Prog. Su$. Sci., 7(2), 71. Westwood, W D,, Ed. (1989). Physics of Thin ~ i i ~ ~ s , Academic Press, New York. Wieczorek, C. (1985). T ~ i nSol. Filn~s, Wittmer, M. (1983). Thin Sol. Films, 1 Wolf, G. IC. (1992). J . Vac. SCE.Techrz Wu, S . IS.and Lin, R. Y (1993) ~ i ~ h - T e ~ n ~ e rOrdered at~re ~nterme~allicAlloys V. Materials Research Society Symposium, 288, 1031 . Yamamoto, T., Takenaka, T., and Rawakami, M. (1996). M a t e ~ i a ~ sr a ~ . ~ a c ~JIM, i o ~37(3), s , 363.
680
Processes and Phenomena
Yang, L.-C. and Rockett, A. (1994). J. Appl. Phys., 75(2), 1185. Yang, Y., Chen, H., Zhou, Y. Q., Mei, X. B., Huang, Q., Zhou, J. M., and Li, F H. (1996). J. Mat. Scz., 31(3), 829. Yu, B. D., Miyamoto, Y., Sugmo, O., Sasaki, T., and Ohno, T. (1998). Appl. Phys. Lett., 7 Yung, E. K. and Turlik, I. (1991). IEEE irrans. Comp. ~ ~ ~ an^^ ~ r Technol., i ~ s 14(3), 549. Zahavi, J. and Pehrsson, P. E. (1986). Photon, Beam, and ted Processes at Surfaces, V. M. P l ~ s m a~ t i ~ ~ u l aC~emical
Donnelly, I. P, Herman, and M. Hirose, eds. ~ a t ~ r i a l s Research Society, Pittsburgh, PA, 173. Zhang, D., Zhang, Z.-D., Chuang, Y.-C., Zhang, B.-S., Yang, J. L., and Du, €3. L. (1995). J. Phys.: Cond. ~ a t t e r7(13), , 2587. Zharmg, H., Zhang,X. J., and %hang, Y. K. (1993). Plat. Sztr~ i n i ~ s h i80, ~ g ,80. Zribi, A., Chromik, R, R., P r ~ s t ~ uR., s , Glum, J., Teed, K., Zavalij, L., DeVita, J., Tova, J., and Cotts, E. J. (1999). 49th Electronic Componerzts and Technology Conjerence, IEEE, 45 1.
Chapter 32 ulk Amorphous Alloys R. B. Schwarz Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, NM, USA
1. Short Overview on the Development of Bulk Metallic Glasses Amorphous (non-crystalline) materials can be obtained for each of the major bonding types (ionic, covalent, van der Waals, hydrogen, and metallic). Ionic and covalent glasses often form when the corresponding melt fails to crystallize during relatively slow cooling, on the order of 1 Ks-I. Naturally occurring glasses such as obsidian were the first ~ them known to man, who as early as 7 0 0 0 0 ~used to make tools. Around 5 0 0 0 ~the ~ Phoenicians discovered oxide-glass making, and near 300 BC they developed the iron-blowing technique. Amorphous metallic alloys are relative newcomers to the world of glasses. In contrast to oxide glasses, metallic melts exhibit far less resistance to crystallization when undercooled and as a consequence do not form glasses at ordinary cooling rates. In 1954, Buckel and Hilsch demonstrated that amorphous thin films could be grown by quenching a metallic vapor onto a cryogenically cooled substrate. About the same time, amorphous Ni,,P, glass was prepared by electrodeposition (Brenner et al., 1950). Then, in 1960, Duwez and co-workers demonstrated that Au7,Si2, glass (throughout this chapter, compositions are given in atom percent) could be obtained by rapid cooling of the melt at rates on the order of 106Ks-'. Metallic glasses rapidly became the subject of extensive research, both because they represent a new state of matter for metallic alloys and because they have several potentially important technological applications. Amorphous metallic alloys can now be prepared by a variety of other techniques, including mechanical alloying (Koch
et al., 1983), mechanical milling of intermetallic powders (Schwarz and Koch, 1986), hydrogenation of intennetallics (Yeh et al., 1983), interdiffusion in solidstate reactions (Schwarz and Johnson, 1983), and heavy-dose electron and ion irradiation (Follstaedt, 1985). Although amorphous alloys prepared by these various methods may be indistinguishable in terms of microstructure and properties, in analogy to oxide glasses the term 'glassy metallic alloy' is usually reserved for amorphous alloys prepared from the melt, those to be discussed in this chapter. High cooling rates are necessary to prevent the melt from crystallizing at T, while it is undercooled from the liquidus temperature, T,, to the glass-transition temperature, Tg. As it is undercooled, the melt viscosity increases monotonically and, once the melt has been undercooled to below Tg, it becomes kinetically frozen in the glassy state. Forming binary metallic glasses (e.g. zrs7Ni3,, Fe8:e,,B,,)requires cooling rates on the order of 106K s-'. Attaining such cooling rates is not easy. Even when a thin melt film is spread over a chilled copper block, the cooling rate is highest near the melt-copper interface but decreases rapidly towards the interior of the melt, which has a lower thermal diffusivity than copper. In practice, a cooling rate of 106Ks-' restricts at least one dimension of the quenched product to be less than about 5Opn1, and thus most binary metallic glasses can only be prepared in the form of thin foils or wires. We shall call these 'rapidly quenched glasses' to distinguish them from the 'bulk metallic glasses' to be discussed in this chapter. Bulk metallic glasses are a small subclass of metallic glasses that can be prepared at relatively slow cooling rates (less than about 103K s-I), enabling the
Infermetallic Compounb: Vol. 3, Principles and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 02002 John Wiley & Sons, Ltd.
682
Processes and Phenomena
tons, rods, and plates with minimum based, ferromagnetic, bulk amorphous alloys is of least a few mni. To date, all bulk significant technological importance because these metallic glasses have at least three components. alloys can have extremely low hysteresis losses, bulk m e t a ~ l glasses i~ can be traced important in sensors, tra~~sformers,motors, and (1974), who showed that other electrical apparatus. 7Cu6Si17and Pd,0Mi,,P20 ulk metallic glasses are easier to prepare in some could be cast in t state by chilling the melt alloy systems (see also chapter 31, by Greer, in the first volume of this series). Table 1 lists intermetallic in water-cooled molds. Later, rehman et al. (1982) noticed that cm-size amorphous Pd,,Ni,oP20 slugs conipositioiis that have bcen quenched into bulk could be formed from the melt by carefully removing glasses with all dimensions exceeding 10 mm. Such tlie surface heteronucleants (oxides, inclusions) and large dimensions could not have been achieved unless cooling the sample at rates of the order of 102Ks--'. In the critical cooling rates were lower than about 1990, resear~hers at Tohoku University started a 100 K s-l. All these bulk metallic glasses have reduced co~prehensiv~ efTort on bulk amorphous alloys, glass-transition temperat~res, Trg= T'J Tl (where discovering several multicornpoiient alloys that can known) that are larger than 0.58. be made amorphous in cm-diameter rods. In 1993, The work by Japanese researchers at the Tohoku eker and ~ o h ~ i s o nfound that Zr,, 2Ti13,8C~12.5-University, and in the USA at the C~liforniaInstitute 22,5 could be formed into a 16-mm diameter of Technology, have been recently summarized (Inoue, 2000; Johiison, 1999) and will not be discussed in great oy s y s t ~ used ~ ~ sto prepare bulk metallic glasses detail here. The present chapter will dwell mainly on Zr-AI-TM, Hf-A1the research conducted at Los Alamos National Laboratory. et al., 1993a, b, 1996a; Tnoue, A1 (Inoue et aE., 1996b; Inoue, Peker and Johnson, 1993), PdCu-P (Inoue et al., ynthesis i-Fe-P (Shen et al., 1999a, (Inoue and Gook, 1995), To form a metallic glass, the melt must be severely ue et nl., 1997b), and Feundercooled. Early studies on crystal nucleation by chwarz, 1999). Here Turnbull and Cech (1950) and Turnbull (1952) demonone or more lanthanides and strated that the main i ~ ~ e d i n i e to n t undercooling a pectively. All these ~~E~ glassy metallic melt resided in oxides and other inclusions that alloys are characterized by having (a) at least three acted as heterogeneous nucleation centers for crystals of AT= Tx--Tg r a n ~ ~ n g lization. Using small d r o p l ~and t ~ emulsion techniques, Characteristic (b) is of these authors were able to isolate or neutralize the cause it opeiis a temperaheteronucleants and obtained relative undercoolings ture window for the p ~ o d u c t ~ oof n near-net-shape AT of about 0.2 T, for a number of elemental metallic metallic compo~entsusing techniques such as forging melts. Perepezko and Paik (1984) performed further and injection molding. The recent discovery of Fedetailed studies using this technique. Other techniques, Bulk metallic glasses with smallest dimension, d, in excess of 1Omm Alloy
Trg
Synthesis method
OS9 0.7 I 0.71 0.58
FXWQFS WQFS FXWQFS MMC MMG WQFS MMC
0.67
*
d ~~)
25 40 72 16 30 14 12
Ref.
We et al. (1996a) Inoue et al. (1996a) Inoue et uE. (1997a) Inoue et al. (1993b) Iiioue and Zhang (I 996) Peker and Johnson (1993) Inoue et d.(1996'0, 1998)
MC =metal mold casting; WQFS = water quenching in fused silica tubes; FXWQFS = fluxing followed by water quenching in d silica tubes. o glass t ~ ~ n s i ~observed. ion
Bulk Amorphous Alloys
including containerless electromagnetic levitation (Herlach, 1991, 1996; Holland-Moritz et al., 1993), and fluxing (Devaud and Turnbull, 1985; Baricco et al., 1996) have been used to extend the range of undercooling and study microstructure developmeiit from undercooled liquids, inagnetic properties of undercooled melts (Wilde et. al., 1996) and the liquid-solid interfacial tension (Shao and Spaepen, 1996). If the hcteronucleaiits in the melt are removed or neutralized, the remaining competition to glass formation is homogeneous nucleation. The most important parameter describing the homogeneous nucleation rate is the ratio Trs= Tg/Tl or reduced gl~ss-transition temperature (Turnbull, 1969). The solid curves in Figure 1 show homogeneous nucleation rates for various values of Trgcalculated using the parameters listed by Spaepen and Turnbull (1976). The nucleation rate becomes negligible for T--+T,because the thermodynamic force for crystallization approaches zero. It is also negligible for T-+Tgbecause as the temperature is lowered, the nucleation rate is kinetically suppressed by the increasing glass viscosity. If we associate a smaller nucleation rate (positive abscissa) with a larger time period during which the undercoo~edliquid can exist without crystallizin then the honiogeneous nucleation rate curves in Figure 1 become equivalent to the customary continuous-cooling-transfo~ation (C-C-T) curves for crystallization. It is then clear that the larger the value of Trg, the smaller the critical cooling rate required to avoid crossing the nose of the C-C-T curve during a continuous cooling experiment, as indicated schematically by the three dashed lines. 1.O 0.8
0.6
T / T, 0.4 0.2
0.0 40
20
0
-20
-40
-60
-80
igure 1 Calculated homogeneous nucleation rate in undercooled mctallic melts for various values of the reduced glas~-transitio~ temperature, TrB.The dashed lines indicate ~ypothetical cooling rates. Formation of metallic glass requires bypassing the nose of the nucleation curve, reproduced with permission of Kluwer Academic/Plenum Publishers
683
Figure 1 suggests that a necessary (though not sufficient) condition for the formation of bulk metallic glasses is a large Trgvalue. Trgcan be increased by increasing Tg and/or by dec~easingT,. Alloying is the main venue for finding melts with large Trgvalues. In general, a change in composition affects T fmuch more than Tg3and experience has shown that glassy alloys are easier to form near eutectic compositions, where 7; is lowest.
Metallic melts usually contain a finite number of crystalline inclusions (oxides, carbides, etc.) that are thermodynamically stable at the melt temperature. Some of these inclusions become potent nucleation centers once the melt is undercooled. A low density of heteroiiucleants may be tolerated if the melt is being cooled at rates on the order of 106Ks-I (as needed to fabricate a 30-pn thick ribbon), since the arnouiit of crystalline alloy that can grow on them is very limited due to the short time the undercooled melt is within the labile regime q--Tg.Potent heteronucleaiits cannot be tolerated, however, if the melt is cooled at slower rates, as needed for the production of bulk metallic glasses. FlLuring is a conimoii method for renioviiig oxide inclusions from melts. For the synthesis of bulk metallic glasses, the flux should remain liquid even below the glass-transition temperature of the glassy alloy being formed. Otherwise, the solidified flux may itself facilitate lieterogeneous nucleation. In addition, the flux must not react with the crucible containing the melt, and the crucible itself must not be a crystallization agent. Drehman et al. (1982) demonstrated the efficacy of B,O, as a flux to purify P d ~ ~ melts ~ i held ~ ~in P ~ ~ amorphous SiO, crucibles, and this Aux has been used since to prepare a large number of bulk metallic glasses: Pd-Cu-P (He et al., 1996a, b, c; He and Schwarz, 1997), Pd-Ni-Cu-P (He et al., 1998), Pd-NiFe-P (Shen et al., 1999a), and ~ e - ( ~ a - C o - C r ~ (P,B,C) (Shen arid Schwarz, 1999). Typically, is first formed by arc melting its coniponents. Chunks of the alloy are then placed in a fused-SiO, tube together with pieces of dehydrated B,O,. Upon heating the crucible, the melt forms at the bottom of the tube and is covered by a layer of the molten flux. The flux wets the fused-silica tube preventin contacting the iiiner walls of the tube, which may contain SiO, crystals. The fluxing process is helped by
684
Processes and Phenomena
cycling the temperature and by evacuating the tube to facilitate the removal of evolved oxygen. Pd-Ni-P glassy alloys prepared by purifying the melt in B20, have oxygen contents less than lppm, as measured by a 160(d,p)170nuclear reaction (Maggiore and Schwarz, 2000). Reducing the oxygen content not only reduces the critical cooling rate for glass formation (thus enabling the synthesis of larger bulk glasses), but also improves some of the glass properties. For example the Krc fracture toughness of bulk Zr41.2Ti13,8CuI25Ni10Be22,5 glass is about 55 MPa mrj2 when the glass contains 800ppm oxygen (Conner et al., 1997; Cilbert et al., 1997), but only about 20 MPa m1j2when the same glass contains 1600ppm of oxygen (Lowhaphandu and Lewandowski, 1998). The removal of oxide inclusions by fluxing works when the partial Gibbs free energy for dissociation of oxygen in the flux is more negative than in the oxide inclusions. The elements tfiat can be fluxed by molten 203 can be found by analyzing the free energy of formation for the following reactions:
where diamond- shape^ (angular) brackets denote the solid state; braces the liquid state; round parentheses
the gaseous state; and square brac~etsthe state of dilute solution, with the suffix denoting the solvent. Figure 2 shows the elements that can be fluxed with B20,. All these elements form less stable oxides than B2°3*
Elements of the group IVA (Ti, Zr, Hf) have a high affinity for oxygen and reducing them would require very aggressive Auxes, such as lithiuni nietstboride. The problem of using such fluxes is finding inexpensive crucibles that do not react with both the molten alloy and the flux. Fortunately, the Group IVA elements have both a large affinity for oxygen and a large oxygen solubility in the molten and solid states. Because of this, alloy melts rich in 2 r or Ti will dissolve less stable oxides, incorporating the atomic oxygen in the melt. If the total oxygen content in the melt is significantly below its solubility limit, then upon quenching the melt to form a glass, the oxygen will most likely remain in solution, Thus, alloy melts rich in Ti, 2 r and Hf are in a sense s ~ ~ ~ ~ ~ x i ~ g . An example of a selfng, bulk-glass forming alloy is Zr, .2Ti1,&U 2.5Ni 2.5 (Peker and Johnson, 1993), which contains about 55 at.% of the elements Ti and Zr. After prolonged melting of the alloy in an arc
Figure 2 Elements amenable to B,O, fluxing
Bulk Amorphous Alloys
685
same composition prepared by the rapid quench in^ of furnace, all oxide inclusions that initially were part of melt. The degradation of properties is most likely due the starting elements, including BeQ, dissolve in the to residual porosity and to oxides, originally on the alloy melt, Upon turning off the electric arc, the melt surface of the powder particles, which are incorporated cools in the copper hearth, but the oxygen in solution into the consolidated product. does not precipitate as oxides. Having no oxide It was mentioned earlier that hulk metallic glasses inclusions to act as heterogeneous nucleation centers, (prepared by melt quenching) have large AT= Tx--Tg and a very low eutectic temperature, the natural values, typically in excess of 40K. Amorphous cooling of the melt in contact with the copper hearth powders of the same composition are easily prepared is sufficient to undercool the melt into the glassy state by techniques such as high-pressu~einert gas atomiza(He et aE., 1996c), as shown in Figure 3. The selftion and mechanical alloying. However, in our fluxing characteristics of the 2;r41~2Ti,,~,C~,2~sNi~o~e22.5 experience, these powders have AT values about half melt have been studied by Kim et wl., (1994), who used those measured in a fluxed and quenched bulk glassy electrostatic levitation to avoid the heterogeneous alloy of the same composition. The AT decrease is nucleation that might otherwise occur at the meltattributed to a decrease in T, rather than to an increase crucible interface. in Tg.Even with a lower AT value, amorphous powder consoIidation has great ~otentialfor the s~nthesisof the ~ o ~ s o ~ i d a ~ i o porosity-free, bulk amorphous alloys ( ~ a w a ~ u et ra ders al., 1994). There is great interest in the reparation of bulls metallic glasses by consolidating amorplious powder precursors. The main advantage of this technique is that it poses no limit on the size and shape of the amorphous product. Early attempts on this synthesis route started from binary amorphous powders prepared by mechanical alloying (Schwarz et al., 1989; Shingu et al., 1990; see also chapter by Seetharaman and Semiatiii in this volume). These were binary amorphous alloys having relatively small AT= Tx-Tg ranges. Hot pressing these powders at temperatures slightly below T, produced low-porosity bulk amorphous allays. However, the properties of these alloys were inferior to those of thin glassy ribbons of the
Figure 3 Ingot of Zr41,2Ti,3J.Zu,2 5Ni,,Be,,,, prepared by arc melting the elements in a water-cooled copper hearth, and letting the melt solidify after interrupting the arc
sses
Pd4,Ni,,P,, was one of the first bulk amorphous alloys discovered. Using water-quenching techniques, Chen (1974) produced amorphous P d 4 0 N i ~ ~ alloy P ~ o rods with diameters of 1 to 3mm. Using thermal cycling and surface etching, Drehman et al. (1982) were able to remove heteronucleants residing at the sample surface and prepared glassy Pd~oNi40Pzo buttons with minor diameters of about 5mm. Using B,Q, as a flux, Kui et al. (1984) were able to increase the dimensions of the buttons to about 10mm. Finally, i ~provi ngon this technique, Schwarz and He (1997) demonstrated that this alloy could be prepared as rods having at least 25mm diameter, as shown in Figure 4. Pd40Ni40P20, although expensive because of the high cost of Pd, is an important scientific alloy, because it is one of the easiest to prepare in bulk glass form and contains only three elements. This alloy is therefore a good model glass system on which to study fundamental properties of bulk glasses. Indeed, a large number of studies have been performed on this alloy. Furthermore, having only three elements facilitates the simulation of its structure by computer modeling (Nicholson et al., 1998, 2000). Figure 5 shows the composition range for the formation of 10-mni diameter Pd-Mi-ES glassy rods using the B,O? fluxing techniaue. The DhosDhorus
Processes and Phenomena
50
_.
L.s..."."._.-..
-150 ~~0
300
400
T
re
300-g ingot of bulk amorphous P d ~ ~ N i ~rod ~P~* 6 ~ i ~ e r e n t iscanning al calorimetry curves for various with 25-mm diameter prepared by fluxing in B,O, and water Pd-Ni-P bulk metallic glasses, reproduced with permission of y uenchiiig Trans Tech Publications Ltd
content needed to form the amorphous phase is close to 20 at.%, whereas the pal1adium:nickel ratio can be varied over a wide range. The metal homogeneity m 10-mm diameter range (25 to 60at.Oh ~ a l ~ a d i ufor rods) is determined by the decreasing TrSvalue as the Pd:Ni ratio moves away from that at the lowest eutectic point, which is close to 1:l. Certainly, the
i
20
40
60
80
Pd
t.%) Composition range for bulk glass formation in the Pd-Ni-P system. Filled circles denote the foilnation of glassy rods with diameters of at least 10mm. Open circles represent compositions at which the rods were crystalline. Squares give compositions that develop ~anocr~stalline phases during I hour at 410 "C, reproduced with peiinissioii of the Materials Research Society
metal homogeneity range widens for smaller diameter rods that can be prepared at faster cooling rates. Di~erei~tial scanning calorirnetry (DSC) is a simple tool for measuring the thermal stability of metallic glasses. Figure 6 shows the DSC traces of the amorphous Pd,,Ni,,P,,, Pd40Ni40P~o and Pd,oNi,4P16 alloys heated at the rate of 20Kmin-". The glass transition temperature, Ts, is defined by the onset of a small endothermic signal in the DSC trace, as indicated in curve (a). The large exotherrnic peak at T, is caused by the crystallization of the glass. As mentioned before, the difference AT= Tx--Tgis of technological importance because the larger AT, the easier it is to prepare near-net-shape amorphous components by forging. Figure 7 shows AT for 10nun diameter glassy Pd-Ni-P rods as a function of Pd coiitent (He et al., 1996a). AT is largest for Pd4oNi4$?,, and decays away from this compositioii. There is a qualitative correlation between the value of AT and the ease of glass formation, but even an alloy with 25 at. ($6 Pd could be easily cast into 10-mm diameter arnorphous rods. Melts containing 20 at.% or less pal l adi u~ could not be cast into a 7-mrn diameter amorphous rod. ~ ~curve (c) in Figure 6, The DSC trace for P d , ~ N i 16, shows a small exothermic peak, labeled A , between Tg and T,. A similar peak has been observed in thin glassy Pd-Ni-P foils prepared by rapid quenching at rates on the order of 106Ks-'. This peak has been given various interpretations. Chen (1976), using X-ray diffraction, attributed this peak to the formation of
Bulk A ~ o r p Alloys ~ ~ ~ ~ s
687
that the two decomposing glassy phases have different atomic packings, with the phosphorus-rich phase being better represented by the Polk (1972) model (a dense random packing of hard spheres, with the metalloid atoms in the larger holes) and the phosphorus-poor phase better represented by the Gaskell (1979) model (a random aggregate of chemically ordered structural 80 units). bX Phase separation is also observed during the I1 annealing o f ~~Z~ Pd-Ni-P glasses. 5 used field-ion microscopy and a three-dimensional atom probe characterization (APFTM) to study the solute distribution in bulk P d 4 ~ N i glass ~ ~ ~inP the ~ ~ascast state and after annealing below and above the 20 30 40 50 60 glass-transition temperature, Tg= 303 "C. Statistical ~ ~ L L CONTENT, A ~ I X~ ~ analysis of the atom probe data detected the presence of chemical short-range order (CSRO) in the as-cast Figure 7 Difference TX-T as a function of palladium concentration in Pd-Ni-P bulk metdlic, reproduced with alloy. All three elements exhibited a preference to be permission of Trans Tech Publications Ltd surrounded by any of the other two elements, this tendency being strongest for the P atoms, and weakest for the Pd atoms. The atom probe chara~terizatio~ also indicated that the palladium and nickel atoms are an fcc(Ni,Pd) crystalline solid solution. The TEM randomly distributed in the as-cast bulk P d 4 ~ ~ i 4 ~ P ~ * obse~ationsof Willman et al. (1 987) corroborated glass. No evidence of deco~positionor clustering was that the first crystalline phase formed upon annealing found in the field-ion images of material annealed for near T, was a fine dispersion of nano-sized, phos48 h at 140 "C, 24 h at 200 "C, '2 h at 300 "C, or 0.5 h at phorus-poor fcc crystallites dispersed in a phosphorusrich amorphous matrix. However, the small exother350°C. However, ~ d ~ ~ glass N iannealed ~ * ~2 h ~at ~ mic peak between Tgand T, has also been interpreted 350 "C and 0.5 h at 390 "C (above T,) showed phase as resulting from two glass transitions (i.e. two separation into two glasses: a phosphorus~rich~pallaconsecutive endothermic peaks), occurring at slightly dium-depleted glass and a phosphorus-depleted/ different temperatures (Chen, 1976; Schluckebier and palladium-enriched glass, in agreement with the earlier Predel, 1983; Jink et al., 1987; Oehring and Haasen, observations of Oehring (1989) in thin ribbons of 1986; Oehring, 1989). More recent investigations glassy Pd36Ni47Pi7. Upon annealing the bulk glass for suggest that peak A is exothermic in nature and results 1 h at 410"C, it developed three nano-crystalline from phase separation in the amorphous state, as we phases: a palladium-rich phosphide, a nickel-rich will discuss next. This phase separation is then phosphide, and a palladium-nickel solid solution. followed by crystallization, which gives a stronger The compositions of these three phases are indicated and sharper signal consisting of two or more exotlierin Figure 5 by square symbols. The analysis also mic peaks. determined that the boron content in the as-prepared ~nsightinto the excellent glass-forming ability of Pdglass was 0.03 at.%, showing that few boron atoms are Ni-P alloys has been provided by atom-probe field-ion incorporated into the melt during Auxing. microscopy ( A P F I ~ ) techniques. Oehring (1989) These two APFIM studies strongly suggest that applied this method to thin Pd-Ni-P glasses prepared upon heating P ~ 4 ~ ~ glass, i 4 ~crystallization P ~ ~ is by rapid quenching. He found that Pd36Ni47P17 preceded by phase separation, which starts within the annealed below the crystallization temperature develT,--T, range. ~resumably, Tg denotes the onset ops compositional Auctuations, up to 1OOnm in size. temperature enabling di~usionin the glass. Due to The two decomposing amorphous phases have the differences in atomic and/or bonding strength, the same nickel content as the matrix, but one amorphous various elements in a multicomponent glass niay have phase is Pd-rich/P-poor and the other Pd-poor/P-rich. different mobilities and thus different Tg. The The latter phase has a composition close to Pd46Ni45P9, Tg= 303 "C, measured by DSC, i s determined by the and this phase is the one believed to crystallize first most mobile element in the glass. Since phasc into an fcc(Pd,Ni) solid solution. Oehring suggested separation may occur even if only one element is A
I
laJ/
688
Processes and P h ~ n o ~ e n f l
mobile, it is not surprising that phase separation should start slightly above Tg (Figure 6). If crystal nucl~ation in ternary (and higher-order) glasses requires two or more mobile species, then 7; will exceed Tg.In binary glasses, however, the motion of one atomic species should also enable crystal nucleation, and thus Tx should be close to Tg. Phase separation in binary metallic glasses, even when thermodynamically favored, may be difficult to observe. This simple kinetic model may explain why ~~1~ metallic glasses having at least three elements have large undercooled Tx-Tgregions.
Figure 8 shows the glass-forming range for 7-mm diameter glassy rods in the Pd-Cu-P system (Schwarz and He, 1997). The metal homogeneity range for 7-1nm ods is narrower than that for 10i-P rods (Figure 5). Similar to the Pd-Ni-P system, the optimum phosphorus content for bulk glass f~rmationis restricted to near 20at.%. However, the data in Figure 8 suggest that the optimum phosphorus composition for glass formation in the Pd-Cu-P system is less than 20at.%. To understand the effect of p h o s pho~ son the formation of Pdbased bulk glasses, Jin et al. (2000a) studied structural
and
electronic
properties of ( P d o . ~ ~ i o . ~ ) ~ ~ - * P x and (Pd0.625CU0.375) 100-xPx glasses as a function of phosphorus content, x. They found that the molar volunies of the P d - ~ i - Pbulk glasses plotted as a function of P content follow a straight line, whereas those for the Pd-Cu-P glasses follow two straight lines of difYerent slopes, intersecting at a critical P concentration, as shown in Figure 9. An extrapolation of these lines t o the pure-phosphorus ordinate determines the partial molar volumes of phosphorus, F p in the glass, given in the last column of Table 2. The molar volumes deduced from the extrapolation of - the lines with negative slopes have comparable Vp values of 7.18, 6.85, and 7.10 cm3mole- I. Similarly, the Tp values deduced from the extrapolation of the lines with positive slopes are also of comparable magnitude, 9.2 and 8.95 cm3mole-'. This suggests that Tp assumes one value (close to 7.1) for all the Pd-Ni-P glasses and for the two Pd-Cu-P glasses with phosphorus contents lower than their respective critical concentration, and a different value (close to 9) for the Pd-Cu-P glasses with phosphorus above the critical concentrations. An extrapolation of these lines to the metal ordinate yields the combined partial molar volume of Pd and Cu in these glasses, weighted by molar fraction of the elements. The partial molar volumes of Pd and Cu were calculated assuming that, as for phosphorus, the partial niolar volumes of the metal atoms take only two values: one for the Pd-Cu-P bulk glasses fitted by the two lines of positive slope, and another for the alloys fitted by the two lines of negative slope. The 100-xpx,
(P&.&u0.5)
c'
8.20
-
8.10
-
,375)100-x'x
6E 8.15 43
8.05 8.00
U
2
4
P
7.9 7.7 7.6
Composition range for bulk glass formation in the Pd-Cu-P system. Filled circles denote the formation of glassy rods with diameters of at least 7mm. Open circles represent compositions at which the rods were crystalline, reproduced with permission of ASM Internation~land The Minerals, Metals and Materials Society
5
16
17
18
19
20
~ ~ 0 S ' ~ O C~NTENT ~~S (at."/") ~ i ~ 9~ Molar r e volumes of Pd-Ni-P and Pd-Cu-P bulk metallic glasses as a function of phosphorus content, reproduced with permission of the Materials Research Society
~
~A ~l o kr ~ h Alloys u~s
689
Table 2 Partial molar volumes of Pd, Gu, Ni and P deduced from extrapolations of density me~surernentsfor bulk metallic glasses as a function of phosphorus content. The table includes the known molar volumes of elemental Pd, Cu, and Ni with fcc structure Alloy ~ h o s p h o ~ ux-range s V p d cm3/mole VcUcrn3/rnole VNi c~l'//nnole Vp cm3/rnole @do sNi0 5) 100- P, (Pd05cu0 5)100-XB\
0%
625CUO 37s)*oo-xpx
fcc Pd fcc c u fcc Ni
16-22 15.5-17.5 17.5-19 17-1 9 19-20
8.95 8.95 8.68 8.95 8.68 8.90
7.56 6.83 7.56 6.83
__
7.10
-
-
calculated partial molar volumes of Pd and Cu are given in Table 2. Figure 9 and Table 2 show that the partial molar volumes of Pd, Cu, and P in the Pd-Cu-P glasses change abruptly at critical concentrations of phosphorus. These changes were interpreted (Jin et al., 2000a) as caused by changes in the local atomic environi~ent(short-range order, SRO). For all the Pdng to the lines with negative slope ial molar volumes of Pd, Cu, and P are approximately 9.0, 7.6, and 7.0 c d mole-', respectively. When the phosphorus content increases beyond the critical phosphorus concentrations, the partial molar volume of Pd decreases by 3.3% to 8.7, that of Cu creases by 10.5% to 6.8, and that of phosphorus i n c ~ e a s eby~ 30% to 9.1, The F p value in the Pd-NI-P glasses, deduced from the single fitted line in Figure 9, is approximately 7.2, thus close to the F p values in the Pd-Cu-P glasses below their respective critical concentrations. These results suggest that the SRO in the Pd-Ni-P bulk glasges is similar to that of the Pcl-Cu-P bulk glasses for phosphorus contents below their respective critical concentrations. Thus, the glasses studied by .Tin et al. ( 2 ~ ~ can ~ abe) classified into two groups: (1) glasses with Fp GZ 7 and (2) glasses with Fp x 9 cm3mol-'. For the first group, the partial molar volume of the metal atoms i s slightly larger in the glass than in elemental form (elemental molar volumes are also given in Table 2), whereas the opposite is true for the second group. If indeed these two groups of glasses have different types of SRO, then the partial molar volume results suggest that the change in SRO is accompanied by a change in the electronic bonding, as discussed next. ir et aE. (1999) performed X-ray photoelectron spectroscopy (XPS) ~ e a s u r e ~ e n ttos elucidate the driving force for a possible change of SRO in the PdCu-P glasses. Figure 10 shows XPS data for
6.66 -
7.18 6.85 9.20 7.10 8.95
__
-
-
-
Pd40Ni40P,o,and for three (Pdo~5Cuo,5)~oo~APx glasses with x = 16.5, 17-5, and 19. It is customary to present the XPS spectra with the zero of the ene Fermi energy, so that the abscissa repi energy. In Figure 10, the abscissa of each curve has been shifted to superi~posethe h i g ~ ~ ~ n e redges g y of the d-band for all XPS curves, and this edge was placed at x=O. By doing this, one can more clearly see the difYerent ~ontributionsto electronic ~ o n d i n gin these glasses. The XPS signal for the Pd4~Ni40P,~ glass has a broad density of states with a sharp high-energy edge, characteristic of a transition metal with a filled, or partially filled, d-band. That the Fermi energy for Pd40Ni40P,oglass is located within the cl-band agrees with the ab-initio electronic calculatio~sof Nicholson eC al. (2000). The XPS spectra for the have similar broad d-type bands but, in addition, have
Pd-Gu-P
1.o
1'5
0
-12
-10
-8
-6
-4
-
ENERGY (sV) Figure 10 XPS spectra of Pd40N~40P~0 and three (Pd0.5C~0.5)100-xPx bulk glasses with x = 16.5, 17.5, and 19. The inset is an expanded view of the top energy of the spectra for the copper-containing glasses, reproduced with permission of the Materials Research Society
690
Processes and Phenomena
high~energytails characteristic of partially filled s and p-states. This indicates that the d-bands in the Pd-Cu-P glasses are full and that excess electrons occupy s and p states of energy higher than that at the top of the dbands. The inset to Figure 10 is an enlargement of the highest energy range of the XPS signals for the Pd-CuP glasses. The inset shows that the phosphorus increase causes an increase in the occupancy of the s- and-pbands. The energy increase is about 0.08eV per atom percent increase in phosphorus. Jin et al. (~0OOa)used these observations to explain the structural changes implied by the changes in partial molar volumes seen in Figure 9. in metal-metalloid glasses has rs for a long time. Some undertallic glasses has been obtained by simulating their structure in computer models. A comparison of measured and calculated atom-pair distribution functions was taken as a measure of the model accuracy. It was found that the models reproduced the measured atom-pair distribution functions only when con~tructed fol~owing special algorithms that favor the development of specific types of SRO. Examples are the algorithm of regor (1977), which biases the alloy against metalloid-metalloid nearest neighbors, and the algorithm of Gaskell (1979), which biases the alloy towards containing a large number of trigonal prismatic units. These units have a metalloi~atom at the center and metal atoms at the corners and near the faces. Clearly, the glass could adopt many other SRO configuration$. Such amorphous polymorphisi~ should be more prevalent in ternary (and higher component) glasses, which have a higher degree of freedom for packing the tetrahedral atom units that are most prevalent in liquids (Schwarz and He, 1997). The SRO that is trapped in the glassy state is that present in the undercooled melt as it approaches the glass-transition temperature. As the melt is undercooled, it seeks the state of SRO that minimizes its free energy at that temperature. Changes in the electronic ~ont~ibution to this energy, brought about by changes in alloy composition, can cause changes in the SRO adopted by the undercooled melt and the glass. The XPS data (Figure 10) show that Pd-Cu-P glasses have filled d-bands. For these glasses, an increase in the phosphorus content causes a rapid increase in the occupancy of the s and p states, raising the electronic contribution to the free energy. The increase in free energy witli iiicreasing phosphorus content niay then force the glassy alloy to perform a polymorphic change
to a SRO structure of lower free energy. The molar volume data suggest that a structural change occurs in glasses at 17.5 at.% P. A similar the (Pdo&uo 5)100-xPu structural change seems to occur in the (Pd,,,&u, 37s)100-xPx glasses at 19 at.% P. The diEerence in critical phosphorus concentration that triggers the transformations is consistent with the difference in the Pd:Cu ratio in these glasses. Classes having a larger Pd:Cu ratio have also a larger density of unoccupied dstates, which can accommodate a larger fraction of the electrons donated by the phosphorus. Thus, in these alloys, the s and p states above the d-band are less populated, and the polymorphic transition should occur at a higher phosphorus concentration, as is observed in Figure 9. In contrast to the Pd-Cu-P alloys, the molar volume data for ~ d ” ~ iglasses - P decrease monotonically with increasing P co~centration.Within the composition range shown in Figure 9, Pd-Ni-P shows no indication of a poly~orphictransition. This may be explained by the fact that the density of states for amorphous Pd,oNi,oP,, is cliaracterized by a partially filled, or just filled, d-band (Figure 10). Thus, in Pd-Ni-P glasses an increase in the phos~horusconcentration does not cause a significant increase in the free energy of the alloy. Increasing the phosphorus concentration by a large amount may eventlxally cause a polymorphic transition similar to that observed in the Pd-Cu-P glasses. However, this transition may occur at phosphorus co~centrationswell beyond that needed for the formation of bulk metallic glasses. The optimum phosphorus concentration for bulk glass formation in the Pd-Ni-P system is at the center of its homogeneity range, X N 19 at.%. For the Pd-Cu-P glasses, the optimum phosphorus concentration for bulk glass formation is close to the crossing points in Figure 9, i.e. 17.5 at.% and 19 at.% for Pd:Cu ratios of 1:1 aiid 5:3, respectively. However, none of the Pd-CuP co~positionsis as good a bulk glass former as P d ~ o ~ i ~ oThe P ~ XPS o . data suggest that the s aiid p bonding at energies above the d-band, present in all the Pd-Cu-P glasses, is less favorable to bulk glass formation than d-type bonding. These studies show that the metalloid concentration can influerice the SRO in inetal-~etalloidglasses and thus can have a significant effect on their glass~formin~ ability and properties.
Since the Pd-Ni-P and Pd-Cu-P systems have wide homogeneity ranges for bulk glass formation, it is
Bulk A ~ o ~ Alloys ~ ~ o ~ i ~
logical to expect that a quaternary Pd-Ni-Cu-P alloy will also form bulk glasses over a wide coniposition range. A quaternary alloy has an additional degree of freedom that may stabilize the melt, and thus improve its glass formability. Figure 11 shows the glass-forming range for 7-mm diameter (Pd-Ni-Cu)*0P20glassy rods. The mini mu^ a ~ o u n of t palladium needed to form a 7-mm diameter glassy rod is about 20at.%, similar to that in the ternary Pd-Ni-P system. Decreasing the size of the glassy product enables a reduction of the palladiLim content. For example, amor~hous2-nimdiameter balls were prepared at the composition p d * ~ ~ ~ ~ N Within i ~ ~ P ,the ~ . Pd-Cu-Ni-P system, P d 4 0 ~ u ~ 0 ~ iseems ~ 0 P ~to0 have the highest glass formability (Inoue et al., 1997b).
There is great techno~o~ical interest in developing bulk ferromagnetic glasses. Soft magnetism and hard magnetism have been reported for bulk amorphous Fe-(Al, Ga)-(P, C, B, Si, Ge) and Nd-Fe-A1 alloys, respectively (Inoue, 1997). It is therefore of interest to study the substitution of Fe for Ni in the Pd-Ni-P glasses. We found that 7-mm diameter amorphous
69 1
rods of Pd-Fe-Ni-P can be easily prepared over a wide composition range (He et al., 199$). Figure 12 shows the glass-foming range for ( P d , F e , ~ i ) ~ O Pbulk ,~ metallic glasses prepared by fluxing the melt with B,O, and cooling the purified melt at rates of about ~ O O I C S - ~(He et al., 1998; Shen elf al., 1999a). The figure shows that starting with substitute better for nickel than glassy Pd40Ni40-.xFexP20 alloys can be formed for x 20, whereas glassy P d 4 0 - ~ N i 4 ~ Falloys e ~ P ~can ~ be formed only for x < 7.5. Palladium and nickel in combination are necessary for this class of bulk metallic glasses since neither Pd-Fe-P nor Ni-Fe-P glasses can be prepared by the ~ u x i n gand water-quench in^ eth hod. Depending on temperature and applied magnetic field, Pd-Fe-Ni-P bulk glasses can exhibit paramagnetic, superparamagnetic, fe~~.omagne~ic, or spin-glass behavior. Superparamagnetisni is of technological importance because it may result in caloric effect (Kokorin et al., 1984; 1992, 1993; Bennett et al., 1994; Chen et al., 1994, 1995) and giant magnetoresistance (Berkowitz et al., 1992; Colzde et al., 1994; Howson et al., 1994; Dimitrov et al., 1995; Fujimori et al., 1995; Hickey et al., 1995; Yu et al., 1995; Wiser, 1996; Madurga et al.,
i
Cu-Ni-P system. The phosphorus concentration is kept at 20 at.%. Filled circles denote the formation of glassy rods with diameters of at least 7mm; open triangles denote partially amorphous rods; open circles denote crystalline rods. Thick black lines indicate the bulk glass formation range in the ternary alloys Pd-Ni-P and Pd-Cu-P, reproduced with permission o f ASM International and the Minerals, Metals and Materials Society
Figure 12 Glass formation range for bulk ( P d , ~ ~ , F e j ~ , P ~ ~ bulk glasses. Filled circles denote the formation of glassy rods with diameters of at least 7 m ; open triangles denote partially amorphous rods; open circles denote crystalline rods. The thick black line indicates the bulk glass formation range in the ternary (Pd,Nij,,P2, system, reproduced with permission of ASM International and The Minerals, Metals and Materials Society
692
Processes and P h ~ ~ o ~ ~ ~ a
1996), which have applications for magnetic refrigeration (Kuz’min and Tishin, 1993) and magnetic Tf recording (~ujiwara, 1993), respectively, The spinglass state appears to be only of theoretical importance. The spin-glass state results from a co-operative n freezing of spins at a well-defined spin-freezing tem~eratureTf ( ~ y d o s h 1993). ? Below this temperature, the spins are in a metastable state that lacks the long-range order clxmcteristic of ferromagnetic or 3 antiferromagnetic states (Mydosh, 1993). The Fe”~ontainingbulk glasses are paramagnetic at high temperatures and become s ~ ~ p e r ~ a r a ~ a g n e t i c below a critical temperature, Td (x). At even lower temperatures, and under a weak applied magnetic field, these alloys become spin glasses, as evidenced by static n and dynamic magnetic measurements. 1 Figure 13 compares the low~temperaturemagnetic and Pd40Ni22 5susceptibility o f amorphous Pd40Ni40P20 Fe17.sP20, both measured at 100e (Shen et al., 1999b). ure 13 §usceptibility of bulk P d ~ N i 4 0 ~glasses Z * (open The susceptibility of ~d40~i40P20 can be described by symbols) and bulk amorphous Pd40NizZ .J?eI7sP,, (filled the super~os~tion of temperature-i~dependent and symbols) as a function of ~ e ~ ~at an e applied ~ ~ field t ~M, r ~ Curie-Weiss ternis. The susceptibility of the Feof 10Oe, reproduced with per~ission of the American conta~ningglasses is more interesting. The zero-fieldInstitute of Physics cooled (ZFC), field-cooled (FC), and field-warmed (FW) susceptibility curves af glassy Bd40Ni2z.sFe,7,sP20 all show a cusp at 29.6IS, which defines the spin-glass error) with increasing Fe content. This increase transition tem~erature,Tf, for this alloy and applied correlates with the observed increase in the paramagmagnetic field. The ZFC, FC, and FW curves supernetic Curie temperature Q, indicating an almost linear impose for T > Tf but the ZFC curve diverges from the increase in the average Fe-Fe interaction force, J,, with other two for T < T,. These characteristics are typical increasing iron content in the glassy alloy. The proportionality between Td and 0 is clearly visible in of spin-g~a~s behavior: below Tf, the spins freeze into random spatial distributions that depend on the Figure 14, where 0 has been plotted with the ri hand-side scale. magnetic field at the time of freezing. All the FeOne unexpected o ~ s e ~ ~ t iino n~~1~ ~ e t a l l i c con~ainin~ bulk ~ d 4 , ~ i 4 0 - , F e ~ Pghsses ~o studied Pd,oNi40-,FexP20glasses, cooled under a finite applied (5 d x 6 17.5) s h ~weda similar low-temperature susfield, is the appear~nce of re-entrant spin-glass ~eptibilityversus temperature behavior. In contrast, behavior . The occurren~eof the r e ~ ~ ~ t r sapn~t ~ - g l a ~ s the ZFC, FC, and FW susceptibility curves of the applied field, as behavior is strongly amorphous Pd40Ni4,P20 (plotted in a 2500-times shown in Figure 15, I ir, and Tf denote the ordinate scale) superi~pose.Clearly, there is no spintransitions from the agnetic to the ferroglass state in this alloy. magnetic, and from the ferromagnetic to the spin-glass Figure 14 defines the magnetic states of bulk states, respectively. The ferromagneti~state was clearly 2o glasses with 5 <x < 17.5, measured identified by Arrott plots (Shen et at., 1999b) and by at a low field of 100e. The spin-glass freezing the lack of time dependence in the ZFC ~ a ~ n e t i ~ ~ t i o n . temperature Tf increases with increasing Fe content. In bulk P d 4 0 ~ i ~ 2 , s F e ~ 7lasses, . ~ P ~ re-entrant o spinThe apparent linear increase in Tf (x) represents a glass behavior is only observed at large applied fields. ‘scaling law’? which has been considered indirect The ferromagnetic-like region broadens with increasevidence for the role of the Ruderman-Mitteling applied field, as shown in Figure 15. This Kasuya-Yoshida (RISKY) interaction in creating the broadening results from a simultaneous increase in T, spin-glass state ( ~ y d o s h 1993; , NCel, 1949). The figure and a decrease in T,. This seems to be the only example also shows that the temperature Td, denoting the to date of a field-induced ferromagnetic-like state transition from the paramagnetic to the superparabetween superpara~agneticand spin-glass states. The magnetic states, increases linearly (within experimental
1
U
Bulk A ~ o r p ~ o Alloys us
693
300 140 120
100
@
8Q
In
1
1 Temperatures Td (open circles) and Tf (closed circles) denoting domains for paramagnetic (PM), superparamagnetic (SPM), and spin-glass (SG) phases in (x = 5-17.5) alloys. The bulk amorphous Pd,Ni,,-,Fe,P,, domain boundaries were measured at an applied field of 10Oe. The parama~~eticCurie temperat~re 0 (open triangles) is pfotted with the right-hand side scale. Straight lines are least-square fits to the Td and T, data, reproduced with permission of the American Institute of Physics
mechanism for the field-induced ferro~agnetic-like state is not understood. It has been theoretically suggested (Chantrell and Wohlfarth, 1983) that for small applied fields Ha, the average interaction energy .Ei~etween pairs of superpara~agne~icdomains increases with H i . Thus, it should be interesting to study: (1) the influence of a high applied field on the value of Ei; and (2) the effect of increased value of Ei on the average value J, of the spin-spin int~ractions. or more information on magnetic phase diagrams see the chapter by Cadogan.
lasticity in crystals, and more recently also in s (Feuerbacher et al., 1997; Caillard et elton, l995), has been interpreted in terms of the non-conserva~ive motion of dislocations,
1
Figure If Magnetic phase diagram for bulk amorphous Pd,,Ni,, ,Fe,, sPzoalloy. Filled circles and triangles denote the transition from s u p ~ r p a r a m a ~ n e t i ~(SPM) ~ to ferromagnetism (FM), and from ferroma~netism to spin glass (SG), respectively, reproduced with permission of the American Institute of Physics
Attempts to explain plasticity in glassy metals in terms of dislocation glide processes have not been fruitful. Plasticity in glassy metals i s better conceptualized in terms of localized shear processes, each involving the motion of a small ~ u ~ b ofe ratoms. In this picture, macroscopic glide results from the cooperative activation of localized glide processes on planes of m a ~ i m uresolved ~ shear stress. Two questions then arise: what are the localized shear processes and how do they get organized? To elucidate the nature of the a t o ~ i s t i cshear defects, researchers have si~ulatedthe d e f o ~ a t i o n of glasses at high stresses using computer (Yitek et QE., 1982) and bubble raft (Argon and Muo, 1979) models. None of these stu~iesreveal~dthe o~erationof longlived, extended, defects akin to dislocations in crystalline materials. The defects responsible for shear in glassy metals a ~ p e a to r be much smaller. ~paepenand Turnbull (1974) proposed that the ~ i c r o s c o ~defects ic are small unidentified groups of atoms that are able to shear in the presence of local excess ‘free volume’. If
694
Processes and P ~ e n o ~ ~ ~ a
excess free volume is necessary for atomic shear, then the yield stress should depend not only on the critical resolved shear stress, but also on the hydrostatic component of the applied stress, or on the stress component normal to the slip plane, as proposed by Donovan (1989). The free-volume mechanism also implies that deformed metallic glasses should contain excess free volume. Experiments performed to verify these predictions have yielded contradictory results (Masumoto and Maddin, 1971; Davis and Kavesh, 1975; Donovan and Stobbs, 1981; Deng and Lu, 1983). Argon (1979) proposed that the deformation defects are small regions cont~iningexcess free volume, which undergo collective, thermally activated, atomic rearrangements in response to the applied shear stress. At temperatures T > 0.68 Tg, these regions are assumed spherical, whereas for 0.55 Tg T < 0.68 Tg,the shear processes are assumed to occur inside disk-shaped voluine elements. This last process resembles the incu~ationof a dislocation loop that does not expand. Recent molecular dynamics simulations (Donati et al., 1998; Qligschleger and Schober, 1999) of undercooled liquids suggest the existence of string-like collective thermal excitations. The average string length increases with decreasing temperature and the string~length distribution is nearly exponential. ~ l t h o u g hthe displac~mentof any given atom along a string is small, the total displacement, integrated over the atoms in the string, can approach the interatomic distance. The interstitialcy model of melting (Granato, 1992) predicts similar string-like ‘defects’ exist in a11 liquids, and thus also in amorphous alloys. These ~ua~i-linear e~citationsare assumed responsible for the excess low-temperature specific heat (Boson peak) measured in metallic and oxide glasses (Granato, 1996). It is not known whether string-like defects exist in bulk metallic glasses and what their role is in plasticity. What is attractive about these defects is that their activation may require changes in short-range order, but little or no excess free volume. Because plasticity in metallic glasses cannot be iiimged in an electron mkroscope, indirect ed to test the presence of free volume and/or string-~ikedefects, and their relation to plasticity, may provide clues as to the microscopic nature of glide processes in metallic glasses. The ~ a c r o s c o ~ ~i c~ a r a ~ t e r i s tof i c splastic flow in metallic glasses are now well known. At temperatures approa~hingTg,and for a slow imposed deformation rate, each element of the material contributes to the strain. This ~eformationmode is denoted as ‘homogeneous’ deformation. At lower temperatures, or for high imposed deformation rates, the deformation
becomes localized on narrow ‘shear bands’ which form approximately 011 planes inclined at orientations of maximum resolved shear stress. This localized d e f o ~ a t i o nis often referred to as ~inhQmogeneous’ deformation. For any given metallic glass system, the transition from homogeneous to inhomogeneous deformation depends on the type of test being performed, test temperature, imposed deformation rate, and the microstructural state of the metallic glass (excess free volume retained during quenching, C
4.I . I r ~ ~ o ~ o ~Flow e ~ e o ~ s At room temperature, most metallic glasses deform in the inhomogeneous regime. Under a constant applied strain rate, the material yields when the applied stress approaches the theoretical limit of the material (i.e. critical shear stresses equal to or larger than about 1/50 of the shear modulus). At this temperature, the plastic strain-to-fracture in either compression or tension rarely exceeds 1 %. Inlioniogeiieous flow is confined to a finite number of shear bands, the rest of the specimen deforming only elastically. From a microscopic viewpoint, a localized shear band is assumed to initiate by the simultaneous and cooperative activation of a large number of microscopic defects, each representing the shearing of small groups of atoms. The localized shear process is highly dynamic, implying that near the shear band the local stresses and strains do not obey Hooke’s law. Adiabatic heating and cavitation seem to participate during the later stages of shear-band f o ~ a t i o n~ r i g et~ al., t 2000). Following localized shear, the fracture surface often has a veined appearance, suggesting that the material within the shear bands deforms similarly to a low-viscosity liquid (Leamy et al., 1972). From a macroscopic viewpoint, researchers have been interested in predicting a yield criterion similar to those successfully advanced for crystalline materials. Yield surfaces may also provide some insight into the i~icromechanisms of plastic deformation in bulk metallic glasses. Two yield surfaces proposed to explain the deformation of metallic glasses are a pressure-modified von Mises yield criterion (Bowden and Jukes, 1972), (01
- 0 2 1 2 + ( 0 2 - 0 3 1 2 + (03- 6 1 ) 2 = 6{k,
- P(0l 3- 6 2 + .3)/312
(3)
and the Mohr-Coulomb yield criterion, Z,
= k, -,.a
(4)
Bulk A ~ ~ o ~ Alloys ~ ~ o u s In equation (9, oI, 02, and o3 are the principal stresses; k, and fi are constants. In equation (4), T~ is the shear stress on the slip plane at yielding, CT, is the stress component in the direction normal to the slip plane, and a is a constant. The von Mises criterion predicts that plastic flow will commeiice when the shear strain energy reaches a critical value. The MohrCoulomb criterion predicts that plastic flow starts when the shear stress on the glide plane reaches a critical value, and that this value depends on the stress component normal to the slip plane. Kiniura and Masumoto (1983) studied the plastic d e f o ~ a t i o nof bulk Pd77.5Cu~Sil~,5 glasses in tension, com~ression,and torsion, and concluded that amorphous alloys obey the von Mises yield criterion with no pressure effect (i.e. p=0). Donovan (1989) tested bulk P d 4 ~ ~glass i 4 specimens ~ ~ ~ ~in uniaxial compression, plane strain compression, plane strain tension, and pure shear. By comparing the yield stresses determined through these various tests with the predictions of e~uations(3) and (4), he concluded that plasticity in bulk metallic glasses could not be explained by the von Mises criterion, even with the addition of the pres~ure term. All his results could be explained, however, by the Mohr-Coulomb equation with ko=0.795rfi:0.025GPa and a=0.113&0.03. The recent development of cm-diameter bulk metallic glassy rods has enabled testing of larger samples in ruck et al., (1994) performed room-temperature compression, tension, and torsion tests on bulk il;r41.25Ti,3,7,Ni,,Cu~2.5Be22.5 glasses. In agreement with the earlier observations of Kimura and Masumoto, Bruck et al. concluded that their bulk glass behaved as an elastic-perfectly plastic solid, best described by the von Mises yield criterion without the pressure term (i.e. fJ = 0). The disagreeme~twith the earlier multi-testing results of Donovan has not been explained satisfactorily and, clearly, further investigation is needed to establish a yield criterion for bulk metallic glasses. Some of the contradictory results reported for the inhomogeneous deformation of bulk metallic glasses may be due to the dynamic nature of the deformation. Under these ~onditions,the actual stress at the yielding t measure. In all previous work it surface is d i ~ c u l to has been assumed that the applied stress at the site of i n ~ o ~ o ~ e n yielding ~ o u s equals the value measured at the initiation of the deformation, which is deduced from the applied load and sample geometry. However, during a fast yielding process the elastic stress along the test specimen is not constant and the stress at the shear site could be significantly different from the applied stress. For example, in a long uniform sample
695
being tested in tension, the axial stress near a localized yield event is given by (Molsky, 1963; Schwarz and Mitchell, 1974; Schwarz and Funk, 1985).
( 0applied ) stress before the initiation of where ~ ~ ~is the the yield process, Z is the acoustic impedance of the material, and (au/&) is the particle velocity. Z = p vs, where p is the density and v, is the lon of sound. The axial stress at the shear site decreases in proportion to the i ~ s r u ~ ~ ~ ~value e o zof ~ sthe motion in the adjacent material. This provides an intrinsic negative feedback mechanism that re~ulatesthe rate of deformation. Equilibrium with. the external world (via the deformation machine) is only established following the multiple reflection of com~ressional and tensile stress waves generated by the abrupt yield process. The time for the est ab~i sh~ent of e~uilibrium may exceed the duration of the localized yield phenomenon. For other ~eformationtests, equation ( 5 ) takes a more complicated form. The point to be made here is that because localized yielding is a dynamic phenomenon, the stress at the shear planes is not constant. The temporal variation of the applied stress may determine the m o r p h o l o ~of ~ the shear defects and control the deformation rate. Thus, measurements of the inhomogeneous yield stress performed in different machines (tensile, co~pression, creep, bending, etc.) may give different results.
4,I .2 H o ~ ~ o ~ e n elow ou~
At temperatures approaching Tg,and for low imposed rates of deformation, bulk metallic homogeneously by a viscous-like proc scopic plastic strain results from the sum of the strain contributions, y, from a large numbe t ransfo~at i onor 'flow defects' of volume tion state theory applied to the flow process (Spaepen, 1977; Taub, 1982) predicts a stress-strain relation of the form
~easurementsof the product (ye by Spaepen and Taub (1983), range fr atomic volumes. Therefore, the shear events can be i~terpretedas either a small number of atoms undergoing an extensive shear tion (yo*l), or a large nuniber of atoms un small shear. c%, is a threshold stress below which no Bow i s possible. Its
696
Processes and Phenomena
existence is attributed to the fact that the local shear process produces atomic disorder, leaving the local atomic structure in a state of higher energy due, for example, to the local destruction of CSRO. No net thermally activated transitions occur if the initial and final states have the same energy. The parameters entering equation (6) depend on the microstructural state of the glass (CSRO, excess ‘free volume’ trapped during sample preparation, etc.). To compare creep tests in metallic glasses, Taub and Spaepen (19’79, 1980) did measurements on isoconfigurational states obtained by prolonged isothermal anneals. These measurements must be done within a short time period, so short that the imposed deformation does not change the internal state of the alloy appreciably. Recent measurements on bulk metallic glasses, discussed below, were conducted to understand the evolution of the deformation in more realistic constant deformation rate and creep tests.
0
10
15
20
Figure 16 Compressive stress-strain curves for bulk glassy ~ d 4 ~ C ualloys ~ ~measured ~ i ~ at ~ room P ~ t~~ m p ~ r a t 225 u r ~“G ~ and 260 “C. Imposed deformation rate = 2 x 10-4 s-’
by the annealing at 260°C. The a n n e a l ~ ncaused ~ a 4% increase in the shear modulus, but only a 0.1% increase in the alloy density. After reaching the peak stress, the metallic Figure 16 shows true-stress vs. true-strain curves for softens, as indicated by the rapid decrease in the bulk ~ d ~ ~ i l O C glass u ~deformed O ~ ~ Oin compression. stress at the imposed constant defor~ation rate. These tests were done on 4 x 4 8mm3 ~ samples Notice that at a plastic strain of about 10Y0,the flow deformed at an initial strain rate of 2 x 10-4 (1,’s) stresses in the as-prepared and annealed samples merge (Jin ef al., 2000b). With increasing temperature, the asymptotically. The slight increase in the engineering yield stress decreases and the strain-to-fracture stress for strains in excess of 10Y0 is probably due to increases, At a test temperat~re of 260°C (about the increase in the sample cross-section.The true stress 40°C below T J , and for the imposed cornpression is difficult to calculate precisely, because at such large rate, the sample is fully plastic. strains it is imposs~bleto avoid non-unifor~deformaAn important characteristic in the stress-strain tion (barreling and/or twisting). The marked increase curves of bulk P d ~ , N i l O ~ uglasses ~ O P ~is~the presence in peak stress caused by annealing suggests that the of a yield point. Jin et al., (2000b) found that the softening following the peak stress (yield point ma~imumstress at the yield point is a function of the behavior) is caused by a strain-induced decrease in initial state of the metallic glass. Figure 16 shows two compression stre~~-straii~ curves for Pd40Ni10Cu30P20 SRO, This niay be a common feature in all metalmetalloid glasses. tested at 260°C. The curve labeled ‘as-quenched’ was Creep tests at various stresses provide additional obtained on a sample cut from a 7-mm diameter, insight into the plastic deformation of bulk metallic water-quenched ingot. Heating the sample to the test glasses. Figure 17 shows true-strain versus time for 60°C took about 10 minutes. For this four const~nt-loadtests per~ormedon bulk P d ~ ~ ~ i ~ , s at the yield point is about 720NIPa. Cu,,P,, glasses at 260°C. All the samples were preThe curve labeled ‘annealed’ is for a sample cut from annealed at the test temperature for 6 hours, as the same ingot, but which had been annealed in argon explained above. The four loads used in these tests at 260°C for six hours, At the end of the annealing are above the plateau flow stress for an applied strain period, the sample was rapidly cooled to preserve the rate of 2 x 10m4(l/s), but below the peak stress for the microstructure developed during the annealing. For annealed sample. Figure 17 shows that creep d ~ ~ o ~ a ple, the stress at the yield point is . Jin et al. (2000b) attribute the tion is preceded by an incub~tionperiod of near zero strain rate, The applied stress aEects both the length of increase in the yield stress to an increase in the initial the incubation period and the strain rate followin chemical s~o~t-range order in the glassy alloy caused
Bulk A ~ o ~ ~ hAlloys o~is
period. Notice that for an applied stress of 9OOM~a, the strain rate reaches 0.5 s-', which does work at the rate of 57W, mostly being dissipated as heat. It is easy to visualize that at a higher applied load, alloy softening due to an increase in power dissipation, combined with a large elastic energy stored in the defo~ing machine, may lead to catastrophic yielding. It also becomes clear that catastrophic yielding can be avoided if one loads the sample (to above -700 MPa, in this case) and waits. ~ i ~ e s u ~ a b lduring y, the incubation period the annealed glass acquires the microscopic flow 'defects' it needs for inacroscopic plasticity. It is also apparent that some of these defects are already present in the as~~uenched glass, which has a lower yield stress than the annealed glass (Figure 16). It is not clear at this time whether the microstructural changes during the in~ubationperiod are associated with an increase in free volume, a destruction of CSRO, or both, During steady-state deformation, the lass deforms uniformly, suggest in^ a uniform distribution of active Bow defects. _
I
697
To demonstrate the ease of homogeneous defomation, a 1.5 x 1.5 x 0.2 cm3 saniple was cut from a glassy Fd4*~i40F20 ingot, mechanically polished, and placed on top of a hardened steel block. The surface of the steel block had small pyramidal indentations previously made with the diamond tip from a hardness tester. The sample and the steel block were heated to 340°C and pressed at 1OMPa for 1 minute. The SE micrograph in Figure 18 shows the smooth flow of the glassy alloy into the indentation hole, forming a pyramid. On cooling the sample to room t e ~ p e r a t u ~ e , it recovered its strength of about 1.7 GPa.
Bulk metallic (Pd,~i,Cu,Fe)**P~* glasses prepared by fluxing and water quenching have been found to be elastically isotropic, and thus their f o ~ r t h - o r d eelastic ~ constant tensor has only two inde~endentmoduli. Table 3 gives the elastic properties of various bulk amorphous alloys measured by resonant ultrasonic spectroscopy (Kuokkala and Schwarz, 1992; Migliori et al., 1993). Only two of these moduli are independent. The table also gives the elastic constants of bulk eneous flow in bulk metallic glasses occurs over amorphous ~d4*Ni40F2* measured by Lambson et al. a wide temperature regime. The results discussed above show that homogeneous flow in bulk P d 4 * N i ~ o ~ u ~ * P(1986) ~ o using the pulse-echo method. The table includes the Debye temperatures, OD, of the glassy alloys, glasses can be achieved at 260"C, which is about 140°C calculated from the values of the elastic constants at below the crystallization te~peratureof this bulk glass. cryogenic temperatures. These 0, values are someWithin this wide temperature range, the glass can be rnolded into complicated forms.
0 Figure 17 True plastic strain rate versus time during compressive creep tests on bulk lassy Pd,,Cu,,Ni,,P,, blloys, measured at 260 *C
Figure 18 SEM micrograph show in^ the form~tionof a micro-pyramid on the surface of a bulk amorphous Pd,,Ni,oP,, specimen after pressing it against a hardened steel block marked with micro-indents. The bar at the lowerright corner denotes 10pm
698
Processes and Phenomena
able 3 Elastic moduli, density, and Debye temperature of bulk metallic glasses prepared by flux melting and water quenching. The moduli were measured by the resonant ultrasonic technique (Kuokkala and Schwarz, 1992). E and B are the Young's and bulk moduli. All moduli are in units of GPa. v is the Poisson ratio. The Debye temperature was deduced from the lowtemperature elastic~constantdata Glassy aHoy
T (K) 4.2 300 300 13 300 13 300 300
P
c,,
(g/cni3) (GPa) 9.455 9.36 9.30 9.46 9.38 9.27 9.20 7.44
246 229 202 223 214 225 216 225
0,
(GPa)
(GPa)
B (GPa)
v
(K)
41.7 36.6 33.2 38.7 35.7 40.1 37.0 59.0
117 103 93 108 100 112 103 156
190 180 158 172 166 171 166 146
0.40 0.405 0.402 0.395 0.400 0.391 0.396 0.322
301
c 4 4
what lower than those deduced from the low-temperature specific heats, as is to be expected.
-
291 -
285 -
-
Ref. Lambson et al. (1986)
Me et al. (1998) He et al. (1998) Vuorinen et ul. (2000) Vuorinen et al. (2000) Vuorinen et al. (2000) Vuorinen et al. (2000) Vuorinen et al. (2000)
reduced thickness of the foils causes a decrease in the core packing density since multitudinous air gaps are left between the layers of the large number of thin foils needed to build up the core. A lower core density requires larger copper coils, and this decreases significantly the transfoimer efficiency. Therefore, there is ~m o r p ho u s ferromagnetic alloys are being investino doubt that the advent of thicker, glassy ferromaggated for potential use in electrical transformers and netic foils would facilitate the i~plen~entation of amorphous-core transformers. motors. Currently, about 1% of the generated electrical energy is lost as heat dissipated by distribuSeveral bulk, iron-based, ferromagnetic glasses have tion t r a i is f o ~er sbefore it is delivered to the end user. been recently developed, as mentioned in the Introduction. Shen and Schwarz (1999) used a Electrical motors dissipate an even larger percentage. fluxing and water-quenching technique to prepare These losses can be substantially decreased (by a factor ~ e - ( ~ o , C r , ~ o , ~ a , S b ) - ~ -bulk B - C ferromagnetic o f about 2/3) by replacing the current devices made with ferroi~agneticcores from crystalline Fe-Si alloys glasses. Because these alloys had to contain a large atom fraction of Fe (in order to have a large saturation by more efficient transformers and motors made from magnetization), the alloy had to belong to the metalamorphous (glassy) materials. Other applications for o ) glasses, where M metalloid class ( ~ ~ o mof~metallic ferromagnetic metallic glasses include switch-mode stands for the metal atom and m for the metalloid or power supplies, magnetic amplifiers, ground-fault nonmetal atom. Starting with the basic composition interrupters, sensors, ~ ~ g n e tshielding, ic and tags for Fe,oB,O, which is known to form a metallic glass at electronic article surveillance (Smith, 1993). cooling rates on the order of 106K/s, the authors Thin ferromagnetic foils, typically 25 to 40 pm thick, added metallic and metalloid elements selected accordhave been c o ~ e r c i a l l yavailable in the US since the 1980s under the t r a d e n a ~ e~ e ~These g glasses ~ ~ are ~ . ing to rules we will discuss in the next section. Iron was substituted by elements such as CO, Cr, MO, Ca, and fabricated by quenching the melt at rates on the order Sb, all of which have very small heats of mixing with of 1OhK/s. Power-line transformers built with these iron in the molten state (-l7 --I, -2, -2 and $. 10kJ/ glasses have higher efficiency than those built using mol, respectively), (De Boer et al., 1988). Part of the crystalline, grain-oriented Fe-3 wt.% Si alloys. Howboron was substituted by phosphorus and carbon. This ever, the com~erciali~ation of transformers based on simple strategy yielded a large number of new, thin-foil metallic glasses has been limited by three main ferromagnetic, bulk amorphous alloys. factors (Raskin and Smith, 1983; DeCristofaro, 1998): The bulk ferromagnetic glasses were prepared by: (a) thinness of gauge, post-anneal brittleness, and stress mechanically alloying the various elements; (b) melting sensitivity. The thin glassy foils are diEcult to handle, the alloyed powders in a fused silica tube; (c) purifying and thus the t r a n s f o ~ e r scannot be built by the the melt with B,O, flux; and (d) water quenching the traditional methods used with conventional 0.3-mm fused silica tube containing the molten alloy. The metal thick crystalline Fe-Si laminae. In addition, the
Bulk Amorphous Alloys
and metalloid contents in these alloys were optimized using the difference Tx-Tg as a gauge for the glass formability. Figure 19 shows two 4 m m diameter rods of amorphous F e ~ 5 , ~ ~ r 4 M 2B5,sC5. 0 4 ~ a ~ Figure P~ 20 shows the thermal stability of this alloy, which is representative of this class of bulk amorphous alloys. The crystallization temperature, T,, exceeds 480°C, which is suffici~ntlyhigh for many consumer applications. The presence of well~separatedglass-transition and crystallization temperatures, AT,: 60°C, indicates that the undercooled liquid is highly stable. This large AT is of practical importance, because on heating the glass above Ts, its viscosity decreases very quickly. Since crystallization does not start until the temperature approaches T,, heating the glass to within the TxTg range enables shaping the glass under a small applied pressure (see also section 4.3). Thus, many fer~omagneticpieces, similar to those shown in Figure 19, could be hot pressed or extruded to f o m the large ferromagnetic laminae needed in the manufacture of transformers and motors. The crystallization enthalpy (proportional to the area defined by the exothemic peaks in the DSC trace) ranges from about '7.0 to 7.8kJ/mole atom. These values are close to the value 7.2 kJ/mole atom reported for the classic, rapidly quenched FeRoB,, glass (Cunat et al., 1983), which can only be prepared in thin foil form. The static magnetization curve for as-cast bulk amorphous F e ~ 5 . ~ C r 4 ~ 0 4 ~5.sC5 a 4 Pshows ~ , extremely low hysteresis. In the as-prepared state the coercivity is 0.04Oe. Upon annealing this alloy for 4 hours at 4OO0C, the coercivity decreases to about 0.0075Oe~as
699
shown in Figure 21. A field of only 0.3 Oe is sufficient to approach the saturation induction. The saturation induction measured with a superconducting quantum i ~ t ~ r f e r e ndevice ~ e at an applied field of 50kOe, is 94.4emu/g (0.9 tesla). The low hysteresis suggests this bulk ferromagnetic alloy could be used in the nianufacture of sensors, high frequency transducers, magnetic shielding, etc. This alloy, however, may not be useful for building the cores of 60Hz transformers and motors because these devices require significantly higher saturation induction.
etallie Glasses
As stated earlier, the formation of a metallic glass by the iindercooling of melt requires bypassing crystallization within the temperature regime T; to Tg.Early experiments showed that glass formation was easiest near eutectic compositions, where the liquidus temperature has a minimum, At this composition, the time the undercooled melt is exposed to the labile regime T , T g is lowest (in general, Tghas a weak composition dependence). Experience has shown that under conditions of v e r ~ cooling, ~ ~ s the ~ main co~petitionto glass formation arises from composition-invariant singlephase crystallization, since crystallizatioii involving solute partitioning is much easier to bypass. This realization led to the so-called To criterion for glass formation. To is the temperature at which the Gibbs free energies for the liquid and solid phases of the same composition are equal. In essence, this criterion states that for those compositions for which To Tg,as the 10
0
-50' 200 Figure 19 Ferromagnetic rods of amorphous Fe,, ,Cr,Mo,Ga,P,,IT, ,C5,reproduced with permission of the A ~ e r i Institute ~ ~ n of Physics
Figure
*
' 300
20 DSC
*
'
400
trace
Fe,,.,Cr,rWo,Ga,P,,B,,,C,alloy
'
.
I
500
for
bulk
I
'
~ 0 0
I
J
700
ferroma~netic
700
Processes and ~ ~ e ~ o ~ ~ n u
B,O, flux. In the Zr-rich alloy, the he~eronucleantsare dissolved in the melt by what could be termed ‘selffluxing” The removal or neutralizat~onof the heteronu~~eants is a necessary but not a sufficient condition for bulk glass formation. An addi~ioiialrequirement, as discussed by Turnbull (1969), is a small T,-Tg labile regime or, alternatively, a large reduced glass-transition temperature, T,/T’. This requires ~tabilizingthe undercooled liquid with res ect to the ~ o t ~ n t i a l l y competing crystalline phases which can be accomplished either by stabilizing the liquid, destabilizing the crystal, or by both effects acting sim~ltaneously. -1 .o -0. 0.0 A large number of solutes helps stabilize the liquid with respect to an ordered compound through the entropy of mixing. The liquid can be further stabilized netic induction B as a function of the internal field H for as-cast bulk amor~housFe,, ~ C r 4 ~ 0 4 &, ~ a 4 ~ rough , ~ ~ ~c~emicaland topological s~ort ~range orderalloy. The cyclic hysteresis curve was measured on a small ing. Short-range ordering in metallic liquids is based toroid wouiid with primary and secondary coils of 35 and 100 on the packing of tetrahedra since this results in a more turns, respectively. The inset shows the magnetic hysteresis efficient volume filling, In singie~com~onent liquids, near the origin. the regular packing of tetrahedra is limited in size by frustration, the geometric inability to form large tem~eratureof the molten alloy is lowered, the alloy tetrahedral aggregates unless some of the interatomi~ will become trapped in the glassy spate before it has a bonds are severely stretched or compressed (Nelson chance to crystdlize. The To criterion does not and Spaepen, 1989). Clearly, geonietrical frustration is consider kinetics since it assumes that partitionless easier to accommodate if the melt contains atoms of crystallization occurs instantaneously, once it is various sizes. t h e ~ i o ~ y n a m i cfavored. ~ ~ ~ l y Explicit consideration of A low TI value (deep eutectic) is usually obtained in the kinetics of nucleation and crystal growth leads to a binary alloys whose elements have a lar lowering of the To curves, and to a widenin beat of mixing and vastly different atomic sizes (atomic composition range for easy glass formation radii differing by more than 15%). Strong atomic agreement with experiment (Nash and Schwarz, 1988). interactions favor atomic-size polytetrahedral cluster The To criterion loses some of its usefulness at the f o r ~ a t i o nin the melt, whereas a large atomic size slow cooling rates needed to form bulk metallic glasses, difference reduces the number of crystalline comsince the undercoofed liquid will then have more time pounds which can form in the undercooled melt. to partition. The ability of the melt to partition during Many ~etal-metalioidmelts have deep eutectics at o . general, the metalcompositions close to M s ~ ~ , In solidi~cationalso enhances the importance of heterometalloid interaction is much stronger than the metaliiucleants such as oxides and other crystalline metal or metalloi~-~etalloidinteraction^, and this inclusions that r e ~ a i nsolid in the melt. In the presence causes the metalloid atoms to be surrounded almost of solute ~artitioningthe potency of the inclusions may exclusively by metal atoms. That the eutectic coinposiincrease, because the melt has then more time and tion is clorse to follows then from t ~ ~ ~ l oAg y . degrees of freedom (due to the large number of typical example of this type of melt is Feg3BI7,which elements) to form a crystalline overlayer that is can be quenched into a glass if cooled at a rate in coherent with the inclusion. excess of 1o5Ks- l . Our experience with ~d-(Ni,Cu,Fe)-P(He et al., A meta~-~etalloid melt can be further stabilized by 1998) and (Zr,Ti,Cu,~i)-e bulk glasses (He et al., the addition of a second metallic element, N , to form 1996c), suggests that a main requirement €or decreasthe alloy (M,N)som20.Common sense dictates that the ing the critical cooling rate for glass formation is the element N should have the following characteristics: removal of heteronucleants from the melt. In these two (1) negative heat of mixing with the metalloid un; (2) alloy classes the removal of heteronucleants is achieved near-zero heat of mixing with element M ; and (3) by different routes. In the Pd-(Ni,Cu,Fe)-P system, the atomic size different from M . ~haract~ristic (1) ensures heteronuclea~tsare reduced and/or dissolved in the ~~~~~~
70 1 (Pd,Nij~,P2, does not make a noticeable imthat element N will not change the basic polytetrahe~ dral order of the initial ~ ~ melt.~Characteristic ~ ~ (2) 2 0 provement, if any, to the glass-forming properties of the ternary glass. Notice, however, that Fe ensures that the melt will be homogeneous and will has slightly negative heats of mixing with molten have no tendency to phase separate in the undercooled Pd and Ni (- 4 and - 2 kJ/mole respectively (De state. Characteristic (3j should reduce the average Boer et al., 1988)), and that in the solid state the atom-level internal stress created by the geometrical Pd-Fe and Ni-Fe systems are fully miscible only , With these characterfrustrations in the ~ * , ~ 2melt. at high temperatures. istics, the third element ives the melt an additional (b) Zr41,2TiinsG~,2.5Nit0 e22.5: This five-component degree of freedom, which enables a more compact glass has two pairs of atoms, (Zr, Ti), and (Cu, atomic packing, making it even more stable. Certainly, Ni), with near zero heats of mixing. In 1979, this argument can be further extended to alloys with Tanner and Ray demonstrated that binary alloys four and more components. of transition metals and e form metallic glasses The addition of many elements not only stabilizes by rapid quenching. The addition of transition the melt by increasin~its density, but also reduces the metals to form pairs of atoms with near zero number of crystals that may compete with glass heats of mixing further lowers the eutectic formation. Greer (1993) has named this a ‘principle temperature and enables the formation of bulk of confusion’. A. low number of intermetallics also favors the formation of deep eutectics, which form 5: This live-component glass away from the compositions of intermetallics. If there rming ability than the ternary are no crystalline compounds in the vicinity of the melt glass LaSSAl2sNi2,. Whereas the ternary glass can composition, then crystallization will require severe be cast into 3-mm diameter rods (Inoue et al., composition fluctuations. Desrk (1997) has estimated 1990), the ~ve-component glass, obtained by that the addition of one component to the melt lowers partial replacement of the nickel by copper and the probability of creating a nucleant fluctuation by a cobalt, can be cast into 9-1n~1 diameter Factor of 10. amorphous rods (Inoue et al., 1993a). Nickel, A better glass f o ~ a b i l i t yis not obtained by simply copper and cobalt have near zero heats of adding more elements to the melt. A few examples, mixing. given below, support the earlier statement that pairs of atoms of different atomic size, but near zero heats of $): These ferromagmixi~g,favor bulk glass formation the most. netic bulk glasses were designed based on the three general criteria stated above, with the additional constraint of having an Fe content ~Z~ ~ Z ~ s s e sThe : binary Pdlarger than 65 at.%, as needed for a high P and Ni-P systems have eutectics near 16 at.% saturation magnetization. From the 3 1 initial P, with eutectic temperatures of 1053 and 1143K, alloy compositions chosen according to these respectively (Massalski et al., 1996). Forniing criteria, 19 were found to form bulk ferromagP d x ~ Por ~ 6NiX4Pl6glasses requires cooling rates netic glasses (§hen and Schwarz, 1999). on the order of 106K/s; and, at these cooling rates, the glassy products are necessarily thin foils or ribbons. Pd and Ni fulfil1 the three criteria discussed above: they have near zero heat of mixing, they both have largely negative heats of mixing with phosphorus, and their atomic radii differ by about 15%. Combining these two In the last decade, research on bulk amorp~ous alloys to form the ternary Pd40Ni40P2,alloy metallic alloys has addressed the synthesis of new lowers the eutectic temperature and raises single-phase metallic glasses. The number of distinct Tr,=O.6S. Fluxing this melt to remove the cZasses of bulk metallic glasses is small, perhaps a heteronucleants enables the synthesis of glassy dozen or so. Finding new classes of alloys is not simple Pd40Ni40P20 rods with diameters of at least 2.5 cm since we know of no general selection rule. Most bulk (He et al., 1996a). The quaternary glass Pd40Ni,,glass compositions have been found by trial and error, GuloP2, is an even better glass former than the quenching a large number of alloy melts until the best ternary Pd40Ni4,P20glass (Inoue et al., 1996a, window in a multi-component alloy phase space could 1997a). The substitution of Fe for Pd or Ni in be established. Computer modeling of multi-compo-
702
Processes and Phenomena
nent molten alloys that incorporate realistic descriptions of a temperature-dep~ndentshort-range ordering may provide insight into the conditions for glass formation, but such research is in its infancy. Bulk metallic glasses are being used to manufacture golf clubs, both in the US and in Japan. Manufacturers claim that the low modulus of elasticity, characteristic of metallic glasses, enables a longer time of contact between the ball and the club. Furthermore, the low dynamic loss, due to the absence of crystalline defects, is supposed to enhance the energy recovery, making the collision more elastic and thus improving the driving characteristics. One major potential application of bulk metallic glasses is in electromagnetic energy conversion devices. As discussed in Section 5, ferromagnetic bulk metallic glasses have extremely low hysteresis losses and thus have the potential of reducing the power losses of transformers and motors. Industrial implementatioi~of bulk ferromagnetic glasses, however, will require further research. Although the ferromagnetic bulk glasses prepared so far in Japan (Inoue 1997; Inoue et al., 1997b) and in the U (Shen and Schwarz, 1999) grainhave lower hysteresis loss than the crys~a~ine, oriented, Fe-3wt.% Si presently used in power transformers, these glassy alloys also have lower for satu~ationma~netization,Ms. For example, bulk glassy Fe6,Cr,Mo,Ca4P, lBsCSis approximately chwarz, 1999), whereas M , for crystalline Fe-3 wt.% Si is close to 1.6T. The problem arises because, to prepare a bulk glass, one has to alloy the Fe with non-magnetic metallic elements such as Cr, MO, and Ca. The metalloid elements (C, P, B) occupy interstitial positions in the random metal-atom skeleton. Since this does not increase the average Fe-Fe atom distance, the metalloids (in total concentration of about 20 at. %) have a much smaller effect on Ms. To increase Ms, one must reduce the amount o f the nonmagnetic alloying elements. But, preparing a metallic glass with higher Fe content will require faster cooling rates. Necessarily, these glass products will be smaller (e.g. mm-size pellets). However, provided these glasses retain a signi~cantTx-Tg window, Fe-rich amorphous, ferromagnetic, laminae could be prepared by consolidating these pellets. A ~o t herroute to increase Ms without sacrificing the hysteresis is by partially crystallizing a ferromagnetic amorphous precursor. This approach is already being explored in Japan, where two classes of ~ow-hysteresis ferromagnetic alloys have been developed under the tradenanies F I ~ ~ and ~ NANQPERM. E T These new classes of soft ferromagnetic materials are character-
ized by 10-25 nanometer-si~edgrains of a (bcc) a(Fe,X) phase, comprising 70-80% of the total volume, homogeneously dispersed in an amorphous ferromagE are ~ alloys E ~of Fenetic matrix. The F ~ ~ alloys Cu-Nb-B-Si, have a-Fe grain sizes of about 1 5 nni, and are optimized for use at high frequencies. They are desig~ed to replace the current, more expensive, cobalt-based amorphous alloys. The ~ A ~ O P E R ~ alloys contain Fe-Zr-Cu-B-Si, have grain sizes of about 25nm, and are optimized for use at low freq~encies (60 Hz), offering the potential for applications in electrical power distribution transformers. Currently, these nanoc~ystallinealloys are fabricated by partially crystalliziiig a ~ o r p h o u sprecursors, cast as 25 pmthick ribbons (Yoshizawa et al., 1988; Suzuki et al., 1990; Makino et al., 1991, 1995a,b). Research opportunities exist for developing nanocrystalline alloys from bulk metallic glasses. The challenge is to first form a Fe-rich bulk metallic glass containing solutes that favor, during the subsequent anneal of the glass, a substantial nucleation rate and slow crystal growth. Partial crystallization during the undercooling of the melt must be avoided, since this would lead to a heterogeneous crystal-size distribution following the anneal. ~ l t h o u g hbulk metallic glasses have very high yield strengths, between 1.5 to 2CPa, their poor ductility limits their potential for structural applications. When deformed in tension at ambient temperature, all bulk s~ngle-phase glasses fail ~at~strophicallyby shear localization. To increase the ductility and toughness, researchers are investigat~ngtwo-phase ~ n t e ~ e t a l l i c alloys based on crystalline inclusions in an a ~ o r p h o u s metal matrix. The role of the inclusions is to arrest crack propagation. The optiiiium characteristics (type, size, distribution) of the crystalline inclusions are not known, because so little is known about the morphology and dyiiamics of the shear bands causing catastrophic failure. TWOmethods that could be followed to prepare twophase amorp~oLis/crystalline alloys based on btilk amorphous compositions are: (1) solidifying melts containing a dispersion of crystalline, ductile, secondphase particles, and (2) precipitating crystalline inclusions from a super-saturated melt prior to queiiching it. For the first method, the crystalline particles must not dissolve in the molten alloy and should not become heteronucleants during the solidi~cationof the melt towards producing the a ~ o r p h o u smatrix, Schwarz and Jin (2000) prepared a 50vol.% composite of commercial tungsten powder and amorphous is ductile in compression P d 4 ~ N i 4 ~This P ~ ~composite .
703 at 225°C (175°C below the crystallization temperature), whereas single-phase Pd-Ni-P is brittle at that temperature. Johnson et al. (2000) have prepared rods containing parallel W wires embedded in ~r,,.2Til,~,Cu,,,SNi,o-EQe22.5 glass- This composite has enhanced mechanical properties under dynamic axial loading. The second method for developing two-phase crystalline/amorphous composites is by precipitation from the melt. Johnson et al., (2000) have found that tungsten dissolves in Zr41,2Ti13,,Cu,7.5Ni,0Be27 melt when heated to about 1200 "C. Tungsten dendrites precipitate upon cooling the melt to produce a glass. Using directional solidification, it should be possible to control the size, density, and morphology of the dendrites and thus prepare composites with optimal mechanical characteristics for specific applications. The field of metallic glasses started in 1970 by serendipity (I?. Duwez was looking for an extended solid solution, not a metallic glass). During the following 20 years, most research addressed thin amorphous foils prepared by rapid solidification, but applications were limited by the thinness of gauge. A new chapter in interi~etall~c alloys is now developing with the advent of bulk amorphous alloys. Although single-phase amorphous alloys have some unique properties, it is becoming apparent that, in relation to intermetallic compounds, even better properties are two-phase, amoroften obtained in. ~ne"structur~, phouslcrystalline compounds prepared by the partial crystallization of bulk amorphous precursors. This synthesis route has a number of important features (Creer, 1995). Because the starting amorphous phase is chemically homogeneous (with no solidification segregation), the crystallized ~icrostructure is very homogeneous. By adding trace elements that are soluble in the melt, but not in the crystal, one can vary both the rate of crystal n ~ ~ l e a t i oand n the rate of crystal growth. Then, a partial crystallization of the quenched amorphous melt can lead to the formation of nanocrystalline intermetallic compounds having improved properties (strength, ductility, superplasticity, hard or soft magnetic properties). Either in the amorphous form (showing Newtonian viscous flow) or in the nanoc~stallineform (showing superplasticity) there are possibilities for shaping the components. This suggests that controlled devitrification of bulk amorphous alloys should become an important synthesis route for intermetallic compounds, having properties optimized by design.
This work was supported by the US Energy, Office of Basic Energy ~ciences,Division of Materials Science. The author thanks Dr. T. D. Shen for a critical reading of the manuscript.
Alamgir, F M., Jam, H., Miller, A. C., Williains, D. B., and Schwarz, R. B. (1999). Phil. Mag. B, 79, 239. Argon, A. S., and Kuo, H. Y. (1979). Mater. Sci. Eng., 39, 101. Argon, A. S. (1979). Acta Metall., 27, 47. Baricco, M., Ferrari, E., and Battezzati, L. (1996). Mater Res. Soc. Symp. Proc., 398, 81. Bennett, L. H., McMichael, R. D., Tang, H. C., and Watson, R. E. (1994). J . Appl. Phys., 75, 5493. Berkowitz, A. E., Mitchell, J. R., Carey, M. J., Young, A. P, Zhang, S., Spada, F E., Parker, F. T., Hutten, A., and Thornas G. (1992). Phys. Rev. Lett., 22, 3745. Boudreaux, D. S. and Gregor, J. M. (1977). J . AppL Phys,, 150; ibid, 48, 5057. Bowden, P. B. and Jukes, J. A. (1972). J. Mater. Sci.,7, 52. Brenner, A., Couch, D. E., and Williarns, E. K. (1950). J . Res. Bruck, H. A., Christian, T., Rosakis, A. J., and Johnson, W. L. (1994). Scripta Metall. Mater., 30,429. Buckel, W. and Hilsch, R. (1954). Z . Phys., 138, 109. Caillard, D., Vanderschaeve, G., Bresson, E., and Gratias, D. (1999). Mat. Res. Soc. Synp. Proc., 553, 301. Chantrell, R. W. and Wohlfarth, E. P. (1983). J. Mag. Mag. Mater., 40, 1. Chen, D. Y., Patel, S . , and Shaw, D. T. (1994). J. Mug. Mag. Mater., 134, 75. Chen, D. Y., Patel, S., and Shaw, D. T. (1995). J . Mag. Mag. Mater., 146, 175. Chen, H. S. (1974). Acta Metall., 22, 1505. Chen, H. S. (1976). Mater. Sci. Eng., 23, 151. Conde, F., Gornez-Polo, C., and Hernando, A. (1994). J. Conner, R. D., Rosakis, A. J., Johnson, W. L., and Owen, D. M. (1997). Scrlpta Mater., 37, 1373. Cunat, C., Notin, M., Hertz, J., Dubois, J. M., and Le Caer, G, (1983). J , Non-Cryst. Solids, 55, 45. Davis, L. A. and Kavesh, S. (1975). J. Mater. Sci., 10, 453, de Boer, F. R., Boom, R., Mattens, W. C. M., Miederna, A. R., and Niessen, K. (1988). Cohesion in Metals ( ~ o r t h - ~ o l l a n d , Amsterdam, 1988). DeCristofaro, N. (1998). Mater. Res. Soc. Bull., 23, 50. Deng, D. and Lu, B. (1983). Scripta Metall., 17, 515. Desrk, P. J. (1997). Metal. Trans. JIM, 38, 583. Devaud, G. and Turnbull, D. (1985). Appl. Phys. Lett., 844.
Dimitrov, D. V., Murthy, A. S., and Hadji~anayis,G. C. (1995). Mater. Sci. Erg., A 204, 25. Donati, G.,Douglas, J. F., Kob, W., Plimpton, S. J., Poole, P. H., and Glotzer, 5. C. (1998). Phys. Rev. Lett., 80, 2338. Donovan, P. E. (1989). Acra Metall., 37, 445. Donovan, P. E. and Stobbs, W.M. (1981). Acta Metall., 29, 1419. Drehtnan, A. J., Gseer, A. L., and Turnbull, D. (1982). Appl.
H., and Klement, W. Jr. (1960). J. Appl. Phys., 31, 1136. Feuerbacher, M., Metzniacher, C., Wollgarteti, M., Urbait K., Baufeld, B., Bartsch, M., and Messerschmidt, U. (1997). ~ a ~ eSci. r . Ertg., A 223, 103. Follstaedt, D. M. (1985). Nucl. Instrum. Methods, 1, S,, and Ohnuma, S. (1995). Fujiwara, H. (1993). J. Appl. Phys., 73, 5757. Gaskell, P. H. (1979). J . Non-cryst. Solids, 32, 207. Gilbert, C. J., Ritchie, R. O., and Johnson, W, L. (1997). Appl. Phys. Lett., 71, 476. Granato, A. V. (1992). Phys. Rev. Left., 68, 974. Granato, A. V. (1996). Physica B, 2 ~ 9 ~ 2 2270. 0, Greer, A. L, (1993). N a ~ ~ r366, e , 303. Greer, A. L. (1995). Iii I ~ ~ e r ~ C e ~( ~~ l~l ~i ~edited o ~ n by ~ 9.s €3. , Westbrook and R. L. Fleisclier (Wiley, Chichester). Chapter 31. He, Y., Schwarz, R. B,, and Archuleta, J. I. (1996a). Appl. Phys. Lett. 69, 1861. He, Y., Schwarz, R. B., and Mandrus, D. G. (1996b). J. He, Y., Schwarz, R. B., Mandrus, D. G., and Jacobson, L. A. (1996~).J . N u ~ z - C ~S~ V OS~. ~ 205-207, S, 602-606. rz, R. B, (1998). Metall. Mater. Res. Sywp. Proc., 455, 495.
Her~~~cli, D. M. (1991). Annu. Rev. at er. Sci., 21, 23. Herlach, D. M. (1996). Mater. Res. Soc. Syinp. Proc., 398, 15. Hickey, B. J., Howson, M. A., Musa, S . O., and Wiser, N. (1995). Phys. Rev. B, 91, 667. Hol~and-Moritz,D., Herlach, D. M., and Urban, K. (1993). P?ij.ts. Rev. Lett., 71, 1196. Howson, M. A., Musa, S. O., Walker, M.J., Hickey, B. J., Cochrane, R., and Stevens, R. (1994). J. Appl. Phys., '75, 6546. r . J I M , 36, 866. Inoue, A. (1995). ~ ~ t eTram. Inoue, A. (1997). Mater. Sci. Inoue, A. (2000). Acta Mater. Inoue, A., Zhang, T., and Masumoto, T. (1990). Mciter, Inoue, A,, ~akarnura, T., Sugita, T., Masumoto, T. (1993a). Mater Trans. , T., Nishiyama, N. 19938). Mater, Trans,
Inoue, A., Nishiyama, N., and Matsuda, T. (1996a). ~ a ~ e r Tram. JIM, 37, 181. Inoue, A., Zhang, T., Zhang, W., and Takeuchi, A. (1996b). Mater. Trans. JIM, 37, 99. Inoue, A., Nishiyama, N., and Kimura, H. (1997a). Mater. Trans. J I M , 38, 179. Inoue, A., Zhang, T., Ito, T., and Takeuchi, A. (1997b). at er. Trans. J I M , Inoue, A., Zhang, T., Takeuchi, A., and Zhang, W. (1998). M e ~ ~ lMater. l. Traizs. A, 29A, 1179. Inoue, A. and Gook, J, S. (1995). Mater. Trans. JIM, 36, 1180. Iaoue, A. and Zhang, T. (1996). Mater. Trans. JIM, 37, 185. Jin, O., Schwarz, R. B., Alamgir, F. M., and Jain, H. (2000a). Mat. Res. Soc. Symp. Proc., Jin, O., Harms, U., Cady, C. M., and Schwarz, R. €3. (2000b). Unpublished results, Los Alamos National Laboratory. Jink, J., Gonser, U,, and Wagner~ H A . (1987). Z . ~ e t a l ~ k d e78, . , 767. Johnson, W. L. (1999). MRS Bulletin, Johnson, W. L. (2000). Personal communication, California ~nstituteof Te~hn~1ogy. Kawamura, Y., Inoue, A (1994). Mater. Sci. an Kelton, K. F. (1995). In J. H. Westbrook and R. L. Fleischer (1Xliley, New York), Ch. 20. Kim, Y. J., Busch, R., Johnson, W. L., Rulison, A. J., and Rhim, W.K. (1994). Appl. Phys. Left., 65, 2136. Kimura, H., and Masurnoto, T. (1983). In Anzorphous Metallic Alloys, edited by F E. Luborsky (Butterworths, London, 1983), p. 187. Koch, C, G., Cavin, 0.B., McKa J.O. (1983). Appl. Pktys. Lett. Kokorin, V. V., Minikov, A. V., Plzys. Met. Metall., 57, 184. Kolsky, H. (1963). Stress Waves in Solids (Dover, USA). Turnbull, D. (1984). Appl.
. B. (1992). J. Sci.I ~ . s t ~ ~ ~ . , M. (1993). Cryogenics, 33, 868. L a ~ b s oE. ~ ,F., Lambson, W. A., M a c ~ o n a l dJ. , E., Gibbs, M. R. J., Saunders, G. A., and Tusnbull, D, (1986). Phys. Rev. B, 33, 2380. Learny, H.J., Chen, H. S., and Wang, T. T. (1972). Metall. Trans., 3, 699. nd Lewandowski, J. J. (1998). ~ c r i p ~ #
Madurgn, V., Ortega, R. a,, Vergara, J,, Elvira, R., Korenivski, V., and Rao, K, V. (1996). ~ a ~ o ~ ~ r ~ at er., 7, 185. Maggiore, C. J. and Schwarz, R. B. (2000). Un~ublished results, Los Alamos National Laboratory. ., and Masunioto, T. Makino, A., Suzuki, K. A., I (1991). Mnter Tram. J I M ,
Schwarz, R. B. and Koch, C. C. (1986). Appl. Phys. Le% 146. Schwarz, R. B. and Sin. 0. (2000). ~ n p u b ~ i s h eresults, d Los Alamos National Laboratory. Schwarz, R. B. and Mitchell, J, W. (1974). PCzys. Rev. B, 3292. Shao, Y and Spaepen, F. (1996). J . Appl. Phys., 79, 2981. Shen, T. D., He, Y., and Schwarz, R. B. (1999a~,J . Mtrter. Res., 14, 2107 Shen, T. D., Schwarz, R. B., and Thompson, J. D. (1999b). J. Appl. Phys., 85, 4110. Shen, T. D. and Schwarz, R. B. (1999). Appl. Phys. Lett., 7 49. uyama, J., Shingu, P. H., Ishihara, K. N., Uenishi, Huang, B., and Nasu, S. (1990). In Solid State Pow~er Processing, edited by A. H. Clauer and J. J. de Barbadi~lo, (TMS, 1990), pp. 21-34. in Glasses: An E~per~mental Smith, C. €3. (1993). In ~ a p ~ d Sl yQ l ~ ~ Alloys, ~ e d edited by Introduction (Taylor & Francis, London). York, 2993), Ch. 19, H. H. Liebermann (Dekke Nash, P., and Schwarz, R. B. (1988). Acta Metall., 36, 3047. Spaepen, F. (1977). Acla Met NCeX, L. (1949). Cornpt. ~ e ~ pari d ,is^, Spaepen, F. and Taub, A. I. Nelson, D. R. and Spaepen, F. (1989). In Solid State Physics, Alkoys, edited by F. E. Luborsky (~utterworths,London, Turnbull (Academic Press, New 1983), p. 23 1. Spaepen, F. and Turnbull, D. (1974). ~ c r Metall., ~ ~ ~ a 6. M., Sheidon, W. A., Wang, Spaepen, I;. and Turnbull, D. (1976). In Pro~eed~ng.s of 2nd Y., and Swihart, J. C. (1998). Metall. Mater. ~ r a n s .2, 9 4 er er national ConJi.rmce on R ~ p ~ dQumched ~y Metals, 1845. edited by N. J. Grant and B. C. Giessen (MIT Press, Nicholson, D. M. C., Sterne, P. A., Swihart, J. C., Tran, J., Cambridge, MA, 1976), pp. 205-227. and Wang, Y. (2000). mad ell in^ Si~ul.Mater. Sci. Eng., Suzuki. K., Kataoka, N., Inoue, A., Makino. A., and 8, 261. Masumoto, T. (1990). Mater. Trans. JIM, Oehring, M. (1989). Z. ~ e t a l l ~ d e80, . , 1. Tanner, L. and Ray, R. (1979). Acta ~ e ~ ~ l l . , Oehring, M., and Haasen, P. (1986). J. de Phys., Taub, A. I. (1982). Acta Metall., 30, 2117. Oligschleger, C. and Schober, W. R. (1999). Phys. Rev. B, 59, Taub, A. I. and Spaepen, F. (1979). Scvipta ~ ~ e ~ a13, l l .195. , 811. Tdub, A. I. and Spaepen, F. (1980). Peker, A. and Johnson, W. L. (1993). Appl. Phys. Lett,, Turnbull, D. (1952). J. Chern. Phys., 2342. Turnbull, D. (1969). Contemp. Phys. Perepezku, J. H. and Paik, J. S . (1984). J. Non-Cryst. Solids, Turnbull, D. and Cech, R.E. (1950). 61&62, 113. Vitek, V., Ssolovitz, D., and Egami, T. (1982). In Prac. Fourth Polk, D. E. (1972). Acta ~ e ~ a l l20, . , 485. Intl. Con$ Rapic~ly ~ u e n c ~ e~d e ~ a l edited s, by T. Raskin, D. and Smith, C. €3. (1983). In Amorphous Metallic Masumoto and K. Suzuki (Japan Inst. Metals, Seadai, Alloys, edited by F. E. Luborsky (Butterworths, London, 1982). 1983), p. 381. Vuorinen~J., Shen, T. D., and Schwarz, R. Schluckebier, C.and Predel, B. (1983). Z. Metall~de.,74, 569. Unpublished results, Los Alarnos Natio Schwarz, R. B., Hannigan, J. W., Shteinberg, H., and Tiainen, Wilde, G., Gorler, G. P., and Wiltnecker, T. (1989). In Modern De~elapmentsin Powder Meta~l~rgy, Phys. Lett., 68, 2953; ibid. 69, 2995. Proceedings of the 1988 Intl. Powder Metallurgy Conf., Willman, N., Mader, W., Wachtel, E., and Predel, B. (1987). Orlando, Florida, June 1988. (Metal Powder Industries Phwvs.Stat. Sol. ( a ) , 10 Federation, Princeton, NJ, 1989), Vol. Wiser, N. (1996). J. Mag. Mag. Ma~er.,15 Schwarz, R. Bi.and Funk, L. (1985). Acta Metall., 33, 295. Wright, W. J., Nix, W., and Schwarz, R. B. (2000). Materiuls Schwarz, R. B. and He, U. (1997). In Proc. Intl. Symp. on Science and ~ n g i n e e r i ~A, ~ g319-321, 229. Me tastable, Mechan~call~Alloyed and Nanocrys talfinc? Yeh, X. L., Samwer, K., and Johnsoii, W. L. (1983). Appl. ~ a t e r i a l s ,Rome, 1996, edited by D. Fiorani and M. Phys. Lett., 42, 242, Magini (Trans. Tech Public, Switzer~and, 1997). Also Yoshizawa, Y., Oguma, S., and Yamauc~i,K. (1988). J. Appl. published in M a ~ e rSci. ~ Forum, 23 Phys., 64, 6044. Schwasz, R. B. and Johnson, W. L. (1983). Pizys. Rev. Lett., Yu, R. H., Zhang, X. X., Teyada, J., and Zhu, J. (1995). Phys. 51, 415. Rev. B, 52, R6987.
Makino, A., Inoue, A., and Masumoto, A. (1995a). Mater. Trans. JIM, 36, 924. Makino, A., Inoue, A., and Masumoto, A. (1995b). NanoStrucr. Mater., 6, 985. Massalski, T. B., Okamoto, H., Supramanian, P. R., and Kacprzak, L. (1996). Binary Allay Phase D~agrams(ASM International). Masumoto, T., and Maddin, R. (1971). Acta Metal/*,19, 725. McMichael, R. D., Shull, R. D., Swartzendruber, L.J., and Bennett, L. M.(1992). J . Mag. Mag. Mater., 111, 29, McMichae1, R. D., Ritter, J. J., and Shull, R. D. (1993). J. Appl. Phys., 73, 6946. Migliori, A., Sarrao, J. L., Visscher, W M., Be$ T. M., Lei, M., Fisk, Z., and Leisure, R. G, (1993). Physica B, 183, 1. Miller, M. R . , Schwarz, R. B., and He, Y. (1999). Mut. Res.
This Page Intentionally Left Blank
Increasing demands for energy-efficient processes, changes in the processes to utilize different energy resources, and renewed interest in newer processes have led to development of materials that possess superior mechanical properties from the structural standpoint as well as adequate corrosion resistance at elevated temperatures in the hostile environ~ents present in these systems. In this vein, intermetallic materials are a class of alloys th t have been under development for several high-temperature structural applications, details of which are discussed in another chapter in this vol~me(Lipsitt et al., 2001). Sulfidatio~can be defined as a form of hightemperature corrosion suffered by metals and alloys when exposed to sulfL~r-bearingatmospheres such as pure sulfur vapor and gases containing hydrogen sulfide, sulfur dioxide, and sulfur trioxide. The corrosion products consist of metal sulfides of the constituent elements in the base alloy. Su1fidation degradation of structural materials is of concern because of the widespread occurrence of sulfur in all fossil fuels and the presence of sulfur-containing species in many con~bustion atmospheres derived from fassil fuels used in the power generation, c h e ~ i c a l ~ p e t r o c h e and ~ i ~process ~l, industries. Sulfidation of structural alloys is generally an irreversible corrosion process, and to minimize the attack, either the fuel needs to be desulfurized by physical and/or chemical means or the alloys must be inherently
resistant to corrosive attack in sulf~r-bear in^ environments. The complex environments that arise during combustion of coal and coal-based fuels in coal gasification and ~~idized-bed ~ o ~ b ~ s t si yosnt e ~ shave ~ ~ c u s e d attention on the problems of high-temperature sulfidation of structural alloys and coatings. Even though sulfidation has long been r e c o g ~ i z ~as d a major degradation process ( ~ a u f f eand ~ a h m e l ,1952), it did not receive serious attention until the advent of advanced power systems usin coal as a feeclstock ( ~ a t e s a n , 1980; Tiearney Jr. and Natesan, 1982; Mrowec, 1995). The process of sulfidation from the standpoint of the~odynamics,kinetics, and mechanisms has been revi ed in several papers (Natesan, 1985; Strafford and atta, 1993; Datta et al., 1998). Sul~dationis similar to oxidation in many aspects; the reaction phase that forms is d e ~ e ~ i n ebyd the relative thermodynamic stability of that phase with respect to others and the activity of the constituent elements of the alloy. In the case of heat-resistant materials such as Fe-, Ni-, and Co-base alloys, the sulfidation kinetics are generally faster than oxidation of the same element and the ~orphologyof the sulfide scales are as complex as those of oxides. The general absence of basic information concerning rates and mechan~smsof s~lfidationof metals and physicochemical properties of sulfides (e.g. thermochemical data, diffusion data, defect chemistry of sulfides) constitutes a major gap in knowledge at present and greatly inhibits a systematic, ~uantitative
~ n ~ ~ ~C o? ~ ~p oe u Vol. ~n ~~~3,~Principles ~ i c and Practice. Edited by J. H. Westbrook and R. 1;. Fleischer. 02002 John Wiley & Sons, Ltd.
708
Processes and ~ ~ e n o ~ e n u
approach to alloy development for resistance to sulfidation corrosion at elevated temperature. In practice, however, except in oil-refining industry applications, materials are subjected to bi- or multioxidant environments in which 0 is one of the oxidants and the usual approach to prevent sulfidation is use of an alloy that can devclop an adherent, protective oxide scale that significantly resists sulfidation attack. On the basis of morphological information developed on a number of commercial engineering alloys, highly alloyed metallic materials, and model alloys mixed-gas environments, three regimes were defined to describe the oxidation/sulfidation behavior of the alloys at elevated temperatures tesan, 3980, 1986). In Regime 1, below the baseal sulfidation potential and the threshold boundary of oxide formation, high-Cr alloys develop outer scales of Cr sulfide with variable amounts of soluble Fe and Co. The substrates cont substantial porosity and internally sulfidizcd Cr. gime 2 conditions, above the base~metal sulfidation potential and at oxygen ~ o t ~ n t i a higher ls than that for threshold boundary, oxide scale formation is favored. The scale thickness varies somewhat, but the most important differences between alloy types pertain to the subscale structure and the shape of the nietal/oxide interface. Regime 3 conditions, above the base-metal sulfidation potential and at oxygen potentials lower than that for threshold boundary for oxide formation, result in scales that are complex mixtures of the base-metal sulfide, Cr sulfide, Cr~depletedalloy, and a p p r o ~ ~ ~oxide. a t e Accelerated li~u~~-p~ corrosion ase occurs if Ni is present in sufficient supply in the alloy to form a eutectic
Ni/Ni,S, with a liquidus temperature of x 645 "C. Zn Regime 3 , com~etitionbetween oxide and base-metal sulfide nuclei is followed by sulfide overgrowth, which generally has a growth rate that is orders of magnitude higher than that for the oxide. Void f o ~ a t i o nin the subscale occurs throughout the sequence, due to outward catioii diffusion. Cr sulfide forms at the sulfide/alloy interface through reduction of the basenietal sulfide by diffusion of the Cr from the matrix. Finally, the sulfides dissociate at the heavily voided interface, and S that is released into the voids diEuses into the matrix, forming i~ternalCr sulfide. In AI-containing structural alloys, the formation and maintenance of an a-alumina phase on the alloy surface is essential to achieve acaeptably low corrosion rates for the alloys. On the basis of a review of the growth kinetics of Cr,O, and A120, scales, it was concluded that for a~proximatelyparabolic kinetics, the rate constant for A1,03 formation would be lower by two orders of magnitude than for Cr,O, formation (Hindam and Whittle, 1982). In general, sulfidation of Al-containing alloys proceeds at a rate that is greater by orders of magnitude than the rate of oxidation (Zelanki and Simkovitch, 1974).
Because the mode of interaction of heat-resistant materials with O/S mixed-gas environments is the formation of Cr-rich oxide or Cr-rich sulfde, it is pertinent to consider the thermodynamic stability of oxides and sulfides of other alloying elements (of
L
Periodic table showing elements that form oxides and sulfides of greater stability than Cr oxide and Cr sulfide
709
Sulfidat ion Beha vior
interest from the standpoint of structural alloys and intermetallics) relative to those of Cr. Elements that form oxides and sulfides of greater stability than those of Cr are indicated in Figure 1. The alloying elements Mn and Si, which form oxides that are more stable than Cr oxide, have a tendency to migrate from the bulk of the alloy to the scale~substrate interface. However, at high Mn concentrations, formation of MnS is a distinct possibility and would lead to enhanced corrosion rates. In the oxide mode of interaction, Mn preferentially segregates in the scale and causes scale spallation, to enhance corrosion rates. content of up to 4wt.% is beneficial for the ation of protective oxide layers in oxidation studies; however, the mobility of Si (due to its much larger ionic radius) is much lower than those of Fe, Cr, and Mn. As a result, formation of an external silica scale is virtually impossible even in high-Si alloys. Si generally segregates in the oxide/alloy interface, especially in c h r o ~ i a - f o r ~ i nalloys. g Refractory metals such as Ti, Zr, Nb, Hf, V, and Ta can form oxides and sulfides that exhibit greater thermodynamic stability than Cr oxide and Cr sulfide. On the other hand, Cr oxide and sulfide phases are more stable than sulfides aizd oxides of metals such as ure 2 shows a comparison of thermogravimetric test data for several refractory metals/alloys with those for conventional alloys exposed to mixed-gas atmospheres with the relatively high partial pressure of sulfur (pSJof 9.4 x 1W2Pa at 871 "C. Results showed catastrophic sulfidation in Alloy 800 and 310 stainless steel; whereas the refractory metals (except Ta) exhibited low rates of corrosion. MO and TZM developed very thin, adherent sulfide scales, and V developed a subscale of V sulfide.
Table 1 Parabolic rate constants for reaction of several metals/alloys in O/S atiiiospheres at 871 "C Parabolic rate constant (g2/cni*/s)
Metal/allo y
5.0 x 10-7 3.5 x 10-7
Cr sulfidation Alloy 800 310 S S Ta Nb MO
3.4 x 10-7 1.7 x 10-7 1.2 x 10-9 7.0 x 10. 2.7 x 10-'l 1.8 x lO-'I 2.1 x 10-l2
'')
v
TZM Cr oxidation
Table 1 lists the parabolic rate constants for the oxidation/sulfidation reactions for several of these metals/alloys and for Cr. The results indicate that Cr or A1 additions may improve the oxidation resistance of an alloy, but that the same additions will not impart significant sulfidation resistance, especially in high environments at elevated temperat~res. Generally, the sul~dation rate constants for structura~ alloy constituents such as Fe, Cr, Ni, and COare, depending on temperature, at least several orders of ma~nitude higher than the correspo~ding rates in oxidizing environments, Several refractory metals, which often are the constituents of i nt e~et al l i c compounds, exhibit low sulfidation rates and may undergo protective scaling in S-containing atmospheres.
3. I
310 ss
Figure 2 Comparison of t~ermogravi~etric test data for several refractory metals/alloys with those for conventional alloys exposed to O/S mixed-gas atmospheres
Here, the philosophy is to modify the defect structures of selected sulfides through the incorporation of stable dopants by using the classical ~ a g n method. e ~ Indeed, in Cr-Mo alloys, doping with Cr was observed to reduce the sulfidation rate; but the improvement was not significant (Strafford and Hanipton, 1971). The main problem with this method is the difficulty in finding suitable elements with appropriate atomic radii and valences. This method is of theoretical interest but has little practical application for developing alloys with sulfidation resistance; however, it may be more effective in the design of overlay coatings.
710
Processes and F ~ e n o ~ e n a
The practical approach to developing sulfidationresistant alloys and coatings is through formation of protective sulfide barrier-layers. The principle is to achieve selective sulfidatioii of B in an alloy AB where aB
Assessment of probability of external scale andor sub-scale formation via
substrate alloy, and factors determined by the sulfide volume fraction. Several conclusions can be drawn from the data in Tables 2 4 : 1. In an alloy based on Fe, Ni, and C O ,a preferential s u l ~ d a t ~ oofn an alloying addition, if judiciously selected, can be achieved. 2. The high melting points for the sulfides of the semirefractory and refractory metals indicate that their rates of sulfidation may be slow due to the anticipated slow rates of diffusion of the cations (e.g. Ti, Zr, Hf, V, Nb, Ta, and MO) in their respective compounds. 3. The behavior of Cr, however, defies the argument in Conclusion 2 above.
4. Sulfur Resistance of ~ n ~ e r m e t a ~
New structural materials based on nickel-iron and nickel alurninides are being developed for application
properties of sulphide scalehub-scale
Figure 3 Factors influencing development of barrier layers to resist sulfidation
S u ~ d ~ t i oBehavior n Tab~e2 T h e ~ o d y ~ ~ data m i cfor su~fidesof Group III-VI metals at 727 "C
Element
Sulfide
A1
AV4
CO
CO,%
Cr Fe Af
Cr,S, FeS Hf5,
MO
MO§,
Nb Ni Ta Ti V
NbS, Nix% TaS, TiS, VA
w
zr
VrlS,
ZrS
-AG" (kJ/mole S,) 378.0 168.0 197.0 196.4 532.7 215.8 295.9 177.9 292.9 363.7 301.4 198.0 311.6
71 1
0.16
Dissociation partial pressure (Pa) 3.4 x 10-'O 5.9 x 10-4 5.8 x 10-l2 8.0 x 10-' 1.0 x 10-4 1.0 x 10-'O 1.9 x 10-3 2.3 x lO-'O 1.3 x 10-15 6.0 x 10-26 5.0 x 10--3 3.3 x 10-20
at elevated temperatures. The nickel-aluminide, Ni,Al, is an interrnetallic compound with superior strength properties (Pope and Ezz, 1984). Oxidation/sulfidation behavior of Ni3AI has been studied extensively to evaluate the kinetics of oxidation and sulfidation, microstructural characteristics of the corrosion product scales, and sulfur resistance of in-situ-forrned and pre-formed oxide scales (Natesan, 1988).
Figure 4 Tliermogravinzetric test data for sulfidatioti of Ni3Al at 875 "C
Figure 4 shows tlierrno~ravirnetrictest data for the sulfidation of Ni,AI at 875 "C in several H,-H,S gas mixtures. Also shown for comparison are weight change data for oxidation of the alloy in an air environment. H,S concentration in the exposure eiivironment ranged from 0.16 to 4.7 vol.%, The weight-change data indicate that at H,S levels up to
Table 3 Self-diffusion coefficients of cations, DMe, in some metal sulfides and oxides _ _
Sulfide CO,-,S
cr,s, Cu, -,s Fe1-ys Ni,-,S
T ("C)
600 720 1000 650 800 800
DMe
(cm'/s>
1.0 x 10-'3 7.0 x 10^-? 1.0 x 10-? 5.1 x 10-5 3.5 x 10-? 1.4 x 10--%
Oxide
T ("C)
A1203
1000 1000 1000 1000 800 1000
CO,
-yo
Cr203 CU,-,O Fe,-,O Nil -p
DMe
(cm"s)
1.0 x 10-l6 1.9 x 10-9 1.0 x 10-12 1.7 x 10-8 1.3 x 10-* 1.0 x 10-'I
Table 4 Physicochemical properties of sulfides of certain Group 111-VI metals
Sulfide
Pillin~-Bedwort~ratio*
Defect structure
Melting point ("C)
2.6 2.4 2.5 2.5 3.5 2.5 2.4 1.1 3.5
n-type P-tYPe nlp-type P-tYpe
1099 1080 1550 1189 2100-2273 1457 796 999 1999-2099 1799-199 1800 1549
1.9
-
n-type n-type P-tYPe n-type n-type n-type n-type n-type
*This is the ratio of sulfide volume grown to unsu~fidizedmaterial consumed in sulfidation.
Processes and P ~ e ~ ~ ~ ~ n c i
712
Sul~dationkinetics of Ni,Al in H,-H,S atmospheres at 875 "C H,S in H,-W,S mixture (vol.'%) 0.16 0.38 0.75 1.50 4.70
Sulfur partial pressure (Paj
Rate constant (g2/cm4/sj
2.2 x 10-4 1.2 x 10-3 4.8 x 10-3 2.0 x 10-2 2.1 x 10-'
2.1 x 10-13 3.8 x 10-*' 9.6 x 10-'* 2.0 x lO-'l 8.7 x lO-'O
*Sulfur pressure that corresponds to Ni-Ni,S, equilibriu~at 875 "C is 2.75 x 10-, Pa.
0.75 vol.%, the reaction rate followed a parabolic behavior. However, at H,S levels of 1.5vol.% or higher, the weight-change data showed that the alloy exhibited accelerated corrosion. The weight-gain data were used to compute parabolic rate constants for the sulfidation of the alloy at various H2S levels in the gas
phase, and the results are listed in Table 5. At low levels of H2S (0.16 and 0.38 vol.%) in the gas phase, the sulfidation rates were similar to oxidation rates in an air environi~ent,As the H2S levcl increases, the sulfidation rate increases and, at w 1.5 vol. % H,S, liquid Ni-Ni,S, eutectic forms and accelerated corrosion ensues. At an H2S level of 4.7vol.%, the rate constant for sulfidation is 8.7 x lO-Io g2/cm4/s,which is ~4 orders of magnitude larger than the rate constsnnt for the oxidation of the alloy. At higher S levels, the weight-change curves show an accelerated rate for longer times, due primarily to a decrease in reaction surface area produced as the reaction front advances deeper into the alloy. Figure 5 shows the ~ o r p h o ~ oand ~ yX-ray analyses at several locations on the surface scale developed on a Ni,Al-rich specimen exposed to an H,-H,S gas mixture containing 0.38 vol.%O H2S. AI-rich regions, indicated
Figure 5 Morphology and X-ray analysis of the surface scale developing on Ni,Ai specimen exposed for 154h at 875 "C to 0.38 vol.% H,S-H, gas mixture
713 by the white areas on the surface, consisted of discrete particles that had no associated sulfur. Plates or needles that contained Ni and S were observed in isolated locations, indicating susceptibility of the alloy to sulfidation attack even when exposed to a mixture. When the H,S content of sed to 0.75vol.%, an external scale of A1 sulfide developed on the alloy. The scale had a tendency to spall; underneath the spalled area, the alloy was enriched in Ni and the grain boundary regions exhibited an Hf sulfide phase. With a further increase in H,S to 1.5 vol.%, a two-layer scale wits observed after 91 h of exposure. The outermost layer consisted of AI- and Ni-sulfide (as shown in Figure 6), which tended to spall from a number of locations on the surface. The inner exhibited regions of enriched sulfide (see 7) and globules of Ni sulfide. AI was virtually absent in the inner layer. This observation suggests that replenishm~ntof A1 in the outer region (if the outer scale cracks) will not be possible and that the scale that reforms after spallation of the original outer scale will be enriched in Ni sulfide. At the longer exposure time of 265 h in a I-2,-1.5 vol.% H2S gas nnlxture, Ni-Ni,S, eutectic forms, thus indicating that the AI-sulfide scale offered virtually no protection against catastrophic sulfidation attack on the allay.
,
s
Iron aluminide intermetall~csare being developed for use as structural materials a n d / o ~as claddin conventional engiiieering alloys. In addition to strength advantages, these materials exhibit excellent resistance to corrosion in both oxygen- and sulfurcontaining environments at elevated temperatures through the formation of slow-growing, adherent alumina scales. In the iron-aluminum s alloys of interest are of com~os~tions F FeA1. The crystal stru l32, while FeA1 has t temperatures of Fe,AI an and densities are 6.72 and 5.56 g C I ~ - respectively. -~, However, the Young's inodulus values for Fe, FeAI are 140.6 and 2 6 0 . 4 ~ P a respectively, , and the stifTer intermetallic has a tendency to be much more brittle. Iron aluminides are of interest primarily because of their much lower cost than that of nickel aluminides, a lower density than stainless steels (with potentially a better strength-to-wei~htratio), and hi~h-temperatu~e corrosion resistance. However, their limited ductilikies at ambient temperatures and a sharp drop in strength eterrents in their use as above 600 "C have been structural materials ( ey et al., 1991). Tn
F~~~~~6 Morphology and X-ray analysis of the surface scale developed on a Ni,Ai specimen exposed for 91 h at 875 "C to 1.5 vol.% H,S-H, gas mixture
Processes and Phenomena
ure 7 ~ o r p h o l o ~ and y X-ray analysis of inner-scale region, shown in box (a), of specimen depicted in Figure 6
general, other elements are added to these alloys to improve their mechanical and/or Corrosion properties in differing environ~ents. Corrosion resistance is generally imparted to Fe aluminides by in-situ development of an alumina scale on the alloy surface. The slower the growth rate of the oxide, the better the oxidation resistance. Thus, alumina scales (which have inherently slower growth rates) offer s ubs ~ ~ nt i al advantages over other oxide scales, especially in single~oxidante n v i r o n ~ ~ n tThe s . alumina scales also act as barriers to the transport of corrosion-accelerating reactants such as S, and they retard the scaling kinetics of the underlying substrate alloys when they are exposed to multioxidant environments. Sulfidation experiments have been conducted on several heats of Fe aluininide at temperatures between 400 and 1000°C (Natesan, 1993). Figure 8 shows therinogravimetric data for ternary Fe aluminide tested in a 1.35vo1.0/0 H,S-H, gas mixture at 650, 875, and 1000 "C. Also shown in the figure are data for 310 stainless steel oxidized in air at 1000°C and sulfidized in the 1.35voi.% H,S-H, gas mixture at 875°C. Figure 9 shows SEM photomicrographs of surfkes of P;e,Al and Type 310 stainless steel specimens after sulfidation. The morphologies of surface sulfides on the Fe,Al seem different at different temperatures, probably because the the~odynaxnic activity of S in the exposure environnient is lowest at 1000°C and highest at 650°C (H,S concentration in
the gas was kept constant). Scales on the Fe3Al consisted of (Fe,AI) sulfides and Fe sulfides, the relative proportion of the former to the latter decreasing with decreased temperature. The scale on the Type 3 10 stainless steel sample was predominantly (Fe,Cr) sulfides with some nodules of Fe sulfide. Co~parativestudies have been conducted on the sulfidation resistance of Fe,Al and several chromiaand alumiiia-for~ingalloys in O/S mixed-gas environments (Natesan, 1993; Natesan and Johnson, 1995). Thermogravi~~tric studies on oxidation of Fe-base 310 SS,875OC,sulfidation -40
E
0
re
40
80 120 Exposure time (h)
160
8 ~ ~ e ~ o g r ~ vtest i ~ data e t ~forc ~r-containing Fe,A1 exposed to 1.35vol.% H2S-H2gas mixture at 650, 871, and 1000 "C. Also shown are data for oxidation (at 1000"C) and sulfidation (at 875 "G) of Type 310 stainless steel
S u ~ d a t i oBehavior ~
715
0.
E
v
-
0.1
-
RV 8 4 i 3
.---+-- GE 1541 Fe-25Cr-20N
--+--
..--a--
Fe-12AI Fs-25Cr-6AI
-0.0 0
2 0
4 0
60
80
100
Exposure Time (h) Figure 15 Weight-change data for several commercial alloys, Fe-12A1, and Fe,Al tested in O/S environment with poz =4.1 x 10-13 and-ps, =9.4 x 10'-,Pa at 875 "C
Figure 9 SEM photomicrographs of surfaces of Fe aluniinide and Type 3 10 stainless steel after sulfidatioii exposure to a hydrogen-1.35 vol.0/0 H,S gas mixture. Fe,Al at (a) 650"C, (b) 875"C, (c) 1000°C; 310 stainless at (d) 875°C.
alloys with differing AI concentrations and Fe,Al alloys showed that a minimum A1 level of 12wt.70 is needed to develop a continuous alumina scale that is resistant to sulfur attack. A detailed comparison has been made of the corrosion performance of aluminaand chromia-for~ingalloys exposed to O/S mixed-gas environments. eight-change data, summari~edin Figure 10, show accelerated corrosion for chromia-forming alloys such as high-purity Fe-25Cr-20Ni and 310 stainless steel. Data for the Fe-2SCr-20Ni alloy show that a Cr content at the high level of 2Swt.Y~does not improve sulfidation resistance of the alloy, which develops a scale (consisting of a mixture of Fe and Cr sulfide) at a very high growth rate. In fact, the presence of Ni leads to the formation of the low-melting Ni-Ni,S2 eutectic, if the test is run for a longer time. Alloys containing Cr and AI (RV 8413, CE 1541, and Fe-25Cr-6A1), which form alumina scales in single-oxidant environments, also exhibit significant corrosion in O/S environments. An addition of 6 wt.% A1 to the Fe-25Cr alloy seems to reduce the corrosion rate somewhat; but the external scale consists of Fe sulfide, which is not expected to offer protection against breakaway corrosion. The alloys GE 1541 and RTJ 8413 are Fe-based alloys with moderate Cr content and z 5-6 wt.% Al. Corrosion performance of these alloys i s not adequate
for service in H~§-containing environments. The binary Fe-12 wt.%Al alloy and Fe,AI (with 13.9 wt.% Al) exhibit superior corrosion resistance in Q/S mixed-gas atmospheres. The scales on these alloys were alumina and contained little S. Figure 11 shows a plot of typical depth profiles obtained by Auger analysis for several of the ~ e r t i n e ~elements t in the scale after sputtering the surfaces for various times. The peak-to-peak height as a function of depth for Fe,Al overlay specimens after exposure in an air/S02 e ~ v i r o n m ~ nindicates t an alumina scale of 0.4 pm thickness. Evaluation of an iron aluminide overlay was conducted on a Type 316 stainless steel substrate
2 24~ 0
L
0
0.2 0.4 0.6 Distance from surface (urn)
0.8
Figure 11 Auger peak-to-peak height data for several elements as a function of depth for the scale developed on Fe,Al overlay on Type 316 stainless steel after exposure in air/ SO, environment at 650 "C
716
Processes and Phenomena
whose surface was coated with a FeAl welding rod (Natesan, 1997). Specimens with the overlay, along with other s t ~ c t u r aalloys, ~ were exposed to O/S mixed-gas environments for up to T28h and periodically retrieved to measure weight changes at interInediate exposure times. eight-change data were obtained for several com~ercialhigh-Cr alloys exposed together with the FeAl-overlay specimens in SjS mixed-gas environments, with and without HCl. Figure 12 shows corrasion-loss data based on parabolic kinetics for several of these alloys. Corrosion performance of the FeAl overlay is comparable to or better than those of most other materials tested.
25 20
15
0
10-14
0.11
*
ICYt4
1.6
4
E 0
0
10
20
30 40 50 xposure time (h)
60
70
Figure 13 Weight-change data for sulfidation of Ti-54AI exposed to different ps, at 900 "C
Information on the high-temperatur~sul~dationbehavior of Ti-A1 interrnetallics is not extensive. Same work has been reported on the response of Ti-A1 based materials, such as Ti-54Al and Ti-48Al-2Nb~2~n, after exposure at 900 "C to mixed-gas environments with a fixed poz (10-l4 Pa) and with various psZ values (0.0016, 0.11, and 1.6Pa) (Datta et al,, 1998). The results from this study indicate parabolic kinetics at all ps2 values for suEdation of the alloys. After some un~ertaintiesat the early stages of sulfidation, grolonged exposure (168 h at 900 "6)indicates that the reatest resistance to sulfidation occurs at the highest ps2 ( F i ~ ~ r 13 e sand 14). In initial stages of e x ~ o s u r ~the , high a ~ n i t yof oxygen for Ti and A1 leads to development of an outer layer of TiO,, beneath which an AI@, layer forms. 0.4
0.0 -0.4
~ o ~ p a r i s oofn corrosion loss data for FeAl overlay on Type 316 stainless steel and several ki&-Cr alloys after exposure in gas mixtures containing H2S with and without HC1
Sulfw diffuses through the TiO, and Al, reaches the substrate/scale interface where the poz is low enough to promote the f o r ~ a t i o nof TiS, and Al2S3(and NbS, in the case o f Ti-48Al-2Nb-2Mn). Clearly, the presence of high psZ in the bulk environment would provide a higher driving force for sulfur migration, that would support a higher sulfur flux and favar formation of sulfides of alloy constitue~ts. Accordingly, a mixed layer of sulfides and oxides develops in the high-ps, atmosphere, thicker than in the 1 0 w - p ~e~vironment, ~ thereby providing higher resistance to ~ulfidation. In summary, the superior sulfidation resistance of Ti-54Al and Ti-48AI-2Nb-2Mn can be ascribed to several aspects of the scaling processes. First, the development of an inner layer of sulfides (TiS,, A12S3, NbS,) provides an effective barrier to cation transport. The presence of refi*actorymetal sulfide NbS, is known to cause significant i ~ ~ r o v e m e nint the sulfidation resistance of the inteirnetallic alloy (Du et al., 1994). ~ u r t ~ e r ~ Qsulfidation re, resistance of Ti-54A1 stems from the formation of a TiAI, layer at the scale/ substrate interface by the diffusion of Ti after development of TiO, and TiS,. Ti released from the dissociation of TiS, is likely to promote a thick TiA1, layer, thus i ~ ~ a rfurther t ~ n resistance ~ to sulfidation. The o x i ~ a t ~ o n ~ s u l ~ dkhavior a t i o ~ of another intermetallic alloy, Ti-SQ.6Al-1.~Mn-2Mo,with duplex. and lamellar microstructure was investigated in H , / ~ , S / H , ~ (ps2= 0.1 Fa, po2= 10-12Pa) at 750 "C and @s, = 0.1 Pa, po2= 10-'4Pa) at 950°C (Du et al., 1994). The weight g a ~ n / t ~ edata fo~owed parabolic kinetics (Kp= 1O-I2g2/cm-4/s) at 750 "C and cubic kinetics at 900°C. The increase in exposure temperature had no
6
5 4
3 "c
O E
2
c3,
1
0
0
40
Exposure time (h) Sulfidation kinetics for Ti-48A1-2Nb-Mn at (a) 750 "C and (b) 900 "C
sign~~canteffect on corrosion behavior, Scaling pattern development resembled that of Ti-Al-Nb-Mn alloy, with an outermost TiO,/inner A1,03 layer and base-element sulfides forming between the oxide layers and the substrate, The enhanced corrosion resistance observed at 900°C compared to that at 750°C follows from the differences in the defect structures. Both 0 vacancies and interstitial Ti ions are important point defects (Kofstad, 1988), with interstitial Ti ions predominating at low PO, and high temperature, whereas 0 vacancies predoi~inateat high pox and low temperature, At low po2,many interstitial Ti ions are expected in TiO, and hence an increase in poz would result in decreased educing the interstitial Ti ions would decrease the transport of Ti through TiO,. Temperature change between 750 and 900°C may not signi~cantlyalter the diffusion of Ti and 0 ions in these alloys. The increased 0 pressure will iiicrease the 0 vacancies in TiO, and thereby increase the inward diffusion of 0 and promote Al2O3,i~partinga slow corrosion rate for both materials.
Higli-temperature structural materials are critically needed for i ~ ~ r o v i nthe g thermal efficiency and reliability of energy-conversion systems and advanced engine systems. The currently available alloys, such as Ni-base single-crystal superalloys, are limited to tem~e~atures of rn 1100 "C. Superalloys derive their intrinsic strength by reinforcement with gamma prime (Ni,Al) precipi~ates,but these tend to coarsen and
ultimately dissolve as the temperature increases beyond 1100°C. Alu~inidealloys based on NiAl, which are currently under development, have the potential for use up to 1200°C. ~owever,many a~plications require temperature ca~abilitie$ that exceed 1400°C. The melting temperature (qn)of a material for structural application^ at be >2OOO"C; so that, at most, 0.75 during service and appreciable high-t~mperat~~re strength is maintained, Of the candidate systems, molybdenum silicides (MoSi, and MO$$,) are particularly attractive because of their high melting points, good mechanical strength, high thermal and electrical ~onductivities,and ~romisingoxidation resistance at elevated temperature, Even though several studies have been conducted on the o on performance and ~echanicalproperties of 2, no information is currently available on its §ul~dationbshavior, Natesan has investigated the sulfidation performance of the M o s ~ i 3 - t y ~intermetallic~ e in a 1.5vol.% H,S-H2 gas mixture and compared the sulfidation resistance with their oxidation performance in air (Natesan and Deevi, 2000; ~ a t e s a n ?2000). Figure 15 shows the t h e ~ o g r a v i ~ e t r itest c data obtained at several temperatures for oxidation and Mo,Si,, The data showed sulfidation of a B-cont~inin~ a protec~ivescaling of the alloy oxidizing conditions, not because o but because the volatilization rate negligible. At te~peraturesof 800 and 1200"C, the curves show a sharp drop in s~ecimenweight for rn 2 h, after which a plateau is reached and the weight changes little during 50 h of additional exposure, indicatin
Processes and P ~ e n o ~ e n a
718 0 . 0 5 , , , , , , , , , , , , , , , , ,
,
,
)
I
, , w -
0
E
0.005
E
I
%
E"
0
-d
a,
p -0,05
tl, c
(0 c 0
E
5 E
-0.1
-0.005 "0.01
0)
a ..
-0.015 -0.15L
0
50 100 150 ~ x ~ o s utime r e (h)
200
0
50
100 450 Exposure time (h)
200
Figure 15 Weight-change data during (a) oxidation in air and (b) sulfidation in 1.5vol.% H,S-H, gas mixture for B-containing Mo,Si, after exposure at several temperatures
protective scale. The morphology of the scale after oxidation at 800 "C consisted of a light-colored Mooz phase and a dark gray Si-rich oxide. After oxidation at 1200 "C, the surface consisted predominantly of Si-rich oxide and almost pure MO particles. The surface layer showed signi~cantcracking and peeling and seemed to rather than be highly plastic, as i n ~ i c ~by t e curling ~ spalling of the oxide layer. Such degradation of the oxide layer can expose interior MO silicide to additional oxidation, and the sequential processes of oxidation and peeling can continue without offering oxidatio~p r o te~t~ofor n the alloy over long periods of exposure.
The tkermogravimetric test data obtained during sulfidatioii of the material showed an a ~ ~ r o x ~ ~ a t e l one order of magnitude smaller drop in s~ecimen weight than in the data obtained during oxidation. It is not clear as to the cause for the initial drop in weight, except that the possible presence of residual ~ ~ i s t u r e in the gas mixture can lead, in the early stages of exposure, to forniation of volatile oxides such as MoOJ andlor SiO (especially in the reducing condition used in the experiments), before the development of sulfide reaction. products. Figure 16 shows SEM photomicrographs of surfaces of containing Mo5Si3alloy after sulfidation at 500,800, and 1100 "C in a 1.5 vol.% H2S-
F i ~ u ~16 e SEM photomicro~raph~ of surfaces of Bcontaiiiing Mo,Si, after sulfidation in 1.5 vol.'% H,S-EI, gas m ~ x ~ u at r e (a) 500"C, (b) 800"C, and (c) 1100°C
S ~ ~ d a t Bi ~o h~ ~ ~ i o r
719
and its Control. CORCON-97 (eds A. S. Khanna, M. E=. Totlani, and S. K. Singh). Mumbai, India, December 3-6, 1997, Elsevier, p. 176. Datta, P.K., Burnell-Gray, J.S., Du, H.L., Dowson, A., and Jacob, M. (1998). Proceedings of World Ceranms Congress & Forum on New Materials. Florence, Italy. Du, E L . , Datta, P.K., Gray, J.S., and Strafford, K.N. (1994). Corrosion Science, 30(2), 99. Haufe, K. and Rahmel, A. (1952). Z . Phys. Chern., Neue Folge, 199, 152. Hindarn, H. and Whittle, D.P. (1982). Oxid. Met., 18, 245. Kofstad, P. (1988). High Ternperatare Corrosion. Elsevier Applied Science, London. mm~ry Lipsitt, H.A., Blackburn, M.J., and Dirniduk, D.M. (2001). ~ i ~ h - T e ~ ~ ~ e rStructural u t u r e Applicutior?s, rnterme~~llic Compounds - Principles and Practice, Volume 3. Wiley. Sulfidation of materials is a potential problem in McKarney, C.G., DeVan, J.H., Tortorelli, P.F., and Sikka, several systems in both power and industrial sectors. V.K. (1991). J. Mater. Res., 6(8), 1779. The sulfidation process is generally an irreversible Mrowec, S. (1995). Oxid Met., 44(1--2), 177. corrosion process in that the sulfur that is released Natesan, K. (1980). Corrosion and M e c ~ a n ~ c ~ehavior al of during reoxidation of the sulfides can penetrate the Muterials for Coal ~ u ~ s ~ c ~ App~icutions. tion Argonne substrate alloy along the grain boundaries, thereby National Laboratory Report ANL-80-5. affecting the mechanical integrity of structural compoNatesan, K. (1985). Corrosion, 41(1 1), 646. nents. This chapter presents i n f o r ~ a t i o on ~ the Natesan, K. (1986). Oxidation of alloys in b~oxidan~s. Proc. sulfidation p e r f o ~ a n c eof several intermetallic alloys, Symp. on Oxidation of Metals and Associated Mas,s claddings, and coatings in pure sulfur gas and oxygen/ Transport. TMS-AIME, p. 161. sulfur mixed-gas environments that are of practical Natesan, K. (1988). Oxid. Met., 30(1/2), 53. interest for processes in power and industrial sectors. Natesan, K. (1993). Proc. 7th Annual Con$ oy2 Fossil Energy Information is presented on the susceptibility of Ni Matermls. ORNL/FMP-93/1, p. 249, aluminides to the sulfidation mode of attack and the Natesan, K. (1997). Proc. 11th Annual Conf on Fossil Ertergy inherent resistance of bulk and overlay coatings of iron ~ a t e r i a l s0RNL/FMP-97/l7 . p. 289. Natesan, K. (2000). Argonne National Laboratory, aluminides to such an attack. Furthermore, data are unpublished work. presented on the sulfidation performance of several Ti Natesan, K. and Deevi, S.C. (~000).Proc. Int. Symp. on aluminides and on recently developed MO silicides, rnterme~allics.for the T ~ i r dMillenni~m.ASM Materials Solutions, Cinciniiati, 1999 (eds S. C. Deevi, C. T. Lm, and M. Yarnaguchi), Elsevier, Vol. Natesan, K. and Johnson, R.N. (1995). In Proc. 2ndInt. Con$ Heat-Resistant Materials (eds K. Natesan, P. Ganesan, This work was supported by the US Department of and G. Lai). ASM International, Materials Park, OH, Energy, Office of Fossil Energy, Advanced Research p. 591. Materials Program, under Cont~actW-3 1 - 1 0 ~ - ~ n ~ - 3 8Pope, . D.P. and Ezz, S.S. (1984). Int. Met. Rev., 29(3), 136. StrafTord, K. N. and Datta, P. K. (1993) Corrosion Sci., 35 (59), 1053. Strafford, K.N. and Hampton, A.F. (1971). J . Less C o m ~ o n Metals, 25, 435. Datta, P. K., Du, N.L., Jenkmon, D., Burnell-Gray, J. S., Tiearney Jr. T.C. and Natesan, K. (1982). Oxzd. Met., 1 and S ~ r a ~ o rK.N. d , (1998). Proc. Intl. Conf. on Corrosion Zelanki, P.D. and Simkovitch, C . (1974). Oxrd. Met., 8, 343.
Hz gas mixture. After exposure at 500 "C, the specimen surface showed isolated regions of MO sulfide, but no gross oxidation. After 800 "C exposure, the specimen exhibited a greater coverage of the surface with MO sulfide, whereas after exposure at 1100"C, it showed almost complete coverage by MO sulfide. These preliminary results indicate that the intermetallic material based 011 MO silicide can develop protective sulfide scales during service in reducing eiivironmeiits at elevated temperatures.
This Page Intentionally Left Blank
ook within. Let neither the peculiar quality of any thin^ nor its value escape thee.
search on applications of existing knowledge yields ter and lower cost products and new fields of use for those products, but only fundamental research can the new knowledge which makes possible radically new things. croli 75
The ~~llibilityof methods means that there is no cookbook approach to doing science, no formula that can be applied or machine that can be built to ~ e n e r ~ t ~ scientific knowledge . . . The skilful a ~ ~ l i ~ a t iofo n methods to a challenging pleasures of science.
i i n e ~ ~ iGn e~o, ~ ~~i an ~ t i tof~ t e t e ~ i a S~ ~ s ien~ and e Atlanta, GA, USA
Intermetallic compound thin films have been synthesized by methods employing diffusion across interfaces of solid-solid or solid-liquid reactants, ion implantation of constituents followed by annealing, or molecular beam epitaxy (see chapters by Ng and Moustakas and by Ramanath et al. in this volume). Bulk intermetallics have been produced by fusion of elemental constituents, by consolidation of pre-alloyed powders (see chapter by Seetharaman and Semiatin in this volume), or through combustion (reaction) synthesis during (or prior to) consolidation of reactants. The high heat of reaction associated with i nt e~ e t a l l i c formation aids the diffusion kinetics and promotes reaction synthesis of intermetallic compounds of high purity and fine grain size. Large volume changes accompanying ~ ~ orm mat - ion ~of intermetallic ~ t alloys ~ during reaction synthesis can, however, lead to retained porosity and other defects that deteriorate the overall properties of bulk solids. Mechanochemical synthesis, involving mechanical activation of precursor powders, has also been used to enhance their chemical reactivity and synthesize inte~etallicsvia controlled reaction mechanisms at lower temperatures and in significantly shorter time scales. Consequently, intermetallic alloys with highly refined and even nanocrystalline microstructures, as well as free from defects associated with uncontrolled combustion reactions, have been formed. ~ e c h a n o c h e ~ i c synthesis al also permits the formation of non-equilibrium phases and compounds with a wide range of solid solubilities. all-milling using vibratory and attritor mills and
shock-co~pression employing high-velocity impact with a gas-gun or explosive devices have been used for mechanochemical synthesis. Approaches involving the use of plasmas, electric-fields, and microwaves, have also been employed for ~eld-activatedreaction synthesis of ultrafine-grained intermetallic alloys. In their overview of synthesis and processing of intermetallics, Martin and Hardwick in Chapter 27, Vol. 1, provided a succinct description of solid-statc reactions, and general characteristics of combustion synthesis, mechanical alloying, shock compression, ion implantation and ion-beam mixing processes. They also described the advantages and disadvantages of these processes, and applications to specific iatemetallic systems documented in the literature until the early nineties. Significant advances in the development of new synthesis and processing tec~nologieshave been made since that time. For example, time-resolved diagnostics have been used to further the understanding of reaction mechanisms by (1999) during combustion synthesis, by Charlot et al. (1999a) during mechanical alloying, and by Thadhani et al. (1997) during shock-induced reaction synthesis. Combining of various processes, such as devitr~fication of mechanically-amorphizedpowder compacts prepared by He and Ma (1996) using pulsed-electrodischarge, or by Counihan et al. (1999) using shock-conso~dation techniques, and development of novel variations of reaction processes, e.g. micropyretic synthesis by Dey reaction and Sekhar (1999) and el~t~c-~eld-assisted synthesis by Munir (2000), has enabled control and optimization of reaction processes for fabricating intermetallics with unique microstructural chara~te~stics.
In~errne~~lli& C ~ i ~ p o uVol. ~ d3,~Principles ~~ and Practice. Edited by J. H. Westbrook and fa. L. Fleiseher. 02002 John Wiley & Sons, Ltd.
724
Resecnrch Techniques
igh~pressuretuning has been used in the search of new ductile iiitermetallics by Badding (1998), for example by introducing pressure-induced structural phase changes from brittl to the more ductile Ll, hanochemical synthesis eniployiiig ball-milling and shock compression, has also been extended to fabrication of various types of functional iiitermetallics, nanocrystalline alloys, and iiitermetallic-ceramic composites by a number of a Torre et al. (19959, Durovic et al. (1993), Kang et al. (1999), and Thadhani (1993 aiid 1999). The understanding of the fundamental process mechanisms of these various non~equilibriu~ processes has beeii signific~~iitly advanced, and critical and coni~arativestudies of their unique characteristics, limitatio~s,and a~plicationshave been documented by
esis
In most solid-state synthesis used for fabrication of
bulk inter~etallics,chemical reactions are carried out between intimately mixed, fine powders, such that product layers foriii at the interfaces and grow via ~ the ~ ~ t e r d i g u s ~ In o n the . case of i s o t h e r ~ areactions, rate of intermetallic formation is a function of the diffusion rates of the respective constituents, and the ~ ~ c l ~ a and t i o growth ~ kinetics of the int~rfacial reaction product(s). In many cases the reaction product (inteimetaliic) is formed in such a way that it is not coherent with the reactan~s.It may also contain many defects and fissures because of volume changes. Hence, calculations of reactant diffusivities for determining the reaction (and alloying) time for intermetallic f a ~ a t i o n become c~mplicated (see Larikov, Chapter 32 in Vol. 1). There is also extensive of intermetallic cornopportunity for surface and boundary diffusion in the involving solid-state case of powder reactants. ales of i04-i06 s (e.g. The measured diffusion coefficients are thus not e.g. combustion synthnecessarily identical with diffusion through singleesis), aiid 10-4-10-6s (e.g. shock synthesis), will be crystal or dense polycrystalline bodies (see Larikov, described. The focus will be on the mechanisms of Chapter 32 in Vol. 1); these values set a lower limit for processes in which non-equilibrium approaches arc the actual diffusion coefficient and the possible used to ther~odynamicallystabilize metastable interreaction time for compoun~formation. For example, metallic phases, or where preconditioning treatments as shown in Figure I(a), the total time for compound are used to activate and enhance the reaction kinetics. formation in a Nb-A1 powder mixture can be he ~ectionthat follows, describes the fundamental calculated to be -107s (or almost 3000 hours) for issues related to defec~-enhancedsolid-state diffusion, 1 pm diameter particles at one-half the melting t of fomiation, and effects of shock-compression, temperature of Al, if compound foi?nation by solidt, the theory and practice of mechanical alloying, state difflision is approximated using an infinite, onealloying mec~anisms,and characteristics and properdimensio~al solid diffusion-co~~plegeometry with ties of mechanically-~lloyedintermetallic compounds concentration-independent diffusion coefficients. Howwill be described. This will be followed by a description ever, if oiie can considers d e ~ ~ ~ t - e ~ solid-state h a ~ i ~ e ~ of the thermochemical princip~esand mechanis~sof digusion using the treatment developed by Whittencombustion sy~ithes~sreactions, densification techberger (1990) for mechanical alloying, then the total niques employed, and inaterials systems studied time required for compound formation is calculated to along with their properties. The effects of inechanobe of the order of 102-103s (or less than an hour), as chemical activation and of electric, plasma, and shown in Figure l(b). The calculation uses a modified microwave fields on combustion synthesis will also be bulk diffusivity term to account for diffusion through hoc~-compressionsynthesis of internietal~ the lattice and the dislocation core, and thus yields an ng the hock-response of powders and alloying time that is about four-to-five orders of magnitude faster than the preceding case involving shock-pro~essingtechniques, shock-consolidation of pre-alloyed crystallinc or amorphous powders, and no enhancement to diffusion. The analytical treatment s ~ o c k - s y n t h ~ §byi ~ and c h e m i ~ ~c~anges l of defect~eiihance~ solid-sta~~ di~usion,in fact, is what will be discussed lly, mechanocheniical forms the basis for mechanochemical synthesis of reaction synthesis of in lic compounds will be intermetallic alloys. described. In each case, the salient features of the Combustion synthesis, also known as self-propa~ating higli-temperature synthesis (SHS) or micropyretic process mechanis~s,unique characteristics, properties of the products, and ben s, in contrast to convcnsynthesis or simply reaction synthesis, utilizes exothermic self-sustaining chemical reactions between the tional methods, will be hi
725
Figure 1 Time for compound formation in a Nb-A1 powder mixture for 1 pm diameter particles at one-half the melting temperature of A1 m d assuming compound forination by (a) solid-state diffusion approxmated using one-dimensional infinitesolid d i ~ ~ s i o n - c o u pgeometry ~e with concent~ation-independentdiffusion coefficients, and (b) considering defect-enhanced diffusion ( ~ h i t t e n b ~ r g e1990) r,
constitue~tsof a powder mixture. These processes have been reviewed by erzhanov (1990), Munir and Aiiselmi-~amburi~ii (1989), Moore (1995), Varina et al. (1998) and Varma (2000). For SHS reactions to become self-sustained, the heat released due to reaction must be at a rate faster than the rate of heat dissipation into the environment (Bordeaux and Yavari, 1990). The resulting heat localization will initiate a reaction at the melting temperature of the low-melting-point constituent (or eutectic temperature) and continue by the dissolution of the solid constituent in the melt and prec~pitationof the intermetallic product, as shown schematically in Figure 2(a). Varma et al. (1998) have demonstrated that in the event of melting of the other constituent, either due to external heating or because of heat released from the dissolution reaction, compound formation occurs via liquid-liquid interaction as shown in Figure 2(b). Formation of a transient liquid phase can aid the densi~cationprocess and enable fabrication of dense c o ~ p a c ts as , illustrate^ by Bose and coworkers (1988). However, in highly exothermic intermetallic-forming systems (e.g. many silicides), the heat of reaction is often s u ~ c i e n to t even melt the reaction products, in which case the final product shows defects reminiscent o f solidification processes. The reaction products often also contain significant porosity, which has been
attrib~tedby Munir (1988) to be due to volume changes associated with reactants converting to denser products, and shrinkage during solidificat~~n froin the melt. The dissolut~oii-reprecipitationreaction mechanism involving the liquid-phase can be inhibited, if enough time is allowed for solid-state, therrnallyactivated diffusion to occur at slow rates of heating as illustrated in Figure 2(c). ~lternatively,as shown by Hida and Lin (1990) and by Lee and Thadhani (1998), use of reactants with large defect concentrations such as those formed due to ball-milling or shock compression, or those pressed to high densities, can result in the reaction being dominated by defect-enha~ced solid-state diffu~ion. onv version of reactants to products then occurs at a rate concomitant with temperature changes (profile 2 in Figure 2(c)), althou~hat any stage melting of u~reactedconstituents will lead to the onset of an insta~taneousreaction (profile 1 in Figure 2(c)). Conversely, otherwise sluggishly reacting intermetallic systems can be made to undergo self-su~tainingreactions in the presence of electric fields as demonstrated by Munir (2000) or microwave fields as demonstrated by Bhaduri (1999b). Dynamic densification employing shock-comp~ession of powders can be used to consolidate rapidly solidified or mechanically-~lloyedpowders of inter-
726
Research T e c h ~ ~ q u e s
Schematic i~lL~strating di~erentcombustion reaction mechanisms at varying heating rates. (a) Reaction initiates at the melting temperatu~eof one constituent and occurs by dissolution of the solid in the melt and pre~ipi~ation of an intermetallic; (b) Melting of both the constituents results in conipound formation via liquid-liquid interaction; and (c)solid-state reaction at slow heating rates. (Varma et al., 1998)
metallic compounds. It can also be used to microstructurally modify and produce a highly activated state of the powder mixture constituents by intimately dispersing (mixing) the reactants and bringing them in close contact with cleansed surfaces, and with significant grain size reduction via fracture and/or subgrain formation. The conibination of defect states and packing characteristics produced in the powder mixtures as a result of such effects can enhance the solidstate reactivity and cause powder mixtures to undergo shock-initiated chemical reactions and phase transformations d u r i ~ gthe ~ i c ~ o s e c o nduration d of the shock state. The unique characteristics of shock synthesis of materials were first reviewed by Duvall (1984) and a discrete description of important processes leading to shock chemistry have been outlined by Graham (1993). Time-resolved measurements of powder compressibility have provided evidence that shock synthesis of
intermetallics via structural and chemical changes can occur during shack compression, suggesting the possibility of shock-induced formation of novel phases and new intermetallic compounds. Shock activation (or modification) of powders and powder mixtures can also be used to generate precursors for subsequent thermal processing. The highly activated and densely packed state of the reactants can lead to accelerated mass transport and cause chemical reactions to occur at substantially lower temperat~~res. The DTA traces shown in Figure 3 illustrate the differences in the y and shockreaction behavior of s t a t i ~ ~ l lpressed densified Ni + Ti powder mixtures. Peaks I, 11, and 111, corresponding to three different exothermic reactions, are observed in both cases. However, in the shockdensified Ni + Ti powder mixture, the reaction initiation temperature for each peak are significantly lowered, due to the enhancement in reactivity and
Novel Synthesis Techniques 140 120 100
80 60
40
100
300
500 700 900 ~ ~ m p ~ r(e) ~ t u r ~
I100
Figure 3 DTA traces showing lowering of initiation temperature for each of the three types of reactions in shock-dens~~edNi + Ti powder mixtures in contrast to statically pressed powders
727
against the container walls, generates an intimately mixed and microstructurally refined powder aggregate, as described by Kuhn et al. (1984). Alloying typically occurs in times of a few to a few tens of hours, via diffusion, accelerated by the presence of deformationinduced lattice defects, and due to localized increases in t e ~ p e ~ a t u rduring e pa~t~cle-ball~ollision~. propagating, Combustion-type, chemical reactions have also been reported to occur during mechanical alloying by Atzmon (1989) and M c C o ~ i (1995). c~ In this case a momentary local increase in temperature of heavily worked and intimately mixed powders, trapped during collisions of balls in the milling process, triggers an instantaneous self-sustaining chemical reaction, resulting in compound formation.
The mechanical alloying process involves milling powder mixtures, blended in the right proportion activation caused by effects of shock compression of (the charge), with steel, alumina, or tungsten-carbide powders. balls (the grinding agent), in a mill (container) made of materials ~imilarto the grin~ingagent. performed dry in an inert gas medium, or wet using a anical ~ ~ l o y i nofg I liquid such as hexane. The liquid medium is typically used to prevent agglomerate formation, particularly in Mechanical alloying represents an ideal example of the case of powders of ductile components, and/or the inechanocheinical synthesis, in which alloy formation formation of dispersed carbide phases. Different is aided by solid-state, defect-enhanced diRusion. The designs of mills are available, depen~ing on the process can be used for synthesizing powders of alloys capacity and efficiency of the milling operation. For with extended solid solubilities; nanocrystalline, disexample, the SPEX' vibratory or shaker mill, comordered, or amorphous structures; and immiscible as monly used in laboratories for 4 10 g of powder, i s a well as intimately dispersed mixtures (e.g. W-Ni, W-Cu ~i gh-e~ci ency mill with collision velocities of the order and Nb-Cu alloys). Commercially, the process has of 10m/s. The planetary ball-mill has rotating vials been employed to make powders of dispersionarranged on a rotatin support disk. It can mill a few strengthened iron-alloys (Incoloy MA 956: Fehundred grams of powders with collisions of balls 203), nickel20wtO~~r-4. SwtO/oAl-O.5wto/~Ti-O*5wto/~Y caused by the centrifugal force. Collision velocities are based superalloys (Mar-M200:Ni-9wt%Cr-5wt%Alhigher in this planetary mill, but the frequency of 2 wt%Ti-12 wt%W-lO wt%Co-1 wt%Nb-1.8 wt%Hf), collisions is lower than that in the SPEX mill. Attritor aluminum-lithium alloys (A~-905~L:A1-4wt%Mg- niills are used for milling up to lOOlbs of powder at a 1 . 3 w t % ~ i - l . l w t ~ ~ C - O . 6 w t ~ 0and O ) , supercorroding time. The powder and the balls are held in a stationary magnesium-based (Mg-5-2Owt~0Fe)alloys (Suryanarcontainer and agitated by a shaft with arms rotating at ayana, 1999). Fabrication of nanocrystalline 250rpm. Hence, the powder charge is subjected to intermetallics is described in detail by Suryanarayana shear and impact forces, although at much lower (see chapter in this v o l u ~ e ) , More accounts of velocities. Mechanical alloying for commercial producmechanical alloying and its applications are described tion is performed using various designs of industrialin reviews by Martin and ardwick in Chapter 27, scale mills, in which up to 3000 lb of powder can be Vol. 1, Lu and Lai (1998), Murty and Ranganathan ~ a l l ~ ~ i l(Kuhn l e d et al., 1984). (1998), and by ~uryanarana(1995, 1996, 1999). In general, the initial powder particle size, powder ~ e c h a n i c aalloying ~ involves ball-milling of powmaterial properties (hardness, fracture toughness, etc.), ders, during which repeated deformation, fracture, and s trade mar^ of SPEX ~ n d u s ~ ~Edison, i ~ s , NJ. cold welding of the particles between colliding balls or
728
Research T e c h n i ~ ~ ~ ~
weight ratio of powder and grinding agents, and milling efficiency (function of diameter of grinding media and rotational speed), influence the milling time required to achieve a specified degree of crystallite size reduction and mechanical alloying, Contamination of powders from the materials of the mill or from the environment during the typically long time-scale of the alloying process, is a major problem that can not only influence the alloy foimation mechanism, but also the structural characteristics and properties of the milled, alloyed material. Aizawa and his coworkers (1999a, 1999b) have developed a high-speed mechanical alloying process that combines compaction and densification as a simultaneous process. The method employs repeated cycles of compression and extrusion as shown in Figure 4, through independently controlled motion of upper and lower punches into the die cavity filled with the powder mixture. Under such conditions, the powder is subjected to hydrostatic pressing and shear strain due to extrusion with ratios of 1.5. The sit^ alloying and densification process allows minimal interaction of powders with the e n v i ~ o n ~ e nand t , therefore, reduces pick-up of contam~nants.Alloying times are s~gni~cantly lower than in other typical methods employed for inechanical milling, and simultan~ousdensi~cationof powders to green densities greater than 85% TMD (theoretical rnaximwn density) can be attained. Scalability of the process has also been demonstrated using a maximum
-
-
-
loading capacity of SO00 kN, with which mechanical1y alloyed compacts of 56 mni diameter and 76 mm height have been fabricated.
3.2 A I I u ~~ ~ ~e ~ ~
~
~
~
~echani calalloying involves a conti~uouscycle o f plastic deformation, fracture, and rewelding of powders, typically occurring as a three-stage process. In. the first stage, particle defor~ation. and interparticle welding dominate the process, as the surface area of constituents increases, and aggregates of lamellnrshaped powder mixtures are formed, During the second stage the large lamellar-like particles are fractured and rewelded into particles of finer lamellae. In the third stage, equilibrium i s reached between ~eformati#n and welding on the one hand, and fracture on the other hand, as the powder homogenizes into a microscopically mixed aggregate, or as alloying is completed. Continued ball milling results in further refinement of the grain size, until either a nanostructured or amorphous coinpound is formed, Chen et al. (1999) have observed evidence of n a n ~ c ~ s t a l l i nB2e NiAl alloy formation (based on appearance of X peaks of B2-NiAI phase) following 8 hours o f ballmilling powder mixtures in a §PEX mill. Figure 5 shows §EM microgra~hsillustratin~the progr~§sion.of mechanical alloying during ball milling; as-mixed Ni and A1 powders of -40pin average initial size are
Schematic of high-speed mechanical alloying and densification process employing repeated cycles of cornpression and extrusion, through independ~n~ly controlled motion of upper and lower punches into the die cavity filled with the powders (Aizawa et al., 1999a)
~
729
SEM micrographs illustrating the p~o~ression of mechanical alloying as (a-f) mixed Ni and A1 powders of 40 ,um average i ~ i t i asize ~ are deformed, fractured, aiid rewelded into smaller and siiialler sized aggregates of intimately mixed and alloyed constituents during ball milling in a SPEX mill. Average size of alloyed aggregates after 48 hours of ball idling is less than 2,um (Chen et al., 1999)
R e s e a ~ c~~e c h n i q ~ e ~
730
deformed, rewelded, and fractured, into smaller and smaller sized aggregates of intimately mixed constituents. The average size of the alloyed aggregates after 48 hours of ball milling is less than 2pm. As shown in Figure 6, the average crystallite size changes from 50 nm with an 8-hour ball-milling time to c 10nm with ball milling for 48 hours. It can be seen that mechanical alloying involves processes that are influenced by many variables that cannot be easily determined. For example, in order to predict the evolution of microstructure during each stage of mechanic^^ alloying, it would be necessary to know the frequencies of effective collisions and the extent of d e f o ~ a t i o nduring each collision. Several attempts have been made to model theoretically the mechanisms and mechanics of the mechai~icalalloying aurice and Courtney, 1994, 1995; Khina et al. 1997; and Lu et al., 1994), and to correlate the phases formed with the process variables based on experimental observations. Courtney and Maurice (1996) have performed i~echani~s- base^ modeling of meclianical alloying, dividing the modeling a ppr ~ a c h into two categories - local and global. In the case of local modeling, they considered a typical collision taking place in a specific device with stipulated process characteristics including impact velocity, frequency, ball size, and charge ratio (ball-to-powder weight ratio), and ~ e t e r m i nthe ~ ~extent to which powder particles deform and the freq~encywith which they cold-weld and fracture in a single collision. In global modeling, they incor~orated local modeling and
-
a
Average crystallite size of mechanically alloyed B2NiAl compound plotted as a function of bull milling time, showing the crystallite size changing from -50nm with 8hour ball-milling time to Cnin after ball milling for 48 hours (Chen et al., 1999)
considered the frequency of ‘efTective’ ball impacts based on impact frequency, angle, and velocity, since all impacts do not generate the same degree of milling. In addition they also considered effects of powder segregation as well as work hardening in g ~ ~ ~ a l modeling of the alloying process. Their model is based on several simplified a s s u ~ p t i ~ nwith s respect to collision velocity, impact frequency, stress distribution in the particles upon impact, and criteria for cold ~ ~ e l d i n gand fra~mentation. Nevertheless they observed a better than an order-of-magnitude agreement between model predictioiis and experimental observations, with the model predicting a slightly larger particle size than the median size measured and somewhat underestimating the milling time for a given median size. More importantly, they were able to use the predictive capability provided by the model to investigate, e.g. how collision velocity and initial material properties affect the processing (or alloying) time. Several other synthesis-based modeling attempts have also been made (Khina et al., 1997; Lu et al., 1994; Magini et al., 1991; ~bdellaouiet al., 1993; and Schaffer and ~ c C o r m i c 1991), ~ , in which a correlation with process parameters, order-of-m~gnit~de estimates of alloying kinetics, and degree of microstructural and morphological refinement have been obtained. Processing diagrams plotting collision energy (based on rotational speed) and frequency (or power) have been constructed for Ni-Zr and Fe-Zr powders, showing areas corresponding to different resulting microstructures, e.g. amorphous, nanocrystalline, or bulk intermetallic (Magini et al., 1991; Abdellaoui et al., 1993). Schaffer and McCormick (199 1) have correlated the kinetics of a number of redox reactions with the charge ratio, ball size, and number of collisions, in addition to identifying the conditions for the occurrence of spontaneous combustion,
Mechanical alloying of a variety of i n t ~ ~ ~ t a l l i c compounds, starting with elemental constituents, has been investigated by a number of researchers. It has been demonstrate^ that intermetal~ics with equilibrium, metastable structures, and non-equilibrium quasi-crystalline states can be formed. Intermetallics with equilibrium structures includin~those based on Cu-Zn brasses, nickelides of Ti and Cu, stannides and germanides of Fe, and aluminides and silicides of Ni, Ti, Nb, Fe, Wf, Re, W, Cr, and MO,are the ones that
731
Novel S y n t ~ e s iTechniques ~~
have been studied the most. Desch et al. (1996) have observed that mechanical alloying of aluminum-rich i n t e ~~etallic compounds of Zr and Hf, results in a metastable L1, type structure that transforms to a stable DO,, structure upon subsequent annealing. On the other hand, TiAl, transforms to a metastable DO,, structure upon mechanical alloying and can be converted to a DOz2phase upon annealing (Descli et al., 1995). Likewise, a metastable, deformationinduced martensite phase has been known to form in mechanically alloyed Cu-Zn powders by McDermott and Koch (1986), as well as in Fe powders ball-milled in a nitrogen atmosphere by Rawers et al. (1995). Other examples include formation of a metastable phase with a simple cubic structure in mechanically alloyed Te-Ag alloys (Chitralekha et al., 1995) and a rhombohedral phase in a Al-Ge alloy (Yvon and Schwarz, 1993). While the peak pressure generated by collisions in ball milling is significantly lower than that required for pressure-induced phase changes, the complex state of stress involving a combination of shear and hydrostatic compression, is sufficient to introduce these structural phase transformations. In fact, mechanical alloying may also be a convenient niethod for syntliesis of high-pressure phases and fabricating ductile intermetallics (e.g. TiAl, with a DO,, or Ll,-phase) as demonstrated by Sahu and coworkers (1997). Ordered intermetallics with nanocrystaline structure (see chapter by Suryanarayana in this volume) and disordered amorphous compounds (Lu and Lai, 1998; Murty and Ranganathan, 1998) have also been synthesized by mechanical alloying. Formation of nanocrystalline or amorphous intermetallic compounds synthesized by mechanical alloying is based on the balance between atomic disordering introduced by ball milling and thermally activated re-ordering. The disordering is dependent on the difference in the energy between the ordered and disordered states, which in turn is known to scale with the enthalpy of formation. However, it is not possible to fully predict whether intermetallics synthesized by mechanical alloying will form a crystalline or amorphous state. For example, iron aluminide is one of the few aluminides that forms a disordered (amorphous) state during mechanical alloying, as illustrated by Schropf et al. (1994). On the other hand, as described earlier, mechanical alloying of Ni and A1 powder mixtures shows only the formation of a nanocrystalline structure and no evidence of amorphization even after 48 hours of ball milling. As shown in Figure 7, the longrange order parameter also decreases with increasing
milling time (Chen et al., 1999). In contrast to aluminides, silicides show a mixed behavior, even within the same material system. For example, mechanica~alloying of co~positions corresponding to disilicides of Nb and Ta forms nanocrystalline NbSiz and TaSi, intermetallics, while that of Nb,§i, and TasSi, compositions forms amorphous compounds. Both silicides of molybdenum, MoSi, and Mo,Si,, generate nanocrystalline structures, while TiSiz and TisSi,, form disordered amorphous com~ounds. Ordered intermetallics can also undergo disordering by ball milling. Depending on the relative free energy values of corresponding states, formation of either ordered solid solutions (e.g. Nb,Al, Ni,Al, etc.), or amorphous compounds (e.g. Nb,Sn, NiTi, NiZr, etc.), or phases with complex structures (e.g. Ni,Sn,, TiSi,) can occur. Disordering for achieving strength increases in ordered intermetallics by mechanical milling, has been described by §uryanarayana in this volume. Zhou and Bakker (1994a, 1994b) used hig~-energy ball milling to produce well-defined, atomically disordered, nanocrysta~lineNi,Si and CO,%, and Li et al. (1997) showed formation of a high-temperature phase FesSi, in addition to FeSi and Fe,Si phases during ball milling for different times. Balogh et al. (1997) have shown that FeB and Fe,B intermetallic powders cannot be amorphized during mechanical milling, unless they are mechanically alloyed with Fe when an amorphous alloy close to Fe80B20composition is formed. On the other hand, mechanical milling of Laves-phase Fe,Hf intermetallics has shown segregation of a phase with a lower degree of chemical disorder but with a structure similar to the original 0.
0. 0
10
0
30
4
Figure 7 Long-range order parameter (S) decreases with increasing ball-milling time during rnechanlcal alloying of Ni and A1 powder mixtures (Chen et al., 1999)
732
Research TPC hn igues
intermetallic, and another phase with a high degree of disorder (Xia er al., 1998). Synthesis of non-equilibrium, q~~si-crys~alline intermetallic phases by ~ e ~ h a n i calloying al has also been reported by Eckert et al. (1990) in a number of materials systems, including Al-Cu-Mn, Al-Cu-MnGe, AI-Cu-Fe, Al-Cu-Ru, Ti-Ni-Fe-Si, and Mg-Al-Pd. In intermetal~ic systems that amorphize at lowintensity milling conditions and subsequently crystallize during milling at high intensities (due to heating), the quasicrystall~nephases are observed to form at intermediate milling conditions. In systems that do not amorphize during mechanical alloying, the quasicrystalline phase forms at low-intensity milling conditions before reaching a stable intermetallic structure h-intens~ty milling. In either case, the icosahedral structure of quasicrystalline phases synthesized by mechanical alloying is observed to be similar to that formed by rapid solidi~cation. ~unctionalintermetallics, such as semiconducting IIIjV compounds (p-type GaAs and AlAs), with nanocrystalline structures, have also been synthesized by ~ e c h a n i ~alloying al at room temperature or under cryogenic conditions. In this work, Matteazzi and Farne (1991) used XRD traces to indicate the formation of the nanocrystalline se~iconduc~ing phases by mechanica~alloying under different conditions, The mechanically alloyed semiconducting intermetallics were either used as precursors for the production of single crystals, or as powders for semiconductor substrates and solar cells.
Co~bustionsynthesis (or SHS) is applicable to a wide variety of intermeta~lic-fo~ing systems, i.e. all combinations of elements with a large negative heat of reaction AHRcan be synthesized. The process involves reactions occurring either by the 'combustion wave~ r o p a ~ a t i o nmethod ' (in which a pressed pellet is ignited at one end), or the 'thermal explosion or volume reaction' method (in which the whole pellet is uniformly heated in a furnace). Characteristics governing the 'combustion-wave propagation' mode are: rea~tion-wavepropagating as a rapidly moving front at velocities of 0.1 to 25 cm/s; temperature of combustion reaction-wave-front in the range of 2000-4000 "C; and re~arkablyhigh heating rates of the order of 10'106K/s. The combustion wave-front propagates either as a steady-state wave of constant velocity, or a nonste~dy-statewave (with 'spin' or 'oscillatory' mode).
The 'thermal explosion' method, which involves a volume reaction occurring simultaneously (or nearly so) throughout the compact, is also referred to as a reaction (or reactive) s i n t ~ r ~process. n~ It i s more appropriate for weakly exothermic reactions that require pre-heating of the compacts. In any exothermically reacting m~terials system, alteration of the energetics of the reaction (e.g. by addition of inert diluents) caii cause the wave-propagation mode to change from steady-state to on-ste~dy-state, and even to complete extinction. Such effects were e ~ a ~ i n by ed Munir (1992), who constructed so-called 'SHS diagrams' (plots of temperature versus diluent aniount), to describe the conditions under which different types of reaction modes can occur.
.E
ase
In order to achieve a self-su~tained tion reaction between two components A and B, the rate of heat release due to the reaction must be faster than the rate of heat dissipation~ The adiabatic temperature (Tad) or temperat~re generated by the combustion reaction, may be greater or less than the melting temperature (Tm) of the reaction products, depending on the enthalpy of formation, AHf, in relation to the heat of melting, AHm, and the c~aracteristics of heat dissipation. Table 1 lists calculated t~ermodynamicdata (Bhaaduri, 1999), ~ n c ~ ~ the ~ i nheats g of formation and melting, adiabatic and melt teniperatures, and the volume change ( A y ) accompanying the reactions in several i~termetallic-forming binary powder mixtures. Rogachev et al. (1995) have c o ~ p a r e dthe c ~ l c u ~ ~ t ~ d adiabatic tem~eraturewith the phase diagram of Ti-Si, and illustrated that Tadexceeds (a) the 1330 "Ceutectic temperat~reof both Ti- and Si-rich alloys for all silicide compounds; (b) the melting point of Si but not of Ti, for combustion of Ti + Si, 5Ti C 43, and 5Ti C 3Si; and (c) the melting point of Ti in the case of 5Ti +3Si, with up to 15wt.% of the Ti5Si3phase molten. Thus, as shown in Figure 8, the reactions occur in two stages of the tem~e~ature and space window: Regian I: 1330"C < T < 1670"C Ti(s) i-Si(s,l) + L ~ ( ~ u t e c ~ i c ) ~i~L2 i ~(Zone i 2 ( ~I)) Region 11: 1670 C < T ~ 2 1 3 0 C " Ti(l) + ~ i ( ~ ~ T i ~ (Zone S i 3 11) (~,~~
Table 1 Calculated values of heats of formation (AHJ and melting (AHnl),adiabatic ( Tad)and melting temperatures (Tm), and volume change ( A V ) for reactions in silicon-based inter~etallics(Bhaduri and Bhaduri, 1999)
ZrSi ZrSi, Zr,Si, TiSi TiSi, Ti,Si, CrSi CrSi, Cr,Si, ReSi ReSi, Re,Si, MnSi Mn,Si, NbSi, Nb,Si, TaSi, Ta,Si, MoSi, Mo,Si, WSi, WSSi, VSi, V,Si,
154 159 572 129 134 138 55 80 222 52 90 156 78 200 137 449 118 333 131 308 92 134 125 459
59 45 60 40 43 179 35 42 33 50 73 47 29 22 69 74 89 72 75 57 85 64 52 49
2909 2228 2739 2349 1808 2528 1215 1252 1233 997 1420 90 1 1553 1100 1863 2222 1671 1772 1872 1672 1471 945 1792 1703
2368 1790 2483 1843 1773 2408 1730 1730 1920 2153 2253 2233 1548 1573 2203 2753 2473 2773 2303 2463 2433 2593 1953 2283
-19.1 -20.8 -15.8 -22.5 -25.9 -16.1 -22.6 -30.7 -18.9 -21.6 -31.3 -18.2 -26.8 -18.0 -24.7 -13.9 -25.9 -18.6 -27.4 -17.7 -27.6 -17.6 -28.7 -18.2
Heating rate also influences the self-sustaining nature of SHS-type combustion reactions, in addition to determining the type(s) of intermetallic-phase(s) formed. For example, Deevi (1991, 1992) has shown that in No + 2Si powder mixtures Mo5Si, is the first phase to form at low heating rates. The Mo-rich phase then becomes enriched with liquid silicon present at a higher temperature to form the final MoSi, product. As the heating rate is increased, approaching the conditions in the 5 S process, the MoSi, phase is observed to form directly via a reaction between the liquid silicon and solid molybdenum without going through an i n t e ~ e d i a t eMo,Si, phase. Thus, combustion synthesis at higher heating rates typically yields a single-phase final product, while slow heating rates often result in multiple reaction products that may be detrimental to the overall properties. The ~echanism of intermetallic formation during combustion synthesis is also dependent on several structural and chemical factors, such as particle size, particle size distribution, density, stoichiometry, phase t r a n s f o ~ ~ t i o i i sand , mass transfer. A major limitation of combust~on synthesis is porosity in the reaction products. The porosity results from the presence of the initial void volume, and from intrinsic sources including the molar volume change
! i
I
Phase diagram of Ti-Si system and schematic illustrating a profile of equilibrium temperature during various stages of reaction (Rogachev et al., 1995)
734
Research T e c h ~ ~ q ~ e ~
(as indicated in Table I), thermal migration, and shrinkage associated with solidification of melted reaction products (Munir, 1988). Statically pressed powder compacts may be used to eliminate initial porosity, but in that case the com~ustion-typeSHS reactions may not be self-sustained due to the imposed thermodynamic and kinetic limitations. Figure 9 illustrates an example of the dependence of the combustion reaction wave velocity and percent conversion (of T i + C reactants to T i c product) on the initial density of powder compacts (Varma et al., 1998). It can be seen that both the combustion front velocity and percent conversion increase with compact density up to -56%, and then both decrease until the reaction extinguishes above a density of -63%. Intermetallic~formingpowders show a similar initial increase in velocity with increasing density, followed by a sudden extinction of reaction beyond a certain compact density. The limitation imposed by the high density can be seen to result from the equation that correlates the heat content, or accumulated heat, in the reacting material with that corresponding to the heat di~sipateddue to cond~ctionand heat generated from the reaction:2
where p is the density, Cp is the heat capacity, 1c is the thermal conductivity, Q is the heat of the reaction, q is the degree of completion of the reaction, T is the temperature, t is time, and x is the distance along which the propagation wave is traveling. The limitations arise due to the thermal conductivity, which is related to the velocity of the reaction front and the density in tlie sample. If tlie thermal conductivity is very high, then the heat accumulated will also be low, which in turn causes the reaction to become sluggish and eventually extinct. On the other hand, a higher initial compact density will raise the thermal conductivity due to which heat dissipation from the reactants will be too fast for the high temperature (and hence, the reaction) to be sustained. Methods to overcome the t he ~ odynam i c and kinetic limitations include activating reactant-powders by balI~~~illing (Hida and Lin, 1990) or by shock compression ( ~amjo s hand i Thadhani, 2000), preheating reactants prior to reaction synthesis (Bose et al., 1988), or performing reactions under externally applied electric (Munir, 2000) or microwave (Bhaduri, 1999) fields. In the preheating technique, however, 2Equation assumes adiabatic conditions, neglecting heat losses and transient effects from ignition surfaces.
OL
40
,
45
I
I
50
It
55
4%
I
65
~ i ~ 9w rEffect ~ of initial density on combustion wave velocity and degree of conversion (Varma et al., 1998)
unwanted phases may result due to the promotion of digerent reaction mechanisms. The applied electric and microwave fields have proven useful by causing reactions to occur even in dense compacts, and enhancing reaction rates, even in several otherwise sluggish intermetal~ic-forming systems. Electric-field activated combustion synthesis of silicide-based intermetallics (of MO, Nb, W) as well as of inter~etallic~ ceramic composites, has been extensively studied by Munir and his co-workers (1997). In this approach, ~ectangu~a~-shaped compacts of the powder reactants are placed between spring~~oaded electrodes across which a voltage is imposed simultaneously with the activation of the reaction ignit~onsource (viz., tungsten filament). Xue et al. (1999) have shown that under the action of the applied field, reactions prop~gateat a velocity higher than that under conditions of no applied electric field. Figure 10 shows the variation of measured combustion wave velocity as a function of the green density of MO+ 2Si compacts in the absence and presence of electric field. It can be seen that without the application of an electric field, synthesis of MoSi, can be accomplished by the propagation of a self-propagating combustion wave in compacts of up to -75% green density. However, in the compacts with higher density, the wave propagates with an unstable (pulsating) mode, until ultimately no combustion wave can be initiated. In contrast, when the coinbustion reaction is initiated under an imposed voltage across the electrodes, the results are markedly different, as seen in Figure 10. With an applied voltage equivalent to 7V/cm across the reactant compact, (a) the combustion wave propagates in a steady-state mode, (b) for any given green density, the velocity i s
Novel Sy~t~esis Techniques
Figure 10 Wave velocity as function of compact green density and electric field strength (Xue et al., 1999)
higher by about a factor of two than the corresponding value when combustion synthesis is carried out in the absence of a field, and (c) the combustion wave can be initiated in samples with densities as high as %0/0 TMD, which represents the highest value to which the reactants were densified. The general trend of velocity with relative density is the same for results obtained with or without applied field. The reaction wave-front velocity also increases with an increase in voltage. For 65% dense compacts, the peak velocity for the 13V/cm results (- 8 mm/s) is about a factor of two higher than that for the 7V/cm results and about a factor of four higher than that obtained in the absence of a field. The influence of electric field on SHS reactions has shown that the contribution of field is thermal in nature, providing additional energy in the form of Joule heating and resulting in a higher combustion temperature and wave velocity (Feng and Munir, 1995, 1997). The degree of activation in the combustion (reaction) zone depends on the current distribution in the reacting sample, which in turn depends on the electrical conductivity of the combustion product. For systems with relatively non-conducting products, Feng and Munir (1995) observed that the current is localized in the reaction zone. In such a case the degree of activation is i~dependentof time, which is revealed by the constancy of wave velocity for any given applied field. However, for relatively conducting products, they found the degree of activation in the reaction zone to decrease with wave propagat~on.In such cases, the wave velocity is constant, but decreases with time of wave propagation. Synthesis of MoSi, belongs to the
735
first category, i.e. where the synthesized product 1s relatively non-conducting. Typical current and voltage changes during wave propagation in the synthesis of MoSi, are shown in Figure 11 (~edevanishviliand Munir, 1998). The voltage decreases with the start of the wave but remains relatively constant during wave propagation through the sample. The calculated resistance, also shown in Figure 11, is constant during wave propagation, which indicates that the current is confined to the combustion zone. Microstructures of the products of electric-fieldassisted combustion synthesis show differences depending on the initial compact green density. The grain size of MoSi, synthesized from reactants with a 95% green density is found to be considerably (an order of magnitude) smaller than that of products of samples with lower initial densities. Xue et al. (1999) have suggested that, in the presence of a field, the grain size is smaller as lon as the product remains in the solid state. In contrast, for systems in which the product contains a liquid phase, the application of electric field results in a microstructure with a larger grain size (> 10 pm).
The inherent drawback of porosity retained in combustion (or reaction) synthesized materials, is that application of concomitant pressure is required to fabricate bulk, fully dense materials. Various approaches of static and dynamic pressure application have been employed, including application of pressure simultaneously with reaction initiation, or immediately
Figure 11 Sample voltage, resistance, and current during wave propagation in the electric-field-activated synthesis experiment with MoSi, (~edevanishviiiand Munir, 1998)
736
Research ~ e ~ h ~ i q u e ~ ~
following reaction completion. Both methods take advantage of the high temperatures generated due to the heat of reaction. However, the reacted sample needs to be isolated from the die, due to the high temperature of the reaction products. Furthermore, in the former case, out-gassing of volatiles (of impurities adsorbed on powder surfaces) may be inhibited, causing formation of blowholes in the densified products. Techniques of simultaneous application of static pressure have employed especially designed dies (Dunmead et al., 1990), or steel dies lined with graphite or BN ceramics (Nishida and Urabe, 1992), or even pressure-transmitting media such as SiO, sand or (Merzhanov, 1990) for pseudo-isostatic pressing of the reaction products. High-pressure gas has also been used for the dens~ficationof reaction products sin~ultaneously with combustion synthesis. The approach termed 'chemical furnace' involves placing the reactant powder compact in an evacuated glass container which is surrounded by a highly combustible powder mixture (such as Ti + C powder mixture) that is enveloped by Ar gas at l00MPa. Ignition of the combust~blepowder mixture acts as a chemical furnace and heats the sample until it undergoes reaction. The reaction product is then simultaneously statically pressed to eliminate residual porosity. This method has been particu~arlyuseful for producing ceramicceramic, ceramic-metal, and ceramic-intermetalli~ functiona~1y"~raded materials (Miyamoto et al., 1990). In the case of d en si ~ cat i ~ nfollowing reaction completion, the time delay (lag) between detection of reaction completion and actual application of pressure is critical for a ~ h i e v i ~maxin~um g density and controlled microstru~ture,The time lag should be long enough to allow expulsion of gases from the sample, but less than the time when the compact cools below its ductile-to-brittle transition temperature (DBTT) (Grebe and Thadhani, 1999). With less than the optimum time delay, the reacted and densified compacts show dela~ination,lower densities (due to effects of 'after-burn'), and coarse-grain microstructures (due to slower cooling rate), while with longer time delays, the SHS reacted product cools to a point where full densification and microstructural breakdown to a fine-grain structure is not achieved. Dynamic methods of pressure appl~cationfollowing reaction completion have been particularly useful, since these provide the opportunity for expulsion of volatile gases from the compact, and the pressure can be applied immediate~y wit hi^ seconds) fo1lowing the SHS reaction and before the products cool below DBTT (Niiler et al. 1988, Rabin et al., 1990; LaSalvia
et al., 1992, 1994; Roke et al., 1992). Another similar method involves SHS reaction followed by extrusion, which has been used by Podlesov et al. (1992) to fabricate a variety of Tic-based cermet electrodes used for electric spark plugs. In this method, the powder mixture compact is placed in a mold, and the reaction is initiated locally by a he ate^ tungsten wire. After the combustion reaction wave propagates through the compact, a relatively low pressure (< 1000 MPa), is applied to the plunger to extrude the hot reaction product through the die,
Combustion (or reaction) synthesis of a variety of ~ntermetallic-forming alloys has been investigated. While the emphasis in most investi~ationshas been focused on unders~dndin~ the process mechanisms and kinetics of reaction synthesis, structure-property correlations have also been studied. The effect of the initial mixture ratio on the SHS reaction synthesis of 3Mi+Al and Ti+3AI powder mixtures has shown formation of single-phase Ni3Al and TiA1, compacts with 70--80% of theoretical density, while mixtures of 2Ni -f- JAl, Mi + AI, Ni + 3A1, 3Ti + Al, Ti + Al, Ti+2A1, and Ti+3A1, all yield major phases corresponding to the initial ratios but also containing varying amounts of other phases. The reaction synthesized materials thus have properties that are not just a function of the intri~sic intermetallic behavior but those of two-phase or multi-phase materials (Song and Wang, 1994). Similarly, combustion synthesis of Zr-Si compounds with Si-rich starting compositions has shown formation of the ZrSi, (disilicide) and ZrSi (equiatoniic) alloy, but with a Zr-rich starting stoichiomet~y,the high-tem~erature Zr,Si3 phase is formed, The thermodynamic stability and phase diagram (Figure 12) topology do not necessarily provide for reliable prediction of phase formation. In fact, in the Zr-Si system, reaction products containing the thermodyna~ically stable Zr3Si2,Zr2Si, and Zr3Si c o ~ p o u n d s(having relatively large heats of formation) are never observed. The Zr,Si2 intermetallic also has almost congruent phase stability in addition to a fairly large heat of formation. On the other hand the inc~ngruenthigh-temperature phase, Zr,Si3, is always obtained as a reaction product of Zr-rich compositions, as illustrated by Bertolino et al. (1999). Inter~etallics containing ternary additives to improve their ductility via formation of more ductile phases (those with a greater number of operating slip
30 40 50 60 70 80 $0 100 agrain of the ZrSi system (Bertollino et al., 1999)
systems), or composites with ductile reinforcements, have also been studied by combustion synthesis. Fu and Shekhar (1998) employed reaction synthesis in Mo-Si-A1 powder mixtures, and with appropriate control of process parameters including degassing, green density, furnace temperature, and reactant particle size, they formed a 94% dense C4O-type molybdenum aluii~~no-silicide internietallic, containing traces of Mo,Si, and A1203 The Young's modulus of the ternary silicide was measured to be 342. and its Vickers hardness was -8.6GPa. It had ture is slightly higher than that of MoSi,. The oxidation resis intermetallic was better than that but it showed a higher rate of oxidation at 1200°C. Reaction synthesis studies by Dey and Sekhar (1999), based on the Ni-A1 system, have shown that the combustibility o f the binary mixture increases at first (due to kinetic reasons) and then decreases with the addition of Nb and Ti as the overall heat of reaction is lowered. The microstructure of the Nb- and Ti-lean Ni-AI alloys contains predominantly the 232phase which changes to a DO, structure with higher amounts of Ti and Nb additives. The hardness, compressive ductility, and fracture tougl~nessof these various alloys are listed in Table 2. It can be seen that the ductility and toughness increase with only fractional additions of Nb and Ti. With increasing additive content, the hardness of the NiAl alloy increases, but its ductility and toughness decrease. On the other hand, its oxidation resistance deteriorates with even small additions of Ti or Nb.
In addition to consol~dation of intermetallic alloy powders, shock compression can be used to synthesize iiitermetallic compounds via clieniical reactions and phase transformations. These phenomena occur as ~anifestations of ~ n h ~ n ~solid-~ta~e ed reactivity caused by changes in reactant material con~guration and defect states introduced during the shock-cornpression process, as well as due to kinetic freezing of high-pressure phases. A variety of intermetallic compounds with equilibrium, non-equilibrium and metastable phases, as well as otherwise im~iscible alloys, have been synthesized by shock compression (Thadhani, 1993). Time-resolved in-situ nieasure~ents have been perform4 to infer the formation of the new phases in the high-pressure shock state, and shockrecovery experiments have been performed to understand the ~hase-formation ~ ~ h a ~ and i s to ~ s characterize the microstruct~reand properties of the synthesized materials, While shock compression is considered an attractive process, particularly since long-tern thermal escursions are avoided and alloys with fine-grain-size microstructures and novel phases can be produced, questions regarding its commercial feasibility still remain to be resolved.
The shock mechanics of ti~e-dependent~ompression of powders (highly porous or distended solids with porosities of tens o f percent), cause wave propa features which are ~ual~tatively d i ~ e ~ e from. nt for the same solid in the fully dense state. It is to be expected that it is these rmation features that can powders, and synthes structural changes in the microsecond time-scale of shock-wave propa~ation.The key to the successful
Table 2 Properties of Ni-A1 alloys with Ti and Nb additions (Dey and Sekhar, 1999) ~-
Vickers hardness
~ o ~ ~ r ~Fracture s s i ~ e ductility toughness (%) (MPAylm) Alloy composition number (VHN)
Research Techn~q~es
738
c..-....
static
~ i ~ 13 ~ r Pressure-volume e compressibility (Hugoniot) curves for solid and porous copper, showing comparison of pore collapse relations obtained using various dynamic snd static models (Tong and Ravichandaran, 1997)
application of these processes is the prediction of the magnitude and duration of the shock pressure required to fully consolidate the powders, or to initiate the chemical/structural changes for synthesis of new alloys/phases. A large amount of energy is dissipated in plastic deformation and crushing of the powder particles during shock compression. Hence, at the micromechanical scale, shock compression of powders is considered to involve the collapse of voids (or pores) under large external pressures. Various void collapse models, based on rate-independent (RI) and ratedepe~dent(RD), as well as perfectly plastic and elasticviscoplastic material considerat~ons,have been developed (see review by Tong and Ravichandran, 1997). Figure 13 shows a comparison of the pre~sure-volLi~e compressibility (Hugoniot) curves for solid and porous copper, and the pore collapse relations obtained using various dynamic and static models. It can be seen that the quasi-static ore-collapse curves, unlike the dynamic pore-collapse curve, are far below the Raleigh line defining the shock pressure required to achieve full density, In-situ measurements of stress-wave propagation have also revealed that the ‘crush-up’ or compression of powders to solid density produces complex wave loading characteristics (Anderson et al., 1994). The measured shock-wave rise times are observed to vary from a few tens to several hundreds of nanoseconds, depending on the magnitude of the
shock wave, Consequently, while rapid~loadingrates at high pressures make it necessary to incorporate ratedependent considerations, long rise-times at low pressures alter the otherwise prompt thermal effects and ~ ~ d r o d y ~ a considerations mic assumed in theoretical treatment of the shock state, A realistic analysis of shock-compression effects is, therefore, extremely complex.
The criterion for full-density cornpaction has been based on correlating the d e n ~ i ~ c a t i shock o i ~ pressure, via the dynamic crush strength of a given powder, to the yield strength (or hardness) of the solid material by a simple factor, as shown in Figure 14 (Meyers et aE., 1993). Alternatively, it can be taken equal to the crush strength measured from quasi-static or dynamic compaction experiments. Consolidation of the powders actually occurs by the energy of the shock wave, preferentially dissipated in collapsing the voids, in shearing the particles during local particle-particle impacts, and in deformation and friction sliding of the particles. The resulting energy partitioniiig between the surface and the interior of the powder particles produces interparticle welding due to localized melting of particle surfaces. Various models have been proposed describing the mechanisms of energy dissipation and partition in^ (Meyers et al., 1993; Schwarz et al., 1984; Courdin, 1986; Vreeland et al., 1986), in 200
160
0
s c-BN
10
20
30
40
60
80
70
80
SO
100
Correlation of shock pressure required for densification plotted as a fbnction of hardness at several distensions (a) defined as the ratio of final and initial compact density (Meyers et al., 1993)
739
Novel Synthesis Techniques order to determine the optimum shock pressure and duration required for obtaining well-bonded, fully dense compacts. However, t e dissipation and partitioning of energy is not only controlled by the shockloading conditions and the green density of the initial powder compact, but also by factors such as the particle size, shape, and surface texture, which are difficult to incorporate in such mechanistic models. Two other major issues need to be addressed to successfully obtain well-bonded, fully dense powder compacts, Micro- and macro-scale cracking and the non-uniformity of microstructure, both occurring due to shock-wave reflections, the resultant generation of tensile stresses, and wave interactions with powder containment and free surfaces. A two-dimensional analysis of wave-propagation effects, combined with rate-dependent models of shock~compaction,are therefore, essential to capture realistically the stress states and deformations experienced by the powders during consolidation. The results of two-dimens~onalanalysis can then be used to design shock compaction fixtures and employ densification parameters such that the powders are subjected to desired stress histories without allowing the radial compressive (re-shock) and tensile (release) waves to enter the compacted material. A cylindrical implosion geometry, employing explosives or a planar-pressure impact assembly with explosive or gun launched projectiles, are the two common inethods employed for shock-compaction and shock-synthesis experiments. Two-dimensional numerical simulations performed on bath these geometries indicate that the actual loading conditions on the powders are often an order of magnitude higher than those predicted by calculations based on one-dimensional wave propagation. Consequently, the recovered coinpacts exhibit various types of cracks and variations in the microstructure of the compacted materials in different regions along the conipact thickness and radius. In recent years, shock consolidation has been applied for making bulk compacts of nanocrystall~ne alloys. Jain and Christman (1994) performed gas-gun experiments and obtained fully dense, well-bonded compacts (32mm diameter by 6mm thick) of nanocrystalline Fe-28A1-2Cr powder prepared by ball milling. The nanophase intermetallic (Fe,Al) had a grain size of -80nm; and, in tension, it exhibited brittle behavior with a failure strength of 0.65 GPa, comparable to a coarse-grained intermetallic with similar composition. In compressio~,the nanophase material exhibited superplastic-like flow during roomtemperature quasi-static deformation to true strains greater than 1.4. The compressive flow strength was
measured to be 2.1 GPa, and no macroscopic strain hardening was observed; however, further refinement of microstructure (to 10 nm grains) was observed. Korth and Williamson (1995) and ~uryanarayanaet al. (1997) used the cylindrical implosion approach with a double-tube (or flyer-tube accelerated inwards with explosives) geometry, and showed the best retention of nanostructure in mechanically alloyed Ti-24A1-11 (Ti(A1,Nb)3. (Ti,Nb),Al) and Fe(N) powders after shock consolidation in contrast to conventional methods of powder compaction. The Ti-alloys showed microhardness values OF 960 DPH (diamond pyramid hardness), while the microhardness of Fe-alloys was 1020 DPH. Nauocrystalline B2-pliase NiAl alloy, synthesized by ball milling of Ni and A1 powder mixtures, has been shock consolidated using the gas gun (Chen et al., 1999). TEM characterization of the nanostructured NiAl compacts showed extensive residual defect structures, including distorted grains, dislocations and shear bands, indicating that non-homogeneous deformation during ball niilling and shock compression is what causes dynamic recovery and generates the nanocrystalline structure. As shown in Figure lS, the niicrohardness of the NiAl compacts varied inversely with grain size; even in the nanocrystalline range, and was observed to be significantly greater than that of conventional grain size NiAl alloys. The Hall-Petcli strengthening was attributed to be dominated by the N
N
8
7
N
Figure 15 Wall-Petch plot of hardness versus grain size for §hock-consolidated nanocrystalline NiAl along with other data for submicron grain-size NiAl (solid points - Chen ei al., 1999; open circles - Smith, 1995; and open squares Nash et at., 1993)
dislocation ledge meclianisin~ similar to the results observed by Jain and Christman (1994) on shockconsolidated nanocrystalline Fe-28A1-2Cr alloy (E;e,Al). Shock consolidation of Ti-Si amorphous powders and subsequent devitrification of the product to form n a n o c r y s ~~lli~e alloys has also been employed. Vamasaki et al. (1994) shock compacted amorphous Tiloo-,Si, (s=10-60 at.%) powder mixtures to produce partially crystallized Ti-Si alloy coinpacts of various compositions. Upon subsequent annealing at 1223K, the compacts fully crystallized into 15-20 nm size crystallites, yielding a microhardness as high as 1700VHN recent work on shock consolidation of mechna amorphized 5Ti+ 3Si and its subsequent crystallization at temperatures in the range of 800” to 1200°C (with 1 to 12 hours hold time) showed formation of a sing1e”ph~seTi5Si3compound with an ultrafine grain microstructure (Counihan et al., 1999; Glade and Thadhani, 1995). The average grain size changed from 50 nin upon heat treatment at 800 “C for 1 hour, to -160nm after 1200°C for 3 hours; however, it remained stable ( 115-125 nm) during annealing at a constant temperature of 1000°C and with increasing heat treatment time from 1 to 12 hours. ~ ~ - scrystallization i € ~ studies performed by heating the shock- consolidate^ samples in the TEM at temperatures up to 900”C, revealed that the rapid rate of conversion of the amorphous to crystalline phase was number density of nucleating crystalvia significant growth of existing crystallites, since their growth is inhibited by the inip~ngement of the crystals. Vickers microhardness ~easurementsshowed values of 1200-1400 kg/mm2 for grain sizes ranging from -60 to 160nm. While these microhardness values are 80% higher than those for microcrystalline, shock~densified Ti5Si, alloy, the fracture toughness values were measured to be -24 ~ ~ a which ~ m ,are ty ical of those of brittle ceramics.
-
N
-
5.3 hock compression of powders and powder mixtures also results in various types of mecha~ical,physical, and chemical efTects. A large number of defects are introduced in the powders due to the kinetic energy of the shock pulse. Extensive plastic deformation, fiuidlike turbulent flow, heating, particle commiiiution, and mixing of constitueiits with fresh and cleansed surfaces are all possible. These eEects significantly alter the
mechanical, physical, and chemical characteristics of powders, thereby influencing their solid-state reactivity and ~niti~ting ‘shock-induced’ structural and chemical phase changes o ~ c u ~ r i ningtime-scales of ~icrosecond duration of pressure e~uilibration,or simply generating shock-activated ‘m~crostructurallymodified’ states. The conditions e~counteredin the shock process are not achieved in any other environment, hence, the shock-compression process has the potential of yiclding novel compounds (formed via shock-initiated chemical reactions), metastable high~pressurephases, and materials with uniquely modified microstructures. Synthesis of high-pressure intermetallic phases is described by Levinsky (see chapter by Levinsky in this volume). The fundamental mechanisms controlling ‘shock-induced’ chemical reactions in powder mixtures and leading to synthesis of new compounds are dominated by processes occurring during the stress-pulse rise time of a few to more than a hundred ~anoseconds and the microsecond~~uration of the peak pressure state. The critical processes have been described by Dremin and Breusov (1968) and by e con~gurationa~ Graham (1989) to i n ~ l ~ dparticle changes caused principally by plastic deform~tion and flow (less often, by fracture and f r a ~ ~ e n t ~ t i o n ~ , mixing of the reactant powders within and around the voids by plastic flow of the reacta~tsand dispersion of fragments of the reactants, enhanced solid-state reactivity of powders, and temperature increases. The deformation and chemical reaction process is one controlled and a~cel~rated by physical variables that are highly heterogeneous. Under such conditions, chemical reactions have been observed to occur in powder mixtures during the high-pressure state before unloading to the ambient pressure in the microsecond time-scale of pressure equilibration. Chemical reactions Can also occur at times after the sample is in the ambient, post-shock state in the shockmodified (activated) material due to bulk (resi~ual) shock-temperature increases in time-scales of thermal equilibration. This post-shock effect can occur by mechanisms similar to the more conventional S type co~bust i on reaction processes descri~ed in Section 4, but it is believed to be dominated by shock-activated solid-state diffusion and theimal quenching. While the two types of chemical reactions - defined respectively as ‘shock-indu~ed’and ‘shock-assisted, chemical reactions (Thadhani, 1994) -have been distingu~shedon the basis of the t infer time period over which they occur, it is d i ~ c u lto from study of samples preserved for post-shock analysis (‘recovered samples’) alone, whether the
Novel Synthesis T ~ c h ~ i ~ u e ~ ~ observed reactions occur during the high pressure loading and unloading conditions, or if they occur after unloading to ambient conditions. Reactions occurring in powder mixtures are generally highly exothemic, and are accompanied by rapid temperature increases, often melting the reaction products. Thus, recovered samples show reaction products with microstructures typical of melted and solidified materials that mask the characteristic features of the shockcom~ressedpowders at the time reaction is initiated. In samples obtained from shock-compression recovery experiments, it 1s not possible to ascertain dirmtly the kinetics or the mechanisms of processes leading to sho~k-inducedchemical reactions. Post-shock microstructural characterization of the recovered materials reveals the final state of the product after e~uilibration with the env~ronment.Furthermore, the large exothermicity of the reaction often results in melting of the products, leaving no evidence of how and when the reaction may have occurred. The final structure simply reveals characteristics typical of a melted and resolidre 16 shows examples of typical compounds formed via shockinduced chemical reactions in powder mixtures of Nii. A uniform dendritic, cellular, or equiaxed microstructure^ with the presence of spherically shaped voids (indicatin~possible gas escape or shrinkage due to volume changes) is typical of a fully reacted material. hock-induced chemical reactions are generally accompanied by relatively small changes in bulk properties. Thus, most conventional pressure and velocity mea$urement systems fail to respond accurately to reaction rate measurements. Rapid temperature increases, which are commonly associated are the only direct with such e ~ o t h e ~ ireaction^^ c change acco~panyingthe chemical reaction. However, it is difficult to d~stinguishthe reaction temperature increase from the heterQgeneoustemperature increases associate^ with shock compres~ionof powders. The most revealing and comprehensive results providing evidence of shock-induced chemical reactions include Hugoniot measurements perfomed by Batsanov et al. (1994) on stoichiometric mixtures of tin and sulfur auges to obtain records of the shock profile. At pressures > 15CPa the measured pressure points were observed to deviate towards i o t indicatincreased volume an the P-V H ~ ~ ~ o nplane, ~xpansion caused by the highly exothemic chemical reaction. tress-wave measL~re~entswith PVDF (Poly-Vinyl DiFluoride) gau used to study shock init~at~on of reaction in Ti-Si
741
Figure 16 Micrographs showing examples of typical m i c r o s t ~ c t u ~of~ s~ n t ~ r ~ ~c to am~~ o~ ui ~cdformed s via shoc~-inducedreactions in powder mixtures of (a) NGAI, (b) Ni-Si, and (6) Ti-Si
powder mixtures of various powder particle msrphologies, at shock pressures up to 5 1997). Compression of the Ti density was observed to occur at -1GPa pressure. With increasing pressure, the compressibility shifted to larger volume and an increase in volume to as much as 20% was observed at 5 GPa pressure, as illustrated in
742
Research ~ e c h ~ i q ~ e ~ ~ 8 7
0.8
R E ~ ~ ~ L ~ M E
i~~~~ 17 (a> Stress versus relative volume for Ti-Si powder mixtures of different morphologies, indicating cnish-up to hill density at - 1 GPa followed by volume expansion, showing evidence of shock-induced reaction in the case of medium morphology powders, while fine and coarse morphology Ti-Si powders remain unreacted (Thadhani et ul., 1997)
Figure 17(a). The volume increase is attributed to the thermal expansion due to sudden temperature increases caused by the chemical reaction. Higher shock-wave velocities were also measured for input stresses greater than 1 CPa. The combination of these results provides conclusive evidence of chemical reactions occurring during shock compression and resulting in the formation of Ti-Si intemetallic c o ~ p o u n d in s time scales of mechanical equilibration. The measurements also showed that the Ti-Si powder mixtures of different morphology have different crush stre~gths( ~ i ~ u 17(a)), re and therefore, reveal different reaction initiation thresholds. Similar results have also been observed by Vandersall and Thadhani (2000) in recent studies of shoc~-initiatedchemical reactions in MO-Si powder mixtures. Figure 17(b) shows densification behavior at P <4 CPa and shock-induced reaction at P=4--6CPa, which is evidenced by data points approaching the reacted product Hugoniot state. No reaction is revealed in experiments at P > 6 GPa, due to premature melting of silicon. t~ of stress and wave speed in The i ~ - s imeasure~ents Ti-Si and MO-Si powder mixtures have illustrated that the ‘crush-strength’ of powders (i.e the stress at which powders densify to full density) is the parameter that most i n ~ u e ~ ~shock-induced es reaction initiation^ irrespective of the physical/themal properties of reactants~pro~ucts. ~onsequently,if agglomeration of
1
1.2
1.4
f.B
1.8
2
2.2
Figure 17 (b). Pressure versus volume plot for Mo-tZSi powder mixture, showing data points following three different trends: {i) P-a d e n s ~ ~ c a t i ~behav~or n at P<4GPa, (ii) reacted powder Hugoniot state, due to shock-induced reaction, at P = 4-6 CPa and (iii) states close to that of unreacted inert powder mixture at P > 6 GPa. Lack of reaction at P> 6 GPa is attributed to premature melting of Si (Vandersall and Thadhani, 2000)
fine submicron-size particles, or fracture and entrapment of coarse particles, or even p r e ~ a t ~melting re of the low-melting point constituent (e.g. Si) occurs prior to the intimate mixing of the reactants at stresses less than the crush strength, then shock-induced chemical reaction will not occur in the microsecond-duration shock state. These results demonstrate that the initiation of shock-induced chemical reactions is influenced by processes that promote simultaneous deformation and intimate mixing of reactants, leading to product formation via a displacive mechanism. Other theories of shock-i~ducedreaction initiation based on the ‘energy threshold’ criterion have also been proposed (Yu, 1995; Montilla 1997; Vreeland, 1998). Meyers and his co-workers (1994) have proposed that if the energy generation due to the chemical reaction is greater than the energy dissipated by thermal conduction, a steady-state reaction can start from local ‘hot-spots’ and propagate into the interior of the particles. Accordingly, critical molten hot-spot regions were calculated, based on a shock-energy threshold corresponding to the mean bulk temperature, which must be above that required to initiate reactions at ambient pressure. However, tlie shockenergy threshold criterion is based on time-scales of temperature e~uilibration corresponding to millisecond times. Thus, the ‘energy threshold’ criterion and the ‘hot-spot’ initiation m echani s~ may be
Novel Sy~thesisT e c h ~ ~ ~ ~ e s
applicable to shock-assisted chemical reactions occurring after unloading to ambient pressure, but not to shock-induced chemical reactions occurring during the shoc~-compressionstate in time-scales of pressure eq~~ilibration corresponding to a time duration of a microsecond. It is well established that the defects generated clue to shock compression can significantly modify and enhance the solid-state reactivity of powders, as described in the reviews by Duvall(l984) and Craham (1993). Brittle ceramics, and even silicon, undergo significant grain size reduction via fracturing or by generation of sub-grain structures during shock compression. Increased mass transport rates are possible in shock-compressed materials due to introduction of defects and creation of new paths for motion of point defects along grain boundaries. Such characteristics play a vital role in enhancing the solidstate chemical reactivity of powders and their mixtures, essentially by creating a shock-modified material. Various attempts have been made to advantageously utilize the enhancement in reactivity of shock-modified materials by post-shock, controlled-rate thermal treatments; successful examples of these include sintering of ~ifficult-to-consolidateoxide and non-oxide ceramics, improving catalytic activity of materials, enhancing the kinetics of nucleation of precipitation-strengthening phases or other types of metastable high-pressure phases (see reviews by Craham and Thadhani 1993, Thadhani and Aizawa 1997, and Thadhani, 1999). Shock modification also provides a state in which compacts of reactive powder mixtures can be used for mechanochemical synthesis via solid-state, defectenhanced chemical reactions.
Mechanically activated reaction synthesis (or mechanochemical synthesis), in which ball milling or shock compressioii are used to enhance the chemical reactivity of powders and increase their ability to undergo solid-state chemical reactions, has received significant interest since the eighties. Mateazzi et al. (1993) have suggested that free radicals, deformed bonds, and ions and free electrons are generated at newly formed, nascent surfaces during impact between colliding balls or during shock Compression, giving rise to a so-called 'triboplasma' (Thiessen, 1986). In addition, intimate mixing and grain-size refinement due to extensive plastic deformation and subgrain formation, lead to defect-enhanced solid-state d ~ ~ u s i o nand hi~hly
743
accelerated reaction kinetics. Mecha~osynthesiscan thus lead to formation of amorphous and nanocrystalline intermetallics, immiscible alloys and compounds with increased solid solubilities. Highly activated states of reactant powders can also be produced by niechanosynthesis, for use as precursors for subsequent reaction synthesis. Figure 18 shows an example of how ball milling of Ni+Al powder mixtures (1 :1 atomic ratio, -325 mesh) results in partial alloying occurring during ball milling. The reaction onset temperature and heat of reaction, measured upoii heating Ni+Al powders in a DTA (differential thermal analyzer), are plotted as a function of the ball-milling time. It can be seen that the heat of reaction (whicli provides a measure of the exothermicity and extent of reaction) decreases with increasing ball-mil~ngtime, and practically no heat release is observed for powder mixtures milled for more than four hours. The reaction onset temperature also decreases with increasing milling time. The extensive cold-working of powder particles and their intimate mixing during ball milling, promote solidstate diffusivity between the powders, ~ausingreactions to occur upon subsequent heating at significantly lower temperatures, until the reaction occurs during ball milling. Similar effects have also been observed by Hida and Lin (1990) during grin~ingof Si0,-A1 therniite mixtures in a high-speed planetary mill, Mechanical activation of powder ~ i x t u r e sby ball milling is also known to i ~ ~ u e n cthe e character of combustion-type SHS reactions. The enhanced reactivity in the ~ e c h a ~ i c a l lmilled y powder mixtures influences the reaction mechanisms and kinetics, which
Figure 18 Variation of heat of reaction and reaction onset temperature during heating of Ni + Al powder mixtures as a function of ba1l"~il~ing time ( ~ h a d ~ a n1993) i,
in turn aRect the composition and microstructure of the synthesized products. It has been shown that the propagation rate of SHS reaction increases with milling time and results in compacts with higher density and extremely fine microstructures, in contrast to those produced by SHS reactions without prior milling (Thiessen, 1986). Figure 19 shows the variation in the propagation rate of SHS reaction as a function of milling time in 3Ni+Si powder mixtures using a planetary mill with two different milling intensities (Lagerboni et al. 1999). It can be seen that the enhancing influence of milling time on the SHS propagation rate is fairly linear for 200rpm samples up to 5 h of milling, but it rapidly decreases with further milling, until no SHS reaction ignition can be observed beyond 7 h milling. The 160rpm milling shows a similar trend but with a lesser influence. The ball-milled and ~ ~ S - p r o c e s s ematerials d consisted of Ni3Si matrix and Ni,,Si12 phase islands that actually underwent phase transfomatio~to Ni,Si upon annealing. Mechanical activation is also known to change the ther~odynamicsof reactions as well as increase the mutual solubilities of reactants. For example, Gras et d.(1 999) have shown that mechanically activated Fe+2Si powder mixtures undergo SHS reactions at 400°C resulting in the formation of equiatomic FeSi and /I-FeSi,. Likewise, Charlot et al. (1999b) have formed a nanocrystalline FeAl intermetal~icalloy (70S reaction in mechanically-activated nano~rystallineNbA1, intennetallic with
grain size of -60nm has also been observed in mechanically activated, and SHS-reacted Nb and A1 powder mixtures by Gauthier et al. (1999). Timeresolved XRD analysis of reactions in this system showed ignition of an SHS reaction following 38s of melting of A1 (at 668"C), and su~sequentalloying within 27s of reaction ignition at a temperature of 972°C. However, it is not known from these studies how the high-temperature phase and the nanocrystalline phases can be retained in products of reactions which generate large temperature incre ~eclianochen~icalreaction synthesis of shockactivated powders has also been de~onstratedby a number of investigators. Use of shock compression for m ~ h a n o ~ ~ e m i activation cal cif powders has an added advantage that a densely packed configuration with 75%--95% dense compacts of reactants in intimate contact with each other is produced. Consequently, it i s possible to control the reaction mechanism and initiation conditions such that the reaction-synthesized products retain a high compact density and are free from defects associated with liquid phase formation. Mecha~ochemicalreaction synthesis of shock-activated (or modified) powder mixtures was first etral. demonstrated on Ni-A1 mixtures by ~ a m m e t t ~ (1 988). The post-shock reaction behavior of 3Ni + A1 powder mixtures, studied using differential thermal analysis, showed that while unshocked powders underwent an exothemic reaction with the melting (eutectic 650 "C, shocked mechanical formation) of A1 at mixtures revealed a dominant 'pre-initiation' exotherm at SS0 "C corres~ondingto a solid-state reaction, prior to the main 650°C exotherm. ~ u r t h e ~ o r e , 3Ni + AI powders o f different morphology always exhibited the single reaction exotherm following the melting of Al, but shocked powders showed the intensity of the solid-state reaction peak increasing with powder irregularity, to the extent that shockcompressed 3Ni+Al mixtures of powders of flaky morphology showed complete reaction occurring in the solid state (Thadhani et al., 1992). The flaky powder morphology mixtures are m~chanoch~mically (shock) activated more than mixtures of powders of fine and rounded-coarse morphology powders due to the more extensive d e f o ~ a t i o nand intimate mixin achieved with the irregular particle morphologies during shock compression, Similar results of enhanced solid-state reaction behavior of m~chanochemically activated, shock-com~resse~ powder mixtures have also been observed in other aluminide- and silicideforming powder mixtures (Dunbar et al., 1993)) and
-
--
I
n of SHS reaction are 19 Variation of p r o p a g a t ~ ~rate as a function of milling time in 3Ni + Si powder mixtures using a planetasy mill with two diff'erent milling intensities (Lagerbom et al., 1999)
-
'345
Novel Synthesis T e ~ ~ n ~ ~ ~ e s have even been used to fabricate simple, near-netshaped components. The reaction behavior of shock-densified Ti-Si powder mixtures ( ~ 2 ~ average O ~ m size) has been modeled based on experimental determination of activation energies (obtained from DTA analysis) and analysis of reacti~n~ineticsbased on reaction fraction measurements performed on shock-densified samples reacted in a furnace at different temperatures and hold times ~ a i ~ J o s and h i Thadhani, 2000). As shown in Figure 20, complete reaction occurs in the solid state at temperatures as low as 1000 "C,and liold times of 5 hours. At higher synthesis temperatures, the reaction almost insta~taneouslyreaches completion, due to being taken over by the combustion-type SHS process as the rate of heat evolution exceeds the heatdissi~ationrate at temperatures greater than 1000 "G. The apparent activation energy for the solid-state reaction in the dynamically densified powder compacts was deter~inedto be 93 kJ/mol to 123kJ/mol, for the Ti-Si compacts made at 5-7 GPa peak shock pressures. A predictive model incorporating heat and mass transport balance and the measured kinetic parameters, has been developed to establish the dominant reactioii mechanisms and to deteirnine how the reaction mechanism is influenced by variables 4
30 min
rnin
1
including: porosity in the dyna~icallydensified powder compacts, powder particle size, and the degree of shock activation. The results illustrate that the particle size and porosity dominate the reaction m e c ~ a n i s ~ (solid-state versus combustion~type)by controlli~gthe temperature above which the reaction may be taken over by co~~bustion synthesis. In cont~ast,reduction of the activation energy via shock compression influences the kinetics of the solid-state diEusion reaction, aiid Ti,Si, intermetallics with highly refined microstructures (<5pm grain size) and a microhardness of 800 kg/mm2 are realized. The predictive model can be applied to any type of reactive powder mixture compact to determine the reaction synthes~scon~itions required for ensuring that bulk synthesis occurs in the solid state. N
Significant advances have been made during the last several years in the development of' intermetallic processing technologies employing reaction synthesis approaches. Time-resolved diagnostics have been used to further the understandiiig of reaction ~ ~ e c h a n i s i ~ s during combustion synthesis, mechanical alloying, and 1 hr
3 hr
8
9
4 hr
5 hr
0.9 0.8 0.7
0.6 0. 0. 0.3
0.2 0.1
0 5
6
7
10
11
1
(time) Con~parisonof reaction fraction data obtained from react~on-syn~esis experiments on shoc~-densi~ed Ti-Si powder mixtures performed in a furnace, correlated with data obtained from Carter's solid-state diRusion and Jo~nson-Mehl-Avrami (JMA) kinetics models using activation energies determined from DTA experiments. The comparison of experiments with model calculations illustrates that solid-state reactions dominate the synthesis process at temperatures up to 1000"C, and at higher t e ~ p ~ ~ d t u rthe e s ,reaction occurs ~nstantaneo~sly by combustion synthesis ( ~ a i ~ j o and s h ~T~adhani,2000)
746
Research Techniques
shock-induced reaction synthesis. Devitrification of mechanically amorphized powder compacts prepared by pulsed-electrodischarge or shock-consolidation techniques, and novel variations of reaction processes including field-assisted reaction synthesis, have enabled process optimization for achieving intermetallics with unique microstru~tural characteristics. igh-pressure tuning has been used in the search of new ductile intermetallics by employing pressureinduced structural phase changes to form the more ductile L 12-type phases. ~echanocheniical reaction synthesis has been extended to fabricate various types of functional intermetallics, nanocrystalline alloys, and inte~metallic-cera~ic composites. The process employs ball iiiilling and shock compression to mechanically activate precursor powders to enhance their chemical ~eactivityand cause reaction synthesis to occur via controlled ~ e c h a n i s ~ats ,lower temperatures, and in significantly shorter time-scales than in conventional processing. ~echanochemicalreaction synthesis also permits the formation of thermodynamically stable non-equilibrium phases and compounds with a wide range of solid solubilities. ~nnovationsin fabrication of inte~metalliccompounds will require combinations of processes, such that specific benefits of various synthesis processes can be exploited in a manner that allows intermetallic alloys to be designed and fabricated with tailored microstructures.
The author acknowledges the National Science Foundation (Dr. Bruce ~ a c d o ~ ~and l dthe ) Army Research OEce (Drs. Andrew Crowson, Ed Chen, and William ullins) for the funding and support for the author's research activities over the last 15 years. The author is grateful to his many collaborators, and in particular raham (forme~lyat Sandia National Laboratories) and Professor Marc A. Meyers (University of California at San Diego) who have been his mentors and guides. The sincere efforts of past and present graduate students is also acknowledged.
Abdellaoui, M., Barradi, T., Gaffet, E. (1993). J. Alloys
. (1999a). J . Alloys Compds.
Aizawa, T. and Tokumitu, K. (1999b). Mater. Sci. Forum, 312, 13-22. Anderson, M. U., Graharn, R. A., and Holman, G . T. (1994). In ~ig~-Pressure Science and Techpzology - 1993, AIP Conf. Proc. 309, Part 2, (eds S. C. Schniidt, J. W. Shaner, 6. A. Saniars, and M. Ross). AIP Press, New York, pp. 1 1 11-1 114. Atzmon, M. J. (1989). In Proc. of T M S S y ~ p o s i u mon Solid State Powder Processing, (eds A. H. Clauer and 1. J. d ~ ~ a r b ~ d i TMS, l ~ o ) ,pp. 173-179. Badding, J. V. (1998). Annu. Rev. Mater. Scr. Balogh, J., Bujdoso, L,, Kerneny, T., and Vi J. Phys. Condens. Matter, 9, L503-L508. Batsanov, S. S., Cogulya, M. F., Brazhinikov, M. A., Sirnakov, G. V., and Maksimov, I. I. (1994). Comb. Expl. Shock W ~ V ~36,S 361-365. , Bertolino, N ., Anselmi-Tarnburini, U Munir, Z. A. (1999). J. Alloys Comp Bhaduri, S. B. (1999). Ceramic Eng. Sc 206. Bhaduri, S and Bhadun, S. B. (1999). In ~ o n - ~ ~ z ~ i l i b r z ~ Processing of Mater~als, C. Surya~arayana (editor), Pergamon Press Materials Series, pp. 289-309. Bordeaux, F., and Yavari, A. R. (1990). J . Mnter. Res., 5(8), 1656. Bose, A., Rabin, B. H., and German, R. M. (1988). Powder MQtdl. fnt., 20, 25. Charlot, F., Bernard, F., Gaffet, E., Klein, D., Niepce, J. C. (1999a). Acta Mater., 47(2), 619-629. Charlot, F., Gaffet, E., Bernard, F., Gras, C., and Niepce, J. C. (1999b). Mater, Scr. Forum, 312-31 Chitralekha, J., Raviprasad, K., Gopal, E. S. R., and Chattopadhyay, K. (1995). J. Mater. Res., 10, 1897. Chen, T., Hampikian, J., and Thadhani, N. N. (1999). Acta Mater., 47(8), 2567-2579. Counihan, P. J., Crawford, A., and T h a d ~ ~ nN, i , N. (1999). Mater. Sci. & Eng., ~ 2 6 7 26-35. , Cuurtney, T. H., and Msurice, D. (1996) Scr@ta Mater., Deevi, S. C. (1991). J Mater. Scr. 24 1-25 1. Deevi, S. C. (1992). Mater Sci. & Dela Torre, S. D., Ishihara, K. H. P (1999). Mater. Scr. & Eng., ~ 2 6 6 3, 7 - 4 3 . Desch, P. B., Nash, P., Schwarz, R. B. (1995). In A~~licatiopzs of Thermodypzamics in the Synthesis and Processing of Materjals, TMS Fall Meet Proc., The Minerals, Metals & Materials Society (TMS), Warrendale, PA, p. 75. Desch, P. B., Schwarz, R. B., and Nash, P. (1996). S c r ~ t a Mater., 34(1), 37-43. A, (1999). Metall. and Mater, Dremin, A. N., and Breusov, 0. N, (1968). Rus.sian Chemicul Reviews, 37 (3,392-402. Dunbar, E., Thadhani, N. N., and Graharn, R. A. (1993). J. of Mater Sci., 28, 2903-2914. Dunmead, S. D., Holt, El., and Kingmaii, D. D. (1990). In ~ o ~ b u s ~ i oand p z P l a s ~ aSynthesis qf H i g h ~ @ m p ~ ~ n t u r ~
Novel S y n ~ h e s ~Techniques s M~~terials,(eds Z.A. Munir and J. B. Holt), VCH Publishers, New York, p. 229 Durovic, I),,Kostic, E., Kiss, S. J., Zec, S. (1998). Journal of Alloys and Comp., 279, pp. Ll-L3. Duvall, G. (1984). Shock-Compression Chemistr.y in Materials Synthesis and Processiag, NMAB Report NO. 414, National Academy Press, Washington DC. Eckert, A., Schultz, L., Urban, B. (1990). Europhysics Letters, 13(4), 349-354. Feng, A., and M u m , Z. A. (1995). Metall. Trans. 26 586 and 587-593. Feng, A., and Munir, 2;. A. (1997). J . Amer, Ceram. Soc., 80, 1222-1230. Fu, M., and Sekhar, J. A. (1998). J. Amer. Ceram. Soc., 81(12), 3205-3214. Gauthier, V., Bernard, F., Gaffet, E., Josse, C., and Larpin, J. P. (1999). Mater. Sci. Eng., A272 334341. Gedeva~shvili,S., and Munir Z. A. (1998). Mater. Scl. Eng. A , A242, 1-6. Glade, S. C., and Thadhani, N. N. (1995). Metall. and Mutls. Trans., 26A, 2565-2569. Graham, R. A. (1989). In Proc. of 3rd Internat~onal Symposium on High ~ y n a m i cPressures, edited by R. Charet, Assoc. Francaise de Pyroteclinie, Pans, pp. 175180. Graharn, R. A. (1993). Solids Under High Pressure Shock Compression: Mechanics, Phjwcs, and Chemzstry, Springer Verlag. Graham, R. A., and Thadhani, N. N. (1993). In Shock Waves in Materials Science (ed. Sawaoka, A.B.). Springer-Verlag, 35-65. Gras, C., Gaffet, E., Bern d Niepce, J. C. (1999). Mater. Sci. & Eng., A2 Grebe, H. A., and Tbadhani, N. N. (1999). J of Mater. Syn. Proc., 7(1), 49-61. Gourdin, U'.H. (1986). Prog. Mater. Sci., 30, 39-80 Hainmetter, W. F., Graham, R. A., Morosin, B., Horie, Y. (1988). In Shock Waves IE Condensed Mutter, (eds S. C. Schmidt and N. C. Holmes). Elsevier Science Publishers B,V., p. 431 He, L., and Ma, E. (1996). J. Mater. Res., 11, p. 72. Hida, G. T., and Lin, I. J. (1990). In Comb~stionand Plasma Synthesis of High Temperature Materials, (eds J. B. Holt and Z. A. Munir). VCH Publishers, p. 246. Hoke, D. A., Meyers, M. A., Meyer, L. W., and Gray, 111, G. T. (1992). Met. Trans., 23A 77-86. Jain, M,, and Christman, T. (1994). Acta Metall., 42, 19011911. Kang, B., Ryoo, H. S., Hwang, W., Hwang, S. K., and Kiln, S. W. (1999). Mater. Sci. cft Eng., A270, pp.330-338. Khina, B. B., Baburaj, E. G., Froes, F. H. (1997). Adv. Particul. Mater. Processes, Metal Powder Industries Federation, Princeton, NJ, pp. 120-128. Korth, G. E., and Williamson, R. L. (1995). Metall. Trans., A, 26A, 2571-2578. Kuhii, W. E., Friednian, I. L., Summers, S., and Sragvari, A. (1984), ASM metal^^ hand boo^^ 7, 56-70.
747
Lagerbom, J., Tiainen, T., Lehtonen, M., Lintula, P. (1999). J. Mater. Sci,, 34, 1477-1482 LaSalvia, J. C., Meyer, L. W., Meyers, M. A. (1992). J. Am. Ceram. Soc., 75(3), 592-602. LaSalvia, J. C., Meyers, M. A., and Kim, I). K. (1994). J. Mater. Syrz. Proc., 2, 255. Li, T., Li, Y.-Z., and Zhang, Y.-W. (1997). J. Phys. Cond. Matter., 9, 1381-1388. Lu, L., and Lai, M. 0. (1998). Mechanical Allqying, Kluwer Press, Norwell, MA. La, L., Lai, M. O., and Zhang, S. (1994). J. Mater. Proc. Technol., 52(2-4), 539-546. Larikov, L. N., Omelianenko, V. V., Tishkevich, V. M. and Falchenko, V. M. (1973). In~titut M e t ~ l l o ~ z i l Kiev, ~i, pp. 1-16. Lee, J. H., and Thadhani, N. N. (1998). J . Mater. Res., 3(11), 3160-3 173. Magini, M., Basili, N., Burgio, N., Ennas, G., Martelli, S., Padella, F., Paradiso, E., and Sushi, P. (1991). Mater. SCE. EBg. A , AlM, 1406-1409. Matteazzi, P., and Farne, G. (1991) r at er. Res. BUZZ.,26(12), 1283-1290. Matteazzi, P., Basset, D., Miani, F. (1993). ~ a n o s t r ~ c t u r e ~ Materials, 2, 217-229. Maurice, D., and Courtney, T. H. (1994). Metall. cft Mater. Trans. A, 25A, 147-158. Maurice, D., and Courtney, T. H, (1995). Metall. & Mafer. Trans. A, 26A, 243 1-2435 and 2437-2444. McCormick, P. G. (1995), Mater. Trans. JIM, 36, 161. McDermott, B. T., and Koch, C. C . (1986). Scripta Metall., 20, 669-672. Merzhanov, A. G. (1990). In C o ~ b u ~ ~ t iand o n Plasma Synthesis of High- Temperat~reMaterials, (eds Munir, Z. A., and J. B. Holt). VCH Publishers, New York, pp. 1-18. Meyers, M. A., Shang, S. S., and Hokamoto, K. (1993). In Shock Waves in Ma~erialsScience, (ed, A. B. Sawaoka). Springer-Verlag, pp. 145-176. Meyers, M. A., Yu, L. H., and Vecchio, K. S . (1994). Acta Met. cft Mat. 42(3), 701-714, 715-729 Miyamoto, Y., Takahura, T., Tanihata, K., Tanaka, I., and Yamada, 0.(1990). Proc, of First U.S. Japan Workshop on C o m b ~ ~ ~ tSynthesis, ion Tokyo, Japan, p, 173. Montilla, K. L. (1997). S h o c ~Wave Processing of Tr~nsitional Metal Silicides, PhD Thesis, California Institute of Technology,. Moore, J. J., and Feng, H. J. (1995). Prog. Ma~er.Sci., 39, 1944. Mukasyan, A. S., Rogachev. A. S . , and Varma, A, (1999). Chem. Eng. SCE.,54(15), 3357-3367. Munir, Z. A. (1988). Ceramic Bull., 67, 2342. Munir, Z. A. (1992). Meta~l.Trans., 23A, 7-13. Munir, Z. A. (1997). Int. J. SHS, 6, 165-185. Munir, Z. A. (2000). Mater. Sci. dt Eng. A., Munir, Z. A. and Anselmi-Tamburini U. (1989). Mater. Sci. Rep., 3, p.277. Murty, B. S., and Rangmatban, S. (1998). Inter. Mater. Rev., 43(3), 1-141.
Narnjoshi, S. N., and Thadhani, N. N. (2000). Metall dL Thadhani, N. I?., Graham, N. N., Royal, T., Dunbar, E., Mater. Trans. B, J1(2), 307-316. Anderson, M. U., and Noltnan, G. T. (1997). J. Appt. Phys., 82(3), 1~13-1128. Nash ,P., Ur, U. C,, and Dollar, M. (1993). Proc. 4 Z n d Intl, Thadhani, N. N., and Aizawa, T. (1997). In pressure Con$ Struct. App1. Mech. A1loying, Vancouver, Canada, Sho~k-C~mp~ession of Solids - W , Response of ~ ~ ~ ASM International, Materials Park, OH, p. 192 P o r o ~ sSolids to Shock Loading, (eds L. E, Davison, M. Niiler, A., Kecskes, L. J., Kottke, T., Netherwood, P., Jr., and Shahinpoor, and Y. Horie), ~~ringer-ve~lag, pp. 257-288. Benck, R. F. (1988). ~xplosive Co~solida~ionof Thiessen, K. P. (1986). J. Chzm. Phys, & Phys., (France), covn~ustion S?nthe~~i~ed Ceramics: T i c and TiB,, 83(11-12), 717-24. (Ballistics Research Laboratory Report BRL-TR-295l), Tong, W., and ~avichandsan,G. (1997). High Pressure Shock Aberdeen Proving Ground, MD. Co~pressionof So1ids: ~ e s p o ~os e~ ~ i Porous ~ h ~ ~ ? 0 1 ~ to ds Nishida, T., and Urabe, K. (1992). fnt. J. SHS, 1, 566. Shock Loading, (eds Davison, L. et al.) pp. 177-204. Podlesov, V. V., Meszhanov, A, G., and Stolin, A. M. (1992). Vandersall, K., and Thadhani, N. N. (2000). In ShockJ . Eng. Phys. T h e ~ ~ o ~ h63, y ~1065 . , and 1156. C Q ~ p r e s ~ t of n co~densed after - 1999, Proc. of APS Rabin, B. H., Korth, G. E., and ~illiamson,R. L. (1990). J. Topical Group ~ e e t i n g , (eds M. D. Furnish, L. Am. Ceram. Soc., 73(7), 2156. Chnbildas, and R. Hixon). AIP, pp. 763-766. Rawers, J. C . , Govier, D., Cook, D. (1995). J. Mater. Syn. Vama, A. (2000). ScientiJic A ~ e r i ~ a283(2), n, 44-47. Proc., 3(4), 263. V a ~ aA, , ~ogachev,A. S . , Mu~asyan,A. S., a n d ~ w a nS.~ , ~ogachev,A. S., Shugaev, V. A,, K h o ~ e n ~I.o A., , Vama, (1998). A ~ Chem. V ~ Eng. A., Kachelmyer, C . (1995). Comhust. Sri. Tech., 109, 53. Vreeland, T., Jr., Mutz, A. Sahu, P. C., Chandra Shekar, M., Yosouf, ET., Govinda In ~ e ~ a l l u r g i c ~a l~ p ~ i c a t i oofn Ss h O c ~Wave and HighRajan, K. (1997). Phys. Rev. Lett., 78, 1054-1057 Str~in-Rate ~ h e n o ~ e neds. ~ , L. E. Mum, K. P. Schaffer, G. B., and McCormick, P. G. (1991). Metall. Tmns,, S t a u ~ h a ~ mand e ~ ,M. A. Meyers, Marcel Dekker, 22A(12), 3019-3024. pp. 231-246 and 247-269. Schropf, W., Kuhrt, C., Arzt. E., and Schultz, L. (1994). Vreeland, T., Jr. (1998). S c r ~ ~Met., ta 3 ~ ~ r ~ e~ t ut ~ al30(12), ., 1569-1574. Whittenberger~J. 2). (1990). In Solid Sta Schwarz, R. B., Kasiraj, P., Vreeland, T., Jr., Ahrens, T. J. (eds A. H. Ciauer, and J. 3 . DeBarbadillo). TMS, (1984). Acta ~ e t a l32, ~ . 1243-1252. Warrendale, PA, pp. 137-155. Smith, T, R. (1995). ~ a ~ eRes. r . Soc. Syvnp. Proc., 362, 245. Xia, S . K., Saitovitch, H., Silva, I?. R. J., Comez, J. A., Rizzo Song, H. Y., and wan^, X. (1994). m at er. & an^^ Proc., Assuncao, F. C., and Baggio-Sattovitch, E. (1998). J. 9(1), 75-87. Phys. Condens. Matter, 10, 3457-3466. Suryanarayana, C. (1995). Inter, Mater. Rev., 40, p. 41. Xue, W.,Vandersall, K., Carrillo~Heian,E., Tha Suryanarayana~C. (1996). Metals and ~ a t e r ~ a2,l p. ~ ,195, and Munir, 2. A. (1999). J. Am. Ceram. Soc. §uryanarayana, C. (editor) (1999). ~ o n - ~ ~ u i l i ~ r i u m 1446. s, Press Materials Series. Processing of ~ a t e r ~ a lPergamoii Yarnasaki, T., Ogino Y., Morishita, ., Fukuota, K., Atou, Suryanarayana, G., Korth, G. E., and Froes, F. W. (1~97). T., and Syono, Y. (1994). ~ f f t e Sci, arzd Eng., ~ 1 ~ 9 1 Metall. and Mater. Trans., 28A, 293. ,4180, 220-223. Thadhani, N. N. (1993). Prog. Mater. Sci., 37(2), 117-226. Yu, L. (1995). Energetics and ~ e c h a n i c s of Shock Thadhani, N. N. (1994). J. of Appl. Phys., 76(4), 2129-2138, Consol~dat~on and S?~thes~s, Ph.D. Thesis, New Mexico Thadhani, N. N. (1999). P/M: Resear~hand Tec~nQlogy Institute of Mining and Tec~no~ogy. Briefs, P / M Scicnce and Technology Briejs: The Latest Yvoii, P. J., and Schwasz, R. B. (1993). J. Mafer. Res. ~ e v e l o p ~ e n tin. ~ Powder ~ e t a l l u r g y and Part~cul~te 239-241. Mater~als, fed. A. Bose). Metal Powder Industries Zhou, G. F., and Bakker, H. (1994a). Acta M e t ~ l l , , Federation, Princeton, NJ, 1(4), pp. 26-30. 3009-3017. Zhou, G. F., and Bakker, H. (199417). J. Phys. Conderts. Thadhani, N. N., Work, S., Graham, R. A., and Hammetter, M a ~ ~ e6,r ,4043-4052, W. D. (1992). J. Mat. Res., 7(5), 1063-1075.
er
are notable in that they are s~ultaneouslyhard, strong, and tough. Therefore, a number of i n v ~ ~ t i ~ a thave io~s been conducted to mimic nature ( ~ ~ ~ and also ~ ~ ~ TVanocrystalline materials are single- or multi-phase to artificially synthesize nanostructured materials and polycrystalline solids with a gratn size of a few study their properties and behavior. These investigananometers (1 nrn = 10m9m = 10 A), typically less tions have clearly shown that one can engineer (tailor) than 100 nm. Since the ain sizes are so small, a the properties of nanocrystalline materials through ni~cantvolume of the microstructure in nanocontrol of microstructural feature^, more speci~cally crystalline materials is composed of interfaces, the grain size. mainly grain boundaries, i.e. a large volume fraction Gleiter (1989) and coworkers synthesized ultrafineof the atoms resides in, or atomically close to, grain grained materials (with a grain size of a few nanoboundaries. ~ onse~ uen tly, nanocrystalline materials meters) by the in-situ conso~idation of nanoscale exhibit properties that are significantly different from, atomic clusters in the 1980s and showed that these nd often improved over, their conventional coarser a i ~ e dpolycrystalline counterparts (~u~yanarayana, materials have properties si~nificantlyimproved over those of conventional grain-sized (> 1 pm or 1000 nm) 995). Materials with microstructural features of polycrystalline or amorphous materials of the same naiioinetric dimeirsions are referred to in the literature chemica~ com~osition. These results s t i ~ u ~ a t e d as nanocryst~llin~ mat~rials(a generic term), nanoconsiderable research in this area and the field of crystals, nanostructured materials, nanophase nano~rystalli~e materials has now become a major materials, nanonieter-sized crystalline solids, or solids identifiable activity in materials science. with nanometer-sized i ~ i cr os t~ ctur features, al NanoThe subject of nanocrystalline materials has attracted stru~tured solids i s perhaps the most accurate the attention of materials scientists, physicists, chemists, description, even though nanocrystalline materials mechanical engineers, electrical engineers, and chemical will be the appropriat~term, if one is dealing with engineers. At present the very broad field of nanossolids with grains made up of crystals. tructured materials includes (i) nanofabrication, (ii) Nanostructured materials are not really very new. n anometr 01 ogy , (iii) functiona1 nanotechnology, (iv) phases were detected in samples of nanomechanical devices and m a c ~ n ~(v) s , molecular y conventional catalytic materials are based on very fine microstru~tures.Nanostru~tures nanotechnology,(V_i) nanoparticles, (Gi) nanostructured mateijals, and (Viii) ‘extreme’ i i a i i ~ t ~ c h n o lSmith o ~ . et formed chemically under ambient conditions can also al. (2000) have briefly discussed the di~erent aspects of be found in natural biological systems from seashells nanotechnology and h i g h l i ~ ~at range ~ d of t ~ h ~ o ~ o g i e s to bone and teeth in the human body. These materials primed to develop in the twenty-first century. The field of na~ocrystallinemat~rials has been reviewed earlier and the reader is referred to a number *See List of ~ o n t r i b u t ~for r s current address, I ~ t e r ~ i ~ ~t ~~l l~i ~p ~ Vol. u n3,dPrinciples s ~ and Practice. Edited by J. H. Westbrook and RI. L. Fleischer. 032002 John Wiley & Sons, Ltd.
of recent review articles and conference proceedings (Suryanarayana, 1995; Gleiter, 1989; Suryanarayana et al., 1996; Edelstein and ~ a m m a r at a,1996; N a et al., 1997; Suryanarayana and Koch, 1999, 2000). It has been claimed that the digusional creep rate, dcldt, of ceramics could be significantly increased by processing them into a nanocrystalline state, and consequently they would behave in a ductile fashion at room temperature (Karch et al., 1987). This has been attributed to the decreased grain size and increased grain boundary di~usivity,according to the relation: de ---BO'QADb dt d3kT
where B is a constant, O' is the applied stress, SZ is the grain boundary volume, A is the grain boundary width, L)b is the grain boundary diffusivity, dis the grain size, k is the Boltzmann constitnt, and T is the absolute temperature. However, it has not been possible to reproduce these results in fully consolidated ceramic samples, nor has it been possible to achieve increased ductility in metallic systems. But, this hope of improving the ductility of the inherently brittle intermetallics was the driving force for the tremendous amount of research and development activity in the field of nanostructured intermetallics during the last few years. In this chapter, we will high~ightcurrent research and progress in the area of nanostructured intermetallics. We will first review the classification and characteristics of nanostructured materials, briefly describe their methods of synthesis, and then discuss the formation, structure, and properties of nanostructur~d intermetallics. Since nanostructured inte~etallicsare synthesized mostly by mechanical alloying methods, the present chapter will emphasize the results obtained by this technique.
Nanocrystalline materials can be classified into three categories: (a) layered or Lamellar structures; (b) filamentary structures; and (c) equiaxed nanostructured materials. A layered or lamellar structure is a one-dimension~l(1-D) nanostructure in which the magnitudes of length and width are much greater than the thickness that is only a few nanometers in size. One can also visualize a two-dimensional (2-D) rod-shaped nanostructure that can be termed filamentary and in larger than width Or this, the length is diameter, which are of naiiometer dimensions. The most common of the nanostructures~however, is
basically equiaxed (all the three dimensions are of nanometer size) and are termed nanostructured crystallites (three-dimensional [ 3-D] nanostructur~s). The nanostructured materials may contain crystalline, quasicrystalline, or amorphous phases and can be metals, ceramics, polymers, or composites. If the grains are made up of crystals, the material is called nanocrystalline. On the other hand, if they are made up o f quasicrystalline or amorphous (glassy) phases, they are termed nanoquasicrystals and nanoglasses, respectively (Suryanarayana, 1995). Cleiter (1995) has further classified the iia~iostructuredmaterials according to the composition, morphology, and distribution of the nanocrystalline component. Figure 1 shows a schematic representation of a hardsphere model of an equiaxed na~ocrystallinemetal. Two types of atoms can be d~stin~uished: crystal atoms with nearest-neighbor configuration corresponding to the lattice and the boundary atoms with a variety of interatomic spacings, diffe~ing from boundary to boundary. A nanocryst~llinemetal typically contains a high number of interfaces (-J 6 x 1025niV3for a 10 nm grain size) with random o~entationrelationships, and consequently, a substa~tialfraction of the atoms lies in the interfaces. The volume fraction of atoms in the grain boundaries can be as much as 50% for 5-nm grains and decreases to about 30% for 10-nm grains and 3% for 100-nm grains, In contrast, for coarse-grained materials with a grain size of > 1 pm, the volume fraction of atoms in the grain boundaries is negligibly small.
Figure 1 Schematic representation of an equiaxed nanocrystalline metal, distinguishing between atoms associated with the individual grains (black circles) and those adjacent to the planes of grain boun$aries (white circles). Apparent dif.fere1lces in structure in various indiv~dualgrains are due to differences m grain orientatio~
The physical, chemical, and mechanical properties o f nanocrystalline materials have been determined
after consolidating the nanostructured powders into ‘dense’ compacts. In comparison to their coarsegrained counterparts, iianocrystalline materials have been shown to have 4-5 times higher hardness and strength, increased diffusivity, reduced density, reduced elastic modulus, higher electrical resistivity, increased specific heat, higher coefficient of thermal expansion, lower thermal conductivity, and superior soft magnetic properties. These property changes have been reviewed by Gleiter (1989), Suryaiiarayana (1995), aiid ~uryanarayanaand Koch (1999). But, it is becoming increasingly clear that the early results on the properties of naiiocrystalline materials are not very reliable, mainly due to insufficient consolidation and consequent presence of significant amounts of porosity, cracks and other discontinuities in those samples. Thus, many of the ‘improvements’ in the properties of nanocry$talline materials have not been reproduced in porosity-free samples. However, the most striking observation made in the early days of research on nanocrystalline materials is the improved ductility and plastic deformation of nano~rystalline ceramic inaterials such as TiO2 and CaF2. Since most intermetallics, like ceramics, are inherently brittle, there was considerable excitement and hope that nanostructure processing of intermetallics could make them usefully ductile. Even though this dream has not been fully realized, it was the motivation for the significant amount of research conducted in this area.
3. ~anocrystallinematerials can be synthesized either by consolidating small clusters or by breaking down the bulk material into sinaller and smaller dimensioiis. Gleiter (1989) used the inert-gas-conde~sation technique to produce nanocrystalline powder particles and consolidated them in-situ into small disks under ultrahigh vacuum (UHV) conditions. Since then numerous techniques have been developed to prepare nanostructured materials starting froiii the vapor, liquid, or solid states. Table 1 lists some of the more common methods used to produce nanocrystalline materials and also shows the dimensio~~lity of the product obtained. Nanostructured materials have been synthesized in recent years by several methods including inert gas condensation, mechanical alloying, spray conversion processing, severe plastic deformation, electrodeposition, rapid solidification from the melt, physical
vapor deposition, chemical vapor processing, coprecipitation, sol-gel processing, sliding wear, spark erosion, plasma processing, auto-ignition, laser ablation, hydrothermal pyrolysis, thermophoretic forced flux system, quenching the melt under high pressure, biological templating, sonochemical synthesis, and devitrification of amorphous phases. In practice, any method capable of producing very fine grain-sized niaterials can be used to synthesize nanocrystalliiie materials. The grain size, ~orphology,and texture can be varied by suitably modifying/controlling the process variables in these methods. Each of these methods has advantages and disadvantages and one should choose the appropriate method depending upon the requirements. If a phase transformation is involved, e.g. liquid to solid or vapor to solid, then steps need to be taken to increase the nucleation rate and decrease the growth rate during formation of the product phase. In fact, it is this strategy that is used during devitrification of metallic glasses to produce nanoc~s~alline materials (Lu, 199Qa). The choice of the method depends upon the ability to control the most important feature of the nanocrystalline materials, viz., the microstructural features (grain size, layer spacing, etc.). Other aspects o f importance are the chemical composition and surface chemistry or cleanliness of the interfaces. Extremely clean interfaces can be produced and retained during processing and subsequent consolidation by conducting the experiments under UHV conditions; but, a good vacuum adds considerably to the cost of processing. On the other hand, there are also methods that can be very Table 1. Methods to synthesize nanocrystalliiie materials Starting phase Vapor
Liquid
Solid
Technique Inert gas condensation Physical vapor deposition Evaporation and sputtering Plasma processing Chemical vapor condensation Chemical reactions Rapid solidification ~lectrodeposition Chemical reactions
Dimensionality of product
3-D 1-D 3-D 3-D, 2D 3-D
3-D 1-D, 3-D 3-D Mechanical a~loyi~ig/~illing 3-0 Devitrification of amorphous 3-D phases Spark erosion 3-D Sliding wear 3-D
752
Research T e ~ ~ ~ i q u ~ s
inexpensive; but the purity of the product may not be high. Inert gas condensation, mechanical alloying/ ~ i l l i n gspray , conversion proces~ing,electrodeposition, and devitri~cationof amorphous phases are some of the most popular techniques used to produce n a ~ o c r y s ~ ~i n~ ~l ien~~e t a ~ ~ i c s .
Vapor condensation has been known to produce very ~ne~grained or a ~ o r p h o u salloys depending on the
substrate t e ~ p e r a t ~and r e other proce$sin~conditio~s. Thus, this technique was originally used to synthesize small quantities of nanostructured pure metals. A number of v a r i a ~ ~have s also been subs~~uently developed. The inert gas condensation technique (Figure 2), ~opularizedby Gleiter (l%B),consists of e v a ~ o r ~ t i n ~ a metal (by resistive heat in^, ra~io-~requency heati~g, sputtering, electron beam heating, laserlplasma heating, or ion sputter in^) inside a ch~mberthat was evacuated to a very high vacuum of about lW7 torr
Sche~aticr~~resentation of the inert-gas con~ensat~on chamber for synthesis of nano~~ystalline ~ateria~s
and then bac~filledwith a low pressure inert gas, typically a few hundred Pascals of helium. The evaporated atoms collide with the gas atoms inside e chamber, lose their kinetic energy, and condense in the f o m of small, discrete crystals of loose powder. Convection currents, generated due to the heating of the inert gas by the evaporation source and cooled by the iiquid-nitrogen-filled collection device (cold finger), carry the condensed fine powders to the collector device, from where they can be stripped off by moving an annular teAon ring down the length of the tube to carry the powder product into a compaction device. Compaction is carried out in a two-stage piston-andanvil device initially in the upper chamber at low pressures to produce a loosely compacted pellet, which is then transferred in the vacuum system to a highpress~reunit where fmal compaction takes place. The scraping and compaction processes also are carried out under UHV conditions to ma~ntaincleanliness of the su~sequentinterfaces) and also nt of any trapped gases (Siegel, The inert-gas conde~sationmethod is probably the one most used to produce different types of metal and ceramic equiaxed erystallites. The crystal size of the powder is typically a few nanometers and the size distribution is narrow. The crystal size is dependent upon the inert-gas pressure, the evaporation rate, and the gas composi~ion.Extremely fi produced by decreasing; either the chamber or the evaporation rate gases such as He rather than heavy inert gases such as Xe (Granqvist and ~ a n o c ~ s t a ~ l ceramic ine materials, such as oxides, carbides, and nitrides, can be produced by modifjiring the above procedure. The modifications may include (a) intro~uctionofa small a ~ ~ o uofn reactive t gas to the inertreacting the particles as they are the as-synthes~edmetal particles as to interact su~sequentl~, or (c) directly evaporating the ceramic material. The last option is not a~waysfeasible since ceramics have normally very low vapor pressures and t ~ e r e f o ~very e temperatL~resare required to evaporate them. ~anostructuredintermeta~licscould also be synthesized using this method. For example, d i ~ e r emetals ~t could be evaporated from more than one source and the cold finger could be rotated to help in achieviug a of the vapor. But achievement of correct stoichiometry may not be easy. Therefore?ma sputter deposition, which provides excellent control of the alloy cornposition, has been used instead of
thermal evaporation to produce nanocrystalline TiAX (Chang et al., 1392). A full description of the technique, the variety of evaporation methods, and the effect of ~rocess variables on the size, size distribution, and ~onstitution of the powder particles can be found, for example, in Gleiter (1989), Siegel (1991), and Buhrman (1976).
Mechanical alloying (MA) produces nanostructu~d materials by structural disintegrat~on of grained structure as a result of severe plasti tion. Mechanical alloying consists of repeat fracturing, and rewelding of powder particles in a dry, high-energy ball mill until the co~positionof the resultant powder corresponds to the percentages of the respective ~onstit~ents in the initial charge, In this process, mixtures of elemental or pre-~lloy~d powders are subjected to high-energy forces while grinding under a p in equipment such as attrition mills, and shaker mills. The m a j o ~ t yof the work on nanocrystalline materials has been carried out in highly energetic, small shaker mills, The process i s referred to as ~ e c ~ u n~~~cZ ~o ~l when i n g one starts with a blended mixture of elemental powders and as ~ e c ~ ~ ~ ~ ~ l when l ~ one ~ starts g with sin~le-componentpowders such as elements or intermetallic com~aunds. While material transfer is involved in mechanical alloying?no material transfer is involved in mechanical milling. These processes have produced nanocrystalline structures in pure metals, intermetallic compounds, and immiscible alloy systems. It has been shown that nanometer-siz~d obtained in almost any material after sufficient milling time. The grain sizes were found to decrease with milling time down to nano~eterlevels, as shown in Figure 3 for a series of a;?-titaniu compositions, lit has been shown that the smallest 9
~~~
mechanical alloying/milling and the characte~isticsand properties of the nanocrystalline rnaterialts thus obtained. Powder contamination (from the milling tools and/or the atmosphere) is usually a matter of concern with this process, es cially when reactive metals and/or long milling times are involved; some remedial measures have been suggested in. recent years ~~uryanarayana, 200 1). ~ u ~ e r o intermet~ll~cs us have
as VC and Cr&2 are made as binders during the sintering steps. Recently, vanadium is introduced into the starting solution itself to achieve a more uniform distribution in the powder mixture (Kear and Strutt, 1995).
I eTi-24AI
+Ti33AI
-R --r.
E 150
+Ti-SOAI
t:
-cn 1
-X- Ti-24Al-11Nb
1
100
c
P
3.4 ~ ~ ~ ~ t r o d e p ~ s i t ~ o ~
U 50
been synthesized by the technique of MA. In fact, almost all the inte~etallicssynthesized directly by MA are nanostructured in nature. In recent years, the process of severe plastic d e f o ~ a t i o nof bulk solids (by the equal-channelangular pressing, torsion straining, and accumulative roll bonding techniques) has been shown to produce ultra~ne-grainedstructures. Even though the grain size is strictly not in the nanometer range (it is usually about 0.3-0.5 pin), there has been a considerable amount of work on the structure and properties of intermetallics produced by these methods (Valiev et al., 2000), essentially due to the possibility of producing bulk materials possessing sub-micron grain sizes and improved properties.
This is a simple and well-established process and can be easily adapted to produce nanocrys~alline~at eri al s (Erb, 1995). Electrodeposition of multilayered (1-D) nietals can be achieved using either two separate electrolytes or, much more conveniently, from one electrolyte by appropriate control of agitation and the electrical conditions (particularly voltage). Also, 3-D nanost~cturedc ~ s t a l ~ t can e s be pre~aredusing this method by utilizing the interference of one ion with the deposition of the other. Erb (1995) and his collaborators have extensively used this process to study the synthesis and properties of 3-D nanoc~stallinematerials. It has been shown that electrodeposition yields grain sizes in the nanometer range when the electrodeposition variables (e.g. bath composition, pH, te~perature,current densityy,etc.) are chosen such that nucleation of new grains is favored rather than growth of existing grains. This was achieved by using high deposition rates, formation of appropriate complexes in the bath, addition of suitable surface-active elements to reduce surface diffusion of ad-atoms, etc. This technique can yield porosity-free finished products that do not require subsequent consolidation. Further, the process requires low initial capital investment and provides high production rates with few shape and size limitations.
This is a commercial process employed to produce nanocrystalline VVC-CO composite powders. This process starts with aqueous solution precursors such as ammonium metatungstate [ (NH4), (HzVVl20 4 0 ) - 4H201 and cobalt chloride [CoC12], cobalt acetate [Co(CH3COO)2],or cobalt nitrate [CO(NO~)~]. The solution mixture is aerosolized and rapidly spray dried to give extremely fine mixtures of tungsten and cobalt complex compounds. This precursor powder is then reduced with hydrogen and reacted with carbon monoxide in a fluidized-bed reactor to yield nanophase cobalt~t~ngsten carbide powder. The tungsten particles are 2 4 0 nm in size. A typical powder particle consists of a hollow, porous 75-pm sphere containing hundreds of millions of WC grains in a cobalt matrix. To prevent grain growth of tungsten, additions of inhibitors such
Many non-equilibrium processing techniques such as rapid solidification from the liquid state, mechanical alloying/milling, electrode position^ and vapor deposition can produce amorphous (glassy) alloys (Suryanarayana. 1999). Controlled crystallization of these amorphous alloys leads to the synth~sis of nanostructured materials (by increasing the nucleation rate and decreasing the growth rate). In fact, the most common method to produce nanocry~tal~ine magnetic materials has been to obtain an amorphous phase by rapidly solidifying a melt of appropriate composition and then ~rystallizingthe glassy phase at a relatively low temperature. The microstructure in this condition consists of nanometer-sized grains in an amorphous matrix, These materials - referred to as FINEMET were first investigated by Yoshizawa et al. (1988), and
0
0
5
10
15
20
25
30
35
milling time (hf --+
igure 3 Variation of grain (erystallite) size, as determined from X-ray peak broadening, in mecbanically alloyed a2titanium aluminide samples as a function of milling time
this technique has now become an established practice to study the structure and properties of nanocrystalline magnetic materials. This simple devitrification method has been commonly employed to study the magnetic properties of nanocrystalline materials because it can produce: (a) porosity-free samples, (b) samples with different grain sizes by controlling the crystallization parameters, and (c) large quantities of material. Furthermore, since no artificial consolidation process is involved (as in inert gas condensation, mechanical alloying, or plasma processing, where fine powders are involved), the interfaces are clean and the product is dense, Additionally, samples with different grain sizes can be synthesized by controll~ngthe crystallization process, affording a way of comparing the properties of amorphous, nanocrystalline, and coarse-grained materials of the same composition (Suryanarayana, 1995; Lu, 1996a). In recent times there have been several basic investigations to define the conditions under which very fine (nanocrystalline) microstructures form, starting from the glassy phases (Lu, 19966; Koster, 1997). A recent development in this area is the study of nanocrystalline materials produced by crystallization of amorphous powders or ribbons when they are subjected to mechanical milling (Trudeau, 1994). Even though the exact mechanism of the process is presently not known, mechanical milling offers a useful route to the preparation of high-purity nanocrystalline materials in. bulk quantities since metallic glassy ribbons are commercially available. Several companies all over the world have started producing nanocrystalline powders, mostly of pure metals or ceramics, but very few (or almost none) intermetallics, on a commercial scale. According to a recent estimate (Rittner and Abraham, 1998), in the United States alone, over 50 small and large companies are active at various levels in the development and production of nanocrystalline materials. More than a dozen of these businesses are involved in the manufacture of nanostructured materials on an industrial scale.
As mentioned above, nanostructured intermetallics have been synthesized mostly by mechanical alloying (MA), even though some have also been produced by electrodeposition and devitrification of the amorphous phases. However, the amount of scientific i n f o ~ a t i o n
available on intermetallics produced by and, therefore, we will now discuss the aspects of formation and properties of nanostructured intermetallics synthesized by MA. Only those intermetallics that are synthesized directly by MA have grain sizes in the nanonieter range. It may also be noted in this context that sometimes the intermetallics may not form directly after MA. In those cases MA fornis only an intimate mixture of the component metals; an additional heat treatment is required to provide the necessary iff fusion for the formation of the intermetallics. The intermetallic formed under these condition^ may not always have nanometer-sized grains. If the heat treatment was carried out at relatively low temperat~ir~s~ the intermetallic may be nanostructur~d. MA has produced nanostructured inte~etallics under three categories: 1. Quasicrystalline phases. 2. Metastable intermetallics. 3. Equilibrium intermetallics.
Quasicrystalline phases are metallic phases that exhibit traditionally forbidden translational sym~netries. Instead they exhibit rotational symi~etries,e.g. Sfold, 7-fold, 10-fold, etc. The reader may consult Suryanarayana and Jones (1988) and 1995) for details of the structure, formation, and properties of quasicrystalline phases. There has only been an academic curiosity in this area since no commercial-scale applications have been reported for materials with quasicrystalline phases. MA was shown to be another technique to produce these phases under nonequilibrium conditions, but with the advantage of having nanometric dimensions. The q~~sicrystalline phases synthesized by MA are ~ummarized in Table 2. It may be noted that all of these phases are o f the icosaliedral type (having the 5-fold symmetry) and that there are no reports of synthesis of phases with other symmet~ies. A number of the latter category (with 10-fold symmetry, etc.) were synthesized by other techniques such as rapid solidification processing (RSP) from the melt (Suryanarayana and Jones, 1988). The icosahedral phases synthesized by MA have been found to be similar to those produced by RSP.
7.56 le 2 ~a~o-qu~sicrystalline phases synthesized by ~ e c h a n i ~ ~ l l1o y ~ ~ g
has also been shown to be capable of synthesizing some metastable crystalline intermetallic phases. i eta stability here refers to the property obtaining when a phase stable at high temperatures or pressures is retained at room temperature and pressure. Alternatively the phase synthesized may not be present in the ~~u i lib r iu diagrams m at either high temperatures or high pressures, i.e. the phase is entirely new. A similar situation has also been obtained by other noning techn~~ues,most not a ~ l y anarayana, 1999). However, the thesized by MA is far less than those synt SP. This is in contradiction to the hypothesis that metastable phases can be formed by cold working, a good example being the formation of deformation-indu~edmetastable E (hcp) niartensite in austenitic stainless steels, Table 3 lists the crystal structure data of the metastable intermetallic materials synthesized by MA. The pressures generated during MA have been estimated to be of the order of 6 GPa (Davis et al., 1988; Maurice and Courtney, 19901, and these should often be su~cieiitto stabilize high-pressure polymorphs of phases at heric pressure. Highpressure polymorphs of (cI28) and Y2S3 (cI28) have been reported to be retained by MA at atmospheric pressure (Han et al., 1991). Similarly, highgenides, such as Cu2-,S synthesized by MA at et al., 1995). Sen et al. (1999) synthesized the high-pressure orthorhombic form of Ti02 (11) at atmospheric pressure by MA at room temperature. Alonso et al. (1991) reported that they could synthesize the high-pressure fcc polymorphs of the lanth~iiidemetals Dy, Cd, Nd, and Sm by ying the powders for about 24 h in a a hardened steel vial and alloy steel balls. It was, however, later recognized by them (Alonso et al., 1992) that these are NaC1-type phases formed by reaction of these reactive metals with oxygen, nitrogen, and hydrogen during milling in a poorly sealed vial. For example, the milled Nd powder
contained 16.2 at.% oxygen, 22.3 at.% nitrogen, and 10.3 at.% hydrogen. Thus, it is not clear whether the high-pressure phases of pure metals, especially the reactive ones, can be s y n t h e ~ ~by~ ~MA. d But, it appears that hi~~-pressurecompound phases of oxides, chalcogenides, etc. can be synthesized by MA, since contamin~tionis not likely to play any role in the formation of these compounds.
4.3 ~
qC
~ u~
~
~~
~
~~ ~
n~ c
Both disordered and ordered intermetallic~ with nanometer-sized grains have been synthesized by MA. The intennetallics synthesized include aluminides (mostly based on t~tanium,nickel, and iron), siticides, composites, and some exotic varieties. Table 4 presents red a listing of the categories of n a n ~ s t r u ~ t ~ intermetallics synt ~esi ze~ by MA, It has been reported, in the Ti-A1 system, that either a supersaturated solid solution or an amorphous phase could be produced on MA of blended elemental Ti and A1 powders, but not the a2 and y i n t ~ ~ e t a l lphases. ic The desired i nt erm ~t a~icould c be obtained in the appropriate composition range, only after a suitable heat treatment. Thus, an~ealingof the mechanically alloyed Ti-24 at.Oh A1 powder for four weeks at 903 M produced the az-Ti3Al (hP8) phase et al., 1992a). Similarly, the y-TiAl (t be produced only after annealing the as-milled amorphous powder for 168 h at 888 K (Frefer et al., 1993). These temperature and time c o m b i ~ a t i ~ n s required for producing the interi~etallicscould vary considerably ~ e ~ e n d i non g the powder particle/grain size. A TiAl3 (t18) phase, however, could be synthesized in a metastable Llz structure directly by MA (Srinivasan et al., 1991). A process control agent ( pound used to prevent excessive cold we powder particles) usually introduces into the powder that can be niinimized by avoiding the use of such a process control agent. Since non-usa the PCA may result in excessive cold welding and foiiiiation of large particles, true alloying also may not occur. This problem was overcome by using a
Table 3 Metastable inte~etallicssynthesized by mechanical alloying
Lattice parameters Alloy system AI-Cu Al-Ge Al-Elf AI-Hf-Fe A1-Elf-Ni AI-Mn Al-Ti AI-Zr Al-Zr-Fe Al-Zr-Ni Cu-I~-Ga-Se CU-Zn Fe-B Mg-Sn Nb-Ge Ni-A1 Ni-Sn Ni-Sn Te-Ag Ti-Si Ti-Si Zr-A1
Phase/C~mposi~ion
Crystal structure
a (nm)
33 at .Yo Cu
bcc rhom bohedral cubic cubic cubic fcc cubic cubic cubic cubic cubic
0.2905 0.767 0.4048
y' (30 at ."/o Ge)
L1, A1,Hf L1, (Al,Fe),Hf L1 (AI,Ni),Hf 18 at .% Mn L1, A1,Ti Ll, A1,Zr L1, (Al,Fe),Zr L 1 (Al,Ni),Zr CuIn, ,Ga, ,Se, Mar tensite Fe,B 33 at.% Sn 15-18 at.% Ge Ni,Al (disordered)
,
,
p' p"
n(cP1j 67 at YOSi C49 TiSi, DOl9Zr,Al
A1,Ti
0.4088 0.4472 0.3967 0.4077 0.4078 0.4081 0.57 -
orthorhombic (oP69?) fcc fee tetragonal simple cubic bcc ort~orhombic hexagonal
1.523 0.45 -
+ 2 TiW2 -+3 TiA1-t. 2 H2 f
On mechanically alloying the above powder mixture for 52 h in a SPEX mill, they were able to obtain 55 vole% of the y-TiAl phase (the remainder being the unreacted reactants); the amount of the y-TiAl phase was increased to 95% on hot isostatically pressing the mechanically alloyed powder at 1023
c (iim) or ~ ( 0 )
-
-
combination of brittle intermetallics or other cornpounds. Accordingly, Suryanarayana et al. (1992b) synthesized the y-TiAl phase by mixing titanium hydride (instead of pure titanium) and the A13Ti intermetallic (instead of pure a l u ~ i n u mpowders ~ in the proper proportion, according to the equation:
b (nmj
-
0.3061 0.415 0.8267
for 5 h, This technique has been subse~uentlyemployed to synthesize the Ti3A1, TiAI, and A13Ti phases by milling the titanium hydride and aluminum powders in Ti/Al ratios of 3:1, 1:1, and 1:3, resp~ctively (~ukhopadhyayet al., 1993). Considerable literature also exists on the synthesis of nanostructured nickel- and iron-alumini~esby Intermetallic phases such as Ni3Al (cP4), NiAl (cP2) and A13Ni (oP16) have been synthesized by MA in the appropriate composition ranges. An interesting observation is that the e ~ ~ i a t oNiAl ~ i c phase i s produced via a combustion synthesis (also known as the selfpropagating high-tei~peraturesynthesis reaction) that occurs during milling (Atzmon, 1990). In this and
Inte~etallicssynthesized by mechanical alloying -
Category
Examples
Alu~inides
t(Ag-Al), ~(Ag-AI),AI,Fe, AlFe, Al,Fe, Al,Fe,, Fe,Al, Al,Mg,, AlI,Mg,,, AI,Mn, Al,,Mn, Al,,Mo, Al,Mo, AI,Mo, Al,Mo,, Al,Nb, Nb2A1, Al,Ni, NiAl, RuAl, Al,Ti, AlTi, Ti,Al, AIZr, Al,Zr,, MnAl, Nb,Al, Ni,Al,, NiAlNb, (Fe,Ni)Al, Ni(Al,Ti), PdAl
Ber y 11ides
NbBe,,, Nb,Be,,
Chalcogenides
Ag,S, Ag,Se, CuInS,, CuInSe,, Cu,_,S, CuSe,, FeS, Nb,$,
Silicides
Cr,Si, CrSi,, a-FeSi,, B-FeSi,, FeSi, Mg,Si, Mo,Si,, a-MoSi,, fi-MoSi,, Nb,Si, a-Nb,Si,, P-Nb,Si,, NbSi,, Ni,Si,, Ni,Si, Ni,Si, Nisi, Nisi,, Pd,Si,, Pd,Si, TaSi,, Ti,Si,, TiSi,, Ti,Si,, V,Si, V,Si,, VSi,, WSi,
Ni, ,, SnS,, Cu(In,Ga)Se,
758
Research Techniques
similar cases, the combustion reaction took place only after ‘interrupted milling’, i.e. milling of the powders for a given length of time, aging the powder at room temperature after stopping the milling, and then resuming the milling operation. For example, in the AI-Ni system, an intimate mixture of A1 and Ni phases was detected after milling the blended elemental AI-Ni mixture for 2 h in a SPEX mill. If the inilling is stopped and the powder is stored at room temperature for 30 min, and then milling i s resumed, it is noted that the NiAl phase forms after just 1 minute of milling due to the occurrence of an explosive reaction (Atzmon, 1990). Similar reactions were also en et al., 1996), observed in the synthesis of M NbSi2 (Lou et al., 1997), TiS2 and Mlissurski, 1994), PbTi03 (Aning et al., 1995), and a few other compounds. In all these cases, the MA operation produces a fine intimate mixture of the two or more phases involved; then an explosive reaction occurs and the c o ~ p o u ~sudden~y d forms. It has been recently reported that the time for the initiation of the explosive reaction can be delayed by 20 to 30 minutes by the addition of t~rnaryelements (di~uents)such as Ti and Fe to an A1-Ni mixture (Liu et al., 1995). It has also been reported in some cases that intermixing a m o ~ ~ sthe t powder particles takes place continuously and metastable phases form prior to the formation of the e ~ u i l i b r i phases, ~ i ~ In other cases, a series of other equilibrium phases form before the actual expected phase forms. or example, the aMoSi2 (tI6) phase formed when the Mo-Si powder in a planetary ball inill for 6 min, (hP9) phase formed after milling onov et al., 1995). Similarly, the Ni2A13 (hP5) phase formed after milling for 17 min of an Ni-65 at%Al powder; but the NiAl phase formed after 125 min; both these phases forming by exothermic reactions (Mikhailenko et al., 1991). Schaffer (1992) has shown that mixtures of intermetallic phases can be produced by MA by choosing the appropriate compositions. If the two phases are in equal proportion, then it has been observed that grain growth can be cons~de~ably hindered. Thus, by having approximately equal proportions of the NiAl and Ni3A1 phases in a nanocrystalline state, it was shown that grain growth could be slowed below 500 “C. The times required to form a particular phase depend 011 the initial co~centrationof the solute in the powder mixture. If the solute content is much less than the exact stoichiometry, the expected phase (although the proportion of the phase in the mixture of phases is less), would form only at much longer
milling times. For example, while investigating the formation of the p’-Al&‘o(cP2) phase iii AI-CO alloys, Sui et al. (1992) noted that it required 10 h to form the p’-AlCo phase in an A1-50 at%Co powder mixture. On the other hand, it required 30 h in A1-28.6 at%Co, 70 h in 81-23.5 at%Co, and 200 h in A1-18.2 at%Co mixtures under identical ~ i l l i n gconditions. This is easy to uderstand since f o ~ a t i o nof a particular phase requires both difbsion and equilibration, each requiring time and te~perature.The times can be probably reduced if the temperat~~re at which milling is conducted is increased; the grain size may, however, be large in that case. But, there have not been any investigations reported that confirm this hypothesis. The time required to form a phase can also be substantially reduced if the ball-to-powder weight ratio used for milling is increased. This has been shown to be true in many cases. Even though the intermetallics synthesized by MA include both the ordered and disordered types, it is not surprising that MA produces disordered phases. This disorder is because MA involves heavy d~formation, which is known to destroy long-range ordering in the lattice (StoloR and Davies, 1966). In some instances, ordered intemetallics have been found to form directly on MA. This has been shown to be particular~ytrue in Al-rich Al-transition metal systems. Some of the examples include A15Fez (oC *) ( ~ u k h o p ~ d h y aetyal., 1994), A13Hf (cP4) (Li et al., 1995a) and (Al,X),Hf (cP4) where X = Fe or Ni (Li et al., 1995a), (tI8) (Suryanarayana et ul., 1994a), A13Zr (cP4) (Srinivasan et al., 1991; Li et al., 1995a; ~uryanarayana et al., 19941.1)and (Al,X),Zr (cP4) wehre X = Fe or Ni (Suryanarayana et al., 1994b; Li et al., 1995b). Reasons for the formation of ordered intermetallics during MA have not been investi ate$ in detail. It may be assumed that a phase will exist either in the ordered or disordered condition, depending upon the balance between atomic disordering introduced by MA and the thermally activated reordering. The reordering is caused by the difference in energy between the ordered and disordered states. Thus, if this difference in energy is small, the MA-produced alloy will exist in the disordered state; whereas if it is large, the alloy will be in the ordered state. It has been shown that the NiAl phase can be produced in the ordered state by MA of the blended elemental powder, and upon milling the ordered NiAl compound, it continues to be ordered. However, the FeAl phase was found to be in the disordered state, both in the as-produced condition by MA and also on milling the ordered compound (Schropf et al., 1994). The ordering energy is related
to and scales up with the enthalpy of formation ( A H f ) ; A H f values for NiAl and FeAl are 72 and 25 mJ/mol, respectively (Schropf et al., 1994); consistent with the above argument. Nanostructured superconducting conipounds such as Nb3Sn (Larson et al., 1977) and YBa2Cu307-6 (Mizutani et al., 1992) have also been prepared by MA. Thus, the capabilities of MA in synthesizing a variety of intermetallic phases appear unlimited.
It has been long known that partially ordered phases are stronger than those wholly disordered or fully ordered (because at a certain value of the long-range order parameter, S, superdislocations separate into unlinked singles). Thus, it is of interest to study the mechanical behavior of materials in various states of partial order. Disordering phenomena of ordered alloys have also been studied to understand the meclianism of disordering and also to produce disordered material that has a higher ductility/formability than the fully ordered alloy. The first observation of disordering of an ordered compound ZnFe204 by mechanical milling (MM) was reported by Erniakov et al. (1982). l is ordering of Fe3Si by mechanical grinding was reported later by Elsukov et aE. (1983). An exhaustive account of this aspect of disordering of i n t e r ~ e ~ ~ l lmay i c s be found in ~uryanarayana(2001) and Bakker et al. (1995). A variety of ordered compounds with the B2 (cP2), Ll2 (cP4), A15 (cP8), B82 (hP6), and 0 (tP30) structures have been disordered by MM. The progress of disordering has been monitored by several techniques, including X-ray diffraction to iiieasure tlie lattice parameter and long-range order parameter, measurement of superconducting transition temperature and magnetic susceptibility (if the compound is superconducting in the initial state), Mossbauer techniques, differential scanning calorimetry, etc. Mechanical milling introduces high energy into the material being processed. This energy can be stored in the material as atomic disorder and/or grain boundaries, i.e.
+
AG(mil1ing) = AG(disorder) AG(grain boundaries) ~G(dis1ocations)
+
The atomic disorder in an intermetallic can be manifested in three dif€erent ways (Bakker et al., 1995). Firstly, the two atomic species involved can occupy the 'wrong' sublattices and this is referred to as
anti-site disorder. This introduces strain into the lattice. This type of disorder was observed in a number of mechanically milled compounds with the L12 (cP4) structure, e.g. Ni3A1, i3Si, Fe3Ge and those with the B2 (cP2) structure, e.g. CoGa and AI triple-defects can be generated. In an equiatornic compound such as CoGa (cP2), for example, the transition metal CO atoms can be substitute on the Ga lattice, and this is anti-site disorder. But, the C a atoms stay on their own lattices. This leads to the presence of vacancies in the CO lattice to ~ a i n t a i nthe stoichiometry. Thus, vacancies on tlie CO-sublattice in combination with CO anti-site atoms in a ratio of 2:l constitute the triple defects. Third, there could also be redistribution of interstitials wherein the interstitial atoms in the octahedral sites are transferred to the tetrahedral sites, e.g. Mn3Sn2 (hP6), Fe3Ge2 (hP6). Additionally, grain refinement increases the grain boundary area and this also raises the free energy of the system. The sum of the energies of these two effects (disordering and creation of grain boundaries) will be the total energy introduced into the material during milling. Mechanical milling of ordered intermetallics has been shown to result in one of the following three transformations (Bakker et al., 1995, 2997). (1) Formation of a solid solution of one component in the other, i.e. the terminal solid solution based on the major component; this has been observed in compounds such as Nb3Al (cP4), V3Ca (cP8), Ni3Al (cP4), Fe3Ge (cP4), Ni2V (0161, and NbAuz (hP3), (ii) Formation of an amorphous phase, observed, for example, in NbJSn (cP8), NiZr (oC8), NiV2 (tP30), and CoZr (cP2), or ~iii)Formationof a different phase with a complex crystal structure, noted in Ni3Sn2 and TiSi2 (oF24). It has been noted that upon milling, the long-range order parameter (S) in the intermetallic is gradually reduced and, in many cases, the material may become totally disordered (S = O), e.g. Ni3Al (cP4) (Jang and Koch, 1990). In other cases, S is reduced with milling time but does not reach S = 0 partial disorder co-exist, e.g. Johnson, 1990) and AlRu (cP2) In other cases, the S value initially does not decrease at all and is maintained at S = 1; but, with continued milling the material becomes amorphous, e.g. CoZr (cP2) (Cho and Koch, 1993). These three situations are schematically represented in Figure 4. Thus, upon
740
Research Techniques
iso order in^ of i~termetal~ics has been studied primarily by X-ray diffraction techniques. the intensity of the superlattice reflections relative to that of the f u n d a ~ e ~ t ar el ~ e ~ t i o nthe s ~ long-range order parameter S is evaluated using the relationship: S2
= Is(dis) /If(dis) Is(ord) /If(ord)
~ ~ ~ H A N I CALLOYING AL TIME
_t
Variation of the long-r~ngeorder parameter, S with milling time. Three different situations have been represented, viz., complete loss of order, partial loss of order, and no loss of order on milling
milling, an ordered intermetallic can transform, with or without complete loss of long-range order, either into a disord~redcrystalline phase (solid solution) or amorphous phase, If the product is a crystalline phase, the material has an extremely fine grain size, usually in the ~anometerrange, Figure 5 s~mmarizesthe types of changes that can occur on milling an intermetallic compound (Bakker et al., 1997).
where 1, and Ifrepresent the inte~ratedintensities of the superlattice and fundaniental re ections, respectively and the subscripts (dis) and (ord) refer to the disordered and ordered states, respectively ~Suryana~ayaiia and Norton, 1998). Additionally, it has been noted that the lattice parameter increases slightly (0.3 to 0.8%) in the disordered state, if the disordering occurs by anti-site disorder, whereas the lattice param~terd~creasesif disordering occurs by the triple defect disorder (Bakker et al., 1995). The change in lattice parameter continues after the compound has completely disordered and has of change formed a solid solution. This measu~~ment in lattice parameter, of course, is not possible if there is a change in the crystal structure due to ~ i s o r d e r i ~ ~ . Mossbauer spectroscopy te~hniquesand ~easurement of superconducting transition temperature and/or magnetic susceptibility have also been employed to
S ~ ~ e ~ a tshowing a i c the expected situation on me~hani~ally m i ~ ~an i ni ~ t e r ~ ~ t ac ~o ~l ipco u n d
study the disordering phenomenon. Recently, Zhou and Bakker (1996) have demonstrated that magnetic measurements are very powerful in determining the nature of disordering if one of the atoms involved has magnetic moments, uring disordering of CoZr (cP2), Cho and Koch 3) did not observe any decrease in the intensity of superlattice reflections (relative to fundamental lines) with milling time before amorpliization as thus concluded that the grain y in this system is high enough to -+ amorphous transition, but not the destruction of long-range order. Zhou and Bakker at ma~netiz~tion in the mechanialloys increased continuously with milling time up to about 40 h (no reduction in the relative intensity of superlattice reflections was observed) and also that the lattice parameter of the disordered nanocrystalline phase increased with milling time. From these two observations, they concluded that disordering of CoZr occurs by atomic disorder and grain refinement (i.e. creation of additional grain boundaries). Thus, a combination of techniques can shed additional evidence and provide an accurate account of the mechanism of disordering. Whether an interrnetallic transforms to a solid solution or an amorphous phase on milling is determined by the relative free energy values of the amorphous and crystalline phases with respect to the energy stored in the intermetallic by mechanical milling. For example, the intermetallic will transform to the solid solution if the enthalpy of the amorphous state as estimate^ by the Miederna analysis (Bakker, 1998) is higher than that of the solid solution. Thus, if AG(mil1ing) 3 AG"-", where AGa+ represents the difference in free energy between the ordered crystalline and amorphous phases, complete amorphization occurs. On the other hand, i~AG(lni11ing)< AGa-c,no amorphization occurs; instead, a solid solution forms. Similar conclusions can also be reached by observing the nature of the phase diagram. 'If the phase diagram shows that the intermetallic transforms to a (disordered) solid solution before it melts, i.e. it is a reversibly ordered intermetallic, then milling of the intermetallic will result in the formation of a (disordered) solid solution. But, if the i~termetallicmelts congruently (irreversibly ordered or permanently ordered intermetallic), then milling o f the intermetallic will produce an amorphous phase directly. This has been shown to be true in a number of cases (Bakker et al., 1995). Thus, introduction of mechanical energy into the system is equivalent to heating the alloy to
+
i
I
T lsar
se
nical
~ ~ ~ 6u Equivalence r e of mechanical energy and t e ~ ~ ~ r a t u r e in the disordering of intermetallic c o ~ p o u n d s
higher temperature (Figure 6). That is, an alloy can become disordered either at higher temperatures or on mechanical milling. pending on the nature o f the phase, either rneltin r formation of an amorphous phase can occur in some cases. It has been noted that among about 700 binary interm a very few compouiids such as reversibly ordered compound order-disorder transformation before melting. There have also been several stud' reordering of disordered phases obtained by ee, for example, Barb et al., 1993). It may also be mentioned in passing that all order states ( S = 0 to S = 1) cannot be accessed by traditional methods of disordering. For example, in equilibrium, the order parameter for Cu3Au jumps discontinuously from 0.8 to zero at the critical temperature. But, mechanical milling can be used to obtain different degrees of order so that the effect of order parameter on structure and mechanical properties of alloys could be investigated.
The yield strength, 6, of crystalline materials cornmonly increases with a decrease in grain size, d, according to the Hall-Petch relationship: 6
= 6, 4-Kd-l/2
where 6, IS the lattice friction stress to move individual dislocations and K is a constant. A similar relation holds for the hardness variation with grain size. Since it has been difficult to produce fully dense, homogeneous, and large enough speciniens of nanocrystalline materials for conventional tensile testing,
762
Research Techniques
microhardness measurements have been used to test whether the Hall-Petch relationship is valid for materials with nanometer“sized grains. There has been considerable discussion in the literature regarding the variation of hardness with grain size, and it has been noted that the Hall-Petch relationship is observed above a critical grain size; below this grain size, the hardness was found to decrease with a decrease in grain size and this has been termed the ‘inverse’ HallPetch behavior ( ~ u r y a ~ a r a y a 1995). ~a, Figure 7 presents the variation of hardness with the reciprocal of the square root of the grain size for yTiAl (tP4) specimens. The data include results from uryanarayana et aL,, 1997; gas condensed powders (@hanget al., 1992), and ingot material (Koeppe et al., 1993). It may be seen that the hardness values obey the 1-Petch relationship for grain sizes larger than ut 30 nm. When the grain size is smaller than this, the hardness decreased with decreasing grain size. A similar relationship is noted for the a2-Ti3AI (hP8) alloys also. The Hall- etch relationship was derived using the concept of dis~ocationpile-ups in individual grains. owever, in very fine-grained materials, e.g. nanocrystalline ~ ~ t e r i a l s the , expected number of dislocations that can pile up within a grain and concentrate stress is less then one, and therefore pileups cannot form when the grain size is less than a critical value, Lc. Consequently, the ~ a ~ l - P e trela~h tionship does not apply to materials with very small grain sizes. Therefore, when the grain size of the ~echanisms(e.g. viscous material is < Lc, w~ak e n~ ng type Row) occur and lead to a decrease in hardness with decreasing grain size, i.e. a negative value for the slope K in the Hall-Petch equation. The value of E, was calculated by equating the repulsive force between the ations and the applied stress using the relation and Wadsworth, 1991):
I, =
3Gb R(1 - V ) H
where G is the shear modulus, b is the Burgers vector, v is the Poisson’s ratio, and H is the hardness. Values r s therefore it is not of I, are only a few n a ~ o ~ e t eand surpr~singthat a negative Hall- etch type behavior is observed at these very small grain sizes. Some researchers do not believe in the negative Hall-Petch behavior (Siege1 and Fougere, 1994). Others (Cbang et d., 1992) believe that the hardness data at small grain sizes are better fit by a l / d than l / J d dependence.
-2 E 8 E
=-s
12000 10000 8000 6000 4000
2000
0
0.06
0.1
0.15
dii2
0.2
0.26
0.3
0.35
(nm”’”)
Figure 7 Variation of hardness with reciprocal of the square root of the grain size in several ~ e c ~ d n i c a ~alloyed ly and consolidated y-TiAl-type alloys
There have been only a few studies of the measurement of yield strength as a function of grain size and/or temperature for nanostru~turedintermetallics. It was reported (Oeliring et aL,, 1995) that the compression yield strength of ~echani~ally-alloyedand consolidated Ti-48 at.% A1 alloys is between 1800 and 2500 MPa, and is approximately three times higher in comparison to coarse-grained alloys. This strength decreases rapidly with temperature and at temperatures >700 ”@, the yield strength of the mechanically alloyed material is well below that of the ingot material of similar composition with grain sizes between 8 and 11 flm. Superplastic behavior was observed in a mechanically alloyed, two-phase Ti-48 at.% A1 alloy at high strain rates, e.g. 3 x 10-3/s to 3 x 10-2/s and at temperatures between 950 and 1050 OC. The strain rate sensitivity exponent, m, of the compact is approximately 0.5, and a maximum elongation of 550% was obtained at a strain rate of 5.6 x 10-3/s (Ameyama et al., 1994). Tensile super~lasticitywas also observed in a fine-grained nickel aluminide inter~etallicproduced by the severe plastic deforniation process (McFadden et al., 1999). y-TiAl alloys produced from MA powder after consolidation are most likely to contain very small flaws such as pores. Further, titanium alloys are very sensitive to interstitial impurities such as oxygen and nitrogen. Thus, a material with a very high yield strength, such as ultrafiiie-gr~ined titanium alurninides, may have a low fracture strength. That is why it has not been possible to observe the much- anticipate^ increased ductility in nanostructured titanium aluminides (or other intermetallics). However, it has
Edelstein, A. S., and Carnmarata, R. C. (eds.) (1996). Nanomater~als:Synthesis, Properties, and Applications, Inst. Physics, Bristol, UK. Erb, U. (1995). NanoStructured m at er., 6, 533. Ermakov, A. E., Yurchikov, E. E., and Elsukov, E. P. (1982). Fiz. Tverd. Tela No. 4, 1947. Elsukov, E. P., Barinov, V. A., Galakhov, V. R., Yurchikov, E. E., and Ermakov, A. E. (1983). Phys. Met. M e ~ a l l ~ g r , , §5(2), 119. Frefer, A., Suryana~~yana, C., and Froes, F. H. (1993). In ~dvancedSynthesis of’ ~ n g ~ n e e r eM~terials d (eds J. J. Moore et al.). ASM International, Materials Park, OH, p. 213. Gleiter, H. (1989). Prog. Mater Sci., 33, 223. Gleiter, H. (1995). ~anostructured at er., 6, 3. Granqvist, C. G., and Buhrman, R. A. (1976). J . Appl Phys., 47, 2200. Han, S. H., Gschneidner, K. A., and Beaudry, B. J. 1991). Scripta Metall. Mater., 25, 295. Hellstern, E., Fecht, H. J., Fu, Z., and Johnson, W. L. 1989). J . Appl, Phys., 65, 305. Hong, S. I. and Suryanarayana, C. (2001). Mater. Ilrans., JIM, 42, 502. Jang, J. S, C., and Koch, C. C. (1990). J. Mater. Res., Karch, J., Birringer, R., and Gleiter, H. (1987). Natu 556. Kear, B. H., and Strutt, P, R. (1995). ~ a n o s t r ~ c t u r Mater., ed Alonso, T,, Liu, Y., Parks, T. C., and McCormick, P. G. 6, 227. (1991). Scripta Metall. Mater., 29, 1607. Kelton, K. F. (1993). ~nternat.Mater. Rev., Alonso, T., Liu, Y., Parks, T. C., and McCormck, P. G. Kelton, K. F. (1995). In Intern~etallicC o i ~ z ~ ~ o Prmciples un~: (1992). Scripta Metall. Muter., 26, 1931. and Practice, Vol. 1, Chapter 20. Ameyama, K., Uno, H., and Tokizane, M. (1994). Koch, C. C. (1993). NanoStructured Mater., Koch, C. C. (1997). N a J i o ~ ~ r u c f uMater., ~ed and Desu, S. B. (1995). Mater. Sci. Koch, C. C., Morris, D. G., Lu, K., and Inoue, A. (1999). Forum, 179-181, 207. MRS ~ulletin,24(2), 54. Atzmon, M. (1990). Phys. Rev. Lett,, 64, 487. Koeppe, C., Bartels, A., Seeger, J., and Mecking, H. (1993). Bakker, H. (1998). Enthalpies in Alloys Miedema’s SemiMetall. Mater Trans., 24A, 1795. Empirical Model, Trans Tech Publications, Zurich, Koster, U. (1997). Mater Sci. Forum, 2 3 ~ 2 3 Switzerland, vol. 1 of Materials Science Foundations, Larson, J. M., Luhman, T. S., and Mernck, H. F. (1977). In 1998. Mu~uf’acture of’ Superconductor Materiuls (ed. R. W. Bakker. H., Modder, I. W., Zhou, G. F., and Yang, H. Meyerhoff). ASM Iiiternational, Materials Park, OH, (1997). Mater. Sci. Forum, 2 3 ~ 2 3 $477, , p. 155. Bakker, H., Zhou, G. F., and Yang, H. (1995). Prog. Mater. Li, W., Suryanarayana, C., and Froes, F. M. (1995a). In Sci., 39, 159. Advances in Powder Metallurgy and Particulate material^^ Baro, M. D., SurinHch, S., and Malagelada, J. (1993). In (eds M. A. Phillips, and J. Porter). Metal Powder ~ e c h a n ~ Alloying ca~ for Structural Applica~ions(eds J. J. Industries Federation, Princeton, NJ, Part I, p. 145. deBarbadillo et al.). ASM International, Materials Park, Li, W., Suryanarayana, C., and Froes, F. H. (199%). In OH, p. 343. SynthesislProcessi?~gof light weigh^ Meta~lic Materials Bokhonov, B. B., Konstanchuk, I. (eds F. H. Froes, C. Suryanarayana, and C. M. Ward(1995). J. Alloys di Co~pounds, Close). TMS, Warrendale, PA, p. 203. Chang, H., Altstetter, C. .I., aud Averback, R. S. (1992). J. Liu, 2;. G., Guo, J. T., and Hu, Z. 0. (1995). Mater. Scz. & Mater. Res., 7, 2962. Eng., A ~ 9 2 1 1 9 577. ~, Cho, Y. S., and Koch, C . C. (1993). J. Alloys dz Co~pounds, Lou, T., Fan, G., Ding, B., and Hu, Z. (1997). J . Mater. 194, 287. Res., 12, 1172. Davis, R. M., McDermott, B., and Koch, C. C. (1988). et all. Trans. A19, 2867. Lu, I(.(1996a). Mater. Sci. Eng. Reports,
been most recently reported that the tensile strength of a fully dense (99.4% of the theoretical density), nanostructured (27 nm), pure copper obtained by elect~odepositionwas 119 MPa (against 70 MPa for an annealed coarse-grained copper sample). The ductility of this copper specimen was as much as 30%, which is large for a na~ostructuredmaterial (Lu et al., 2000). This high ductility lias been attributed to the minimization of artifacts such as flaws, contamination, residual stresses, etc. This observation rekindles the hope that an increased tensile ductility may still be possible to be achieved by producing fully dense and defect-free nanostruct~redinter~etallicsamples. The d has subject of ductility of ~ a ~ o s t r u c t u r ematerials been recently reviewed (Koch et nl., 1999; Wong and Suryanarayana, 200 1). Increased toughness has been, however, observed in fine-grained, mechanically alloyed y-TiAl specimens containing 50 mol. YO NiAl, as indicated by both microhardness and small punch testing (McMinn and Mao, 1995).
~
K. (1996b). In Processing and Properties of ~unocrystallineMater~als(eds C . Suryanarayana et al.). TNS, Warrendale, PA, p. 23. Lu, L., Wang, L. B., Ding, B. Z., and Lu, K. (2000). J. Mater.
Lu,
Sillith, C. D. W., Davies, G., and Saxl, 0. (2000). ~ a ~ ~ r ~ ~ World, S( l), SO. S~inivas~n, S., Desch, P. S ~ h ~ aR. r ~B., (1991). Scriptu Metat!. Mater., Stoloff, M.S., and Davies, R. C.(1966). Prog. Mater. Sci., 13, ., Shull, R. D., and Nash, P. (eds) (1997). 77. hem is try and Physics of Na~~ostr~ctures and elate^ Sui, H. X., Zhu, M., Qi, M., Li, G. B., and Yang, D, Z. Non-equilibriu~Materiuls, TMS, Warrendale, PA. (1992). J. Appl. Phys., 71, 2945. Maurice, D. A., and Courtney, T. H. (1990). Metall. Tram., Suryanarayana, C.(1995). Inter. Mater. Rev., G21, 289. ~uryanarayana,C. (1996). Metals & ~ u t e r ~ a l2,s ,195. McFadden, S. X., Mishra, R. S., Valiev, R. Z., Zhilyaev, Suryanarayana, C. (ed.) (1999). N ~ ~ - ~ q ~Processing ~ l i b ofr ~ ~ ~ A. P., and Mukherjee, A. K. (1999). Nature, 398, 684. M u t e ~ ~ u lPergarnon, s, Oxford, UK. McMinn, N. A,, and Mao, X. (1995). Scripta Metall. Mater., Suryanarayana, C. (2~00).Prog. ~ u t e rSei., , 33, 1915. Suryanarayana, C., Chen, G. I-I. Mikhailenko, S. D., Kalinina, 0. T., Dyunusov, A. K., (1992a). Mater. Sci. & Eptg., Fasman, A. B., Ivanov, E., and Golubkova, G. B. (1991). §uryanaraya~a,C., Sundaresan, R., and Froes, F. H. (1992b). Siber. J. C h e ~ .No. , 5. 93. Mater. Scs. Le Eng., AI Mizutani, U., Imaeda, C., Murasaki, S., and Fukunaga, T. Suryanarayana, C., Zhou, E., Peng, Z., and Froes, F. H. (1992). Mater. Sci. Forum, 88-90, 415. (19948). S c r ~ t aMetall. Marer., 30, 781. Mukhopadhyay, D, K., Suryanarayana, C., and Froes, F. H. G., Li, W., and Froes, F. N,(1994b). Scripta (1993). In ~ e c h a n i c aAlluying l for StruGtural ~ ~ ~ l ~ c a tSuryanarayana, ~ o ~ s Metall. Mater., 31, 1465. (eds J. J. deBarbadillo et al.). ASM International, Suryamrayana, C., Singh, J. and Froes, F. H.(eds). (1996). Materials Park, OH, p. 131. ~ r o c e ~ and s i ~Prop~rt~es ~ of ~ u n ~ c ~ " ~ ~ ~t u~ltle~rn~ea l s . Mukhopadhyay, D. K., Suryana~ayana,C., and Froes, F. H. (1994). Scri'pta ~ e ~ a lMa~er., l. 31, 333. TMS, Warrendale, PA. Wadsworth, J. (1991). Scripta Metall. Suryanarayana, C., Kor Froes, F. H. (1997). Metall. Mater, Trans 1, F., Pfullmann~Th., and B o ~ a n n R. , Suryanarayana, C., and 8). ~ ~ t e r n uJ.t . Rapid (1995). Appl. Phys. Left., 66, 941. ~ u l i d ~ ~ a3,~253. io~, Ohtani, T., Motoki, M., Koh, K., and QIishima, K. (1995). Suryanarayana, C., and Koch, C . C . (1999). In Note Mater. Res. Bull., 30, 1495. ~ ~ u i l ~ b rF ~ r o~c em~ s s ~ ~ gof ~ a t e r ~ a l s (ed. C. Radev, D. D., and Klissurski, D, (1994). J. Alloys d5: Surya~arayan~). Pergamon, Oxford, UK, p. 313. Co~pounds,206, 39. S u r y a ~ ~ ~ . a y aG., i i ~ ?and Koch, C. C . (2000). Hyperjpie Rittner, M. N., and Abraham, T. (1998). JOM, 50(1), 36. Interactio~s,130, 5. SchafTer, G, B. (1992). S c r ~ t aMetall. Muter., 27, 1. Suryana~~yana,C., and Norton, M. G. (1998). X-Ray Schropf, H., Kuhrt, C., Arzt, E., and Schultz, I., (1994). ~ s ~ r ~ cA~ Practical i u ~ : Approach, Plenum, New York, Scri'pta Metall. Mater., 30, 1569. NY. Seki, Y., and Johnson, W. L. (1990). In Solid State Powder Trudeau, M . L. (1994). In ~ a ~ o p h u sMater~als: e Synthe~is, ~ r o c e s s ~ n(eds g A. H. Clauer, and J. J. deBarbad~llo). P~ope~tie~s, A p ~ l ~ c a t ~ o(eds n s G. C, Hadjipanayis, and TMS, Warrendale, PA., p. 287. R. W. Siegel). Kluwer Acad. Publishers, Dordrecht, The Sen, S., Ram, M. L,, Roy, S., and Sarkar, B. K. (1999). 3. Netherlands, p. 153. Mater. Res., 14, 841. Valiev, R. Z., ~s~amga~iev, nd A~exandro~, I. V. iegel, R. W. (1991). In P~ocessingof Metuls and Alloys, (2000). Prog. Mater, Sci., vol. 15 of Materials Science and Technology - A Yen, B. K., Aizawa, T., and . (1996). Mater. Sci. Le Comprehensive Treatment (ed. R. W. Cahn). VCW Weinheim, Germany, p. 583, Siegel, R. W., and Fougere, G. E. (1994). In ~ a n o ~ h ~ e ~ a ~ e r i a (eds. l s G. C. Hadjipanayis, and R. W. Siegel). Zhou, G. F., and Bakker, H. (1996). S c r ~ t aMuter., 3, 29. Kluwer Acad. Pub., Norwell, MA, p. 233.
a
Simulations at the atomic level are beginning to play an important role in materials science, specifically in the area of intermetallic materials. In general, these atomic simulations take on two flavors. The first includes calculations of a small number of atoms, typically less than a hundred, with periodic boundary conditions used to simulate a bulk solid. These socalled ab initio or first principles calculations use a highly accurate Hamiltonian (energy functional) and are useful for assessing phenomena such as phase stability, point defect structure and energetics, or electronic properties, As input all that is needed are the atomic numbers of the el e~ e nt sbeing modeled. Their major limitation is that they are extremely computationally intensive, hence they cannot address many problems of interest in materials science that encompass mechanisms that depend on a large number of atoms and have important thermal or temporal effects, We will not discuss this type of calculation further here, but the reader is referred to Chapter 3 by Carlsson and Meschter of Volume 1 for further discussion. The second class o f calculations has the potential to address many atoms, occasionally up to many millions, in arbitrary geometry. These semiempirical calculations use an approximate or model Ha~iltonian that frequently has parameters taken from ex~eriment.The calculations are called semiempirical since the form of the Hamiltonian is derived from our basic understanding of physics and chemistry. Since these calculations are orders of magnitude faster than the first principles calculations described above, and scale linearly with the number of atoms, it is possible to calculate many properties not
accessible to first principles calcuIatio~s.The main limitation here is the accuracy of the potentials. A major advance in semi-empirical potentials occurred a nuniber of years ago when the importance of many-body interactions in metals was recognized. In simple terms, a many-body interaction represents the phenomenon that the bond strength and length between two atoms depends on the local environment of the bond, i.e. atoms not directly involved in the bond. The embedded atom method (EAM) was developed to capture this phenomenon (Daw and Baskes, 1983) and Chapter 4 by Voter in Volume 1. It is well known that the EAM is able to reproduce physical properties of many metals (Daw et al., 1993). A number of ~ n t e ~ e t a l l i chave s been sucGessfully modeled with the EAM and related potentials, especially the Ni/Al system (see Baskes (1995) for a compilation of potentials for this system). Unfortunately the use of the EAM is restricted to materials in which angular bonding is unimportant (Carlsson, 1990). proposed by (Baskes, was developed to extend the application to materials with all types of bonding. In the body of the chapter below, we will first review the EAM and MEAM formalism. Then a number of recent applications of atomistic simulations to a few ~ n t e ~ e t a l lsystems ic will be summari~ed.The applications have been chosen to be in order of increasing angular forces, from the Ni-A1 system, which has minimal angular forces, to the Ti-A1 system, where angular bonding is somewhat important, to the Mo-Si system where angular bonding is critical. We will of conclude with a view of the future for the si m u~~t i on intermetallics.
I n ~ ~ r ~ e ~ aCompounds: llic Vol. 3, PrincipEes and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 02002 John Wiley & Sons, Ltd.
766
Research T e c h ~ i ~ ~ e ~
The EAM was developed by (Daw and Baskes, 1983, 1984) over a decade ago. During this period of time the EAM and related methods, e.g. the N-body potential (Finnis and Sinclair, 1984), the glue-method (Ercolessi et al., 1986), the embedded defect method (Pasianot et al., 1991; Farkas et al., 1997) and the modified EAM (MEAM) (Baskes, 1987, 1992; Baskes et al., 1989; Baskes and Johnson, 1994), have become the mainstay of se~i-empirical atomistic calculations for intermetallics. The following sections will review the AM formalism with specific details given for multicomponent systems, which are, of course, necessary for intermetailics. Since a number of very similar methods will be described, the connections will be emphasized by the equation numbering system. E~uationsspecific to EAM will have plain numbers; the equivalent Nbody, MEAM, and embedded defect equations will have numbers with a suffix of N, M, and ED, respectively.
The pair potential has also taken on a number of functional forms. In the original EAM (Daw and Baskes, 1984) the form of a screened Coulomb potential was used:
where the screened charge Z was taken to be a spline or polynomial. More recently (Chen et al., 1989; Aiigelo et al., 1995; Valh6 and Farkas, 1997; Baskes et al., 1996, 1997) the form of a Morse potential has been used for each pair interaction. The energy of a ~ o n a t o m solid i ~ of atoms of type ti under a homogeneous d e ~ o ~ a t i as o na function of the nearest-neighbor distance R may be approximated by the universal equation of state (UES) of Rose et al. (I 984)
E; ( R ) = -Ef, (1 f a*)e-'* where
a" = a,,
od (EL4 The EAM has recently been reviewed by Daw (1993) and by Voter in Chapter 4 of Volume 1. We will include the basic forma~ismhere for completeness. The total energy, E, of a configuration of atoms in the EAM has been taken (Daw and Baskes, 1983) to be given by an approximation of the form:
where the sums are over the atoms i and j . In this approximation, the embedding function F', is the energy required to embed an atom of type ti into the background electron density at the site of atom i, pi; and q51,t, is the pair interaction between atoms of type ti and tj whose separation is iven by RU. In the EAM, p i is given by a linear superposition of ~pherically-ave~ag~d atomic electron densities, pt,:
(E
- 1)
and
where EtI, rt,, f i t , , and Bll are the cohesive energy, nearest-neighbor distance, atomic volume, and bulk modulus, respectively, of an atom of type ti, all evaluated at equilibrium. y apply in^ equation (1) and solving for the e~beddi ngenergy we obtain: -qlii(R)) = E';",(R) -
r:~ 1
~ q 5 ~ * ~ (7) ( ~ ~ ~ )
where the sum is over neighbor shells containing y21 atoms per shell and i q is the ratio of the lth-neighbor distance to the nearest-neighbor distance. The background electron density in equation (7) is given by:
ljm = E y 2 ~ ~ t ~ ( ~ l R ) 1
where the sum is over ail atoms j , not including the o ~the EAM atom at site i. In the original f o ~ u l a t i of (Daw and Baskes, 1984), the atomic electron densities were obtained from the atomic data tiables of Glementi and Roetti (1974). More recent f o ~ u l a t i o n s(Chen et al., 1389; Angelo et aE., 1995; Valh6 and Farkas, 1997; Baskes et al., 1996, 1997) have used analytic forms with free parameters.
(4)
(8)
In the N-body forinulation (Finnis and Sinclair, 1984; Ackland and Vitek, 1990) the total energy is given by equation (1) and the ernbedding function is taken as the negative square root of a term siniilar in form to the background electron density, i.e. Ft,(li,) = -4 f i where
(7N)
767
Defect and Atomic Process Simulatioiis (2p*J)
(1 1)
1
and Qi, takes on the meaning of a hopping integral rather than an electron density. Note that Qi, depends on both the i and j types, while in the equivalent expression in the EAM (equation (2)), p depends only on the typej. The mixed hopping integral is taken as the geometric mean:
where the tiJ are parameters that depend on the angular momentum state 1 as well as the type of atom tj. For a monatomic material tf = t l . The function G(r) has taken on a number of functional forms, the most widely used being: G(r) = d
The hopping integrals and pair potentials are usually evaluated as splines.
An important addition to the EAM was made by Baskes (1987, 1992) and Baskes et al. (1989), which allowed application to materials, e.g. silicon, silicides, or TiAl, in which bond-bending forces are important. The basic fornialism for the MEAM is the same as the EAM, i.e. the total energy is given by equation (1). The major difference is in the calculation of the background electron density. The angular or bond bending forces are captured in what are called partial electron densities that depend on the relative positions of three atoms through Legendre polynomials PI. These densities at the site of atom i are given by:
where p: is an atomic density for an atom of type tj and angular momentum state 1 and Ojik is the angle between atoms i, J , and k with atom i at the vertex. Since PO = 1, the E = 0 partial electron density py is equal to the background electron density defined above in equation (2) for the EAM. For MEAM we now define the background electron density at site i by: 2it
= PPGFi)
(2M)
where G(r;) captures all of the angular dependence and
where are weighting factors that depend on the environment of site i. We define these weighting factors by a simple average, weighted by the I = 0 atomic densities:
m
(12)
By choosing G(0) = 1 and G’(0) = 112, MEAM reduces to EAM in the limit of no angular forces, and the elastic constants do not depend on the f~nctional form of G. The atomic electron densities are assumed to decrease exponentially, i.e.
where the decay lengths, p:,, are constants and p! is a density scaling parameter, all of which depend on the type of atom ti. In the MEAM the embedding function is taken as:
where Atl is an adjustable parameter, which depends on the type of atom tt, and Zb, is the number of first neighbors in what we call the reference phase. To obtain the pair potentials we use a reference phase for both the pure elements and also for binary combinations of the elements. The reference phase for the elements is usually chosen as an equilibrium phase, e.g. fcc for Ni, but this choice i s not necessary. The reference phase for a binary alloy is usually chosen as a simple crystal structure that exists in nature, e.g. L12 (cP4) (Ni3A1) for Ni/Al, but again this choice is not necessary. The key point is that we need to know properties of the reference pliase, either from experiment or from first principles calculations, in order to determine the pair potentials. By using the universal equation af state (UES) for the pure element as above, and assuming first neighbor interactions only, we obtain the following expression for the pair potential: #tlr,(R)=
2 ZtI
- F~,(P!(R))I
( 3 ~ )
where p: (R) is the background electron density for the reference structure. The expression for the mixed pair potential is, in general, much more complicated, but in the case of an AB alloy with only opposite-type nearest neighbors, e.g. the B1 (cF8) or B2 (cP2) structure, the cross-pair potential is given by:
768
where E&, is the alloy UES and Z,,, is the number of nearest neighbors in the alloy structure. The embedded-defect (ED) potential (Pasianot et al., 1991; Farkas et al., 1997) is related to MEAM. In this method a function of p: is added to the total energy term, rather than to the background electron density. The total energy is given by:
The unction $is taken to be linear,
We discuss below a number of applications. There are three compon~ntsto any such calculation that must be addressed to assess the credibility of the results: 1) the technical details of how the simulation was performed (system size and ~eometry,initial conditions, boundary conditions, convergence criteria, etc.); 2) the interatomic potentials; and 3) the interpretation of the results, All three components are critical; failure in any of the three areas will lead to unreliable results, independent of the quality of the other two components. For component one, there must be a direct link to the real (experimental) world. Some important questions to be asked are: How close is the simulation to the e~perimentbein rnodeled; is there enough detail in the si~ulationto capture the appropriate rnechanis~~s? For example, in the case of dislocation mobility, is it s u ~ ~ i e to n t model a straight dislocation in a pure material, or must the model include three"dimensiona1 effects (kinks and jogs) or the presence of impurities? For component two, the interatomic potentials must be able to capture the relevant physics of the mechanisms being studied. For intermetallics, a large number of issues are related to ~ e c h a n i c aproperties. ~ Thus at a minimum, the potentials should represent the elastic properties of the material being studied. Potentials that do not represent the elastic properties may well d e s c ~generic i~ ~aterials,but are not useful for predicting the mechanical behavior of engineering materials. In addition to equilibrium properties, the potentials should describe configurations away from equilibrium, e.g. metastable phase stability. Grain
boundaries, dislocation cores, APBs, and surfmes, which are key to mechanical behavior, have properties far different from the bulk. If the potentials are to describe the structure and properties of these defects accurately, they must be carefully developed and compared to an e x ~ e r i ~ e n t aorl first principles database. As will be seen below, a large number of interatomic potentials, of vastly disparate quality, l~ave been developed for intermetallic systems. To try to keep them straight, potentials will be identified with the authors' initials. For completeness, results will be presented in the next section even €or ~ o t ~ ~ tthat ~als we feel are inadequate. The reader is cautioned to regard results from these potentials with caution. For c o ~p o n e n three, t the issues are similar to those encou~teredin an ~ x ~ e ~ i m~eo~ st i.t~ p o r t a ~do tl~~ the data justify the conclusion or are there alternative explanations?
There are, most likely, more calculations of the Ni/Al system than of all of the other intermet~llicsystems combined. Baskes (1995) recently reviewed eight semiempirical potentials of the EAM type that have been developed to describe the Ni/AI system. A summary af these potentials is given in Tables 1 and 2. Table 1 describes the source of the fu~~ctions as des~ribedabove in section 2.1. As indicated in the table, the authors' initials identify the potentials. The a simple, three-~arameter,empirical potential that is frequently used to represent the potential energy of diatomic molecules. Table 2 presents the details of the database used in the function determi~ati~n, Note that essential~yall of the functions use the lattice constant and cohesive energy of both Ni3AI and NiAl to determine the functions. Other quantities fre~uently used are the elastic constants and fault energies. All of the potentials are successful in reproducing the e x p e ~ m e ~ t alattice l const t for both Ni3Al and NiAl. The FD, VC, and F VT potentials have very poor predictions of the elastic constants for Ni3Al; and the FD, VC, and RB potentials have very poor elastic constants for NiAI. In summary only the RWP, VAC, BAM, and LG potentials yield reasonable agreement with experiment for both Ni3A1 and NiAI. It is r e c o m m e n ~e that ~ future calculations should use only the potentials that reproduce experiment. Many of the older calculations used the earlier, less reliable potentials; and the results from these calculations are suspect.
able 1 Source of functions for the Ni/A1 EAM potentials
Potential
FD VC RWP VAC RB
BAM FMVT LC
Reference
Electron density
(Foiles and Daw, 1987)
(Clementi and Roetti, 1974) empirical empirical cubic spline (Clementi and Roetti, 1974) empirical empirical empirical
(Voter and Chen, 1987) (Rao et al., 1991) (Vitek et al., 1991) (Rubini and Ballone, 1993) (Baskes et al., 1996) (Farkas et al., 1995) (Liidwig and Gumbsch, 1995)
A signi~cantamount of work has recently been p e r f o ~ e dto calculate ~oint-defectconcentrations in ~ i o i i - s t o ~ ~ h ~ o€32 n ~NiAl e t ~ ~(Mishin and Farkas, 1997a; Hagen and Finnis, 1998). There has been signi~cantconfusion in the d e ~ n ~ t i oofn point-de~ect energies in binary alloys, but the recent paper by Hagen and Einnis (1998) has helped to clarify a number of issues. In €act three recent potentials by Yaii et al. (1996) (YVC), ishin and Farkas (E997a) F), and Ludwig and Gumbsch (1995) (LG), were developed with the objective of having nickel vacancies, rather than a ~ ~ atoms, ~ on i the n nickel ~ sub-lattice as the dominant defect in Al-rich subs t o i ~ h i o ~ e tNiA1, r i ~ as found in experiment (Bradley and Taylor, 1337). It is worthwhile to suniniarize here the basic concepts of calculating point-defect energies and concentrations. We will limit the discussion here to binary alloys A,BI-, where x is near 0.5. The concepts are easily generalized to other compounds, e.g. AB3. There are
Embedding energy Like pair potential
Unlike pair potential
using UES
screened charge
geometric mean
using UES using UES square root using UES
Morse Morse cubic spline screened charge
Morse Morse cubic spline geometric mean
using UES using UES using UES
Morse Morse modified Morse
Morse linear comb~nation Morse
four defects that must be calculated, a vacancy on an A site (v-A), a vacancy on a site (V-B),an A atom on a B site ( A d ) , and a atom on an A site ( R A ) . The energies of these d cts are not unique, i.e.: they depend on the reference of ener y and energy partitioning between the two types of atoms. These four energies are used to calculate three energy differences that are independent of the zero of energy and how the energy is partitioned. These relative energies are used oltzmann factors to calculate the con~ntratioiisof the point defects. Let us start with a lattice containing N A sites and N I3 sites. We create each of the four defects, v-A,v-B, A B , and B-A and calculate the &-A, Eo, EA-B, -E=B-A where the calculation i s always performed for the stoichiomet~c alloy, hence the number of A and energies are given by:
2 Database for d e ~ ~ ~ i n aoft the i ~ alloy n potentials, Quantities fit include lattice constant (a), cohesive energy (E), elastic constants (c), vacancy formation energy (Ev),ordering energy (DE), fault energies (faults), point defect ~nerg~es (p.d.), and other phases e.g. Ni,Al,
Alloy Database Potential FD VC RWP VAC RB AM FMVT
Wi,AI
NiAl
a,E,faults a,E,e,E,,DE,faults
a,E a,E a,E,c,faults a,E a,E &,E a,E,c,fauIts f
LG
other phases ~,~,c,faults,p.d.
Research Techniques
770
The so-called 'raw' energies of Mishin and Farkas (19974 or those of Foiles and Daw (1987) are obtained from equation (14) by subtracting NEi,, where Ei, is the energy of an AB stoichiometric unit.
The energies of Hagen and Finnis (1998) are obtained from equation (14) by subtracting off energies based on the number of A ( N A )and B ( N s ) atoms:
+
where EA E s = Ei, and the Ns are obtained from equation 14(a-d) , Now consider the following relativc energies, each of which are between configurations with the same number of A and B atoms: the sum of v-A and v-B relative to perfect lattice, &!Zv; two 9-A relative to the sum of B A and perfect lattice, AEA; and two v-B relative to the sum of A-B and perfect lattice, AE,.
The relative energies represent: 1) the creation of two vacancies from perfect lattice; 2) the formation of two vacancies on the A-sublattice by removing a B atom from its anti-site and an A atom from its site and putting them back into the bulk as a stoichiornetric unit; and 3) the equivalent process on the B-sublattice. The first process controls the number of thermal defects while the latter processes control the number of
constitutional defects that are present in a nonstoichiornetric intermetallic. In Table 3 the relative energies are presented for a number of recent potentials that were developed using data for NiAl. For BAM and YVC, the relative energies at a number of temperatures are also given, There are minor differences between numbers presented for the YVC potential by Yan et al. (1996) and Hagen and Finnis (1998). Note that there appears to be signi~canttemperature dependence of the defect energetics, and that the two potentials where temperature dependence is available predict the opposite direction of this dependence. In order for nickel vacancies to be predominant in sub~stoichiometric NiAl at 0 K, A&, must be less than zero, Note that the LC and W C potentials were fit to satisfy this condition at 0 IS. All of the potentials have A E A ~ significantly greater than zero, hence it is expected that for x > 0.5 the anti-site defect will dominate. Taking the above energies and an entropy contribution that includes only the configurational component, we may approximate the free energy. By minimizing such a free energy expression ( ~ a g e nand Finnis, 1998) with respect to the number of defects, we can determine the defect concentrations as a function of temperature. The result of such a calculation is shown in Figure 1 for x = 0.48. Similar curves are obtained for other sub-stoichiometric coiupositions. For the potentials where temperature-dependent defect energetics are known (BAM and YVC), a linear interpolation of the temperature dependence has been used. We see that at elevated temperature (relevant to the experimen~s) all of the potentials shown give the nickel vacancy as the predominant defect, in spite of the fact that AENi is not always less than zero. For the RWP, FMVT, and which have significantly larger AENi, the anti-site defect is pi-edominant at all temperatures as expected. A number of factors govern these results. Most important is the effect of entropy. Since we are
able 3 Relative defect energies (eV) for a number of Ni/Al potentials at temperature T (K)
T (K) AEv 'ENi
A%,
MF (Mishin and Farkas, 1997a)
LLC (Liu et al., 1997)
0 2.46 -0.49 2.87
0 3.10 -0.09 1.20
LC (Ludwig and Gumbsch, 1995)
0 2.64 -0.08 2.49
(Yan 0 2.66 -0.04 3.27
YVC al., 1996)
et
1000 2.87 0.06 3.28
1400 2.96 0.13 3.31
RWP (Rao BAM et al., (Baskes et al., 1996) 1991) 0 2.30 0.28 2.77
1000 1.81 0.05 2.43
0 2.61 0.70 3.02
FMVT (Farkas et al., 1995)
RB (Rubini and Ballone, 1993)
0 2.44 0.86 2.88
0 2.44 0.90 2.51
77 1
Dgfect and Atomic Process ~ i ~ u l a t i o ~ s comparing the relative stability of two vacancies with a single anti-site defect, the entropy of the vacancies, which is higher than that of the anti-site defect, can significantly affect the free energy at high temperatures. Secondly, the relative defect energy is teniperature dependent. It is iniportant to use the relative energy at the appropriate temperature, as was demonstrated by Hagen and Finnis (1998). Finally, the vibrational entropy should be of some importance. Even though it is not included here, it seems reasonable that it also stabilizes the vacancy defects relative to the anti-site defects. All of the potentials predict that the anti-site defect is predominant for x > 0.5. The lesson learned here is that forcing the 0 K relative defect energetics to be less than zero was probably not a good criterion for determining potentials. Perhaps requiring this relative energy to be close to zero is appropriate, but the temperature dependence and entropy effects must also be considered. The group of Farkas and co-workers has published extensive calculations of B2 NiAl (Mishin and Farkas, 1997a, b, 1998; Mutasa and Farkas, 1998) and Ll2 Ni3Al (Farkas and Ternes, 1996; Farkas and Cardozo, 1998). Their calculations use the VC potential for Ni3Al. For NiAI, they use a recent modification of the FMVT potential (Mishin and Farkas, 1997a) (MF) that does not change the equilibrium properties of NiAL Using the M F potential they investigated the energetics of point defect migration in stoichiometric 8 2 NiAl ( ~ i s h i nand Farkas, 1997b, 1998). By calculating defect concentrations for non-stoichiometric alloys as determined from the discussion above, they conclude that for Ni concentrations below 52at.%
-
00
-
I
igure 1 Ratio of the concentration of nickel vacancies to the concentration of aluminum atoms on nickel sites for B2 Nio,48Alo.52 as predicted by a number of recent Ni/AI EAM potentials
Ni, diffusion is controlled by second-~eighborvacancy jumps, i.e. Ni atoms remain on their own sublattice. In contrast, above 52at. % Ni, the Ni atoms move by an anti-structural bridge mechanism (A Chang, 1993). These results are in at least qualitative agreement with the meas~rementsof Hancock and McDonnel (1971) who see a peak in the activation energy for self-diffusion of Ni in NiAl at the stoichiometric composition. These calculations rule out tlie 6-jump cycle mechanism (Elcock and McCombie, 1958) and the 4-ring mechanis~ (Zener, 1952) as viable mechanisms for Ni diffusion in B2 NiA1, since the predicted migration energies are high. Liu et al. (1997), using an EAM potential (LLC) that they derived in a way similar to VC, but fit to the properties of NiA1, calculated the binding energy of di-vacancies. They found that di-vacancies on the A1 sublattice were never bound. but di-vacancies on the Ni sublattice were bound by 0.2 eV at a distance of 2 a0 and di-vacancies on opposite sublattices were bound by 0.1 eV at a distance of f i j 2 a 0 , where c d ~ is the NiAl lattice constant. At all other distances tlie di-vacancies were not bound. In contrast, Mishin and Farkas (1997a) using the MF potential, found that both two A1 vacancies and two Ni vacancies were bound by 0.1 eV at a distance of ao. Such disagreement brings into question the predictive ability of EAM potentials with respect to point defect binding. Calculations with the BAM potential yield bin two Ni vacancies at a distance a0 of 0.2 eV at 0 repulsion of 0.1 eV at 1000 K. Thus we see that defect interactions in NiAl are strongly temp~raturedependent, as noted above for relative defect energies. The interaction of two Ni on A1 anti-site defects at a distance uo is repulsive by 0.15 eV at Q E; and by ~ 0 . eV 4 at 1000 K. From these numbers we may conclude that point defect binding appears to be uniniportant in NiA1, especially at the high temperatures of usual interest. N
-
-
-
-
3.1.1 Grain Boundaries
A number of studies focused on grain boundary properties. Farkas and Ternes (1996) calculated the interaction of vacancies with the C = 3 (1 12) grain boundary (in a E: = 3 boundary one-third of the atoms are in sites that are common to both crystals) in Ni3Al using the VC potential. Two boundaries were investigated, one Ni-rich and the second Al-rich. The maximum segregation energies for the two boundaries were similar with values of 0.6 eV for A1 and 0.5 eV for Ni vacancies. Very s i ~ i ~results ar N
N
772
Research ~ e c h ~ i ~ ~ e ~ ~
were found by Yan et al. (1996) using the "YVC potential for stoichioinetric C = 5 (210) and Z: = 13 (510) boundaries in NiA1. Binding of 0.5 (0.8) eV for A1 vacancies and 0.3 (0.4) eV for Ni vacancies was found for the C = 5 and (E = 13) boundaries, respectively. Yan et a/. (1996) used the YVC potential to study the strength of stoich~ometric[OOl] syrnmetrical tilt boundaries in NiAl. There are three boundary terminations: Ni/Ni, Ni/AI, and Al/Al. They found that the ideal grain boundary cohesive strengths were in the order Ni/Ni > Ni/AX > Al/Al. Thus it appears that processing techniques that are able to reduce the number of weak (Al/Al) boundaries would tend to reduce the susceptibility of NiAl to intergranu~~r cracking, It is interesting to note that the common Ni3Al that has excess Ni has the effect of reducing the number of 141-rich boundaries and thus improves the fracture properties. However, Baskes et al. (1996) using the BAM potential, found that in the presence of as little as 40 at.ppm hydrogen, these Nirich boundaries trapped significant amounts of hydrogen and were reduced in strength by 15%. Boundaries closer to exact stoic~iometrydid not trap nearly as much hydrogen and were not embrittled. For a further discussion of the aspects of hydrogen embrittlement, see the recent review (Liu et al., 1997). N
N
3.1.2 ~ i ~ l ~ c a t i ~ ~ , ~ Gurnbsch and co-workers have published a number of recent papers on dislocations in NiAl using the LG potential (Schroll et al., 1998a, b; Cumbsch and Schroll, 1999). They find that on the most frequently observed (100) { 01 1 slip system both edge and screw eierls stress of 0.1 GPa while the had approximately twice that barrier. These results are qualitatively similar to previous work using the RWP potential that predicted edge and screw dislocation barriers of 0.2-0.3 GPa and a mixed dislocation barrier of 1 GPa (Rao et al., 1991). Field et al. (1991) also observe (100) (010) dislocation motion in a soft orientation. Screw dis{ OlO} are predicted to cross-slip to (01 1} while mixed dislocations have a Peierls 0.15 CPa. Schroll et al. (1998b) also found that for the (111) { 01l} slip system, the edge dislocation had a low Peierls barrier of 0.1 GPa, but the screw dislocation was highly non-planar and hence had a much higher barrier (2 GPa). In contrast, the WP potential predicted similar behavior between screw dislocations w t h a barrier of e Guni~schgroup also substantiated the N
N
N
experimental (Mills and acle, 1993) and atomistic 5 ) with respect to decomcalculations (Mills et a!., position of the ( 110) edge dislocation into two (100) dislocations. In contrast to the calculations, the Schroll et al. (1998b) calculations needed an applied tensile stress to initiate the decomposition. In addition Schroll et al., (1998a) found that the ( 110)(21 1} edge dislocation d e c o ~ ~ o s einto d two mixed ( 100) dislocations, and the ( 1 11) (0 11} edge into two mixed dislocations dislocation decompo with (100) and (1 10) rgers vectors. Further calculations (Gumbsch and Schroll, 1999) show that both Ni anti-site defects and Ni vacancies interact with an assortment of dislocations to increase the Peierls barrier, but this increase is not sufficient to explain the increase in critical resolved shear stress with decreasing temperature observed in o~-stoichiometric NiAl (Baker, 1995).
3.2
~~~~~~~
Calculations in the Ti/Al system are inherently more difficult than the Ni/Al system since angular bonding is more important. The angular forces manifest themselves in the negative values of the Cauchy pressure in ) TiAl. The Cauchy pressure is the difference two elastic constants, c12 and C.Q.S. Most materials have a positive Cauchy pressure and are well described by EAM. Calculations in the Ti/Al system have emanated from the interatomic potentials of four groups: Chen et al. (1999) (CYL) who use the MEAM formalism, Vitek et al. (1997) (VCSIY) who use the N-body formalism, Rao et al. (1 99.5) (RWSD) who use the EAM formalism, and Farkas (1994) who has an EAM version (Fl) and an ED version (F2). The quality of the potentials may be assessed by examinin Figure 2. Here the ratios of the calculated properties (lattice constants (a, c), cohesive energy (.E)$and elastic to the experimentalvalues for Llo TiAl constants (Cc)) are shown. All of the potentials give reasonable values for the lattice constants and cohesive energy, but only the potentials that include angular dependence (CYL and F2) reproduce the elastic constants with a degree of accuracy. Only the CYL potential may considered to be a quantitative representation of TiAl. Paidar et al, (1999) compared the energetics of large deformations of TiAl using the VGSIY o and potential to that of LDA ab i ~ i t ~calculations concluded that the pote~tialsadequately mimic the first principles calculations. However, the results show far from quantitative agreement. Siegl et al. (1997) have investigated an ordered twin boundary in TiAl
Defect and Atomic Process ~ i m ~ E a t i o ~ ~
C
C
C
C
773
C
al for various potentials for L1, TiAl F~~~~~2 Ratio of the calculated to e x p e r i ~ e ~ tproperties
using high-resolution electron microscopy. They found that the local atomic structure at the boundary predicted using the VCSIY potential is in significant disagreement with the expe~mentalstructure. They attributed the disagreement to the covalent d-bonding between Ti atoms, which is not included in the central potential. It seems clear that the potentials of choice for calculations in the Ti/Al system must be those that include angular dependence. It is interesting to note that many properties do not depend on the an contribution, Farkas (1994) e potentials F l and F2, which compared results fr did not contain and contained angular forces, respectively. She found that the planar defects (APB and stacking fault) had less than a 5% difference in energy for the two potentials in Llo TiAl and a hypothetical Ll2 TiA13, the e ~ ~ i l i b r i uTiAl, m structure being DO22 (t18). Differences in c / a ratio and cohesive energy were also small. It is possible to trace the insensitivity of the planar faults and c / a ratio to the embedded defect method. In this method only the E = 2 component of the partial electron density i s used. It is known that the E = 3 component controls t ing fault energy and c / a ratio in pure materials and Johnson, 1994) in alloys (Baskes, and is also an importan 1999). It is expected that angular bonding would not affect the cohesive ener Ito and Vitek (1998 (modified to yield c / a = 1 to model coherent interfaces), have investigated possible segregation to a number of interfaces usin the Monte Carlo (MC) technique. They found no segregation takes place to ordered twins. In contrast, the calculations predict
that in Ti-rich lamellar TiAl, there is signi~cant segregation of the excess Ti at 120" rotational faults and pseudotwins. The local structure at these interfaces is the same as a thin layer of DO19 (h Ti3AI. The con cent ratio^ of Ti at the interface at K was found to be 53-55.5%, which is in good agreement with the e x p e r i ~ ~ n tvalue al of 55% (Inui et al., 1996). Point-defect energies in TiAl have been calculated using a number of potentials. Due to the complications mentioned above for NiAl, only stoic~iomet~c defects will be discussed here. Results are given in Table 4 where 'vacancy' represents the energy necessary to create both a Ti and an A1 vacancy, and 'anti-site' rep~esentsthe energy to switch an AI with a Ti atom. The angular dependent potentials (CYL and F2) give results in reasonable agreement with the ~ s t - ~ r i n c i ~calculationy les while the central potential gives smaller defect energi (1999) carried these calculations further U version of potential I21 (no angular calculated the mechanis~sof diffusion for both Ti and AI. It was found that at low temperature both Ti and A1 diffuse by the vacancy mechanism. At higher te tures the anti-st~ctural bridge ~ e c h a ~ s m becomes competitive for Ti, leading to the experimen observed non-Arrhenius behavior. high temperatures, the 3-jump an become more important. The calculations predict that the activation energy for AI diffusionis higher than that for Ti diffusion, in a ~ e e ~ ewith n t e x p e ~ m e ~"he t . Ti activation energy (2.40-2.45 eV) is in reasonable agreement with experbent (2.6 e'v), but the predicted AI activation energy (2.8 e'v) is s i ~ i ~ c a n tlower l y than the observed
Research Techniques
774
value (3.7 eV) inclusion of angular forces would improve the a It has been found that ternary B2 alloys of Nb-Ti-Al with high Ti content ( w 40%) exhibit at least 20% ductility at room temperature (Shyue et al., 1993). In order to study the ductility mechanism in these alloys, the Ti/Al potentials were extended by Farkas and Jones (1996b) to include the ternary addition Nb. These potentials used the EAM-based F l potential (no angular forces) and added Nb-Ti interactions in two different ways, since there is no accepted experimental value for the heat of mixing in the Nb-Ti system. The first potential is based on a n tive heat of mixing (FJA), and the second based a positive heat of mixing (FJB). Because of the crudeness of these potentials, the results should be considered more of a parameter study than a true representation of the ternary alloy. It is suggested that in the future, the heat of mixing be determined from first-principles calculations as done in the AljSi system (Gall et al., 2000). The potentials were used to calculate site occupancy, elastic constants, and fault energies of two ternary €32 alloys with Ti concentrations ranging from 1040% (Jones and Farkas, 1996). Both potentials showed that A1 atoms prefer to lie on a single sublattice, while Ti atoms lie on both sublattices. The calculated bulk modulus using potential FJA agreed with first principles calculations (Papaconstantopoulos, 1993), but potential predicted a bulk modulus about 40% low. In contrast, the predicted shear modulus using potential FJB agreed reasonably well with experiment (Hou et al., 1993), but the FJA potential predicted a shear modulus that is about twice the experirnental value. Thus it seems clear that neither potential represents the elastic properties of Nb-Ti-A1 alloys very well. It was also found that Ti additions energy significantly. Relaxed (1 10) APB energies were calculated to be 163 mJ/m2 (FJA) and 336 mJ/rn2 (FJB) for the 10% Ti alloy and 31 mJ/m2 (FJA) and 120 mJ/m2 (FJB) for the 40% alloy. The results for FJA are in good agreement with
Table 4 ~toichioinetricpoint defect energies (eV) in L1, TiAl for a number of %/A1 potentials compared to LDA at 0 K . A vacancy defect ~ ~ c l uadTi ~ sand an A1 vaca~cy CYL F2 (Chen et aE., (Farkas, S999) 1994) vacancy anti-site
3.63 1.14
2.73 0.90
FS (Farkas, 1994)
LDA (Yoo and Foo, 1993)
2.32 0.71
4.41 1.44
the experiniental values of 15 rnJ/m2 for the 40% alloy to 150-180 mJ/m2 for the 10% alloy (Shyue et al., 1993). Additional calculations have been performed to examine the dislocation cores (both (100) and (111) Burgers’ vector) of the B2 Nb-4O%Ti-l5%Al alloy (Farkas and Jones, 1996a). Since the predicted dislocation core structure using either potential is found to be similar to other B2 alloys, the authors conclude that dislocation-core effects are not the dominant factor in determining the enhanced ductility in these alloys. Farkas (1998) calculated the propensity for fracture in a pair of similar alloys with 16Yo and 33% Ti. It was found that the (110) surface energy decreased significantly with increasing Ti content. Thus a simple Criffith model would predict a lower brittle fracture stress for the higher Ti alloy. The results show that the 16% alloy fails in a brittle manner similar to other brittle B2 interrnetallics. In contrast the 33% alloy showed dislocation emission from the crack tip and enhanced ductility similar to bcc Fe. The difference in ductility was attributed to enhanced dislocation emission, which seeins to overcome the surface-energy effect. ons side ring the quality of these potentials, these results must be considered to be more representative o f a generic alloy than of the specific alloys used in the ~ x p e r i ~ e n t $ .
It is only recently that interatornic potentials have been developed for the Mo/Si system (Baskes, 1999). There are two significant reasons: angular bonding and complex crystal structure. As mentioned above, the Cauchy pressure is a good indicator of the extent of angular forces, and these pressures for C l l b (t16) MoSiz are highly negative. In ad~ition,in contrast to the Ni/Al and Ti/Al systems, where relatively simple crystal structures (B2, Ll,, Ll,) occur, only more complex phases (C11b MoSi2, A15 (cP8) Mo3Si, D8, (t132) MogSi3) appear in the Mo/Si phase diagram. Even though calculations in complex crystal structures are possible, development of the potentials and interpretation of the results is usually difficult. Baskes (1999), using the MEAM formalism, chose the B1 structure as a reference phase. Even though this system does not exist in nature as an equilibrium structure, this choice simplifies the determination of the potentials. Previously determined parameters were used for pure MO (Baskes, 1992) and Si (Baskes, 1997), and the four parameters that generate the cross-potential were indirectly determined by using the experimental values of the heats of f o ~ a t i o nand lattice constants of the
775
Defect and Atomic Process S i ~ u l a ~ i o n s three stable Mo/Si compounds and the bulk modulus of MoSi2. The model is very successful in predicting the relative phase stability for the Mo/Si system. In Figure 3 the formation energy at 0 K of a large number of phwes (A15, Cllb, D8m, B1, B2, C40 (WS), D88 (hl”l6), Llo, and L12) is shown as a function of com~osition.The bold lines connect the predicted stable phases. With the exception of two czses, the agreement with experiment is perfect. The model predicts that Mo3Si (A15) is unstable by 0.1 eV with. respect to decomposition into MO and NogSi3. It also predicts that MoSis (Llz) would be a stable phase, again by N 0.1 eV. The disagreement with experiment can be eXPlained by entropy effects, Or more likely, insu~ciencje~ in the model. It is especially encouraging that the hi&er s m e t r y B1, B2, ind Lc0 phases are predicted to be less stable than the experimentally observed phases. The agreement with the experimental lattice constmts is poor. For all three phases the predicted lattice constant ranges from 3-1 2% above the experimental results. However, the c / a ratios and internal atomic coordinates are in good agreement with experiment. The predicted elastic ~onstantsare reasonable (< 20% deviation from experiment) for MosSi3 (Chu et al., 1998), but poor for MoSi2 where the predicted value of c~ is about a factor of 8 too small. Predicted fault energies are in good agreement with experiment for MoSi2, the only case where experimental data exist. The predicted energy of the 1/4 ( 1111{ 110) SISF is 209 mJ/m2 compared to the experimental values of 261-365 mJ/m2 (Evans et al., 1993; Ito et al., 1995) and the 1/6(331~{013}APB energy is predicted to be 1313 mJ/m2 compared to the experimental value of N
I
,
824 mJ/m2 (Ito et al., 1996). The predicted formation energies of stoichiometric point defects are presented in Table 5. For Mo5Si3 multiple sites are possible for defects and the lowest formation energy is given. Defect energies appear quite high in MoSi2 and ~ o 3 S i , but low in MogSi3. It would only take 4 eV to create eight vacancies in MogSi3 and 0.6 eV to create an antisite defect. Such defects are expected to be important in a radiation-damage environment. These potentials are clearly just a first attempt to model the Mo/Si system. To be useful for quantitative prediction, much better potentials must be developed. N
Table 5 Predicted stoicliiotnetric point defect energes (eV) in the Mo/Si system at 0 K. A vacancy defect is defined as removing a stoichiometric unit, e.g. a MO and two Si VaCancleSfor
anti-site
MoSi,
Mo,Si
Mo,Si,
6.2 5. I
9.6 4.8
4.0 0.6
re What can we expect from atomistic calculations in the next 5-10 years? Let me first try to answer this question by predicting what we will not have. In spite of the rapid progress in potentials and in computer power, atomistic calculations will not be reliable enough to design new intermetallics. The reason is simple. To design an intermetallic we need multicomponent phase stability infoimation and mechanical, thermal, oxidation, etc. properties of multiphase materials. Even though I believe we will
S
Figure 3 Formation energy relative to elemental MO and Si of various compounds as a function of stoichiometry. The bold line connects the predicted stable phases. The arrows indicate the change in energy necessary to obtain perfect agreement with the ~xperiment~l phase d ~ ~ g r a ~
776
Research Teehni q w s
be making significant contributions in the phasestability arena, we are not even close to calculating properties of engineering materials. Here are my thoughts on how atomistic calculations will progress and how they will impact the intermetallics field: 1, ~ n ~ u l a r l y - d e ~ e nmany-body d~nt potentials will replace the central many-body potentials, just as many-body po ten tials replaced pair-po ten tials in the last decade. These potentials will be developed routinely for multicomponent systems using first-principles calculations as a data source for mixed atom interactions where experimental i n fo r ~atio nis lacking. 2. These potentials will become quantitative in their prediction of t h e r m o d y n ~ ~ iproperties, c new crystal structures, and defect energetics and geometries. 3. Using these accurate potentials we will begin to understand deformation mechanisms in complex crystal structures. We will be able to utilize this information as we currently utilize our underof deformation in fcc and bcc materials to design multiphase materials. 4. Environmental elements, e.g. 0, S, H, will be modeled as well as the intermetallic components. Using results from quantum chemistry calculations, we will begin to understand the role of surface defects in the environmenta~degradation of properties. These calculations should be able to guide the experimental& in the development of surface alloys and coatings to prevent high temperature oxidation. 5. Large (> 106 atoms) calculations of multiphase polycrystalline materials will begin to give us insights of how boundary deformation can be an important source of ductility in inherently brittle i~iterm~tallics, via dislocation generation, boundary sliding, etc. 6, ~e c h an is ticinfo~inationgleaned from the atomistic calculations will be used as input to mesoscale models which have the promise of being able to predict plastic properties of complex microstructures.
This work was supported at Los Alamos National Laboratory by the US DOE under contract W-7405ENG-36.
Ackland, G. J., and Vitek, V. (1990). Many-body potentials and atomic-scale relaxations in noble-metal alloys. Phys. Rev. 3 , 41(15), 2032633. Angelo, J. E., Moody, N. R., and Baskes, M. I. (1995). Trapping of hydrogen to lattice defects in nickel. Modelling Si~ul.Mater. Sci. Eng., 3(3), 289-307. Baker, I. (1995). A review of the mechanical-properties of B2 compounds. Mat. Sci. Eng. A, 192/193, 1-13. Baskes, M. I. (1987). Applicat~on of the embedd~d-atom method to covalent materials: a semiempirical potential for silicon. Fhys. Rev. L A . , 59(23), 2666-9. Baskes, M. I. (1992). Modified embedded~atom potentials for cubic materials and impu~ities.Phys. Rev. hr, 2727-42. Baskes, M. I. (1995). Atomistic in~erme~allics .Acta Me ta~lur~ica Baskes, M. I, (1997). Calculation Ad-dimers on Si (001). Mtxklling Simiul. M a m . Sri. &g., 5(2), 149-58. Baskes, M. I. (1999). A t o ~ s t i c pQtentiais for the ~olybdenum-siliconsystem. Mater. Scl. Efigr, A , 2 2), 165-8. Baskes, M. I., Nelson, J. S., and Wright, A. F. (3989). Semiempirica~ modi~ede m ~ e d d e d - a t opotentials ~ for silicon and germanium. Phys. Rev, B, 40(9), 6085-1 00, Baskes, M. I,, Angelo, J, E., and Moody, I?. R. (1996). Atomistic calculations of hydrogen i n t e ~ ~ c t i owith ~s Ni3Al grain boundaries and Ni/Ni3Al interfaces. In Hydrogen Eflects in Materials (eds A. W. Thompson, and N. R. Moody). ~ a rre n d a l e ,PA, The Minerals, Metals & Materials Society, 77-90. Baskes, M. I., Sha, X., Angelo, J. E., and Moody, I?. R. (1997). Trapping of hydrogen to lattice defects in n i c k e ~ . ~ o d e l ~ iand n g Simul. in m at^^. Sci. Eag., 5(6), 651-2. Baskes, M, I., and Johnson, R. A. (1994). Modified embedded atom potentials for hcp metals. ~ o d e l ~ i nSi~ul. g Mater. Sci. Eng., 2(1), 147-63. Bradley, A. J., and Taylor, A. (1937).Proc. Ray. Soc. A, 1
56. Garlsson, A. E. (1990). Beyond Pair ~ o ~ e ~ ~ini Ele~ental als ~ransit~on Metals and ~ e m i c o ~ ~ u c ~Academic ors, Press, Boston. Chen, D., Yan, M., and Liu, Y. I;. (1999). A modified embedded-atom potential for Llo y-TiAi. S c r ~ t aMater., 40(8), 913-20. Chen, S . P., Srolovitz, D. J., and Voter, A. F.(1989). Computer simulation on surfaces and [OOl] s ~ m e t r i c tilt grain boundaries in Ni, AI, and Ni,Al. J. Mater. Res., 4, 62. Chu, F., Thoma, D. J., McClellan, K., and Peralta, P. (1998). Mo5Sij single crystals: physical ~ropertiesand mechanical behavior. Mater. Sci. Engr. A , 261( 1--2), 44-52. Clementi, E., and Roetti, C. (1974). AfoomicData and Nuclear Data Tables, Academic Press, New York.
Defect and Atomic Process ~ ~ ~ z ~ ~ l ~ t i o ~ s Daw, M. S., and Baskes, M. I. (1983). Semiemp~rical,~ u a n t u m mechanical calculation of hydrogen embrittlement in metals, Phys. Rev. Lett., Daw, M. S., and Baskes, method: derivation and application to impurities, surfaces, and other defects in metals. Phvs. Rev. B, 29, 6443. Daw, M. S., Foiles, S. M., and Baskes, M. I. (1993). The embedded atom method: a review of theory and applications. Mater. Sci. Rep., 9(7-8), 251-3 10. Elcock, E. W., and McCombie, C. W. (1958). 6-jump cycle. Phys. Rev. B, 109, 6. Ercolessi, F., Tosatti, E., and Parnnello, M. (1986). Au(100) reconstruction in the glue model. S21vf. Sci., 177, 314. Evans, D. J., Court, S. A., Hazzledine, P. M., and Fraser, H. L, (1993). Dislocation dissoc~ationin the inte~etallic compound MoSi2. Phil. Mug. Lett., 67(5), 331-31. Farkas, D. (1994). Interatomic potentials for Ti-AI with and hout angular forces. ~odellingS ~ ~ umlat. er. Sci. Eng.,
777
Hancock, G. F., and ~cDonne1,B. R.(1971). Plzys. Stut. Sol.
rzeorski, T., and Mishin, U. (1999). Selfdiffusion in y-TiAl: an exper~mentalstudy and atomistic calculations. In~er~etallics, 7, 389-404. Hou, D. H., Yang, S.-S., Shyue, J., and Fraser, M.L. (1993). Investigation of B2 and related phases in Ti-modified NbA1 alloys. MRS P~o~eeclings, 3 Inui, H., Kishida, K., Kobayashi, M., and Yamaguchi, M. (1996). Phil. Mug. A, 74, 451. Ito, K,, Inui, H., Shirai, Y., and Yamaguchi, M. (1995). Plastic deformation of MoSi2 single crystals. Phit. Mag. A , 72(4), 1075-97. Ito, K., and Vitek, V. (1998). An atoniistic study of segregation to lamellar interfaces in non-stoic~ometric TiAl alloys.Acta. Muter., Ito, K., Yano, T., Nakanioto, T., Inui, H., and Yamaguchi, M. (1996). Plastic deformation of MoSi2 and WSi2 single crystals and direc~iona~~y solidi~edMoSi*-based alloys. interwietallics,4, S 11 9 413 1. 998). Atomistic shulatioiis of fracture in the B2 Jones, C., and Farkas, 19. (1996). Embedded atom simulation phase of the Nb-Ti-A1 system. Marer. Sci. Eng. A , 249, of the B2 phase in Nb-Ti-Al. Comp. M ~ ~ eSci., r. 249-58. Kao, C. R., and Chang, Y. A. (1993). On the compositio~~l Farkas, D., Mutasa, B., Vailhe, C., and Ternes, K. (1995). dependencies of self-diffusion coefficients in Interatomic potentials for B2 NiAl and niartensitic phases. Mo~ellingS i ~ ~m lat. er. Sci. Eng., 3, 201-14. intermetallic c o ~ ~ o u n dIntermetallics, s. 1(4), 237-50. Farkas, D., VailbC, C., and Panova, J. (1997). Empirical Liu, C., Stringer, J., Mundy, J., Horton, L., and Angelini, P. angular-dependent potentials for intermetallics. J. Phase (1 997). Ordered iiiteriiietallic alloys: an assessment. Equil., 18(6), 530-5. rnter~etallic.~, 5(8), 579-96. Farkas, D., and Gardozo, F, A, (1998). The multiplicity of Liu, Z., Lin, D., and Chen, D. (1997). Atsinistic simulation possible grain boundary structures in Ni3AI. rnter~etall~cs, of point defects in NiAl alloy. Trans. ~ o ~ ~ e rMet. r~z~s 6, 257-68, Soc. China, 9(2), 128-31. Farkas, D., and Jones, C . (1996a). ~ i s ~ o c a t i ocore n structure Ludwig, M., and Gui~bsch, P, (1995). An empirical in the B2 phase of Nb-4O%Ti-l5~~Al. Comp. Mater. Sci., interatomic potential for B2 NiAl. M ~ ~ e l Simul, l ~ ~ g 6, 295-302. Mater. Sci. Eng., 3(4), 533-42. Farkas, D., and Jones, C. (1996b). Interatomic potentials Mills, M. J., Angelo, J. E., Daw, M. S., Winbur~,J. D,, and for ternary Nb-Ti-A1 alloys. Mo~ellingS i ~ u l Muter. . Sci. Miracle, D. B.(1995). Fine-structure of a[01I] dislocations and the mechanical-properties of NiAl in the hard Farkas, Cr., and Ternes, K. (1996). Atomic study of the orientation. Mat. Sci. Interaction of lattice with grain boundaries in Mills, M. J., and Miracle, (1993). The structure of Ni3Al. inter~etallic,~, a( 100) and a( 110) disl Field, R., Lahrman, D arolia, R. (1991). Room Mater., 41, 85-95. temperature deformation in ‘soft’ orientation NiA1 single Mishin, Y . , and Farkas, D. (1997a). Atomistic simulation of crystals. ~ ~ g h - T e ~ p e O~ ~r t ~~ rinter~e~allic e~ ~ e ~ Alloys point-defects and diffusion in B2 NiA1. 1. ~ o i ~ t - d e f e c t IV (eds L. Johnson, D. Pope, and J. Steigler). Mater. Res. energetics. Phil. Mug. A , 7 Mishin, Y., and Farkas, D. (1997b). Atomistic simulation of Finnis, M. W., and Sinclair, J. E. (1984). A simple empirical point-defects and diffusion in €32 NiAl, 2. Diffusion N-body potential for transition metals. Phil Mug A , %(I), mechanisms. Phil. Mag. A , 75(1), 187-99. 45-55. Mishin, Y., and Farkas, D. (1998a). Atomistic simulation of Foiles, S. M., and Daw, M. S. (1987). Application of the point defects and diffusion in I32 NiA1. Scripta Muter., embedded-atom method to Ni3Al. J . M a t ~ r Res., . 2, 5. 39(4/5), 625-30. Gall, K., Ho~stemeyer, M. F., Baskes, M. I., and van Mutasa, B., and Farkas, D. (1998). Effect of ordering energy Schilfgaarde, M. (2000). JMPS, in press. and stoichiometry in C = 5 boundaries in B2 Compounds. Gumbsch, P., and Schroll, R. (1999). Atomic aspects of the Metall. and Mater. Trans. A , 29, 2655-68. deformation of NiAl. inte~~etallics, 7 , 447-54. Paidar, V., Wang, L. 6..Sob, M., and Vitek, V. (1999). A Wagen, M., and Finnis, M. W. (1998). Point defects and study of the applicability of many-body central force chemical potentials in ordered alloys. Phil. Mag. A, 77(2), potentials in NiAl and TiAl. Modelling Sirnul. Muter. 447-64. Sci. Eng., 7 , 369-81.
Papacons~antopoulos,D. (1993). First Principles Calculantions of El~c~ronicS t r ~ c t ~ ~Total e, Energy and Elastic P ~ o ~ e r t ~Naval e s , Research Laboratory. Pasianot, R., Farkas, D., and Savino, E. (1991). Empirical many-body interatoinic potential for bcc transition-
Einpiricd interatomic potentials for Llo TiAl and B2 NiAl. Mat. Res. Soc. Proc., 213, 125-30. Rao, S . I., Woodward, C., Simmons, J., and Dimiduk, D. (1995). Mater. R m Soc. Symp. Proc,, 364, 129. Rose, J. H., Smith, J. R., Guinea, F., and Ferrante, J. (1984). Universal features of the equation of state of metals. Phys. Rev. 3, 29(6), 2963-9. Rubmi, S., and Ballone, P. (1993). Quasiharmonic and molecular dynamics study of the martensitic transfori~a~ion of Ni-A1 Alloys. Phys. Rev. B, 48( I), 99111. Schroll, R., Finnis, M. W., and Gumbsch, P. (1998a). Energies of defects in ordered alloys: dislocation core energies in NiAl. Acte Mater., 46(3), 919-26. Scbroll, R., Vitek, V., and Gumbsch, P. (1998b). Core properties and motion of dislocations in NiAl. Acta Shyue, J., Hou, D., Aindow, M., and Fraser, H. (1993). Mater. Sci. Eng. A, 170, 1.
Siegl, R., Vitek, V., Inui, H., Kishida, R., and Yamaguchi, M. (1997). ~irectionalbonding and asymmetry of interfacial structure in i n t e ~ e t a ~ lTiAk i c combined theoretical and electron microscopy study. Phil. Mag. A , 7§(5), 1447-59. ValhC, C . , and Farkas, D. (1997). Shear faults and dislocation core structure simulations in B2 FeAl. Acta Mater., 45(1 11, 4463-73. Vitek, V., Ackland, G. J., aiid Cserti, J. (1991). Atomistic modeling of extended defects in metallic alloys: dislocations and grain boundaries in L12 compounds. Mat. Res. Soc. PYOC.,186, 237-51. Vitek, V., Girshick, A., Siegl, R., Inui, H., and Yamaguchi, M. (1997). In Properties of C o ~ p l eInorga~jc ~ Solids (eds A. Gonis, A.-M. Meike, and P. E. A. Turchi). Plenum, New York, 355. Voter, A. F., and Chen, S. P. (1987). Accurate int$ratomic potentials for Ni, Al, and Ni,Al.Mat. Res. Soc. Symp, Proc., 82, 175. Yan, M., Vitek, V., and Chen, S. P. (1996). Acta. Mater., 44, 4351. Yoo, M., and Foo, C. (1993). Bonding mechanisms and point-defects in TiA1. Inter~~~etallics, 1(1), 59-63. Zener, C. (1952). In ~ m p e ~ f e c t i ~innNearly s Perfect Crystals (eds VV. Shockley, J. H. Ho~lomon,R. Maurer, and F. Seitz). Wiley, New York, 289.
Bell Laboratories, ~ ~ r r Hill, a y NJ,
s ~ e ~ ~ r t mofe nElectrical t and Computer Engineering & ~ ~ o t u n i Center, cs oston ~ n i v e r s i tM ~ ,A , USA
Molecular beam epitaxy (MBE) is a thin-film deposition process in which thermal beams of atoms or molecules react on the clean surface of a singlecrystalline substrate, held at high temperatures under ultrahigh-vacuum conditions, to form an epitaxial film. Thus, contrary to the chemical vapor deposition (CVD) processes where chemical reactions play an important role, the MBE process is a physical method of thin-film deposition. The vacuum requirements for the MBE process are typically better than 10-l0 Torr. This makes it possible to grow epitaxial films with high purity and excellent crystal quality at relatively low substrate temperatures. Additionaily, the ultrahigh-vacuum environment allows the study of surface, interface and bulk properties of the growing films in real time, by employing a variety of structural and analytical probes, Although the BE deposition process was first proposed in 1958 (Gunther, 1958) its implementation had to wait for the development of the ultrahigh vacuum technology, when it was successfully applied for the growth of epitaxial GaAs films (Davey and Pankey, 1968). The development of the MBE process in its present state was primarily motivated by the desire to study new quaiitum phenomena in semiconducting synthetic T h r o u ~ ~ o uthis t chapter periodic, deposited layer super in the MBE structures that are referred to as su~erlattic~s literature will be indicated by ‘superfattices’ in quotes to distinguish them from convention~l, equilibrium, ordered compounds.
‘superlattice’ structures’ (Esaki and Tsu, 1970). The demonstration of such phenomena required the growth of ‘superlattice’ structures with atomically abrupt and perfect interfaces and control of the layer thicknesses down to a single monolayer. The development of crystal-growth techniques, which led to such a degree of lattice and interface perfection, has been attained by the contributions of many disciplines of science and technology. Although the original focus of the MBE process was to grow materials and devices of the 111-V family, the method has proven very successful in the growth of other semiconductors, insulators, metal and intermetallic compounds. A partial list of materials grown by MBE is presented in Table 1, and the reader is referred to a number of review articles and books on the subject (Ploog, 1980; Gho, 1983; Gossard, 1982; Chang and Ploog, 1985; Parker, 1986; Davies and Williams, 1986; Foxon and Harris, 1987; Foxon and Joyce, 1990). In this chapter, we review progress in the deposition of inte~metalliccompounds by the MBE process. Research in depositing intermetallic-co~~poundthin films by the MBE method was motivated by the desire to obtain epitaxial metallic films on 1x1-V semiconductors such as GaAs. Such epitaxial films do not contain grain boundaries, which contribute to contact degradation through diEusion processes. Thus, they can potentially form stable contacts that can withstaiid high-temperature device-processing steps. Section 2 gives a more detailed descripti~nof MBE technology and of analytical tools compatible with the ultra-~gh-vacuu~process environment. Section 3 addresses the growth processes, s t r u c t ~ eand properties
~ n t e ~ ~ eC t ~~ li ln i~~~ u Vol. n d s3,~ Priaciples and Practice. Edited by J. €3. Westbrook and 0 2 0 0 2 John Wiley & Sons, Ltd.
Research T e c h ~ i q ~ e ~ ~
780 Partial list of ~ ~ t e r i agrown ls by MBE 111-v GaAs GaSb GaP AlAs AlSb InP InAs InSb (A1,Ga)As (In,Ga)As (A1,In)P (Ga,Tn)P (Ga,As)Sb (In,Ga)(As,P) (Al,I~,Ga)P (Al,~a,In)As (Al,Ga,In)Sb
111-N
IV
11-VI
Insulators
Metals
Ge Si SiGe
ZnS ZnTe CdS CdSe CdTe CHg,Cd)Te
CaF2 SrF2 BaF2
&-Fe bcc-Co hcp-CO A1
a-Sn
of various internietallic compounds grown by MBE. Finally, Section 4 deals with device applications of inte~etallic-compoundtKin films grown by MBE.
Intermetallic compounds CoGa NiGa CoAl NiAl FeAl AuGa2 PdTe Fq(A1,Si) ErAs LuAs YbAs Er(P,As) (Sc,Er)As
A schematic diagram of an MBE growth chamber is shown in Figure 1. The facilities in this chamber are ca~ableof forming and m o ~ i t o r i ~the g ultr~hi~hvacuum environment, of heating and mo~itoringthe temperature of the substrate fairly accurately, of generating and determ~ningthe intensity of the molecular or atomic beams, of controlling composition profiles through beam interruption, and of studying surface and interface phenomena during film growth. Modern MBE deposition systems are designed to The primary pumping of the growth chamber is produce high-quality materials and devices at high accomplished with a combination of either storage throughput. The requirement of maintaining an ultra~ s t i t a ~ si ~ ~ ~ n, and c r y ~ ~ ~ ~ high v a c u u ~ e n v ~ r o n ~ e nwhile t s i ~ u l t a n e o ~ s l ~~ u m (ion, facilities) or th~oughput (diffusion or turb improving the throughput was addressed through the the nature and vapor molecular pumps) depend design of MBE systems consisting of multiple chambers pressure of the evaporants. Ad~itionally,the sources separated by gate valves. All ~ommercially available and the substrate, which employ a large heating load, equi~ment is const~cted with at least three such are surrounded by a cryopanel, usually cooled by chambers. The first chamber serves for sample introliquid nitrogen. This secondary pumpi~gminimizes the duction and is capable of medium high vacuum (10-6 unintenti~nal incorporatio~ of im~urities into the to 10-* Torr). The second chamber is capable of growing film. Such a combination of pumping, after ultrahigh vacuum and acts principally as a buffer a typical system bakeout of a p p r o ~ i ~ a t e 24 l y hours at between the introduction and growth chambers. This 250 "C, results in base pressure below the X-ray limit chamber is also used for substrate preparations, such of the ionization gau e (< 2 x l O - l I Torr). The as outgassing or sputter etching, and for accomprincipal impurities, as monitored by a residual gas modation of surface analytical facilities such as analyzer, are Hz, H20, CO and CO2 at partial Auger electron spect~oscopy(AES), secondary ion pressures typically less than 5 x 10"'l3 Torr. mass spectroscopy (SIMS), X-ray photoelectron The design of the substrate holder a l l o ~ ssubstrate spectrometry (XPS), and ultraviolet photoelectron rotation during film growth and additional motions to spectroscopy (UPS). The third chamber, the growth fa~ilitatesurface analysis and beam flux monitorin chamber, is capable of ultrahi h vacuum (< l0-lo The substrate, which is usually held to a molybden~m Torr), and its design criteria greatly depend on the block with indium solder, is heated ra~iativelyand its nature of the ~aterialsbeing deposited.
~ o l e ~ uBeam l a ~ Epitaxy
78 l
HE
S
.f
1 Schematic of an MBE growth chamber
temperature monitored with an optical pyrometer. All parts of the substrate holder are fabricated with ultrahigh purity and refractory materials. The uniform substrate heating in such designs results in thickness and doping uniformity between 0.5 to 1% over a 2inch wafer. The most common method of creating molecular rowth is through the use of Knudsen e~u$ioncells. In ideal Knudsen cells, the orifice diameter should be less than the mean free path of the vapor molecules within the cell, and the beam flux can be calculated from the e~uilibriumvapor pressure, using kinetic theory. In practice, however, the molecular beam sources are not ideal Knudsen cells, since they employ large apertures, which are necessary to achieve enha~cedgrowth rates and better compositional unifo~~ities. Thus, the beam fluxes are usually measured with a nude ionization gauge, placed at the location of the substrate. The crucibles employed in Knudsen cells can be made from a variety of materials. Pyrolytic boron nitride (PBN) appears to be the preferred material for the growth of 111-V compounds.
The temperature of the crucible is controlled to within f l “C. There are a number of alternative sources for creating molecular beams. Most p r o ~i n e n tamong them are electron beam evaporation and gas sources. Electron beam evaporation is commonly used for low vapor pressure materials such as silicon and ref~actory metals. Gas sources have been developed in several laboratories and implemented in commercia~MBE systems. Such gas sources include for example AsH3 and PH3 to produce the group V elements (Panish, 1980; Calawa, 1981). These gases are thermally ‘cracked’ to the dimers As2 and P2 before they reach the substrate. Thus, the process i s not a chemical vapor deposition. The current tendency is to develop gas sources for the group I11 elements usin sources. The process of using such gas sources is called MO-MBE in analogy to MO-CVD. Control over the film com~osit~on and doping profile is attained by incorporating a mechanical shutter in front of each source. Since the flow of molecules or atoms from the source to the substrate is
782
Research T e c h n i q ~ ~ ~
in the molecular rather than the hydrodynamic flow regime, positioniiig a shutter in front of a source will effectively stop the beam from reaching the substrate. In fact it is because the flow regime is molecular that the process is called molecular beam epitaxy. Since the growth rate of the MBE process is inherently slow (--J1 monolayer/s), shutter actuation times of a fraction of a second are required to produce compositionallymodulated materials with interface smoothness of one atomic layer. While the majority of surface analytical probes are accoiiimodated in the preparation chamber to avoid possible contamination by the evaporants, reflection high-energy electron diffraction (RHEED) is routinely used in the growth chamber to monitor and control the growth process. It is generally recognized that this method has played a major role in the development of the MBE process. RHEED consists of a well-collimated, monoenergetic electron beam, which is directed at a grazing angle of about 1" toward the substrate. The primary electron beam has an energy of between 10 to 20 keV, resulting in an energy component perpendicular to the substrate of about 100 eV. Thus, the penetration depth of the incident electron beam is approximately equal to a few atomic layers. As a result, a smooth crystal surface acts as a two~dimensionalgrating and diffracts the electrons. The diffraction pattern is formed on a fluorescent screen placed diametrically opposite to the electron gun. RHEED is routinely used to study thermal desorption of oxides prior to growth, to control the initial stages of epitaxial growth, and to study surface reconstruction as a function of growth parameters. The removal of oxides from a substrate is the first step prior to epitaxial growth. In some cases this step takes place in the preparation chamber in order to avoid contaminat~ngthe growth chamber just before the initiation of epitaxy. However, for 111-V compounds, such as GaAs, oxide desorption takes place in the growth chamber, in the presence of arsenic overpressure, in order to prevent surface fractionation after the removal of the oxides. A clear RHEED diffraction pattern indicates that all oxides have been removed. Another potential application of RHEED i s the study of surface topography. The RHEED pattern from a smooth, single crystalline surface is expected to have the form of a series of streaks running perpendicular to the surface of the crystal, which is consistent with a two-di~ensional diffraction. On the other hand, tlie RHEED pattern froin a rough surface is expected to be spotty, since the p~netration of
electrons through surEace asperities results in threedimensional diffraction. RHEED is also used to study surface reconstruction. For example, the (100) GaAs surface, used for MBE growth of GaAs, reconstructs to different configurations in order to lower its free energy. Since surface reconstruction leads to lower symmetry than that of the bulk crystal, extra diffraction lines are expected in the RHEED pattern. Besides these RHEED applications, it has been observed that when the g r o w t ~ is i~itiated,the intensity of the RHEED features shows an oscillatory behavior. Current thinking is that these intensity fluctuations are related to both crystal growth and electron diffraction phenomena. The influence of diffraction in the intensity enhancement is probably due to the multiple scattering originating from the beam's penetration into the solid. Early ~nterpretatio~ of the RHEED intensity oscillations is that the thin-film growth proceeds in a layerby-layer mode and thus the period of the oscillations corresponds to a rnonolayer growth (Foxon and Harris, 1987). These results can be used to calculate growth rates and composition of ternary and quarternary 111-V compounds. Figure 2 shows how the composition of the (A1,Ga)As can be calculated from the growth rates of their components of Ca and Al in (Ga,Al)As (Moustakas, 1988). From the previous discussion, it is apparent that compared to the other methods of thin-film epitaxial growth, molecular beam epitaxy has some unique advantages which can be su~marizedas follows: (a) The growth rate is generally low, approximately 1 monolayerfs. This allows compositional and doping profile changes to be specified w i t ~ atomic i~ dimensions through the actuation of mechanical shutters; (b) the growth teniperature is relatively low and thus interdiffusion between layers of different composition is negligible; (c) the MBE growth mechanism leads to atomically smooth surfaces; (d) the ability to study growth phenomena in real time provides o~portunities for scientific innovation and quality control in the production environment; and (e) all steps of the MBE deposition process can be fully automated. These favorable features, together with significant progress in the design of modern MBE systems, have led to the growth of films with excellent thickness and doping u~iformities,and excellent crystal quality over large area substrates, Currently, the commercial MBE systems are designed for up to 4-inch wafers. The successful use of these features has led to the fabrication of GaAs microwave devices such as varactor diodes, impact ionization avalanche transit time ( I ~ ~ A T diodes, T) mixer diodes, and Schottky
~ Q l e c u l Beam a ~ Epitaxy
783
phenomena in situ. The application of this method to the growth of intermetallic compounds is discussed in the following sections.
I
l
5
l
~
l
I
IQ 15 Time (Seconds)
~
20
l
~
Figure 2 D e t e ~ i n a t i o nof the composition of (A1,Ga)As from the growth rates of the GaAs and AlAs components, as determined from RHEED intensity oscillations. (Reprinted from Moustakas, 1988)
barrier field-effect transistors (FETs), as well as to the fabrication of optoelectronic devices such as optical waveguides, light emitting diodes, and heterostructure injection lasers. Furthermore, progress has been made in the growth of epitaxial structures with lateral dimensional control using mechanical masks made of either refractory metals or silicoii. Linewidths down to 1 pm have been reported (Tsang and Ilegems, 1977). Additionally, by appropriate motion of the masks with respect to the substrate, three-dimensional patterns and tapered structures were fabricated (Tsang and Cho, 1978; Tsang and Ilegems, 1979). This type of lateral dimensional control is required in optoelectronic devices made of GaAs/(Al,Ga)As multilayers. The multichamber design of the MBE systems also facilitates the formation of metallic contacts. Ideal semiconductor-metal interfaces have been formed by epitaxial growth of single crystalline A1 onto (001) GaAs (Cho and Dernier, 1978). Progress has also been made in the fabrication of ohmic contacts. The area of 'superlattice' structures and device concepts based on such structures 1s one of the most active fields in scientific research today. The MBE process has contributed significantly to this new class of artificially modulated materials. Such structures have been fabricated from semiconductors, metals, insulators, and combinations of these materials (Chang and Giessen, 1985). In conclusion, MBE has emerged as a practical growth method for a variety of materials and devices and as a unique scientific tool to study thin film growth
As discussed in the introduction, there is significant incentive to develop stable metallic contacts to 111-V compound semiconductors. Such contacts need to fulfil1 the following characteristics: (a) grown epitaxially to the semiconductor thin film and thus devoid of grain boundaries, and (b) be stable at high temperatures ( 6~ 0 ~ 9 0' 0"C).~Such metallic are ~ ~ ~ contacts ' ' likely ~ to find applications in integrated microelectronic circuits. In the early stages of the field AI, Ag and Fe were found to form single-crystal1ine thin films on GaAs when deposited by molecular beam epitaxy (Cho and Dernier, 1978; Massies and Linh, 1982; Prinz and Krebs, 1981). In addition, Au, Ag and Nb, when deposited in situ on p-GaAs, formed very low resistance contacts (1-4 x 10-7 sZ-crn2). Aluminum also forrns ohmic contacts to n-CaAs with contact resistivity 2.5 x 10-6 (Hong et al., 1994). However, the utility of such metallic contacts was found to be limited due to the low melting point of A1 (660 "C). On the contrary, CO and Fe when deposited onto GaAs under high vacuum conditions were found to have a high degree of crystallograp~ictexture but were not stable upon high-temperature annealing due to reaction with the substrate. To overcome these problems, a new research direction was initiated in the late 1980s with the development of metallic contacts using intermetallic compouiids (Lince and Williams, 1985; Sands, 1988; Chambers, 1989; Wowchak et al., 1989; Hong et al., 1991; Kamigaki et al., 1991; Chambers and Loeb, 1992; Tanaka et al., 1992). The various interrnetallic cornpounds which were found to form stable interfaces with relatively good lattice match to 111-V compounds can be divided, based on their crystal structure, into the following categories. The first group has a cubic CsCl crystal structure and includes intermetallic compounds such as NiAl, NiGa, CoAl, CoGa, FeA1, etc. (Harbison et al., 1988; Palmstrom et al., 1987; Sands et al., 1988; Cuivarc'h et al., 1987). The lattice mismatches are usually 1-3% larger than one-half of the lattice constant of GaAs. A complete listing of various intermetallic compounds and their relative mismatch to Al,Gal-,As, Ino.53Gao.47As or InP and InAs is shown in Table 2. To obtain a better latticematch of these binary interrnetallic compounds to
~
~
Research T e c ~ ~ i ~ u e s
784
GaAs, a third element can be added to the binary compounds. Such work has been reported by Hong et al. (1 99 1) who developed Fe3(Al,Si) metallic contacts to GaAs. Using high resolution XRD and TEM, Hsieh and co-workers reported that a high-quality ternary film of ~ e 3 A I 0 , ~ 3 Scan i o . ~be~ grown, perfectly lattice-matched to GaAs (Wsieh et al., 1992)' The second group consists of rare-earth monoarsenides which have the NaCl structure. Members of this group include ErAs, LuAs, YbAs, Er(P,As) and (Sc,~r)As.(~almstromet aE., 1988). Of interest is also the class of metal-mononitrides (MN) where M can be Sc, U, or any of the rare-earth metals. These intermetallic compounds also have the rock-salt (NaCl) structure,
CsCl ~ t ~ ~ c t ~ r e
Of the various interme~alliccompounds with the CsC1 structure, NiGa and CoGa form stable epitaxial films on GaAs. The formation of these compounds by the deposition of Ni (or CO)and subsequent reaction with GaAs leads to mixed phases containing interfacial t ran sit ion-metal arsenides. ~ingle-p~ase films of NiGa or CoGa can be formed by co-deposition or by deposition of alternating layers of Ga and the transition metal. The substitution of A1 for Ga leads to i~terrnetall~c compounds such as NiAl (or CoAI) which have the same crystal structure and lattice misfit with GaAs but significantly higher melting point as indicated in Table 3. Thus, NiAl forms a stable
intermetal~iccompound for high-tem~eratureprocessing. A potential problem might be the formation of an interfacial (A1,Ga)As layer due to the exchange reaction between A1 and Ga. However, such an interfacial layer would not affect the epitaxial relationship between NiAl and GaAs because AlAs and GaAs are miscible with a lattice parameter mismatch of only 0.1%. On the contrary, such an (A1,Ga)As interfacial layer was found to be beneficial, because it increases the barrier height for electron trans~ort,as will be discussed later. Thus, NiAl has been the most widely studied intermetallic compound for epitaxy on GaAs or AlAs, In early work, Sands (1988) fabricated NiAl by electron-beam evaporating lOnm of Ni onto GaAs followed by 34 nm of AI and 10nrn of Ni. The sample was then capped with sputtered AlN to prevent loss of As from GaAs during the high-temperature annealing step. After the sample was annealed to 850 "C, a NiAl c o ~ p o u n dwas found to have formed, Similarly~CoAl can be formed by replacing the Ni source with a COone. The growth of NiAl or CoAl using MBE requires ~igh-temperature~ n u d s cells e ~ for metals with hi melting points such as CO and Ni. Such cells employ double-wall PBN crucibles (Tanaka et al., 1992) and are designed to have very low o u t ~ a s s i npressure ~ (< 100' Torr) at 1400 "C (Kamiga~iet al., 1991). An alternative method of beam generation is the use of resistively~heatedtungsten boats to evaporate Ni and A1 (~hambers,1989). In this case, each boat is also equipped with a shutter and a dedicated quartz crystal to monitor the beam Buxes of the elements accurately.
Candidate intermeta~liccompounds for stable and epitaxial contacts to 111-V cornpound semiconductors ~nterm~~allic ~ o ~ p o ~ n aoa ~ CA? CoAl CoGa ReAl NiAl NiCa
FeAl RhAl MnAl IrAL OsAl IrGa RhGa RuCa RuAl Villaxs and Calvert (1985). Fleischer (1993).
2.861 2.878 2.88 2.886 2.887 2.909 2.97 2.976 2.98 3.001 3.004 3.01 3.01 3.03/2.992b
'?40Misfit with AI,Ga,-,As (Odxd 1)
% Misfit with InP or
+ 1.2 + 1.8 + 1.9 1-21 + 2.1 + 2.9
-2.6 -2.0 -1.9 -1.7 -1.6 -0.9 + 1.2 + 1.4 -I- 1.6 i2.3 + 2.4 + 2.6 + 2.6
~~0.53Gao.47~s
OO /
Misfit with InAs
-2.0 -1.8 -1.6 -0.9 -0.8 -0.6 -0.6 + 0.03/ - 1.2
r .Beam Epitaxy ~~~~~~~a
785
GaAs/AlAs/CoAl/AlAs/GaAs. Duri The epitaxial growth of intermetallics by COAL, the Knudsen-cell temperatur also takes advantage of the i~ situ diagnostic tools were about 1350 "CC and 960--1130 "C, respectively. available, as discussed in the previous section, rowth rate of CoAl was found to be between examples being reflection high-energy electron 0.061--0.091 pm/h, depending on the A1 composition. photoeniission (XFS), Attempts at rowing AlAs and GaAs layers on top of' py (AES), low-energy the CoAl layer resulted in layers with high densities of and electron energy loss stacking faults and twins. , in each case without exposing Another intermetallic compound which has been the sample to air. K a ~ g a k iand CO (199 1) have idei~tified grown epitaxially on a 111-V semiconductor i s AuGa2 on GaSb (Lince and Williams, 1985). The epitaxial the growth of NiAl thin three tempe~dturereg relationship between AuGa2 and GaSb was found to be films on AlAs by MBE. They reported that NiAl (OOl)AuGa2~~(00l)~a~b and ~ I O O l A u ~[100]CaSb. a2~~ does not form below 300 "C. Above 300 "C, NiAl was formed, but other ~ntermetallic Ni-A1 com~ounds In addition to being nearly lattice"~atched,AuGaz is more stable (up to at least 573 K) compared to pure (Ni3A12, NiAl3, Ni3Al) appeared as well. Above 40Q "C, remarkable i m p r o ~ e ~ e nof t s the c ~ s t a l ~ ~ n . i t yAu films, which react chemically and consume a large a ~ o u nof t the GaSb s u ~ s t r ~upon t e annealing. Thereand epitaxy were found, and the best NiAl films were grown at 600 "Cwith no interfacial disorder that could be attributed to i~terdi~usion or film discontinuity due to balling up. Chambers (1989) investigated the growth of 1991). The sticking Coefficient of Te2 molecular beams on semi-insulating G a A s ( ~ 1 ) and reported was found to decrease with substrate tem~eratureand creation of a Schottky barrier with a barrier hei become zero above 470 K. However, w ~ e na Pd beam 0.9 eV. He interpreted the high barrier height as d is si~ultaneouslyintroduced, the sticking coefficient the formation of an ultra-thin (A1,Ga)As interfacial of Te2 was found to be finite even. above layer with a h~gher-energybandgap. Chambers and t oeb (1 992) investigated the number of different compounds such as P chemistry and b~nd-ben~ing at the epitaxial PdTe(001) and PdgTe4(010) were found aAs(O01) interface. X-ray photoelectron heteroepitaxially on 2 -MoS2 depending on ratio of Pd/Te and the substrate te~perature. di~ractionallowed angular distributions to be measUsing RHEED intensity oscillations to nionitor the ured after the first NiAl bilayer was deposited. It was found that the layer of Ni atoms is in direct contact growth, Wowchak and co-workers (1989) observed with the GaAs substrate and the A1 layer sits on top. layer-by-layer growth of FeAl on In They found that the SUM of the Schottky barrier height The films were grown at a substrate t e ~ p e r a t ~ofr ~ values for n- and p-type GaAs is, to within experi200 "C and were stable up to at least 600 "C. The Fe mental error, equal to the GaAs bandgap. Thus, the source in this case consisted of a ~.020-inchFe wire conclusion. is that the Fermi level i s pinned around 0.33 wrapped around a resistively heated 0.030~in~h diato 0.36 eV above the valence band maximum. meter W wire, By changing the composition Tanaka and co-workers (1 992) have ~nvestigated Fe,A11-, films from x=O.40 to 0.68, the R buried CoAl i nt e~ et al liccompounds in the form of intensity oscillations begin to show multiple f ~ e ~ u e n c y o nepitaxial ~ntermetal~ic compo~ndson CaAs Table 3 ~ o ~ p a r i s of Intermetallic compound NiGa NiAl CoGa CoAl a
Villars and Calvert (1985). Hansen (1958). Elliot (1965).
Lattice parameter" (maxim~m)
4
2.887 2.886 4 2.878 4 2.861 A
Misfit with GaAs (maximum)
Melting point
2.1 Yo 2.1Yo 1.8% 1.2%
1220OCb 16'38 "Cb 1207 OCc 1645 'Cb
786
Research Techniques
components, indicating a transition from single layer to bilayer growth modes. Noh and co-workers (1996) examined the strain of Fe& and Fe3Si thin films grown on a GaAs (001) substrate. Fe3Si and Fe3A1 have lattice mismatches of 0.17% and 2.5%, respectively, to GaAs. Using synchrotron X-ray scattering e~periments,the in-plane and out-of-plane lattice constants were measured for Fe3Si and Fe3Al 0x1 GaAs. The internal strain of Fe3A1 was found to be completely relaxed near 500 "C, while the Fe3Si was still fully strained at temperatures up to 600 "C. The authors suggest that the strain relaxation may be caused by the interdiffusion of atoms between Fe3Al and GaAs, Transmission electron microscopy (TEM) studies of the 15 nm thick Fe3Al films grown on GaAs at 250 "C showed the presence of misfit dislocations, with the dislocation density increasing with film thickness as a result of strain relaxation (Hsieh et al., 1992). In contrast, a fully strained Fe3Si layer of 60 nm thickness was observed up to a growth temperature of 500 "C.
3.2 The second group of in t e~ e t a l l i ccompounds have the NaCl crystal structure and consist of rare-earth arsenides and metal-mononitrides. It has been shown that a direct metal-seiniconductor contact cannot be formed by rare-earth metals on CaAs because the interface is thermodynamically unstable (Waldrup, 1985). As a result, dissociation of GaAs at the interface region between the rare-earth and GaAs forms an interfacial layer of rare-earth arsenide. These compounds have a snialf lattice inismatch to GaAs (w 1%), have high heats of formation, very high melting points (> 2500 "G) (Hanks and Faktor, 1967; Moffatt, 1984; Shunk, 1969; Elfiott, 1965; Hansen, 1958), and are elect~icallyconduct in^. These qualities make the rareearth arsenides attractive for Schottky barrier contacts with GaAs for the reasons previously discussed. Forming latti~-matched he~erost~ctures between com~oundsof ionic bonding, such as compQundshaving the NaCl structure, with compounds having covalent bonding, such as the diamond and zincblende structures, is a scientific~llychallenging problem. Pahstrom and co-workers (1992) observed destabilization of GaAs surface reconstruction patterns due to surface charge dist~butioninduced by the deposition of Scl -,Er,As. Palmstrom and co-workers (1988) also deposited ErAs on (100) GaAs using MBE at a growth temperature of 450 "C. After the deposition of ErAs, a GaAs capping
layer was grown in order to reduce the contamination of the ErAs films. The ErAs surface was found to be highly reactive to 0 and C contamination even at 1 x 10-s rnbar. However, the GaAs overgrown layer did not wet the ErAs films well, resulting in a rough surface morphology. Initial transport measurements by Hall effect showed mobilities of 80 and 360 cm2/V s at room temperature and 1.35 K, respectively. The resistivity of the ErAs filrns at rooin temperature was 70 psZ-cm and decreased with decreasing te~peraturereaching 17 ps2-rn at 1.5 E;,indicating metallic behavior. Richter et al. (1988) investigated the growth of YbAs on Si-doped (100) GaAs by MBE using high-purity ytterbium evaporated from an effusion cell. From the XPS characterization of the YbAs films, there was no evidence of the presence of any other phases such as Yb3Asz, indicating that the film was composed entirely of YbAs. The XRD data showed that the lattice mismatch of YbAs with GaAs is about 0.8%. For the ~etal-mononitride$~ ScN has been grown by MBE by Moustakas and co-workers (1996). The films were grown on (0001) sapphire substrates using plasma-assisted MBE. The ScN films were found to with crystals have the NaCl structure (a = 4.5 oriented along the ( 111) direction perpendicular to the substrate. XPS studies indicated that the films are stoichiometric with no evidence of free Sc metal. Films 1 to 2 pm thick have a deep red color and transmission measurements indicate a fundamental absorption edge at 2.1 e'V. These results are consistent with ScN being a semiconductor, rather than a semimetal, as earlier literature implied using films grown by other methods. Dismukes and ~ o u s t a k a (1 s 996) have proposed that ScN can form lattice-matched heterostructures on zincblende structure GaN which also has a lattice that growth constant of 4.5 fi. The authors spec~~lated of GaN or AlN on top of ScN will force these materials to grow in their NaCl structure. AlN can exist in its NaCl structure only under very high pressures.
A)
The ability to grow an intermetallic compound on a 111-V semiconductor and vice versa opens the avenue to monolithic vertical integration in microchips. Current integration is mostly two-dimensional since the technology to grow a high crystalline q u a ~ t y semiconductor on a metal is lacking. The research effort of growing intermetallic thin films by MBE also seeks to answer the question whether a high-quality 111-V compound, such as GaAs or AlAs, can be grown
~ o l e c u l aBeam ~ Epitaxy on top of an i n t e ~ e t a ~ lfilm. i c Potential applications of such buried metal structures in semiconductors include electronic devices such as rnetal-base transistors and metallic quantum well devices as well as optical devices incor~oratingburied metallic optical mii~ors, gratings, etch-stop layers, ohmic contacts and Schottky contacts. Reviews of s e ~ c o n ~ u c t and o r optical applications of intermetallic compounds have been given by Masumoto et in Chapter l5 Of vO1ulne and England and Arakawa in Chapter 18 of Volume 2. Early efforts to fabri~ate~ e t a l ~ b atransistors se used the point contact structure of an n-type Si whisker as the emitter with a Au-coated (metal-base) n-type Ge as the collector (Attala and Kahng, 1962; Geppert, 1962). The development of epitaxial silicides (Tung, 1984) resulted in Si/CoSiz/Si metal-base transistors (Rosencher et al., 1984, 1986; Hensel et al., 1985). However, metal-base transistors suffer from a fundamental limitation in the current gain due to quantum. at the semico~ductor-metalinterm e c h a ~ reflection ~~l face (Crowell and Sze, 1966: Sze and Gumrnel, 1966). More recently, resonant enhancement of the tunneling current in A l A s / ~ i ~ / ~metal-base As quantum wells has been observed (Tabatabaie et al., 1988). With the advances in the MBE growth of epitaxial semiconductor/metal/semiconductor structures (Harbison et al., 1990; Sands et al., 19901, a three-ter~naldevice, in which tunneling from an n+ GaAs source through an AlAs tunnel barrier into a q u a ~ state t ~ in a 3.3 nm metallic NiAl quantum well drain was modulated by a third gate on the other side of the well has been demonstrated (Tabatabaie et at., 1989). Attempts at growing CaAs/(Al,Ca)As quantum wells over epitaxial CoAl layers on GaAs by MBE have beer1 reasoiiably successful (Goodhue et al., 1992). The cross-sectional TEM studies of the quantum-well regions showed well-~efinedquantum wells and barriers with slight undulation due to the residual surface roughness from the initial overgrowth of GaAs on the 10 nm CoAl layer. Photoluminesceric~measurements at 77 K revealed narrow (15 to 20 meV) excitori peaks with brightness 50 times lower compared to high-quality CaAs/AlGaAs quantum wells grown directly on CaAs substrates. The proposed device application is a multiple-quantum-vvell spatial lightmodulator utilizing the CoAl metal layer as both an electrode and reflecting mirror. "*
Atalla, M., and Kahng, D. (1962). Devicc Res. Con$, IEEE Trans. Electron Dev., E
787
Calawa, A. R. (1981). Appl. P h ~ sLett., . Chambers, S. A. (1989). J Vac. Sci. Tec Chambers, S. A., and Loeb, V. A. (1992). J . Vac. Sci. ~echIZo1. A , 10, 1940. Chang, L* L.3 and Giessen, B* C. (eds) (1985). S~~.t~e~ic ~ ~ u d ~ ~ lStructures, ated Academic Press, New York. Chang, L. L., and Ploog, K. (eds) (1985). ~ o l e c ~ l Ba ie~a ~ Epitaxy and ~ e t e r o . ~ t r u c t ~Martinus ~es, NijhoE, Boston. solid Fil,ws, 10, 291. Cho, A. y.(1983)+ Cho, A. Y., and Dernier, P. D. (1978). J. Appl. Plzys., 3328.
_ _ _ .
Crowell, 6. R., and Sze, T M. (1966). J. Appl. Phys., 37,2683. Davey, J. E., and Pankey, T. (1968). J. Appl. Phys., 49, 1941. Davies, G. J., and Williams, D. (1986). In Tec~nologyand Physics of Molecular Beam Epitaxy (ed. E. H. C. Parker). Plenum, New York, vol. 2, p. IS. Dismukes, J. P., and Moustakas, T. D. (1996). 111-V itr ride ~ a t e r i a l sand Processes (eds T. D. ~ o ~ ~ s t a k J. a s P. , Dismukes, and S. J. Pearton). ECS Proc., Elliot, R. P. (1965). Constitution of Binary Alloys, First Supple~ent,McGraw-Hill, New York, p. 321. Esaki, L., and Tsu, R. (1970). IBA4 ~ e s e a r ~c ~e v e l o p ~ e n t , 61 Fleischer, R. L. (1993). Acta Me~all.Mater., Foxon, C. T., and Harris, J. J. (eds) (1987). ~ o l e c ~ l aB re a ~ Epitaxy, North Holland, Amsterdam. Foxon, C. T., atid Joyce, B. A. (1990). Growth and Character~zationof Semic~IZ~u~fors, (eds R. A. Stradling, and P. C. Klipstein). Adam ~ i ~ g eNew r , York, p. 35. Geppert, D. V. (1962). Proc. IRE, Goodhue, W. D., Le, H. Q., Johnson, G. D., and Bales, J. W, (1992). J . Vac. Scz. Technol. B, 10, 783. Gossard, A. C . (1982). In Treatise OIZ Material Science and Technology: Preparation and Properties of T ~ i Films n (eds K. N. Tu, and R. Rosenber~).Acadesnic Press, New York, vol. 24, p. 13. Guivar'c, A., Guerin, R., and Secoue, M. (1987). Electrun. Lett., 23, 1004. Gunther, K. Z. (1958). Z. ~ a t u r ~ o r s13%. c ~ , 1081. Hanks, R., and Faktor, M. M. (1967). Trans. Faraday Soc., 63, 1130. Hansen, M. (1958). Constitution of Binary Alloys, McGrawHill, New York, p. 80, 119, and 750. Harbison, J. P., Sands, T., Tabatabaie, N., Chan, W. K., Florez, L. T., and Keramidas, V. G. (1988). Appl. Phys. Lett., 53, 1717. Harbison, J. P., Sands, T., Ramesh, R., Tabatabaie, N., Gilchrist, H. L., Florez, L. T. (1990). J. Vac. Sci. ~echnol.B, Hensel, J, C . , Levi, A. F. J., Tung, R. T., and Gibson, J. M. (1985). Appl. Phys. Letf., 47, 151. Hong, M., Chen, H. S., Kwo, J., Kortan, A. R., Mannaerts, J. P., Wier, B. E., and Feldman, L, G . (1991). J. Crys. Growth, 111, 984. Hong, M., Vakshoori, D., Mannaerts, J. P., Thiel, F. A., and Wynn, J. D. (1994). J . Vac. Sci. Technol. B, 12, 1047. I
788
Research Techniques
Hsieh, Y.-F., Hong, M., Kwo, J., Kortan, A. R., Chen, H. S., and Mannaerts, J. P. (1992). Inst. P h p . Conf. Ser. No. 120; Chapter 2, 95. Kamigaki, K., Yuda, S., Kato, H., Ishida, M., Terauchi, H., and H i y a ~ ~S. u ,(1991). J. Appl. Phys., 69, 2196. Lince, J. R., and Williams, R, S. (1985). J. Vac. Sci. Technol.
Rosencher, E., Badoz, P. A., Pfister, J. C., d’Avitaya, F. A., Vincent, G., and Delage, S. (1986). Appl. Phys. Lett., 271. Sands, T. (1988). Appl. Phys. Lett., Sands, T., Marbison, J. P., Chan, Chang, C. C., Palmstrom, C. J., and Keramidas, V. G. (1988). Appl. Phys. Lett., Sands, T., Warbison, J. P., Tabatabaie, M., Chan, W. K., Massies, J., and Linh, N.T. (1982). J. Crys. ~ r o ~ t56, h , 25. ~ i l c h r ~ H. s ~ , L., Cheeks, T. L., F1 M o ~ ~ t W. t , G. (1984). The ~ a n ~ o~f Binary o o ~ Phase Keramidas, V. G. (1990). Surf: Science Diagrums, Gemum, Schenectady, NU. Shunk, F. A. (1969). C o ~ s ~ o~f E~~ n~ u ~r ~~ o n M o u ~ t a k aT. ~ , D. (1988). M R S E ~ ~ l e t i 13, n , Nov 29. ~cGraw-Hill,New Yark. Moustakas, T. D., Molnar, R. J., and Dismukes, J. P. Sze, S. M., and Gummel, €3. XI. (1966). Solid State Eletron., 3, (1996). In 111-V Nitride ~ ~ ~ t e rand i aProcesses ~s (eds T. D. 751. Moustakas, J. P. Dismukes, and S. J. Pearton). ECS Tabatabaie, N., Sands, T., Harbison, J. P., Gilchrist, €3. L., $roe., 197. and Keramidas, V. G. (1988). Appl. Phys. Lett., 53, 2528. Noh, D. U, Y., Je, J. H., Hong, M., and Mannaerts, Tabatabaie, N., Sands, T., Harbison, J. P., Gilchrist, H. L., J. P. (1996). Appl. Phys. Lett., 68, 1528. Cheeks, T. E., Florez, L. T., and Keramidas, V. G. (1989). Palmstrom, C. J., Tabatabaie, I?., and Allen, S. J. Jr. (1988). Technical Digest qf the ~ ~ ~ e r n a t ~i ~l ~nc~t rl oDevice n meet in^ (IEDM), December 3-6, 1989, ~ a s ~ i n ~ t o n , Palinstom, C . J., Chang, C. C., Yu, A., Galvin, G. J., and D.C., IEDM 83, 555. Mayer, J. W. (1987). J. Appl. Phyys., 62, 3755. a , Ikarashr, N., and Tanaka, M., Sakakibara, H., ~ i s h i n ~ gT., Palm§trom, C. J., Cheeks, T. L., Gilchrist, H, L., Zhu, J. G., Ishida, K. (1992). Surf. Set., Garter, C. B., Wilkens, B. J., and Martin, R, (1992). J , Tsang, T. S., and Cho, A, Y. (1978). J , Appl. Phys., 32, 491. Vac. Sci. Technal. A , 10, 1946. Tsang, T. S,, and Ilegems, M. (1977). Appl. Phys. Lett,, 31, e ~ . 127, 2729. Pamsh, M. B. (1980). J. ~ l e c ~ r o c hSoc., 301. Parker, E. H. C. (ed.) (1986). The Tech~ologyand Physics of Tsang, T, S., and Ilegeins, M. (1979). Appl. Phys. Lett., 3 ~ o l ~ c ~Beam l a r Epitaxy, Plenum, New York. 792. Ploog, JS. (1980). In Crystal Growth, Properties and Tung, R. (1984). Phys. Rev. Lett., 5 H. C. Freyhard). S~ringer-Ver~~g, Villars, P., and Calvert, L. D. (1985 ~rys~allographic Data,for ~ n t e r ~ e t a lPhases, l~c American Prinz, G. A., and ICrebs, J. J. (1981). Appl. Phys. Lett., 39, Society of Metals, Metals Park, OH, Vols. 2 and 3. 397. Waldrup, J. R. (1985). AppL Phys. Lett., Richter, H. J., ~ m i t R. ~ ,S., Herres, N., S e e ~ n ~ ~ n n - ~ g g e ~ e ~ r to, w c h a kA. , M., Kuznia, J, N., and Cohen, P. I. (1989). J. M., and Wennekers, P. (1988). Appl. Phys. Lett,, 53, 99. Vac. Sci. Technol. B, 7, 733. Rosencher, E., Delage, S., Chmpidelli, Y., and d’Avitaya, Yata, M., Nakamura, K., and Ogawa, I(.(1991). J. Vac. Sci. ~ e c h n oA, ~ . 9, 3019. F. A. (1984). Electron. Lett., 20, 762.
ecoming a scientist because you crave factual certainty and thirst for a meaningful vision of human life is like an Archbishop so you can meet girls.
isdom is knowing what to do next, skill is know in^ how to do it, and virtue is in doing it.
The most exciting phrase to hear in science, the one that heralds new discoveries, is not but . ‘‘That’s funny . . . .” “
I
er ia
ct
th
economic impact. And, if a l u m i n u ~were adopted as the default material for automotive structures, intermetallics would still play a vital, but unseen, role in the ~ e a s u r i n gthe economic impact of a class of materials autoinotive industry. Pure alunninum, as with most such as jnterrneE~Zzj~‘~ to the US economy is fraught pure metals, has m a1 or poor mechanical properwith di~culties,~conomiesare built as much by the ties. Only through ing are the strength, stiffness, multiplicative contributions of various sub-sectors as and toughness of modern alloys achieved. At a they are built by their summation. In this light, the microscopic scale, many of these alloying elements contribution of a particular good or service is a ‘precipitate’ from a solution of the parent metal, function of the value or contributions of the goods and forming intermetallic precipitates that strengthen or services surrounding it and linking it to the overall toughen the parent metal. economy. In multiplying these contributions, it is true Intermetallics, as with all basic materials, are not that if any one link goes to zero, the entire product consumed in their raw state. Materials pass through a goes to zero. Thus, it can be argued that all basic number of stages in the conversion froin ‘raw’ materials - intermetallics, included - ]nave a value extraction to finished good, and the demand for equal to the values of all the chains of the economy these raw materials is ‘derived” from the demand for that utilize these materials. Therefore, in its broadest the products that they eventually are used to produce. definition the economic impact of intermetallics is the A typical, representative chain. showing the stages that economic value of the aerospace, electronics, automost materials pass through in becoming end-use motive, construction, etc., industries that utilize products is shown in Figure 1. intermetallics and depend upon them at smze stage The most obvious, measurable, and incontrovertible of their production activities. economic impact of a material occurs in the early It is impossible to define with any rigor, the stages of this chain. For the intermetallics, coiiveneconomic contribution of intermetallics to the US tional thinking would limit attribution of economic economy. Consider galvanized steel, for instance, the n ~parts ing, value to the s y ~ ~ ~ e sri ~s ,~ n i ~ g / ~ Z ~or mainstay of the automotive industry. In response to ~ ~ stages, rand further ~ ~ limit~ this measure to~ consumer demand for more long-lived vehicles, the goods that are substantially or entirely made from automotive industry be an using galvanized steel intermetallics. The authors assert that this perspective extensively throughout the vehicle structure in the systematically undervalues their contribution. Never1970s. The surface of galvanized steel includes intertheless, the purpose of this chapter is to frame the metallic iron-zinc compounds that render the steel more corrosion resistant. ~ i t h o u this t interm.etal1ic~ economic contribution of inter~etallics within a perspective, and the authors will eventually adopt car bodies would ‘rust-out’ in as little as five years as this narrower perspective and attempt to quantify the they did before the adoption of galvanized steel, or direct, ‘first-order’ e nomic impact of inte~etallics they would be made from. more expensive material on the US economy. fore doing so, however, we will systems such as a~uminum,with a substantial negative ~ n t ~ r m e t a lCompounds: li~ Vol. 3, Principles and Practice. Edited by J. H. Westbrook and R. L. Fleiseher. 0 2 0 0 2 John Wiley & Sons, Ltd.
792
Figure 1 This figure represents the stages that materials pass through, typically, in their development, end-use, and disposal. Often, there are many additional stages that span one or more of the activities shown in this figure
present inte~etallicsin a grander perspective - one that, like a politician, takes credit for everything that is good that surrounds it. To accomplish this objective, this chapter will rely on economic statistics tabulated and published by various public and private institutions, with references provided.
2.
‘
easur~ment
Starting from the top, the gross domestic product of the United States in 1997 was declared to be $8.1 trillion by the US Bureau of Economic Analysis. This number is arrived at by an estimation method that is not entirely above controversy or dispute, but is generally accepted as the value of the goods and services produced in the United States. Clearly the contribution of inte~etallicsto the total economy is but a tiny fraction of this total. But, at least the challenge of estirnating the size of this contribution has been bounded on its top. The economic impact of interrnetallics on the US economy is no more than $8.1 trillion. Furthermore, the first dissection of the economy traditionally distinguishes between the value of ‘goods’, the value of ‘services’, and the value of espectively, in 1997 these were 2.9, 4.4, and $0.7 trillion. Intemetallics are most closely associated with ‘goods’, although there are many ‘services’ (for instance teaching, consulting, financing,
and research) that benefit from this class of materials, In an effort to refine the ‘top-down’ estimate of their economic contribution, we will ignore the contribution that intermetallics make to the services and structures sectors of the economy. Given that these two sectors represent 57% of the economy, and include many activities associated with the research and development of new ‘goods’, it can be argued that focusing on the ‘goods’ sector alone is a deliberate underestimation of economic impact, However, without a heroic effort, it seems impossible to establish a ~ e a n i n ~ festimate ul for the economic impact of intermetallics on ‘services’ and ‘structures’. Thus, a ‘lower upper-bound’ for the value of intermetallics in $2.9 trillion. Several key inte~etalliccompounds are closely associated with sub-sectors of the economy. For instance, the iron-zinc compo~ndsand the automotive sector, or nickel aluminide and the jet engine sector. Table 1 identifies some of these key sectors and lists the value of shipments from these sectors, according to the 1997 Economic Census of ~ a n u f a c t u ~ e r s . Each of these NAICS codes is further broken down into a number of contribu raft engine and engine - a subsector that relies heavily on inte~etallics.~~vertheles$, to with a top-down breakdown of the ‘goods’ sectors of the GDP, identifying the true contribution of intermetallics would again be fraught with dificulties.
~ o ~ ~ I ~ e rImpact cia1 Table 1 Key sectors and value of shipments NAICS code Sector of the economy 333 334 335 336 339
Value of shipments ($109)
Machinery manufacturing 270 Computer & electronic product 438 manufacturing Electrical equipment, appliance, & 112 component manufacturing Trans~ortationequipment ma~ufacturing 572 ~ ~ s c ~ l l ~m~~enou uf ~sc t u r(i i~~gc ~ u d ~ n g101 jewelry, dental supplies. sporting goods)
From a ‘bottom-up’ perspective, there are no convenient, comprehensive sources of information for estimating the economic impact of intermetallic compounds. Intermetallics represent such a large and varied group of materials that impact so many industries that the authors’ current approach to estiii~ating their role in the economy would be impossible to implement when including all of these compounds. Instead, a sample group of materials was chosen to represent the intermetallics’ influence on the US economy. Table 2 presents a listing of this group of intermetallic compounds and the estimated economic factors used to compute their impact on the US economy. To understand the table, the following definitions are first in order. ~ p p l i c a t i o-~the field of use of the intemetallic, in this case limited to the direct, ‘first order’ uses of intermetallic compounds. ~stimatedc o n s u m ~ t ~-othe ~ es t i ~ at edquantity of int~rmetall~csconsumed annually in the United States in the specified a ~ ~ l i c ~ t i For o n . ferrocerium, consumpti~n data were not available and a sim~lifyin~, but perhaps incorrect, assumption was made that exports ininus imports was equivalent to ~onsumption. ~ ~ ~ t ~ m price a t -ethe ~ estimated price of the intermetallic compound as it is sold by a material supplier or a parts former to the industrial or professional custom~rfor use in the a p ~ l i c u t i o ~ , Value of , ~ ~ z ~ ~- ethe n tproduct s of the estimated c o n s u ~ ~ t i oand i 2 the e s ~ i ~price. ~ t e ~ Value of sector - the value of the goods or services in the adjacent, well-~efined‘downstream’ economic sector that iiicorporates the intermetallic compound in the designated a ~ ~ l i c ~ t i o For n. instance,
012
the U S Economy
793
intermetallic silver-tin ama~gamsare directly used for dental fillings. The vnlue qf seceor in this instance is an estimate of the value (cost) of all dental filling procedures, including the value (cost) of the dentist’s services and the value (cost) of the intermetal~ic consumed in this procedure. Intermetallic ironlzinc compounds, on tbe other hand, are used to make galvanized steel which, in turn, is sold into many enduse sectors of the economy. In this instance, the value of alvanized steel sales. sector is defined as the val Because there are intermediate links between the intermetallic and its true end-use, the value of sector is sometimes difficult to define. For instance, gallium arsenide plays a vital role in the telecommunications and data Co~Linications sectors of the economy, valued in excess of $1 trillion. However, the exact linkage and the strength of link (see below) in this instance are hard to ~uantify.CaAs is more directly tied to the $100 billion telecommunications e~uipmentsector and even more closely coupled with the GaAs semiconductor device sector of the economy. While CaAs sem~conductors are not a true enduse sector, the va~ue of sector for defined by the value of GaAs semiconductors, ~uantifyingthe direct contributio~of GaAs to the whole telecommunications industry is virtually impossible. Intermediate ‘end-uses’ were used for many of the intermetallic sectors, accepting that this approach sy~tematicallyunderstates the value, but recognizing that to go beyond this point would be folly. ~ t ~ e n g qft h l i ~ k a judgment of the import~nceof intermetallics in the application to the value qf the sector. The strength of link is expressed as a percentage. This percentage attem~tsto answer the question, ‘Of the total value qf sector, what percentage of this value is directly attributable to inte~etal~ics?’. The s t r e ~ g to~f l i n ~is judged to be high if the sector depends substantially on the intermetallic compound. It is judged to be low if a readily available substitute for the intermetallic exists. Economic i ~ ~ a-cthe t sum of the vahe o f s h i ~ ~ ~ ~ ~ and the value of sector multiplied by s t r e ~ g ot ~~ l i ~ ~ k . This column contains the authors’ best estimates of the total direct first-order economic impact of intemetallics in the sector. E c o n o ~ i c leverage - the total e c o ~ o ~ i ic~ p u c t divided by the total value o f s ~ ~ m ise used ~ t ~as a measurement of the ‘leverage’of intermetallics in the defined sectors. A large ratio indicates that
794
~ i s ~ e l l ~ Topics ~~#us ~ s ~ ~e c~o nao ~t factors i e~ ~ Inter-
Application
Estimated
Estimated
metallic c o n s ~ ~ ~ t i o nprice ~ a ~ ~ o u (‘metric ~ d stpy) ($/kg)
~ l ~ FeZn,, 3 aFeZn,, ~ steel (1,2,3) FeZn,, Dental AgzHg,, ~ ~ l(43) i ~ ~ Sn,Hg, s Ag,Hg,, 1R detectors 1nSb (637) ~ e ~ i c o n ~ ~ c t CaAs ors & LEDs 3 a t t ~ ~ ~ LaNi,, NiMH ~ l e ~ t r o d(8,9) e$ Pyrophoric CaFe, alloys (2) Shape ~ e i ~ NiTi o ~ ~ alloys (10) Superalloys (1 1) Ni,AI ~ a ~ n e t o s t ~ ~Tb,-,Dy,Fe, tiv~ devices (12> upe er conductor Wb,Sn wire (I 3) Permanent Co,,Sin, magnets (14) Nd,Fe,,B ~
Value of Value of s ~ i ~ ~ e sector ~ t s ($106) ($106)
Strength of link (%)
Economic Vol. 2 impact Econoinic reference ($106) leverage chapter
1.12
851
8309
25
2929
3.4
24
151
1333
20 1
5000
10
70 1
3.5
27
0,080
8767
0.70
6800
5
34 1
487
15,18
89.0
4495
400
3200
50
2000
6.0
15
5500
20
108
6745
2.5
278
2.6
21
1778
4.25
7.6
1600
10
168
22
31
250
33
8.3
1500
25
383
46
25
11364 0s
13 2500
150 1.3
600 25
50 50
450 I4
3 22
3 17
18
750
14
135
25
47
3.5
16
6776
100
677
19000
5
1627
2.4
14
~758 800 ~
~
Total value of shipments
$24f9
Total economic impact Economic leverage
$8936 3.69
our^^^: 1. US Geologic Survey, Mineral Cornmodjty Summaries 2. Mtttal Statistics 2000, 92th edition, published by American Metd Mwket 3. am er^^^ Iron and Steel ~ ~ ~ ~ t i t ~ ~ e 4. A ~ e r ~ c aDental n Assocation. 5. Sullivan-Schein, dental supply company 6. Andy Johnson, DERA Malvern, Great Malvern, UK 7. Mrs Martin Lamb, ~ ~ r k e tDirector i ~ g of m at^^ T ~ c h n o l o ~Ltd, y UK 8. E. t e e Huston, Manager of Hydride Alloy Technology at Moltech Power Sy‘sterns 9. Yushinori Toyoguchi, Mals~shitaElectric Industrial Co. Inc. 1% Toiiy Anson, Ansan Medical Ltd, UK I I . Gem Maurer, Speaal Metals Xnc. 12. John Snodgrass, Etrena Co. Ames, Iowa 13, Eric Gregory, I n t e ~ ~ ~ g n e tGeneral ics
14. Mark Benz, General Electric R&Dlc
int~rmetallicc o ~ p o ~ ~ have n d s an economic inipact that is well beyond their value as raw materials. A low ratio corresponds to situations where s ~ ~ ~ s ~ ~ nall t i athe l l yvalue of the i n t e ~ e t a l ~ is i ~ c ~ ~ t uinr the e ~ ‘raw’ state.
tially oxidized and, in the process, they protect the steel from corrosion. The US Geological ~urveyestimates that 56% of the 1.4 million tons of zinc consumed in the United States in 1999 were used for ~ a l v a n i z i ~ ~ . They also reported the average price of zinc to be 5 1 $/lb or $lf12/kg. The zinc-iron ~~iter€~etallic c o ~ p o ~inn ~ galvanize^ s steel play a vital role to the estimated $1.4 billion of galvanized steel sold in the United States, but it does not seem appropriate to assign all of this value to the In galvanizing steel, a thin layer of zinc is added to the inter~etal~ics. First of all, the steel in galvanized steel surface of steel sheet, and part of this zinc reacts and a n the ~ l yvalue and, c u to t~~~ ~ ~ ~ ~ e~~r ne ~~ t ~ c ~or ~ ~~ ~oe u ~~ ddearly ~~ ,~ c~o n~t ~ ic~ ~ st ei s~ ~ ~ ~ c to secondly, there are substitutes such as aluminuni or ecause zinc and its intermetalliccompounds are more plastics that could be used in the absence of ~ a l v a n i ~ e ~ reactive than the u n d e r l y i ~steel, ~ ~ they are preferen-
795
~ o ~ ~ ~Impact e ~ on c the i ~USl Economy steel, but at a higher cost to the economy. For these reasons, the authors assigned a value of 10% as the ~ t ~ e of~ Zink g t between ~ zinc intermetallics and galva~of $141 ~ million. ~ c nized steel, yielding an econom~c~ Of course, galvanized steel is the ‘backbone’of the US automotive industry, and plays a key role in the construction equipment, indust~almachinery, and appliance industries, to name just three. Combined, the value of shipments from these industries is roughly a trillion dollars, for which inter~etallicsdeserve some credit.
Silver amalgam - silver, copper, and tin, chemically bonded by mercury into a hard, stable material consisting of several intermetallic compounds - has been used for more than 150 years as a dental filling. Its estimated consumption, over 150ktpy, is fairly impressive when you consider that a ‘single spill, procedure - the filling of a small cavity - consumes only about 0.0004 kg of the amalgam compound. The consumption of 150 ktpy indicates a staggering number of procedures performed annually in the United States; over 100 million per year. Silver amalgam-gains its economic impact, not so much by the value of the material used in the fillings themselves, but by the value added of the professional dentists performing the procedures. While silver amalgam has a relatively high price ($1333/kg) and high value of shipments ($201x 106/yr),its economic impact is ‘leveraged’ 3.5 times because of its vital role in an estimated $5 billion per year dental fillings industry. However, silver amalgam is not the only material used for filling cavities and, increasingly, polymer composite materials are used. Consequently the economic link between silver amalgam and the dental fillings industry is assumed to be only 10%, a value that signifies these materials are important, but not truly essential. Interestingly~composite resins have found use as an alternative to silver amalgams, not because of their effectiveness, but because of the unsubstantiated fear that the mercury component in the silver amalgam is harmful. While there has been no proven link between mercury found in silver amalgams and any of the degenerative conditions associated with mercury ingestion, the fear of possible h a m has been enough to dissuade some of the public.
There are approximately 450 000 metric tons of nickelbased superalloys consumed annually in the United
States, for which approximately 25% of the mass is the intermetallic compound Ni,Al. ickel alurninide contributes high-temperature cree strength, a critical performance attribute for the use of these alloys in jet t engines, high-temperature process equipment, and elsewhere. In jet engines, modern nickel-based superalloys are not truly ‘enabling’ - there are substitute materials available, albeit at a hi h cost in terms of operating efficiency - but there is a strong link between these materials and the $22 x log dollar per year jet engines industry. Nevertheless, it is not easy to quantify this link, or siniilar links to the ultimate enduse sector. So, instead, the value ofsector for Ni,A1 was defined as the value of superalloy ingot or billet as sold by a specialty metals producer. The economic impact and economic leverage of these materials downstream is much larger.
InSb is an intermetallic conipound that is relatively new and only now beginning to find application as a semiconductor material. The first commercial application for InSb is as an infrared detector, owing to the material’s intrinsically narrow band gap. As with most newly developed materials, InSb’s costs are high compared with many of the alternatives, but its performance is unparalleled in certain applications and so, like many new materials, it is finding its first ‘commercial’ uses in the military. Of the three major producers of InSb worldwide, only one is domestic. This supplier estimates that it sells ap~roximately1000 wafers of InSb per year at an average price of nearly $700 per wafer, of which half is exported outside the
Permanent
Shape Memoryd Alloys
I
Figure 2 The economic impact of selected i ~ t e ~ e t a l l i cons the US economy in the year 1999
United States. Ignoring the non-US producers, based on these numbers, the value of shipments within the US economy for InSb i s $700000 per year. This figure is trivial relative to the size of the domestic economy. However, it becomes more noteworthy when viewed as a niche enabler of the IR detector industry. As the demand for InSb detectors increases during the next three to four years, the manu~acturing technology is expected to progress to include 5” wafers at considerably lower prices per square inch than the now available 2” and 3” wafers. As this technology becomes more affordable, InSb will become a inore popular material for use in commercial applications and should have a on the total US economy. more s i g ~ i ~ c a nimpact t Today, the authors estimate that the strength of link between InSb and IR detectors is only 5%. Even with this small linkage, the economic leverage is tremendous.
for some time*Today, the value of the GaAs wafers sold into this application is around $140million, As fabricated ICs, this value rises to $1.9 x log, an economic ‘leverage’ of 13.6,Again, as a critical link in the teleco~munications and Internet technologies, the ‘top-down’ economic impact of GaAs is im~easurable.
By a similar set of analyses, the ‘bottom-up’ economic impact of intermetallics 011 the US econoniy was estimated. In each case, the value of internietallics in state - as the galvanization of steel, or dental ‘spills’, or CaAs wafers - was first estimated, e value of these materials at the next or final h was stage in their use. A subjective s t r e ~ ~of ~ link the ‘raw’ state with the ‘end-use’ stage, and the total economic impact was measured from these two. The bottomup as~ssmentestimates that the direct economic impact of inte~etallicsis $8.9 x lop. Broken Today, galliuni arsenide has two main fields of use, as down by sepent, this assessment i s shown graphically a light emitting diode (LED) and as an integrated in Figure 2. While the top-down ass~ssmentwas only circuit substrate. There are other potential uses for this able to assert that the economic impact was no greater intermetallic compound, but these are small in than $2.9 x 10l2. Several orders of magnitude separate comparison. these estimates, and the reader, at this point, may be As a light emitter, GaAs is used to produce red and inclined to ask, ‘What is your best estimate?’ orange LEDs and laser diodes. The total annual To assert that the entire manufactured goods production of GaAs wafers for this purpose represents economy depends upon intermetallics is foolishness, a rapidly growing business of approximately $260 MM. While the sudden loss of these materials would, As fabricated LEDs, the value o~shipmentsrises to $1.3 undoubtedly, disrupt many supply chains, alternatives billion, an economic ‘leverage’ for this intermetallic do exist. We could go back to filling our teeth with material of 5.0. As electrical components in fabricated gauze, or invent a new ceramic or polymeric filling electronic devices, e.g. cell phones, the economic impact material, as in Fact, we have already done. We could do of GaAs LEDs is immeasurable. Suffice it to say that without MRI imaging or find a way to make these GaAs LEDs are to the telecommunications industry machines work without superconducting i nt e~et al l i c what neurons are to our central nervous system: an wires. Life would go on. essential link in the chain. By the ‘top-down’ measure, On the other hand, our estimates of the direct, firstCaAs enables modern optical telecommunications, and order economic impact of nine specific intermetallic therefore has an economic impact measured in trillions compounds grossly understat~the true impact of these of dollars. For purposes of this ‘bottom-up’ analysis, it materials. Not only are there other direct uses for is assumed that the economic impact is 50% of the value inte~etallicsthat have not been ~ u a n t i ~ e but d , there 33s that this material enables. are many indirect uses for intermetalli~s- for instance, circuit substrates~GaAs is the preferred as strengtheners in metal alloys - and there are many material for high frequency applications; circuits operatindirect economic benefits derived from their use GHz. Historically, these frequencies were needed only in high cost, microwave c o ~ u n i c a - improved fuel efficiency, better medical diagnostics, etc. So, what is the economic impact of inte~etallicson tions systems, radar systems, and high performance the US economy? Somewhere around $10 x 109 seems military electronics. Increasingly, cell phone and broadlike the correct order of mag~itude.To be much more band ~ I n t e ~ e tappl~cations ) are requiring circuits specific would be absurd but, if you need an exact operating this fast. As a consequence, the d e ~ a n dfor number, use $10.1 x log dollars , . . and be sure to GaAs semiconductors is growing at the rate of between 30 reference these authors. and 40% per year, and will con~nueto grow at this rate
L a ~ o ~ ~ t oof rCy o ~ ~~ h~ e t~ ies t~r~y~, s t i t ouf t~~ h e ~ i ~ ~ l Chinese Academy of Sciences, Beijing, ChiEa
The purpose of this chapter is to provide easy and quick access to comprehensive and evaluated data sources on intemetallic compounds (IMCs). There is no need to point out the importance of data sources on IMCs for materials scientists, materials engineers, and others who have interests in IMCs. Over recent decades a substantial effort has been made in the collection and compilation of evaluated data on IMCs, and so far users have been well rewarded by a variety of data sources. Due to the limited space, it is imp~ssibleto cover all the data sources on IMCs, An important principle in the bibliographic-filtering process is that: Not all, but only the best and evaluated data sources will be inclu~ed.Particular emphasis has been placed on directions to Data Compilations, rather than on Technical Reviews of related topics. In this chapter the data sources on IMCs are divided into four categories: crystallography, t h e ~ o d y n a m i c properties, phase diagrams, as well as physical and miscellaneous properties. The related books and computer software in each category are listed in the following sections. They represent, to the best of our knowledge, the most comprehensive, up-to-date, and evaluated data sources of IMCs.
Villars, P. and Calvert, L. D., Pearson’s Handbook of ~ r ~ s t a ~ l o ~Data r a ~ for ~ i cI n t e r ~ e t a ~ l iPhases, c
Second Edition, Vol. ISBN: 0-87 170-416-l(set), ASM International, terials Park, Ohio, 1991. ~ompilationof detailed crystallographic data for over 25000 unary, binary, ternary and multinary IMCs. Daarns, J. L. C., Villars, P. and van Vucht, J. H. PJ. es Types for ~ n t e r ~ e t a l l i c 0-87 170-421-8(set), AS Pai*k, Ohio, 1991. companion set to the Pearson’s Ha~dbook. Includes a detailed graphical description for most structure types. k ~ d i t i o n2-vol. ) Villars, P., Pearson’s H a n ~ b o ~Desk set, ISBN: 0-87 170-603-2, ASM International, Materials Park, Ohio, 1997. Contains the high quality crystal data updated through 1995 for 27686 compounds, derived from the Atlas and the Pearson Handbook (2nd Edition) mentioned above. Eckerlin, P. et al., and Hellwege, EL.-H., et al., eds, Structure Data of ~ l e ~ eand ~ t s ISBN: 3-540-05500-2, 1019pp., Vol. Crystal and Solid State Physics Numerical Data and Functional ~ e l a t i o n s ~ i pins Science and Technology, New Series, Hellwege K.-H (Series ed.), Springer-Ve~la~, Berlin, 1971. Volume III/G and its supplement volume I below) contain structure data and related information for elements and i n t e ~ e t a l l i cphases. Eisenmann, B., Schafer, H., Structure Data of Elements and I n t e r ~ e t a l ~ ~Phases, c Vol. (Supplement to volume 1 ), Group 111: Crystal an dolt-Bornstein. and Solid State Physics.
I n ~ e ~ n ~ e t aCmq,m.mdy!r: ll~c Vol. 3, Principles and Practice. Edited by J. €3. Westbrook and R. L. Fleischer, @2002 John Wiley & Sons, Ltd.
798
~iscellaneousTopics
Numerical Data and Functional Relationships in Science and Technology, New Series. Edited by Hellwege, K.-H. and Hel1wege, A. N., SpringerVerlag, 1986-1988. Three subvolumes: S u b ~ o l u ~a:eElements, Borides, Carbides, Hydrides, 1988, ISBN 3-540-17814-7, 458 pp; ) Tellurides; Subvolu~eb: S u ~ d e sSeleni~es, Part bl: Ag-AI-Cd-S . . . Cu-Te-Yb, 1986, ISBN 3 540-1541 1-6, 504 pp; Part b2: Dy-Er-Te . . . Te-Zr, 1986,ISBN 3-54016402-2, 492 pp. Hafner, J., Hmlliger, F., Jensen, W. B., Majewski,
Physik Daten, 16-7, Fachinfor~ations~e~trum, Karlsruhe, Germany, 1988. Hellner, E., Schwarz, R., and Pearson, W. B., Structure Type ~ e s c r ~ t ~ ofor n ' sI n t e r ~ e t a l ~P~~ca s e s in the Hexagonal and R~ombohed~ul Systems, Physik Daten, 16-8, Fachinfo~ationszentru~, Karlsruhe, Germany, 1990. Hellner, E., Schwarz, ., and Pearson, W. B., Structure Tvpe Descriptions for ~ n t e r m e t a l ~Phases ~c in the ~ r t h o r h o ~ b~~yc s ~ ePhysik ~ s , Daten, 16-29, Fachinformationszentrum, Karlsruhe, Germany, 1992b. Hellner, E., Schwarz, R., and Pearson, W. B., : 0-444-87478-X, Structure Type D e s c r ~ t i o ifor ~ s ~ n t e r ~ e t a l lPhases ic Elsevier Science, Northin the Monoclinic Systems, Physik Daten, 16-10, 382 pp. Coinpilation of the experimental data on the Fac~informationszentrum, Karlsruhe, Germany, structures of binary c o ~ ~ o u n d sTheoretical . 1993. advances in understanding the quantum-~echaiiical a CODATA Directory of Data Sources for Science origins of structural bility are also presented. and Technology, Chapter 1 - C~yst~1~ography, Hellner, E., Gerlich, , Koch, E., and Fischer, W., CODATA ~ u l l e t ~No. n ~24, 1997. The Oxygen Framewor~in Garnet and its Occurrence CRYSTMET, Structure and Powder Database for Metals and Internietallic Compounds, Totb in the Structures of Na3A12Li3F121Ca,A12(OH) 12, RhBi, and Hg3Te0,, Physik Daten, 16-1, ~ n f o r ~ a t i o nSystems Inc., National Research Fachinf o ~ a ~ i o n s z e nu~t r, ~ a r l suhe r Gemany, Council of Canada (NRCC). Collection of 1979. crystallographic structure data for metals and Hellner, E., E., and Reinhardt, A., The intermetallic co~pounds. ~ o ~ o g e n e o u s eworks of the Cubic Crystal ca~a~a.~o~/~. Struct UreS, Physik Daten, 16-2, a ICSD Inorganic Crystal Structure Database, jointly Fachinforinationszentrum, Karlsruhe, Germany, produced by FIZ Karlsruhe, Germany and The 1981. National Institute of Standards and Technology (NIST), USA. FIZ Karlsruhe: P.O. Box 2465,76012 Hellner, E., and Sowa, H., The Cubic Structure Karlsruhe, Tel: (3.49 7247) 808 555, Fax: (t-49 Tvpes Describ~din Their Space ~ r o u p swith the Aid of Framewor~s, Physik Daten, 16-3, 7247) 808 131, Email: NIST: 100 Bureau Drive Fachinformationszentrum, Karlsruhe, Germany, MD 20899-3460. Contains complete structural 1985. info~mationfor inorganic compounds, i ~ ~ l u d i n ~ H~llner, E., Schwarz, R., and Pearson, W. B., compound name, molecular fomula, crystal ~~ztroductionto an organic Crystal C h e ~ i s t r yU, group, unit cell parameters, atomic Physik Daten, 16-4, ~ a c h i n f o r ~ d t i o n s ~ e n t r ~ ~ msymmetry , coordinates, and tempe~aturefactors. The Karlsruhe, Germany, 1992a. database is also available on CD-ROM. Hellner, E., and Pearson, W. B., F ~ a ~ e w o r for ks Intermetallic Phases with Structures in Space Groups Canad~an S ~ i e n t ~ ~c u ~ i e~ ~~ t a~ bca s Service, e of the #-Stem of I 4 ~ Physik ~ ~Daten, ~ ,16-2.5, CAN/SND, Canada Institute for Scientific and Fachinfo~ationszentrum, ~arlsruhe, Germany, Technical Inf~rmation(CI§TI), National Research 1986. Council of Canada (NRCC), Ottawa, Canada. Tel: Hellner, E., and Pearson, W. B., Structure Type (613) 993-3294, Fax: (613) 9.52-8246, e-mail: Phases in the Space ~ ~ ~ c r ~ tfor i o Inter~etallic ns c ~ n ~ ~ ~ ~ n r Provides c . c a . a complete suite of Groups I 4 / ~ and ~ m~ 4 ~ / and ~ cTheir ~ i subgroup^^, international crystallographic databases and an Physik Daten, 16-6, Facliinformationszentrum, integrated search system. [ ~arlsruhe,Gemany, 1987. Hellner, E., and Pearson, W. B., Structure Type e PDF-2 Powder L)@raction FiLe Database, Descriptions for r ~ ~ t e r ~ ~ e ~Phases a l l i c in the Space International Center for Di~ractionData (ICDD). Groups I 4 ~ l a ~and d I 4 ~ l a cand ~ Their Subgrou~~s,
Data Sources
12 Campus Boulevard, Newtown Square, PA 190’733273 USA, Tel: 610-325-9814, Fax: 610-325-9823, E. The world’s largest and most complete collection of X-ray powder
tal Data, NIST Standard Reference Data, Standard Reference Data Program, National Institute of Standards and Technology. This database contains rel~able c h e m i ~ l , physical, and crystallograp~c information of solid state materials including inorganics, organics, iii~e~metallic§,metals, alloys,
graphic plotting program of the Cambridge Structural Database, Displays both molecular and packing diagrams, Capable of performing intra- and inter-molecu~ar geometric analysis. Diagrams may be rendered in wireframe
I
Backhaus, K. Q., Grell, H., and Fichiner, K., Database of OD (0r~er-Di.sorder) ~tructures,Z. Inst. fiir Phys. Chemie, Berlin. Contains information on the crystal structure of substances showing polytypisni and stacking disorder. Bibliographic information, characteristic features of the whole set of polytypes, description of certain individual polytypes, and additional useful information are provided for each substance. Crystal Lattice S t r ~ c t ~ r e sa, web page on the internet, the Center for Computational Materials Science at the Naval Research Laboratory (NRL). OfTers graphical representation of the structures for ://cs~-.rrww.nrl.navy.mil/lattic~/].
Other h e l ~ fdata ~ l sources concerning crystallography are those dealing with phase diagrams. Crystallographic information of LMCs, such as structure type, lattice parameters, and references, etc., can often be found in them. For further information, see the various data sources listed in section 4 of this chapter.
Kubaschewski, O., and Catterall, J. A., T h e r ~ ~ o c h e ~ ~Data i c a l qf Alloys, Pergamon Press, London, 1956. Collection of thermoche~icaldata and experimental methods for binary and ternary alloy systems.
799
Hultgren, R., Desai, P. D., Hawkins, D. T., Gleiser, M., and Kelley, K. K., Selected Values of the ~ ~ e r ~ ~ o d y nProper~ie~~ a~ic of Binary Alloys, American Society for Metals (ASM), Metals Park, Ohio, 1973. Critical evaluations of thermodyna~ic information on metals and alloys. ~ubaschewski,Q., Alcock, C. B., and Spencer, P. J., ~ a t e r i a l sT h e r ~ o c h e ~ i ~ s 6th t i ~ yedition, , ISBN: 0080-41889-9, Pergamon Press, Qxford, New York, 1993. Includes theoretical funda~ental§, experimental methods, and estimation methods of th~rmodynai~ic properties. T h e ~ ~ o d y i i adata ~ ~ i of c some IMCs are also presented. Adachi, G. Y., Imanaka, N., and Fuzhong, Z. Rare earth carbides, chapter 99 of ~ a n ~ b o oonl ~ the ~ ~ y s i and c s C h e ~ i s t r yof Rare ~arth,s,volume 15, Gschneidner, K. A., and Eyring, L., (eds.), ISBN: 0-444-88966-3, Elsevier Science, NorthHolland, Amsterdam, 1991. Provides properties of rare earth carbides, including the~odynamics, phase diagrams, crystal structures and physical properties. Colinet, C., and Pasturel, A., ~ ~ e r ~ o ~ y n a ~ properties of metallic systems, chapter 134 of Volume 19: Lant~ianides/Actinides: Physics-If, Gschneidner, K. A., Eyring, L., Lander, and Choppin, G. R., (Vol. eds.), ISBN: 0-444 9, Handbook on the Physics and Chemistry of Rme ~ a r t h ~ s , Elsevier Science, ~orth-Holland, Amsterdam, 1994. Covers the thermodynamic properties of lanthanide and actinide ~ n e t a ~ ~ i ~ systerns. adelung, 0. et al., eds. ~ ~ a s e ~ q u i l i b r i ~Crystullo~raphic , and T ~ e r ~ o ~ y n a ~ i c Data of’ Binary Alloys. Vol Macroscopic Properties of Bornstein. Numerical Data and Functional Relatioiiships in Science and Technology, New Series. Editor-in-Chie~ Madelung, Q., SpringerVerlag, Berlin, 1991-1 995, multivolume. Contains thermodynamic data on binary intermetallic compounds. For more bibliographic information, see the concise introdu~tionto this multivolume series in section 4 of this chapter. oom, R., Mattens, W. C. M., ~ i e d e m a A. , R., and Niessen, A. K., Experimental and predicted enthalpies of alloy formation (in transition metal alloys), Chap. 111 of Cohesion in ~ e t a l s(p. 95-657) , ~or~h-H ol l and, ISBN: 0-44487098-9, 1988,758 pp. Presents a complete collection of heat of formation data on binary metallics that contain at least one transition metal.
800
Miscellu?zeous Topics
Cuminski, C., and Calus, Z., Intermetullic C o r n ~ o ~ n d sin Mercury, 1SBN: 0-08-037206-6, 1992, 290pp. Solubility Data Series Vol. 51, J. G. Osteryoung and M. N. Schreiner, eds., Pergamon, Oxford. Co~pilationof the solubility of various IMCs in mercury. ., and Bever, M. B., Thermodynumic ~roperties,chapter 3 in ~ n t e r ~ e t u l l iCompounds, c Westbrook J. H., ed., John Wiley $: Sons, New York, 1967. Sluiter, M., Com~ilationof ‘1st princ~les’Formution Erzthnlpy Data, a Web page on the Internet. Provides formation entbalpy data and crystallographic on of some omic Energy Review, Special issues published by the International Atomic Energy Agency, Vienna. Includes t h e r ~ o d y ~ a mdata, i c such as heats of formation, standard entropies, etc., of some IMCs. (see
inorganic substances and for some organic substances. Cheynet, B., T ~ e r ~ o d y n a m P~operties ~c of ~norganic Solids, A Literuture ~ a t u b a s eCovering the Period 1970-1987, ISBN: 0-444-88036-4, Physical Sciences , Elsevier Science, North-Holland, 989, 2402pp. (in 2 vols.). Over 25 800 references covering > 13400 chemical systems, The references encompass all the the~odynamical properties of inorg~nic compounds, gases, metals, alloys and solutions. Thermochemical Database for Light Metal Alloys, In Concerted Action on M Science, Ansara, I., ed., ECSC~EEC-EA~C, 1s and Luxembourg, 1995, ~ a ~ f ~binary f f n alloys tabase, Covers binary systems with Fe, Cr, Ni, n, Al, Si and Cu, [ht Group Ilf- F‘ ~ i n u r ys e ~ i c o n ~ u c t odatabuse, r presents the 15 possible binary systems between the group IT1 elements Al, Ga and I As and Sb.
No. 1, 1966; SSOL: SGTE solufion dcitubase, assessed data for Lavrentev, V. L. et al., Niobium: Physico-Clzmicul condensed phases covering unary, binary, ternary Prop~rtiesof its Compounds and Alloys, Special Issue and quaternary metallurgical systems. [ht No. 2, 1968; www~~~t.kth.s~/tc/tdb/ . et al,, Tantalum: PhysicoCOST 507 Eight alloy datubuse, critically assessed Chemical ~ r o ~ e r t i eofs its Compo~ndsand Alloys, thermodyna~icdata for ~ulticomponentlight alloy Special Issue No. 3, 1972; Spencer, P. J., ~ e r ~ l l ~Puh ~y s:~ c o - C ~ e ~ i ~ a l systems based on Al, Mg or Ti. [http:ll ,CO .ult/npl/elubs/cost507. ht Properties of its C o ~ p o u n d sand Alloys, Special ~etullurgicul T h ~ r ~ o d y n ~ ~Data ic 3ffn~ (MTDAT A ) , National Physical Laboratory al., T h o r i u ~ :~ h y s i c o - C h e ~ i c u l MTDS, Teddington, Middlesex, TWl 1 OLW, UK. pounds and Alloys, Special Issue Thermody~amicdata for about 2500 i n ~ r ~ ~ n i c No. 5, 1975; substances, including IMCs, f ., et al., ~irconium:Physico-Chemical npl/crn~t/rnt~~~~/]. Properties of its Compounds and Alloys, Special Issue No. 6, 1976. a1 Tables, National Bureau of Standard Reference Data, Caithersburg, MD 20899. ein, Tel: (301) 621-2228. The JANAF database contains temperature-dependent ical data on inorganic compounds. T. B,, Bipiary Alloy ~ h u s e~ i u ~ r u m2nd s, W,, ~ I S T - J AT h~e r~ ~ ~~ c h e ~ ~ c a Massalski, l Edition, Vol. 1-3, ISBN: O-g7170-403-X(book), 0Tubles, 4th Edition, Parts I and 11, Journul qf 87170-562-1(CD-ROM), ASM International, ~ ~ y s i c and u l ~ h e m i c a Referencr. l Data, Monograph Materials Park, Ohio, 1990. Classic c o ~ p i l a t i oof~ 9, National Institute of Standards and Technology, phase diagrams on binary alloy systems. 4700 binary America~Chemical Society, LSBN: 1-563-96831-2, alloy phase diagrams covering nearly 3000 systems. 1998, 1951pp. Provides temperature-dep~ndent Tables of invariant equilibria, ~rystallo~raphic data, values for chemical thermodynamic properties of
Data Sources
80 1
Vol. 35(1), 1990, ISBN 3-~32120-00~0; Vol. 3 critical commentary, and primary references are 1990, ISBN 3-932120-0 1-9; included for each diagram. Villars, P., Prince, A., and Okarnoto, H., ~ a n d ~ ofo o ~ Vol. 36(1), 1991, ISBN 3-932120-02-7; Vol. 3 1991, ISBN 3-932120-03-5; Ternary Alloy Phase ~ ~ a g r a m10-vol. s, set in print, Vol. 37(1), 1992, ISBN 3-932120-04-3; Vol. 37(2), ISBN 0-87170-525-7, ASM International, Materials 1992, ISBN 3-932120-05-1; Park, Ohio, 1995. Over 15000 phase diagrams covering Vol. 38(1), 1993, ISBN 3-9321207380 ternary systems. Contains phase diagrams, related 1993, ISBN 3-932120-07-8;Vol. crystallographicinformation, and references. 3-932120-15-9; Ternary Alloy Phase Diagrams CD ROM, ISBN 0Vol. 39(1), 1994, ISBN 3-932120 08-6; Vol. 87 170-GO1-6, ASM International, Collection of over 1994, ISBN 3-932120-09-4; 15000 ternary metatlic phase diagr Vol. 40(1), 1995, ISBN 3-932120-10-8; Vol. 40(2), 10-vol. printed set. Full set CD 1995, ISBN 3-932120-11-6; Vol. -ROM (in~ustrial/heat resistant alloys; solders, brazes, and copper alloys; light metal structural alloys; electronic materials/ semiconductors; precious metals; rare earths/ actinides; carbides/~~trides). 22-1. ~ o o3:~Alloy Phase Baker, H., ASM ~ a n ~ Volume EfEenberg, G., ed., The Red Books on C D - ~ O ~ , D i a g ~ ~ISBN ~ s , 0-87170-381-5, ASM International, ISBN: 0-87170-6Gl-X, 1998 pro~ucedjointly by Materials Park, Ohio, 1992, 512pp., Collection of MSI and VINITI, published by ASM Internati~~al, phase diagrams of the most commercially important alloy systems, selected from the 3-vol. binary set and 10-vol. ternary set published by ASM. Note: in contrast to the diagrams in Massalski (1990), (see Petzow, G., and Effenberg above) the diagrams here are presented primarily in Series, published jointly by VC wt.%, not at.%. CD ROM: ASM Binary Phase Diagrams, ASM VCH Pu~lishers,New York, Group Ternary Phase Diagrams (Windows and DOS), TAPP, ESM Software, Hamilton, constit~tionaldata and phase diagrams on ternary of phase diagrams of binary and ternary alloy alloy systems. 15 volumes of this series have now systems based on the ASM compilations. been made available (see below). More topical P h ~ s e D o-~the Phase Diagram Delivery Program, mo~ographsare under preparation for the future. ASM Inte~~ational, Materials Park, Ohio. Provides Vol. I : Ag Systems Ag-Al-Au to A ~ - C U - P1989, , 3services to hunt for specific phase diagrams of the 527-26941-X, G 12pp. nterested in. Contact: Vol. 2: Ag Systems Ag-Cu-Pb to Ag 527-26942-8, 624 pp. ~ ~ a g r ~of m ~s e t a l l i c Vol. 3: A1 Systems Al-Ar-0 to AISystems, comprehensive collection of constitutional 527-27888-5, 444 pp. data and phase diagrams of unary, binary, ternary Vol. 4: A6 S y s t e ~ sAl-Cd-Ce to A l ~ C ~ - R 1991, u , 3and higher order alloy systems. From 1955, the 'Red 527-27889-3, 652 pp. Book' series were originally publighed annually in Vol. 5: Al Systems AI-Cu-S to Al-Gd-Sn, 1992, 3ssian language by VXNITI (the Russian 527-27890-7, G95 pp. Institute of Scientific and Technological Vol. 6: A1 S y s ~ e A ~ sl " G ~ to - ~Al-~ Information, the former All Soviet Scientific and 527-28349-2, 492 pp. Technical Information Service). Starting with vol. 35 Vol. 7: Al S y s t e ~ sAI-Mg-Se to A ~ ~ ~ i - 1993, T a , 3(covering publications from the year 1990), the ' 527-2837O-G, 497 pp. Book' became available in the English Vol. 8: A1 systems At-NE'-T'b to Alproduced jointly by VINXTI and MSI ( 527-2904G-X, 489 p ~ . Science International Services, GmbH, Stuttgart, Vol. 9: As Sy~temsAg-Al-As to As-Ge-ZB, 1994, 3Germany and now edited by C. Effenberg). The 527-29038-9, 472 pp. volumes of the English series now available through Vol. 10: As Systems As-Cr-Fe to As-& ASM Lntl are as follows. 527-29037-0, 582 pp.
802
~ i s c e l l a n e o ~Topics s
Vol. 11: A s Systems As-ln-Ir to As-Yb-Zn, 1994, 3527-29232-2, 586 pp. Vol. 12: Au Systems Au-B-CO to Au-Ge-La, 1995, 3527-29233-0, 484 pp. Vol. 13: Au System Au-Ge-Li to Au-Tl-Zn, 1996, 3527-29234-9, 488 pp. Vol. 14: Li Systems Ag-Al-Li to Ge-Li-Nd, 1995, 3527-29367-1, 458 pp. Vol. 15: L i Systems Hf-Li-N to Li-V-Zr, 1995,3-52729368-X, 453 pp. Predel, B., and Madelung, 0. et al., ed., Phase E ~ ~ ~ i l i b rCryst~~/lographic ia, and T / ~ e ~ ~ m o d yData n~ic of Binary Alloys. Vol. , Group N:Macroscopic Properties of Matter. Landolt~Bornstein.Numerical Data and Functional Relationships in Science and T e c h n o l o ~New ~ Series. Editor-in-Chie~Madelung, O., Springe~-Verlag, Berlin, 1991-1995, 5 s u b v o l ~ e s . Compilations o f phase diagrams, thermodynamic properties, and crystallographic data of the related binary alloy systems. S~~bvolume a: Ac-Au . . . Au-ZP, 1991, ISBN: 3-54015516-3, 511 pp. Subvolume b: B-Ba ... C-Zr, 1992, ISBN: 3-540551 15-8, 403 pp. Subvolume c: Ca-Cd .. . CO-Zr, 1993, ISBN: 3-54056072-6, 466 pp. S~bvolumed: Cr-Cs ... Cu-Zr, 1994, ISBN: 3-54056073-4, 354 pp. S u ~ ~ o l u me:e By-Er . , . Fr-MO, 1995, ISBN: 3-54058428-5, 337 pp. de Fontaine D., Calculation of Phase Diagrams, chapter 10 in this book.
Kassner, M. E., and Peterson, D. E., Phase Diagrams of Binary Actinide Alloys, ISBN: 087 17O-553-2, ASM International, Materials Park, Ohio, 1995, 489pp. Tanner, L., Okamoto, H., Phase Diagrams ofBinary Bery//!~m A//oys, ISBN: 0-87170-303-3, ASM International, Materials Park, Ohio, 1987, 229 pp. S~bramanian, P. R., Chakra~arti, D. J., and Laughlin, D. E,, Phase Diagrams of Binury Copper Alloys, ISBN: 0-87 170-4844 ASM International, Materials Park, Ohio, 1994, 512 pp. Okamoto, H., Massalski, T. B., Phase Diagrams of Binary Gold Alloys, ISBN: 0-87170-249-5, ASM International, Materials Park, Ohio, 1987, 343 pp. Manchester, E;. D., Phase Diagrams of Binary Hydrogen Alloys, ISBN: 0-87170-587-7, ASM Internatio~al,Materials Park, Ohio, 1999.
0
e
e
White, Charles E. T., and Okamoto, H., Phase Diagrams of Indium and Their Engineering Ap~liCutions~ ISB 0-87 170-438-2, ASM k, Ohio, 1991, 338pp. International, Mater Okamoto, H., Phase Diagrams ofBinary Iron Alloys, ISBN: 0-87170-469-2, ASM I~ternati5na1,Materials Park, Ohio, 1993, 472pp. Nayeb-Hashemi, A. A., and Clark, J. Diagrams of Binary ~ a g n e s i u mAlloys, ISBN: 087 170-328-9, ASM International, Materials Park, Ohio, 1988, 370pp. Nash, P., Phase Diagrams of Binary Nickel Alloys, ISBN: 0-87 170-365-3, ASM Intern~tional,Materials Park, Ohio, 1991, 394pp. G a g , S. P., Venkatraman, M., Krishaamurthy, M,, and Krishnan, R. Phase ~ i u g r a m s of ~ i n a r y T ~ n t a l u Alloys, ~~ ISBM: 81-85307-12-2, Indian Institute of Metals, Calcutta, 1996, 268 pp. Murray, J. L., Phase Di~grumsof Binary Titanium Alloys, ISBN: 0-87170-248-7, ASM International, Materials Park, Ohio, 1987, 345 pp. Nagender Naidu, S. V., and Rama Rao, P., Phase D i a g r a ~ sof Binary T ~ n g s t eAlloys, ~ ISBN: 8185307- 10-5, Indian Institute of Metals, Calcutta, 1991, 326pp. Smith, J. F., Phase Diagrams of Binary V u ~ u d i U ~ Alloys, ISBN: 0-8’7170-354-8, ASM International, Materials Park, Ohio, 1989, 375 pp. Prince, A. A., Raynor, G. V., and Evins, D. S. ~ h a s e Diagrams of Ternary Gold AlZoys, ISBN: 0-904357-503, The Institute of Materials, London, 1990, 512pp. Raghavan, V., Phase ~ i a g r a m sof Ternary Iron Alloys, Part 1, ISBN: 0-87170-230-4, ASM and Indian Institute of Metals, Calcutta, 1987, 219pp. Raghavan, V., Phase ~ i a g r a m sof Ternary Iron Alloys, Part 2, ISBM: 81-85307-00-8, Indian Institute of Metals, Calcutta, 1987, 360pp. Raghavan, V., Phase Dia~ramsof T e r ~ a r yIron Alloys, Part 3, ISBN: 8 1-85307-00-9, Indian Institute of Metals, Calcutta, 1988, 220pp. Rivlin, V. G., and Raynor, 43. V., Phase Diagrams of Ternary Iron Alloys, Part 4: Phase ~quilibriain Iron Ternary Alloys, ISBN: 0-901462-34-9, The Institute of Materials, London, 1988, 485 pp. Raghavan, V., Phase ~ ~ u g orf aT e~r n~ r yIron Alloys, Part 5, ISBN: 81-85307-04-0, Indian Institute o f Metals, Calcutta, 1989, 387pp. Raghavan, V., Phase ~ i u g r a m sof Ternury Iron Alloys, Part 6, 2-volume set, IS and 81-85307-13-X, Indian Institute o f Metals, Calcutta, 1993, 1294pp.
Data Sources Rogl, P., and ~ ~ e n b e r gG., , Phase ~iagrarnsof Ternary ~etal- or on-~arbon Systems, ISBN: 087 170-660-1, ASM Internatioiial, Materials Park, Ohio, 1998, 480pp. Gupta, K. P., Phase rams of Ternary Nickel Alloys, Part 1, IS 8 1-85307-07-5, Indian Institute of Metals, C Gupta, E;. P., Phase ~ i a g r a m o,f~ Ternary Nickel Alloys, Part 2, ISBN: 8 1-85307-1 1-3, Indian Institute of Metals, Calcutta, 1990, 247 pp. Raghavan, V., Garg, S. P., Venkatraman, M., Krishnamurtliy, N., and Krishnaii, R., Phase ~ i a g r a m , ~of ternary Iron Alloys, Indian Institute of Metals, Calcutta, 1996, 406 pp. Tomashik, V. N., and Grytsiv, V. I., State Diugrams of Systems base^ on Semiconductor Compounds, Naukova Dumka, Kiev (in Russian). 1982.
Tlzermo-Calc, Thermochemical Databank for Equilibria and Phase Diagram Calculations, ~e p a rtmen tof Material Science and Engineering, Sweden, Contact: Bo Sundman, Tel: + 4 9140 or 621 1, Fax: +46 8 100 41 1 E-mail: te these.A software package diagram calculations, applicable to any thermodynamic system in the fields of metallurgy, alloy development, mat semiconductors etc. [ht F*A*C*T, Facility for the Analysis of Chemical
T ~ e r ~ o d y ~ a ~a ijoint c s , research project between McGill University and the Ecole Polytechnique de Montreal and Thermfact Ltd. Software package for treating th~rmodynamicproperties and calculations in chemical metallurgy. Applicable in diverse fields of chemical thermodyna~iics.Otfered in the three : FACT-Win. FACT-Web and FACT Ono~yrn~~.ea/faet /~aet,htrn]. MTDATA, Metallurgi~al Thermocheniistry, the NPL Databank for Materials Thermochemistry, National Physical Laboratory, Teddington, Middlesex, TW11 OLW, UK. A software package for the calculation of phase equilibria and thermal properties of i~ulticomponent,multiphase systems, using a wide range of databases of thermodynamic data. ~h~p://www.npl.eo.u~/npl/cmmt/rntdata/]. ~ ~ n ~ Compu~herm h a ~ , LLC, USA. 437 S. Yellowstone Dr. Madison, WI 53719, (608) 274-1414, E-mail: i n f o ~ ~ ~ m Windows-based software program for calculating
binary
803 phase
d i a g r a ~ $ and
ther~o~y~a~ic
C o m p ~ i t e r - A i Learning ~ e ~ ~ Guide, ISBN: 0-90 1-716111, Ashgate P u b ~ i sC~oi~~p a~n y1992. , This software package illustrates the dynamics of ~icrostructural evolution during the solidification of metals and its relationship with the appropriate phase diagrams.
S
Wijn, H. P. J., ed., ~ a g n e t i ~roperties c o f ~ e t a l sVol. , 29, Group 111: Condensed Matter (former Crystal and Solid States Physics). Landolt-Bornstein. Numerical Data and Functional Rel at i ons~p~ in Scien Technology, New Series. Sp~nger-Verla~, 1986-, Multivolume. Includes data on the magnetic, and some non-magnetic properties of metals, alloys and metallic compounds which contain at least one transition element. Adachi, K. et al., S~bvolurne a: 3d, 4d arzd 5d ~lernents,Alloys and Compo~nds,1986, IS 15904-5, 653 pp. Booth, J. G., et al., Subvolurne b: Alloys avld C o ~ ~ o u n ~ s of d-Elernents with ~ a i Group n Ele~ents.Part 1 1987, ISBN 3-540-17094-4, 528 pp. Fruchart, D., et al., Subvolume c: Allqys and C o m p o ~ onf ~d-Elements witl~~ a i ~n r o u pElements. Part 2, 1988, ISBN 3-540-17744-2, 306 pp. Achiwa, N., et al., Sub~ol~irne dl: R ~ y d r i d e sand M u t ~ aAlloys, l 1991, 7,393pp. Burzo, E. et al., Subvolume d2: Compouads Between Rare Earth Elements and 3d, 4d or 5d ~ l e ~ e n t1990, s, ISBN 3-540-51288-8,545pp. Kaneko, T., Subvol~rneel: C o m p o u n ~of~Rare ~ a r t h Elements with Main Group ~lements.Part 1, 1990, ISBN 3-540-18936-73,519~~. Chelkowski, A., et al., Subvolume e2: C o m p o u n ~of' Rare Earth Elements with Main Group Elements. Part 2, 1989, ISBN 3-540-50338-2,440~~. ~
804
~ ~ , ~ c e l l a n e oTopics us J., ~ ~ n ~ b oono kMagnetic : 8-444-82232-1 1995, 708 pp.
Suski, W., and Troc, R., S u b v o l ~ eJ2: ~ctinide Com~oundswith other Elements. 3-540-55114-X, 423 pp. ., ~ a g n e t i c~ropertiesof ~ e t a l s , d s , 0-387~-Elements,Alloys and C o ~ p o u ~ ~ ISBN and Technology, EditorSpringer-Verlag, Berlin, on magnetic and related properties derived from the Groap 111 slabvolumes
uschow, KL. H. J., Handbook on ~ a g n e t i e ~ a t e r i a ~10, s , ISBN: 0-444-82599-1, 1997, 682 pp. Buschow, K. H. J., Handbook on ~ a g n e ~ i c ISBN: 0-444-82956-3, 1998, 730 pp. . H. J., Handbool~ on ~ a g n e t i c 1999, 570 pp. ~aterial,s,12, ISBN: 0-444~50249~1, Buschow, K. H. J., ~ ~ r ~ a nmagnet e n t ~ a t e r i abased l on tetra~onal rare earth c o ~ p o ~ ~of d the s type etism and ~ a ~ n e t i c
~ a g n e t i c~ r ~ ~ e r t i e s inter~etallic c o m p o u ~ ~ s Chapter , 138 of Handbook on the Phys and Chemistry of Rare Earths, 20, ~schneidner . A., and Eyring L., eds., ISBN: 0-444~82014-0, sevier Science, NorthHolland, Amsterdam, 1995. Presents the magnetic propert~es of 3d ~ ~ g n e t iand s ~ lantha~id~ magnetism. e Schultz, L., and Katter, M., Intrinsic Properties and , K., Misawa, S., Shiga, Coercivity of Sm-Fe-TM Phases, Chapter 10 in Editor: Wijn, H. P. J., ~ ~ p e r ~ a g nHard ~ t s , Magnetic ~ a t e r i a l s ~Long, 6. J., and Grandjean, F., eds*,JSBN: 0792310926, Kluwer Academic, 1991, pp. 227-259. 0 , E., and ~ ~ r c h i ~HI. a yR., r ~~ ~ y ~ i c a l ~ ~ o ~ e r t ,Fe,,B-hased alloyas;Szytula, A., and Leciejewicz, J., ~ a g ~ e t i~roperties c of ternary inter~etallic e o ~ ~ o ou~ t~h ~eRT2X2 s type, chapters 82 and 83 of and boo^^ on the Physics and Chemistry of Rare 12, Gschneidner, K. A., and Eyring, L., ~ a r t h s Vol. , eds., ISBN: 0-444-87 105-5, Elsevies Science, NorthHo~land,A ~ s t e r ~ a m 1989. , Amsterdam, 1980-1999. ook on Magnetic Materials, Engdahl, G., and Mayergoyz, I. D*, ~ a n ~ of~ o o ~ G iant Magne tostrict ive ~ ~ t ~ r i (aE l~se c t r o ~ ~tgism ne Series), ISBN: 0-122-38640-~,Academic Press, 1999, 386pp. Covers the physical origin of giant ~a~net ost ri ct ors~their ~ a n u ~ a c t u ~ i n gand ~ e ~gy ,a and ~ ~~ ~r ar i ~ processes - ~ e ~ under ~ ~ ~ operation. s , ISBN: 0-444-87106-3, on Magnetic ~ a ~ e r i a l , 4, Nakamura, Y., and Franse, J. J. M., eds., ~ a g n e t oif ~ in~termetullic ~ c o ~ p o ~ n dJournal s, of . J., Wohlfarth, E. P., ~ a n ~ b o oonk Magnetism and Magnetic Materials, 70, 1987, Magnetic ~ a t e r i a l s ,5, ISBN: 0-444-87477-1, 1990, 462 pp. ~ e r ~ a n e~ ~a tg ~ e t s , H. J., Han~book on ~ a ~ n e t i c Society Proceedings, 0-444-88952-3, 1991, 6 6 6 ~ ~ . . J., Han~book on ~ a g n e t i c Kouvel, J, S., Magnetic Properties, Chapter 27 in : 0-444~89853-0,1993, 676 pp. ~ n t e r ~ e t a l Compounds, li~ Westbrook, S. H., ed., J., Handbook on Magnetic John Wiley &, Sons, New York, 1967. : 0-444-81974-6, 1995, 542 pp. of
Data Sources
805
electronic materials. Complete list of data publications and the tables of contents for each t Data, H-47 11 TECH ngbeach Parkway, Bay PUBLICATIONS, 4 Village, Ohio, 44140, Issued in two or more volumes per year, beg in 1974 (Vol. 8), a series which con~inues,Volumes 1-7 were part of a series entitled 'Di~usionData' which was begun in 1967 by Eusion Information Center, Columb-cts, Ohio. H., ed., Dz~usionin S : 3~540-50886-4,Vol. Croup III: Condensed Matter, Springer-Verlag, Berlin, 1990, 747 pp., contains difhsion data for solid metals and alloys, including IMCs. Allen, C, E., et al., Beke, D. L. eds, Group 111,Vol. 33, Subvolume a: Difksion in Semiconductors, ISBN: 3-~40-60964-4,Sprin~er-Verl~g, Berlin, 1998. Wiley, J. O., ~ o b i l i t yof Holes in III-V Compo~nds, Se~~iconductor~ cind Semimetals, Willardson, R. K., eer, A. C., eds., Vol. ,Academic Press, 1975, p. 134. Larikov, L. N., Geichenko, V. V., and Fal'chenko, V. M., D ~ ~ ~ s iProcesses on in Ordered Alloys, translated from the Russian, Arnerind Publishing Co., New Delhi, 1984, 176pp. Hagel, W. C., DiJfusion, Chapter 20 in lnterrneta Compoun~s,Westbrook, J. H., ed., John Wiley Sons, New York, 1967.
Electronic Properties Information Center (EPIC), CINDAS, Purdue University, 2595 Yeager Rd,, W. Lafayette, IN 47906-1398, Tel: (765) 494-9393, Fax: (765) 496- 1175. Electrical, electronic, magnetic, and optical properties of electrical and electronic materials, many on IMCs, Electronic Materials Properties at the Institute of Inorganic Russian Academy of Sciences, Novosi~irsk.Tel: 73832-355-950. Three subsidiary databases: thermodynamic properties; structural properties; and physical properties. (Westbrook, 1997). o ~o nkf r a r eOptica~ ~ ~aterials, Klosek, P., ~ a ~ d ~ of Marcel Dekker, 1991, 624pp. McDaniel, T. W., and Victora, R. H., ~ a ~ d b oof o ~ ~agneto-Optical Data ~ e c o r d i n~~a t e r i a l ~Sub~, Systems, and Techni4ues, ~ o y e s 1997, , 967 pp. The Databook of ~ a t e r i a l for ~ s ~ n f r u r eDetectors, ~ Japanese Electronics Industry Develop~ent Association, 1979. Willardson, R. K., and Beer, A. C., et al., eds., Se~iconductor'sand S e ~ i - ~ e t a Series, ls Academic Press, rnultivolumes, 1966. Known as 'Willardson and Beer' Series. Cover a variety of fields in semiconductors and sernimetals. ls Malik, R. J., ed., Ill- V Semiconductor ~ a t e ~ i uand Devices, ISBN: 0-444-87074-1, Elsevier, 1989,728pp, Adachi, Sadao. Physical ~roperties of HI- Y Semicond~ctor Co~pounds, IS C o ~ p ~ t eSemiconductor r Databases, A , Crystal Wiley, New York, 1992. Provides numeric and graphical information on many of the Growth of' 4-6 Solid Solution S e ~ i c o n ~ u c t o r sB, ; C, semiconducting and material properties of In Optical Data on 2-6 Compound Semico~ducto~s; Str~cture~sand Properties of 2-4 I n t e r ~ e ~ a l l i c Inl-xGaxAsyPl-y,and Ino.53Cao.47As. Compounds; I), Crystal Data of ~ntermetallic e Li, H. H., Refractive Index of ZnS, ZnSe, and ZnTe C o ~ ~ o u n d ' sElectronic . Material Lab., Dept. of and its Wavelength and Temp ture Derivatives, J. ~ a t e r i a lScience, Faculty of Engineering, Tohoku Phys. Chern. Ref. Data, Vol. NO. 1, 1984, 103University, Ammaki, Aoba, Sendai 980, Japan. Tel: 150. Refractive index data of ZnS, ZnSe, and ZiiTe (81) 022-222-1800 ext 4463. (Westbrook, 1997). were selected, compiled and analyzed. ~Omputer-Aided ~ a t e r i a l s Synthesi~s-Co~pound Moss, T. S., ed., andb boo^ oy1 ~emiconductors,4 Electronic S e ~ ~ i c ~ n d u c t o r s (CAMS-CS), Vols., Elsevier Science, North-Holland, Amsterdam, Properties of 15 compound semiconductor mixed 1993-1 996. crystal systems. Electrotechnical Lab., 1-1-4, Basic Propert~e~ of Landsberg, P. T., Umezono Sakura-Mura, Ibaraki 305, Japan, Se~iconductors, and book on S e ~ i c o n ~ u c t o rIs, ~ Contact: Masato Yamazaki, Tel: (81) 0298-54-5455. ISBN: 0-444-88855-1, 1993, 1218 pp. The ~lectronic ~ a t e r i a l s Info~rnution Service Balkanski, M., ~ p t i c aProperties l ( ~ ~ Data ~ Reviews S ) series, The ~nstitution of and boo^ on Se~niconduc~o~.~, 2, I Electrical Engineers (IEE), Herts, UK. Comprises 3, 1994, 874pp. s other books on a variety of s e~ i c on~ uct orand Mahajan, S., ~ u t e r i a l s Properties , and Pre~aration,
806
~iscellaneousTopics I
~ a n ~ b oiz o So e~~ i c o n d ~ c t o r3,s ,ISBN: 0-444-88835~aterialsResearch Society Proceedings, Vol. 36, 7, 1994, 2398pp. (in 2 volumes). SBN 0-444-89910-3, Elsevier, 1993, 320pp. Hilsm, C., Device ~hysics,~ f f n d b o o kon Semi~iqu~dlsolid Equilibria for III- V Alloy -88813-6, 1996, 1 2 4 4 ~ ~ . ~ e ~ i e o n ~ u c t oPhysical rs, etallurgy Lab., Faculty >fEngineering, Tohoku University, Aramaki, Aoba, Madelung, O., et al., ed,, Semiconductors, Vol. 17, Fendai 980, Japan. Tel: (81) 022-222-1800 ext 4456. Group 111: Condensed Matter, Landolt-Bornstein, Clontact: K. Ishida. ata and Functional Relationships in Science and Technology, New Series. SpringerVerlag, Berlin, 1982-1985, multivolume, Presents the basic physical and related properties of the various families of semiconductors. Birnberg, D., et al., Editor: 0.Madelung, S u b v o l z ~ a: e P~ysicsof Group IV ~ l e ~ e nand t s 111-V C o m p o ~ n ~ , IFlukiger, R., and Klose, W,, eds,, $upercond~ctor.s: 1982, ISBN 3-540-10610-3,642pp. Transition Temperature~s and C h a ~ a ~ t ~ r i ~ a tofi o n Broser, I., et al., Editor: 0. Madelung, Subvolume h: P~ysicsaf II-VI and I-VII Compounds~S e ~ i ~ a ~ e t i c Elements, Al~oysand Compounds, Vol. CIL: Condensed Matter, Landolt S e ~ ~ c o n d u c t o1982, ~ s , ISBN 3-540-11308-8, 543 pp. Numerical Data and Functional ~ e l a t i ~ n ~ h in ips Freyland, W., et al., Editor: 0.Madelung, Subvolume Science and Technology, New Series. Springere: Physics of Non- ~ e t r a ~ e ~ rB~olnldye ~E~l e ~ e n t sand Verlag, Berlin, 1990-1994, multivolume. nary Compounds I, 1983, ISBN 3-540-11780-6, Authors: Braun, €3. F., Capone 11, D. W., Flukiger, 533 pp. Clasen, R., et al., Editor: 0. Madelung, ~ ~ u b v o l u ~ ~ R., e ~ et al., Subvolume a: Ac . . . Na, 1990, ISBN 3-54017621-7, 661 pp. ~ h y s i c s of Non- ~etrahedrully ~ o n ~Binary e ~ Authors: Flukiger, R., Hariharan, S. Y., Kuntzler, Compoun~ II, 1983, IS N 3-540-12160-9, 5 6 2 ~ ~ . R., et al., Subvolume bl: Nb, N b A l . . . Nb-Ge, 1993, Goodenough, J. B Editor: 0. Madelung, ISBN 3-540-55522-6, 284 pp. S u ~ v o l u ~g:~ ePhysics of Non- T’etrahedrully Bonded Authors: Flukiger, R., Hariharan, S. Y., Kuntzler, Binary C o ~ p o u nIII, ~ ~1984, ISBN 3-540-12744-5, R., et al., S~bvolumeb2: Nb-H . . . Nb-22, Nd . . . Np, , et al., Editor: 0. Madelung, S ~ b v o l u ~ e 1994, ISBN 3-540-57541-3 3 6 6 ~ ~ . Suenaga, M., and Clark, A. F., eds. ~ilamentaryA15 of T e r ~ r yC~mpounds,1985, ISBN 3Su~erco~ductors, ISBN 0-30-640622-5, Plenum, New 540-13507-3, 565 pp. York, 1980, 368pp. delung, O., et al., ed., Se~iconductors,Vol. 2 ~ a b e r ~ e i e H.-U, r, Kaldis, E., and Schoenes, J., up 111: Condensed Matter, Landolt”Bornstein, x i s . , High T’, S~percon~uctor ~ a t e r ~ ~(2lVols. s , set), Numerical Data and Functional Relationships in In European Ma ials Research Society Science and Technology, New Series. Springer~ r o c e e ~ ~ n g Vol. s, , ISBN 0-444-88884-5, Verlag, 1987-1 989, multivolume. A. supplement Elsevier, 1990, 1604pp. and extension to Vol. 17 (see above). * Durnas, J,, Neumuller, €3. W., Braun, H. F., ., et al., Editor: 0. Madelung, Seminoz~enko,V. P., and Tretyakov, Yu. D., eds., ~ubvolume a: I ~ t ~ ~ Proper€ies n ~ ~ i c of Group IV High T, Supe~cond~ctors, In Eur ~ l e ~ e nand t . ~111-V, 11-VI and I-VII C o ~ p o u n ~ s , Research Society Proceedings, Vol. 1987, ISBN 3-540-16609-2, 451 pp. 81529-5, Elsevier, 1993, 762pp. ~ ~ m e r l a a nC., A. J., ef al., Editor: M. Schulz, Vincenzini, P., High Tem~erature~~percoiz~uctors, Su~volumeh: I ~ p u r ~ t i eand s Defects in Group I Y Materials Science Monographs, Vol. 70, ISBN 0Elements and III-V Compounds, 1989, ISBN 3-540444-89061-0, Elsevier, 1991, 996 pp. 17917-8, 776 pp. Roberts, B. W., Supercon~uctiveProperties, Chapter Coutts, 71. J., Kazrnerski, L. L., and Wagner, S., o kH., ed., 29 in I ~ t e r ~ e € a lCl ioc ~ p o u ~ d~se, s t b ~ o J. m Diselenide f o r Phofovoltaic Wiley, Inc., New York, 1967. ials Science Monographs, Vol. Su~ercondueting ~ a t e r i a l ~ sata abase, National 7, Elsevier, 1986, 640pp. Research Institute for Metals (NRIIM) 2-1 , Sengen 1-chome, Tsukuba-shi, Ibaraki 305-0047 Japan, Triboulet, R., Wilcox, W. R., and Oda, O., eds., 81 298 592000. CdTe and R e l a ~ eCd-Rich ~ Alloys, In European
+
Data: Sources
Rare-Ea:rth Information Center New~s,a quarterly journal, Issued one volume per year, beginning in 1966 (‘vol. I), Rare-Earth Information Center, Energy and Mineral Resources Research Institute, Iowa State University, Ames, Iowa. Szytula, A., and Leciejewicz, J., Handbook of Crystal ~tructuresand ~ a g n e t i cPro~ertiesof Rare Earth I~termetallics,ISBN: 0849342619, CRC Press, Boca Raton, Florida, 1994. Compilation of crystallographic, physical, and magnetic data on rare-earth intermetallic compounds. Gladyshevskii, E. I. and Bodak, 0. I., Crystal C~~emistry of the Intermetallic Com~oundsof the Rare E ~ ~ ~t eht a l sVyshch. , Shkola, Lviv, 1982. Gschneidner, K. A. Jr., et al., eds., Handbook on the Physics and C h e m ~ ~ sof t ~ Rare y Earths, rnultivolume series, Elsevier Science, North-Holland, Amsterdam, 1978-2000. Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 1: ~ e t a l sTSBN: , 0-~4-85020-1,1978. Cschneidner, K. A. Jr., and Eyring, L., eds., Volume 2: Alloys and ~ntermetallics,ISBN: 0-444-85021-X, 1979. Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 3: N o n ~ ~ e t f f l Compounds-I, lic ISBN: 0-444-85215-8, 1979, 664pp. Gscbneidner, K. A. Jr., and Eyring, L., eds., Volume 4: ~ o n - ~ e t a ~Compounds-11, lic ISBN: 0-444-852166, 1979, 602pp. Gschneidner, K. A. Jr., and Eyring, L., eds., Vohme 5, ISBN: 0-444-86375-3 1982 7 0 2 ~ ~ . Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 6, ISBN: 0-444-86592-6, 1984, 5 7 4 ~ ~ . Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 7, ISBN: 0-444-86851-8, 1984, 5 8 0 ~ ~ . Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 8, ISBN: 0-444-86971-9, 1986, 3 8 2 ~ ~ . Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 9, ISBN: 0-444-87045-8, 1987. A. Jr., Eyring, L., and Hufner, S., High Energy Spectroscopy, ISBN: 0-444-87063-6, 1988, 6 1 2 ~ ~ . G$chneidn~r,K. A. Jr., arid Eyring, L., eds., Volume 11: ~ w o - ~ u n d r e d - y e aImpact r of Rare Earths on Science, ISBN: 0-444-87080-6, 1988, 594 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., V o l ~ m e 12, ISBN: 0-444-87105-5, 1989, 486 pp. . A. Jr., and Eyring, L., eds., Volume 13, ISBN: 0-444-88547-1, 1990, 4 7 4 ~ ~ .
807
Gschneidner, K. A. Jr., and Eyring, I.,., eds., Volume 14, ISBN: 0-444-88743-1, 1991 Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 1.5, ISBN: 0-444-88966-3, 1991,530pp. Gschneidner, K. A. Jr., and Eyring, L., eds., Cumulative Index, Volumes 1-15, I 89965-0, 1993, 522 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 16, ISBN: 0-444-89782-8, 1993, 604 pp. Gschneidner, K. A. Jr., Eyring, L., Lander, G and Choppin, G. R., eds., Volume 17: an than Actinides: Physics - I, ISBN: 0-444-81502-3, 1993, 788 pp. Gschneidner, K. A. Jr., Eyring, L., Choppin, G. R., Lander, G. H., eds., Volztme 18: ~ a n t h a : ~ ~ i d ~ s l Actinides: Chemistry, ISBN: 0-444-8 1724-7, 1994, 692 pp. Gschneidner, K. A. Jr., Eyring, L., Lander, G. H., and Choppin, G. R., eds., l ~ m e19: ~anthanidesl 0-44~-82015-9,1994, Actinides: Physics - II, IS 718 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 20, ISBN: 0-444-8201~-0,1995, 478 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., V ~ l u m e 21, ISBN: 0-444-82178-3, 1 9 9 5 , 4 3 4 ~ ~ . Gschneidner, K. A. Jr., and Eyring, L., eds., ~ o l u m e 22, ISBN: 0-444-82288-7, 1996, 816 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 23, ISBN: 0-444-82507-X, 1996, 664 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 24, ISBN: 0-444-82607-6, 1997, 600 pp. Gschneidner, R. A. Jr., and Eyring, L., eds., Volume 2.5, ISBN: 0-444-82871-0, 1998, 508 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., V o l u ~ e 26, ISBN: 0-444-50185-1, 1999, 5 7 6 ~ ~ . Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 27, ISBN: 0-444-50342-0, 1999, 572 pp. Gschneidner, K. A. Jr., and Eyring, L., eds., Volume 28, ISBN: 0-444-50346-3, 2000, 432 pp.
0
Fournier, J. M., and Cratz, E., ~ r a n s p ~ r t p r o p e ~ r i e s of rare e a r t ~and ~ actinide intermetallics, chapter of Handbook on the Physics and Chemi.~t~y cf Rare Earths Volume 17: Lanthanides/ Act inides: Ph))Lsic,s--I, Gschneidner, K. A., Eyring, L., Lander, 6 . H., and Choppin, C. R., (eds.), IS N: 0-444-81502-3, Elsevier Science, North-Holland, Amsterda~, 1993. Includes transport properties of rare earth and actinide intermetallics.
Mi~~cellaneo~s Topics
808
anovskii, R. E., Ther~ophysicalProperties 1's and Alloys, Metallurgia, MOSCOW, Russia,
5.9
Grin, J. N. and 6ladyshevskii, R, E., Gffllides,
1970, 5OOpp. Fomenko, V. S., Handbook of ~hermionic Properties: ~lectronic Work Functions and ~ i c ~ a r d s oConst~nts n of Elements and ~ompounds, Plenum, 1966, 151pp. ugdale, J. S., The Electricul Properties of Metals nd AZloys, The Structures and Properties of Solids , ISBN 0-7 131-2524-1, 1977, Edward Arnold,
Yatsenko, S. P., Gffllium- rnteraction,s with Metals, Nauka, Moscow (in Russian), 1974,
Samsonov, 6.V. and Bondarev, V. N., G e r ~ a n i ~ e s , Metallurgia, Moscow, 1968. Cladyshevskii, E. I,, Crystal C ~ e ~ i s t orf y Silicides and G e r ~ a ~ ~ dIzd. e s ,~ ~ t ~ l 1 u ~r o~ sicyo ~w 1971. ,
Schroder, K., ed., CRG Handbook o f Electrical ~esistivitiesof ~ i n a r ~y e t a l l i cAlloys, ISBN 0-84933520-5, CRC Press, 1983, 456pp. Schroder, K., ed., Electronic, Mugnetic and Thermal a Engineering Propert~esof Solid ~ u t ~ r i Electrical and Electronics Ser., Vol. , ISBN 0-8247-7024-2, Dekker, 1978. Moruzzi, V. L. Janak, J. E., and Williams, A. R., C~lculated~ l e c t r o n ~Properties c of ~ e t a l s ,ISBN 0-08022705-8, Pergamon, New York, 1978,
.) T h e r m o p ~ y ~ ~ c Pf f ~l o p e r ~ ~ eofs ies of Selected Topics in Solid State Physics, Vol. 18, Elsevier, 1986, 348 pp. , and Ziesche, P., ~lectronieProper aterials Science Mono~raphs,Vol. LSBN 0-444-98838-6, Elsevier, 1990, 196pp,
Yatsenko, S, P,, Indium - Properties ~ p p l i c ~ t i o nNauka, s, Moscow (in
., and Desai, P. D., ed., Beryllides, and ~ i s c e l l a n ~ oR~l/oys9 s 1. 111, Properties of Intermetallic Alloys, CIND Purdue University, 1994. Samsonov, G. V., ~ery/lides,Naukova Dumka, Kiev. 1966.
and
5.12 e Sarnsonov, G. V., and Perminov, V.
P., ~ a g ? z e s i ~ e s ,
Naukova Dumka, Kiev, 1971,
0
Payne, J. E. and Desai, P, D. ed., Alumi~ides,Vol. I, Properties of rnr~rrnetal~ic Alloys, CINDAS, Purdue University, 1994. Noebe, man, R. R., and Nathal, M. V., Ph-ysica chanieal Properties of the 132 C o ~ p o u n d ,NiAl, Intl. Mater. Reviews 38, 1993, 193-232. ope, D. P., and Liu, C . T. (ed.), Whang, S. High ~ e ~ p e r a t ~Aluminides re and Intermetallics, Elsevier Science, ISBN: 1851668225, 1993, 746 pp.
s
Gilp, B. F,, and Desai, P. D., ed., Silicides, Vol 11, ~ r o p e r ~ i oef s ~ ~ t e r m e ~ f fAlloys, l l j c CIN U~iversity,1994. Samsonov, C. V., Silicides and Their Use in Irzdustry, Ukr. Acad. Sci. Press, Kiev, 1959. Tung, R. T., Maex, K., and ~ e l l e ~ r iP.~ W., i , eds,, Silicide Thin Films - ~abrication,Properties, und Applications, Materials Research Society, IS 1558993053, 1996, 648 pp. Maex, K., and Van R U S S U M,, ~ , ed,, ~ropertiesof Metal Silicides, EMIS Data Review Series, No. 14, 1995, ISBN: 0852968590, 349 pp Harris, G. L., Prop~rtiesof Silicide~,EMIS Review Series, No. 13, 1995, 300pp. Gladyshevskii, E. I., Crystal emistry of' Silici~es and G e ~ ~ a ~ iIzd. ~ eM s ,e t ~ l l ~
Begin9 General Research Institude for ~ o ~ f e r r ~ u s ~ e t a l (ALTi. s N i A I and others).
Data Sources
809
Doyama, M., and Yabe, M., eds., ~ a t a b o oof~ Intermetallic Compounds, Science Forum, Inc., 1 pp. (in Japanese). Contains s applied in a variety of fields, such as superconducting, semiconducting, magnetic, shape memory, and refract0 Samwer, K., von Allmen, Stritzker, B., eds., ~ e t u s t a and Properties, In Society ~ r o c e e d i ~ g , ~ , Winter, M., WebEle W ~ a useful , source on many properties of the
~ a t i o n u l~ e s e a r c hInstitute f o r ~ e t ~ (data l ' ~ on creep, Eatigue, superconductivity), Tokyo. RACE (nuclear, superconduct~ng,III-V, and other special i~aterials),~niversityof Tokyo. ~ndustrialTechnology Research Ikzstitute (electrical, mechanical, and chemical properties of alloys), Hsinchu, Taiwan. ~nter~ationul Register qf ~ u t e r i a l s Database Munugers, CODATA, Paris, France. ~ u r e s ~ u r y ~ a b o ~ ~ t o r (high y temperature superconductors and other IMCs), §ERG, Warrington, UK. ~ o y o h u s hUniversity ~ of Tech~ology(alloy design, calc. of phase diagrams, TiAl), Japan. Dept. of ~ e t u l l ~ r g j c aEng. l Hokkaido University (calc. of phase stability of IMCs), Japan.
elevant databases and
1998 JANAF Tables; TAPP dat
~ and Lide, David R., CRC ~ u ~ d b oof oChemistry Physics, A ready-reference book of chemical and physical data, published annually, CRC Press, Boca Raton, Ann Arbor, London, Tokyo. Tlie section ~ropertiesof Solids contains some useful data on physical properties of IMCs, such as properties of semiconductors, diffusion data for semiconductors, optical properties of metals and semiconductors, t ~ e r malconductivity of metals and semiconductors as a function of temperature, properties of supercond~ctors,etc. §authoff', G., Inter~etullics,I§ VCH Verlagsgesellschaft GmbH, D-69541, Weinlieini, 1995, Emphases were laid on those intermetallic phases which have been applied as structural or functional materials or which are curreiitly the subject of niaterials developments. The Ti-Al, Ni-A1, Fe-Al, copper phases, A1 5 phases, Laves phases, beryllides, rare earth phases, and silicides are reviewed. The crystal structures, phase diagrams, and physical as well as mechanical and corrosion behavior are included in this book. Welsch, G., and Desai, P. D., Oxidation and Corrosiotz of Inter~etallicAlloys, ISBN 0-93168260-6, Purdue Research Foundation, West Lafayette, Indiana, 1996, 430 pp. Provides critical assessments of the performance of intermetallic alloys in high temperature environments. Contains detailed information on application problems, properties, and processing of intermetallic alloys including aluminides and silicides.
o
Optical Constants; C r y s t a l O ~ c ~ 98 (numerous software packages to visualize crystal images); NIST Crystal inorganic and organic crystals), Ic Advanced Materials & Processes Technology I n f o r m u t ~ ~ nAnulysis Center ~ ~ ~ ~ 201~ Mill Street, Rome, NY 13440-6916, Tel: (315) 3397117, Fax: (315) 339-7107, Reports and databases CO
I
on the T~ermopl~ysicul, Ther~~rudiative, ~ l e c t r o nElec., ~~~ Prop. of Refractory Borides, Carbides, and Silicides, 1986, Product Code: AMP-172; Optical, ~her~oradiative,T~ermophysical~ and ~ e c ~ a n i c a l~roperties of ~ ~ ~ c uC r a y ~ ~ i T e l / ~ r i ~1994, e , Product Code: AMP- 188; Oxidution and Corrosion of Inte~metullic ~ l l o y s , 1996, Product Code: AMPT-1; Texture De~)elop~nekztand Anisotropic Properties .from High Te~~perature~ o r ~ i t z ~ ~ e t a l sand Intermetallics, 1986, AMPT-2; Propert~esof ~nterme Eloys, Volumes I , 11, UI, 1990, Product Code: 70; Properties qf Interme lloys, Three Volume Electronic Version, 1996, Product Code: AMP:! erc cur?^ C a d m i u ~ Tell~ride ata abuse, 1994, Product Code: A ~ ~ 3 9 4 . o Green, A. J., Tanovic, B., Jones, I, and Fretwell, Mtls Sci on C ~ - ~ ISBN: O ~ 0,~ ~ 2 8 3 6 6 0C2 ~
~
810
~isceZZaneou,~ Topics
ress, 1998. ~esignedfor teachers and students of materials science, ~ e ~ a l l u r gengineeri~g, y, and other related disciplines. Collection of completely The authors gratefully acknowledge Dr. Jack interactive learning modules, including: Westbrook for providing useful materials on the data to ~rystallogra~hy; Introductio~ to Introd~~ction sources concerning IMCs. Electrons in Crystals; Int~oductionto Point Defects; Introduction to Phase Di Thermodynamics of Phase Diagrams, etc. w ~ .w c~c~re~.c~m~ Properties Database (on NiTi Wawrousek, H., Westbrook, J. H., aad Crattidge, W. (1989). Data Sources of Mechanical and Physical Properties of 1 Institute of Technology, ~ngineering Materials, Physik Dnten, No. 30-1, ‘als ~ e c h n o l o ~ y S, 100 44, Stockholm, Fachinformations~entru~, Karlsruhe, Germany. Sweden, +46 8 7908320, Westbrook, J, H. (1997). Sources of Materials Property Data Schlapbacb, L. ed., ~ y ~ ~ o gine nI n t ~ ~ ~ e t a l Z i ~ and Infor~ation,ASM ~ a ~ ~ ~ Volume o o l z20: ~ a ~ e r i a l s e ~ ~ I, ~ Electronic, o ~ Thermodynamic n ~ ~ and Selection and Design, Dieter, G . E., Vol. Chair, p. 491. Crystallographic Properties, Preparation, SpringerWestbrook, J. H., ed. (1967). Intermetallic Compounds.,Wiley, Verlag, Berlin, 1988. New York, ~ y ~ r o gine ~I ~ t e r ~ e ~ a l Z i c Westbrook, J. H., and Fbscher, PI. L., eds. (1995). e and Dynamic Properties, r ~ t e ~ ~ e t ~~lol?i ncp o ~ n dPrinciples s: and Practice, John Wiley & Sons, Inc., New York, Chichester. -Verlag, Berlin, 1992.
It i s possible to divide the problem of the design of new materials (including metallic materials) into two parts: (1) searching for new substances with predefined properties and (2) development of optimum conditions for the production and treatment of new niaterials. This chapter is devoted, for the most part, to solution of the first problem and describes the use of computer program systems, called ‘artificial intelligence’ systems. The necessity for such a materials design stems from several reasons. Principal among them are the following: - most simple binary metal systems are well investigated, but substances based on them no longer supply the needs of industry; - searching for and research on new ternary and
higher-order substances, many of which have unique properties, require considerably more time and expenditure than do binary systems; - while searching for new substances and materials through investigation o f all 5- 6 - . . . component
alloys, the materials scientists would use, just for Laboratory experiments, practically all the rare metals of our planet; - the now common computer design of new machines,
devices, buildings, etc. uses only databases (DBs)’ ‘A ‘Key Tenns List’ with definitions appears at the end of the chapter .
on properties of existing materials. The further development of automated design must address the development of systems for cornputer design of new materials with predefined properties;
- science, as with any institution of a human community, cannot expand indefinitely. increasing complication of materials requires either an increase of expenditures for research, or conversion from an extensive approach to an intensive one; -
the computer design of intermetallic substances is the most powerful way of speeding materials science investigations in this area.
At present the search for new inor carried out, for the most part, on the basis of the experience and intuition of researchers. The problem of a priori prediction of compounds that have not yet been synthesized and evaluations of their properties is one of the most difficult problems of modern materials science. Here the term ‘ a priori prediction’ means predict in^ yet unknown substances with predefined properties from only the properties of constituent components - chemical elements or more simple compounds. The following methods offer possibilities for solution of the last problem: - quantum-mechanical methods (Pettifor, 1983; Shah and Pettifor, 1993; Chelikowsky et al., 1993; Cohen, 1986, etc.); - empirical criteria (iiicluding two-dimensional diagrams) (Darken et al., 1953; Girgis, 1983; HumeRothery and Raynor, 1962; Laves, 1956; Mathias,
~ i ~ ~ e r m e t aCompoundc l~ic Vol. 3, Principles and Practice. Edited by J. H. Westbrook and R. L. Fleischer. 0 2 0 0 2 John Wiley & Sons, Ltd.
812
~ i . ~ c e l l a n e oTopics ~s
1955; Rabe et al., 1992;Villars~1995;Vozdvizhens~ii, 1975, etc.); -
multi-dimensional classifying rules (Chen, 1988; Chen et al., 1999; Gulyev and Pavlenko, 1973; Jackson et al., 1998; Kiselyova, 1987, 1993a, 1993b, iselyova et al., 1977, 1989, 1998a, 1998b; Kiselyova and Burkhanov, 1987, 1989; Kiselyova and Kravcheiiko, 1992; Kiselyova and Savitskii, 1977, 1979, 1981, 1982, 1983, 1984; Kutolin and Kotyukov, 1978, 1979a, 1979b; Kutolin et al., 1978; Manzanov et al., 1987;Pao et al., 1999; Savitskii et al., 1968, 1977, 1978, 1979, 1980, 1981, 1982a,b, 1990; Savitskii and Gribulya, 1985; Savitskii and Kiselyova, 1978,5979,2983,1984,1985; Talanov and Frolova, 1981;Villars et al., 2001; Vozdvizhenskii and Falevich, 1973; Yan et al., 1994; Zhou et al., 1989).
These and other approaches are briefly reviewed in another chapter in this volume by Naka and Khan who also present their application to the design of some niulticonstituent intermetallics. The quantu~-mechanicalapproach to calculation of i n t e ~e t a lliccompounds has been the most attractive for niost physical metallurgists using the methods of modern physics in their researches. However, over the course of the past half-century the achievements of quantum mechanics in a priori calculations of complicated intermetallic compounds evoke little enthusiasm, even among the most ardent followers of this approach. The low precision of results of calculations for known phases does not allow prediction of new substances and, at best, only makes it possible to explain known facts. This situation strengthens doubts as to the promise of the quantum-mechanical approach to the design of new metallic substances. Moreover, analysis of the results of quantum-mechanical calculations calls into question whether the differential and integral equations used are adequate to the complexity of condensed metal systems. Einpirical criteria for classification of known substances and for the subsequent a priori prediction of alloys, not yet investigated, are most commonly used in materials science. There is a common tendency in empirical sciences, because of the complexity of objects to be investigated, to substitute classification schemes for computational models. Some examples are: Laves’ rule (Laves, 1956) for predicting the crystal structure types of some intermetallic compounds, the Humeothery criterion (~ume-Rotheryand Raynor, 1962) and Darken-Gurry diagrams (Darken et al., 1953) for predicting mutual solubility of metals, the Mathias
criterion (Mathias, 1955) for predict~ngnew superconductors with the A15 crystal structure type, etc. (Girgis, 1983; Rabe et at., 1992; Villars, 1995 and Vozdvizhenskii, 1975). Frequently these rules are named for their founders and are the result of labour-intensive analysis of experimental data but are not a consequence of any theoretical calcu~ations. Moreover, in most cases theoretical physics cannot even explain the reason for the successful implementation of such rules. The principle of the development of empirical criteria of this type is a search for such properties of the chemical elements or analytical functions of these properties which would allow one to find a one- or two-dim~nsionalspace, in which it would be possible to divide known substances into distinct domains. The advantage of this approach consists doubtlessly in its simplicity and the ability to visualize the results with the help of one- or twodimensional plots. The essential shortcomings, from our point of view are the following: - these criteria quickly lose their reliability when new data do not easily fit within the framework outlined by the classification rule; -
the laboriousness of the development of the criteria;
- the criteria do not take into account the whole set of properties of the chemical elements (or simple compounds) which determine membership of a given ce substance in a certain class, a c i ~ c u ~ s t a ~that s about intersection of classes, Search for mul~idi~ensional classifying rules has become possible, using computers supplied with special programs for data analysis. This approach is a natural evolution of the above-mentioned empirical approach. The application of computers and programs to searching for ~ u l t i d i ~ e n s i o nregula~ties al in large volumes of data has allowed sharp reductions in the time of development of new criteria and revision of old criteria with the advent of new data. The solution of the problem of multidirnensionality of the experimental data array to be analyzed is limited primarily by the progressively higher capabilities of computers and programs. The advantages of the simplicity in using one- and two~diniensionalcriteria became iminaterial once compact computers were developed, which allow one to instantly predict new substances using multidimensional regularities. The excellent modern computer graphics allows visualization, in accordance with the user’s desire, of any section or projection of the mu~tidimensioIia1property space. Thus the classification programs expand the investigator’s possibilities
Computer Design of ~ a t ~ r i awith l s AI
for solution of the problem of searching for multidimensional regularities in large volumes of information. The search for multidimensional rules, connecting possible formation of compounds and correlating their properties with the properties of the chemical elements, is based on the use of databases on inorganic substances and materials and programs for searching for complicated regularities.
At present thousands of DBs in materials science and chemistry are in operation in the world (Andersson et al., 1985; Ansara, 1991; Bale and Eriksson, 1990; tyaryov et al., 1999; Drago and ~ a u f m a n ,1993; Eriguchi and Shirnura, 1990; Golikova et al., 1989; E40 and Li, 1993; Kiselyova et al., 1996; Kravchenko et al., 1991; Savitskii et al., 1984; STN, 1993; White, 1985; Yudina et al., 1996; Zemskov et al., 1995, 1998a,b, etc.). Only a small part of them are confined to purely bibliographic ~nformation,i.e. contain the abstract or the full texts of publications in a certain subject field as the basic document entries. However, the total i n f o ~ a t i o ncontent of such b~bliographic giants as the databases of CAS, VINITI, ~ Q ~ P E N D E etc., X, containing hundreds and even thousands of gigabytes of data, is highly competitive with the total information content of many factual databases on materials,
\
3
ure 1 Distribution of databases on properties of inorganic substances and ~ a t e r i a lover ~ subject scope: 1, themodynamic ~roperties;2, ~ n g i n e eproperties; r ~ ~ ~ 3, chemical and p h y s i c a ~ - c h ~ ~properties; ica~ 4, crystallographie and crystal chemical properties; 5, physical (electrical, magnetic, optical, etc.) properties; 6 , other properties
813
i.e. those computer-aided systems that contain infomation about the properties of substances and materials. The overwhelming majority of factual DBs contain information about properties of organic substances. However, a great number of large databases on metals and other inorganic substances and materials are maintained in the world, Shown in Figure 1 is the distribution of databases by subject scope of the information contained, The majority include information about thermal, en~ineeringand physical-che~icalproperties of inorganic substances and materials. In recent years the tendency has been toward cooperation in the develop~entof DBs and the integration of already developed international levels including coo frameworks of CODATA and UNESCO. This is because of the opportunities thereby to remove duplication and to cut down the considerable expenses for development and maintenance of such databases. Many DBs are accessible by remote access with the use of telecominunication networks (Drago and Kaufman, 1993; STN, 1993; Degtyaryov et al., 1999, etc.). The increase in the number and infor~ationcontent of DBs on substance and material properties is a natural tendency of the information age. However, the use of DBs only for information service does not reflect the requirements of society for the acceleration of scientific and technical progress and for the substitution of expensive experimental investigat~ons by cornputer simulations. The problem of information processing for DBs on the properties of metals and other inorganic substances is particularly acute. The attempts to supply these information systems with programs of thermodynamical calculations, statistics, and so on (Andersson et al., 1985; Ansara, 1991, etc.), do not allow good predi~tionsof the properties of inorganic compounds from ‘first principles’ using only the information of those databases. We began to develop DBs on terials and material properties in the late seventies y that time it had become clear that the software of the simplest information retrieval systems did not lend itself to subsequent computer simulation and also that this was an extremely archaic kind of information service, It has now been made obvious that it is n e c e s ~ a rto ~ develop DBs with Complicated structures, directed towards both computer simulation and information service. Just such principles were assumed as the basis of the information systems of inorganic substances and material properties that we are developing. The basic ideas, forming the foundation of our databases, are the following. First, databases are
8 14
~ i s c e l l a ~ e oTopics ~s
divided into two classes: DBs containing the most common information about inorganic substances and physical-chemical systems, and DBs that include the most detai~edi~formationwhich was collected and assessed by experts about indus~riallyvital substances (Figure 2). A database of the first type is a DB on the properties of inorganic compounds (Figure 3) containing i ~ f o r m a ~ iabout o ~ t h e r ~ a land crystal chemical parameters of compounds (Kiselyova, 1993b, 1997; Kiselyova et al., 1996; Kiselyova and Kravchenko, 1992; Savitskii et al., 1984). Databases of the second type are, for example, a DB on phase diagrams of systems with semiconducting phases (Zemskov et al., 1995, 1998a,b) (Figures 4-6) or a DB on properties of
single crystals of acousto"optica1, electro-optical and nonlinear-optical compounds (Degtyaryov et al., 1999; Golikova et al., 1989; Kravchenko et al., 1991; Yudina et al., 1996) (Figure 7). Secondly, we have completely departed from the philosophy of biblio~raphicDBs, whose factual elements for each document are data on the publication (paper, monograph or handbook) and in which any i n f o ~ a t i o nabout the substance concerned is added only as supple~ent.Such a philosophy increases the time for data retrieval for specific substances or systems, especially in the case of very large DBs, and does not provide full relevance (i.e. extraction of all documents which are appropriate to a certain request). The basic document of our DBs is
815
Computer DesigB of Materials with AI
cumulative infomation about a particular system or substance, the key identifiers of which are the symbols (or atomic numbers) of the chemical elements forming the system (Figure 2). The references are collected in a separate DB (or relational table) with consecutive . The databases (or tables) on the properties of the compounds (or systems) contain only the n u ~ b e r sof the references. Thus DBs (or the tables) on the properties of compounds (or systems) intersect each other at fields of constituent components (the ~uantitativecomposition of compounds and/or types of crystal structures) and with databases (or tables) of the references - at reference numbers. Such a DB structure optimizes fast data retrieval and preserves full relevance. These principles are used as the basis
+
for the following DBs developed and maintained by us:
1. A DB on the properties of ternary inorganic cornpounds ‘PHASES’ (Kiselyova, 199313, 1997; Kiselyova et al., 1996; Kiselyova and Kravchenko, 1992; Savitskii et al., 1984) contains the following i nfor~at i on (Figure 3 ) on more than 39000 ternary compounds in more than 16000 systems including the elements from H(1) to No(102). Information is extracted from more than 12000 publications. The retrospective data cover more than 70 years. The DB has been developed for IBM PCs operating under MS-DOS. The database management system (DBMS) is DATAREAL (Kiselyov, 1991). This DB i s popular among
System (Atomic Nwmfiers of Chemical Elements) A
Figure 3 Structure of database ‘PHASES’ on ternary inorganic cornpounds
~
1
SYSTEM __ -
I
J
816
~ i s ~ e L ~ a n e oTopics us
Russian users; therefore development of a new version of the DB with Tnternet-access has begun. Some of the data have been assessed by experts. Total data assessment by experts is a feature of DBs on the properties of substances and materials for electronics, which we began to develop during the 1990s. 2 . A database of phase diagrams of systems with intermediate semiconducting phases ‘DIAGRAM’ (Zemskov et al., 1995, 1998a,b) contains i n fo r~atio n on the most important pressuretemperature-concentration phase diagrams of s~miconducting systems evaluated by qualified experts and also on the physical-chemical properties of the intermediate phases. The figures of the d ~ a g r a ~are s only resented after critical assessment, statistical optimization (using expert evaluation of the data of different researchers) or thermodynamical self-consistency. Figures show the i n f o r ~atio nthat is stored for every binary system (Figure 4) and every ternary system (Figure 5); Figure 6 shows the structure of the DB on references. The DB ‘DIAGRAM” includes detailed analytical reviews for each system - rninimonographs of a sort - that reflect the extent of investigation of the system. Thermodynamic parameters and c ~ ~ p u t a t i o n amodels, l which were utilized for thermodynamic self-consistency or statistical optimi~ationof data of the different investigators, are stored separately. Apart from information about se~iconductingsystems, this DB contains data about some binary metallic systems that are ~ o n s t i t u ~parts ~ t of ternary systems with semiconducting phases, and also data about crystal structure of the chemical elements. The database of phase diagrams now contains ~nformation on several tens of semiconducting systems. Data retrieval is carried out by dialog-based menus and screening forms, ~ p d a t i n gof the database is carried out every has been developed for the IBM PCs operating under MS-DOS (DBMS = DATAREAL). We also developed a new Internetversion of this DB. 3 . A database of crystals with significant acousto-, electroand nonlinear-optica~ properties, yaryov et al., 1999; Golikova henko et al., 1991; Yudina et al., 1996) contains information on crystals of the most important substances of this class as evaluated by experts. The information contained in the database about the properties of the crystals
Figure 4 Structure of DB on binary semiconducti~gsystems
is displayed in Figure 7. In addition, the database includes extensive graphical in~ormation about the properties of the substances. At present data on several tens of substances are stored in this data~ase.A version of this DB for ~ ~ t e r n e t ~ u sise r s now in the making. In addition to a DB in Russian, a version in English has also been developed.
Doubtlessly databases on the properties of substances and materials open new avenues for i n f o ~ a t i o n
Solvus) ~ u r ~ ~ c e s ,
Fi~ure5 §tr~ctureof DB on ternary se~i~onducting systems
C ~ ~ ~Design p ~ otf ~e a~t e ~ i awith l s AI
S
service for specialists. However, it is but one of the aspects of the new i n f o ~ a t i o ntechnology. Rational use of stored data implies their processing with the purpose of searching for regularities that could be applied: to the prediction of new substances with predefined properties; to the development of the technology of synthesis of new materials; and to the
817
prediction of the bebavior of materials under the effects of various factors, i.e. automation of the practice of materials science. This problem can be decided easily in those rare cases where there is a good analytical description of the regularities to be sought, and the specialist needs only to insert the necessary information from the database into one of the chosen models. Among computer-aided systems of this kind are numerous databases on thermody~amic properties of substances provided with programs of thermodynamic calculations: IVTANTHERMQ (Ansara, 199l), T H ~ ~ ~ Q C ~ (Andersson et al., 19851, etc. However, the majority of materials science problems cannot be formalized with the ~ ~ p ~ ~ c aoft i only o n those simple algebraic structures that are used, for example, in thermodynamics. Prediction of new substances with predefined properties, interpretation of spectral information, selection of substances for certain purposes, development of optimal technological processes for the synthesis of materials, separation and identification of substances, etc. belong to this class of intractable problems. All
i~Mre7 Structure of database ‘CRYSTAL’ on crystals with acousto-, electro- and no~line~r-optl~al pro~~rties
818
~ i s c e l l ~ n e o uTopics s
these problems are presently solved only from the experience and intuition of the investigators. One of the most effective ways to automate these fields is by the application of the ideas of artificial intelligence and knowledge bases. What are the problems that appear with the intellectualization of materials databases ( G l a~ un, 1995; Gladun and Vashchenko, 1995; Kiselyova, 1997; Popov, 1987; Pospelov, 1988; Pospelov and Pospelov, 1985; Zagoruiko, 1999)? First, it is necessary that the computer understand the professional language and the statement of the problem of the user, i,e. the intelligent system should have two sorts of knowledge: knowledge of language and knowledge of the field in which the user works. Knowledge of the first sort is realized at the level of a conversationffZ, or linguistic processor. It can be a system of special programs or a complex including both hardware and software. As a e of language is stored directly in the conversional processor. The knowledge about the subject field is stored in a special ~ n o ~ l e base. ~ge The terms ‘knowledge’ and ‘knowledge base’ as applied to computer information appeared in the 1970s during the development of artificial intelligence systems. What are the distinctions between ‘data’ and nowle ledge'? Data are values used for solution of problems and metadata (Westbrook and Grattidge, 1991a) for descriptions of objects, situations, phenomena, and coniiections between objects. Knowledge is info~mation a ~ o u tthe processes of solution, the regularities which, applied to data, generate new information (Gladun, 1995). Particular features, distinguishing knowledge from data in connection with their representation in computers (Pospelov, 1988; Pospelov and Pospelov, 1985), are the following: inter~or i n t e r ~ ~ e t ~ b i z which ity makes it possible for the cornputer to understand' the information to be input at a substantial level; a v a i l ~ ~ i l oif~ ystructure which provides a computer with the ability to form a hierarchy of concepts, to introduce new generalized concepts and to decompose concepts into constituent subconcepts and the relations between them; availability o ~ c o n ~ z ~ c t i that o n s provides a computer with the possibility of including the connections not only between concepts but also between the facts, processes and phenom~na;activity is the feature that relates the computer to ~ o sapiens ~ o and is connected with actions leading to a realization of procedures that can be useful for the solution of certain problems (for example, the detection of a contradiction between the prediction and an experimental Fact becomes the stimulation for overcoming it and forming new
knowledge). It is, however, impossible to demark an accurate division between knowledge and data. For example, interpretabil~ty i s inherent to relational databases, and structuring is an i~tegralpart of all modern DBs. But availability of connections and activity have no parallel in a methodology of DBs. Secondly, it is important to have a system that converts the description of a source problem into a running program. A complex of software for the solution of this problem is called a ~ r o ~ sr cf~ fe ~~u l e r , or simply, a s c ~ e ~ ~During ~ e r . its work the scheduler continwally contacts the knowledge base, getting from the base the knowledge of the application domain, the methods for solution of tasks, and i n f o ~ a t i o nabout the possibility of an automated combination of programs from some of the basic programs which are stored in the knowledge base. A special system named the monitor realizes a control by interaction of all subsystems. The complex of conversati~nalprocessor, knowledge base, scheduler, and monitor forms the intelligent ~ n t e r f ~ cofe the computer (Pospelov, 1988; Pospelov and Pospelov, 1985). Expert systems are the most widespread kind of artificial intelligence systems. They date back about It0 years and are intended to solve very complicated problems in particular application domains with the use of large volumes of special knowledge of a high quality. The latter are extracted from various sources, namely: books, papers, scientific and technical documents, domain experts, etc. uch knowledge also includes some procedures, strategies, empiric regularities, and so on. This knowledge is represented in a special manner and is stored in the knowledge base. It will be noted that expert systems use models based on special formalisms of artificial intelligence (Gladun, 1995; Gladun and Vashchenko, 1995; Popov, 1987; Pospelov, 1988; Pospelov and Pospelov, 1985; Zagoruiko, 1999). Unfortunately, many developers doii’t take this aspect into consideration. These developers use the fashionable term expert system for the definition of their program systems that use a conversational mode similar to the natural language. True expert systems are artificial intelli~encesystems that use knowledge represented as rules, frames, or semantic networks (Gladun, 1995; Cladun and Vashchenko, 1995; Popov, 1987; Pospelov, 1988; Pospelov and Pospelov, 1985; Zagoruiko, 1999). A particular feature of expert systems is a subsystem of explanations that is a constituent part of these systems. It controls the work of the scheduler and describes its functioning in a condensed form that is convenient for the specialist. It fosters trust in the work of the intelligent system and
above, a knowledge base, a co~versatio~al processor answers for the user the ~uestions and monitor (see Figure 8). The system is evel lope^ one or another solution is accepte choice. A particular feature o f arti~cialintelligence systems is an ability for automatic searching for regularities and use of them for prediction. Thus expert systems are systems with ‘poor? in e because they use only those regularities (kno which are extracted from the experts. As our e x ~ r i e n c eshows, any attempt to extract rules, connecting the formatio~of a certain compound to the properties of its coiistituent r~ists elements, from cliemists or physical ~ ~ e ~ ~ l l uhave which completely classifies known physical-chemic~l been unsuccessful, because they prefer only to estimate systems, data for which are processed by the c o ~ p u t e r . authenticity of data concerning the existence of the During the prediction process the coinputer receives en compound or t values of the particular only the atomic numbers of the el properties of the phase, hile f ~ ~ ~ l a tthe i nspecific g tions of simple compounds. A brief review of the data features of this applica~iondomain, we abandoned the analysis methods from e point of view of their idea of making the chemist or physical metallurgist applicability for searchin for rules in the i n f o r m ~ t i o ~ outline the rules o f the fornation of metallic and other in databases on the prop i n o r g a ~ c substances with desired properties and decided only to make use of their expert assessment of the data intended for computer learning. It should be noted that an expert system is a passive system that is not capable of obtainin new regularities or of use in the prediction of phases and to forecast the searching for c o ~ i ~ ~ a d i c t ibetween o n ~ knowledge and facts. This circumstance, in ~ o ~ b i n a t i owith n the culties of the extraction of knowledge from the experts, were the reasons for prior failures of the removal of these restrictions and realization of what are called, ~ ~ r ~ ~ e r working with the com~uteror with the i n f o r ~ a t i o ~ ~ predicting s y s t e ~ .It provides application domain also with info~ation-predic~~ng system. employment of a lin~uisticprocessor in the s o f t ~ a r ~ or some softw~re-hardwaresupport will allow the sys~emto ~nderst oriented language of the user. The monitor controls the computation process and 1s now inalting a version of a partner system - an provides the interface between the functio~alsubsysinformation-pr~dicting~ computer-ai~edsystem. tems as well as teleco~municationa c ~ ~tos sthe s~stem. s y s t ~i s~ intended for data retrieval on k In addition, the monitor signals wheiiever new expericompounds, the prediction of hypothetical inorganic mental data contradict exi compounds, not yet synthesized, and the forecasting of Such contradictions will be elimin new data in the computer learnin rule in the knowledge base. ser r e ~ u e s information t~ c as discussed pro~erties o f ~ n o r g a ~ i compounds
If data about this phase are
820
Miscellaneous Topics
Figure 8 Schematic diagram of an information-predicting system
~ o i n p ~Design t ~ r of Materials with AI
they can be extracted and used for further studies. If no information about the compound is stored in the database, or if the i n f o ~ a t i o navailable is incomplete, the computer, in response to the user’s request, determines whether the rule (network) corresponding to the desired property for a compound of a certain type is present in the knowledge base. If the phase is present, the database supplies the appropriate set of component properties to predict the desired characteristic, If the knowledge base does not have the desired rule, then examples for the computer learning process are searched for in the database. The correctness and representativeness of these examples are estimated once more by the user; and, if the sample is found adequate for computer learning, the learning and prediction subsystems process them in turn. The resultant prediction is received by the user, while the classifying rule thus formed is stored in the knowledge base. The above example is the simplest of the problems that can be solved by an informationpredicting system. A more complicated problem would be, for example, predicting all possible phases in ternary and multi~omponentsystems, combined with the estimation of their properties. Although the previous problem can be solved by real-time processing, the latter problem requires much more time. The principles underlying the info~ation-predicting system as developed at present have been tested successfully earlier on the prototype system supported -6, second-generation computer urkhanov, 1989). The employment of powerful personal computers will make it possible to build a version of the system that can be operated by users at large.
82l
(Zagoruiko, 1999). In this case the models of pr~cesses or objects to be investigated are known to ~ a r a m e ~ r i ~ accuracy, i.e. sets of the most important features and a general view of the dependencies known, but it is necessary to ca exponents and other parameters, data for the behavior of the objects processes. This approach does not rigour of computations as in the conversion from one parameter set to another, because assumptions the values of unknown parameters and their eh with available experimental data, connecting input an target features, are required. o ~ e v e rthis , approach gives good results in research on comp~icatedchemicaltechnological and metallurgical processes, simulation of kinetics or diffusion, etc. The emerging field of cybernetics with its concept of ‘a black box’ has allowed an approach to the solution of even more complicated problems in which the investigator has only a set of experim~ntaldata with a description of input and output features, and neither the form of the models nor the degree of influence of the input features on the subjects of interest are known. Such problems are deci~edmost e~ective~y with various methods of artificial intelligence. The challenges faced by physical metal1 and technologists include: prediction of chemical compounds, evaluation of their properties, development of models of very phase technological processes, etc. view, for example, of the theor results are not sufficiently ri~orously substantiated from physical theories, nor from the rigor of the
ods la Classical applications of mathematics to n~tural-science domains are associated with the calculation of one parameter of an object or phenomenon from knowledge of others. Examples are: Ohm’s law, Fick’s law, etc. Each model of the object i s described by some analytical expressionin the form of an equation or set of equations or i~equaliti~s that are approximations to reality. use of accurate
More ‘inexact’ from the point of view of the theoretical physicist, another approach is connected with a solution of problems with identified models
alternative could be to abandon the offered computational approach and use only e~perimental~ e t h o d s ; that is absolutely incorrect under present conditions of the intensification of scientific research. By convention, the problem of data analysis by artificial intelligence methods can be divided into three parts: automatic clas~ification (or pat~ern reco without computer learning); computer learning to classify objects (or coi~puter learning in pattern recog~ition); for the most important classification (or conjunctions of sets of feature values).
e first problem is f o ~ u l a t e das follows: it is sary to divide a set of objects, specified by sets of features, iiito classes, such that points inside each are close to one another in feature space. ithms to solve this problem are most iolation of the hypothesis of compactiiess, the reali~dtionof which is a general prere~uisitefor a correct solution of each of the three types of problems ve. The objects o f each class should form clusters’ in feature space. The problem of cla$sification is the following: it is necessary to determine such clusters aiid to construct a dividing ,the use of which will allow determination ership of objects to the classes found. The most justi~edc l a ~ ~ i ~ c a t iofo nalgorithms of pattern recogi~itionwithout computer learning is given in the orofeyuk (1971). A set of algorithms for an a ~ ~ t o ~ classification ~tic a mo~iographs by Arkad’ev lications of these methods in materials science and e, among the most interesting: auto-
use the system of com~uterlearn in^ un, 1995; Gladmn and Vashchenko, represents the initial information about known physical-chemical systems as growing A pyramidal network is an ac having no vertices with one entering arc. If the processes of concept f o ~ a t ~ are o n determined in the network, then the ~yramidalnetwork is designated as a
set of values of the component properties with an ii~dicationof the class to which the system belongs. The nearby values of properties of com~onentsare united into one interval using a special program or the experience of the researcher. The c o n c e ~ t - f o ~ a t i o n process consists of the analysis of vertices in the network that is built and the choice of those vertices that are the most typical for each class. These vertices become the checking vertices. The resultant concepts (classifying regularities) can be stored in computer and printed out or read in the form of learned, in which the intervals of values o f the coltnpoiient
problem - computer learning for pattern recognition - is formulated as follows: let a feature numbers of the elements or designations of simple compounds. The values of the properties of the t is known that it contains a fixed appropriate elements or simple ~ o i ~ a i n the s , boundaries of which are automatically extracted from the ~ ~ n k n o wand n ~ there are no rules for the definition of the growing pyramidal network embers ship of iz particular point to one or another can easily obtain the necess~rypredic~ions. ~ o ~ ~Ini cn~ m . ~ u t learning, er the points, which are The third problem - the selection of the most sampled randomly from these domains, are entered iiiiportant classifying features - has a dual purpose. and possibly relevant information First, it is necessary to m i n i n ~ the ~ ~ in~tial e feature set in g to one or another domain is given. order to reduce the time of data processing using he purpose of computer learning is the construction of pattern recognition algokt d leave oiily the (most nly the points shown important) dividing features, ondly, most practical to these domains. The problems are aimed at se those features which iction consists of an are the most typical for the given class and distinguish ~nterpola~on and extrapolation of the available data of small volume to the entire set. This this class from others. ~ l g o r ~ t ~ofmtXis s kind can be classified into two classes: algorithms of a ~ r from the previous one only in the rigid weighting and algorithms of a posteriori weighting. In classes. The most successful classification the first case the process of constructin kind and examples of algorithms used by ~ a g o r ~ k(1999). o regularity is carried out before the evaluation of the iven in the mono importance of the features, and on i t h s of computer g are widely applied to other class. The classification of a ~ncludingour investigations aimed at the importance of f~aturesis made possible by the type redicting the ex is ten^ of new inorganic compounds and evaluating their ere a specific a l g o r ~ t ~ of criteria of importance for class division. Our experience in the application of various is used: to teach the coniputer how to process data for methods of artifi~~al intelligence to data analysis concept formation.
~
23
~ ~ ~ ~Design ~ u qft ~ea rt e ~ i with a ~ sAI shows that it is impossible to assess the i ~ por t anceof an individual feature over the whole range of its values. As a rule, some feature is of i ~ por t a nc efor classification, only in a certain range and for certain kinds of substances. In most cases, chemical phenomena are d e t e rm ~ n e~ by a set of features and thus, it makes sense to speak only about the importance of sets of features over specific intervals. e now explain this final proposition. Illustrated in Figure 9 is an example of classes of objects on of objects into narrow ranges of the . The features are not ~ e ~ e important ~ a ~ ~ for y the separation of classes, if a mixture of classes is observed practically everywhere over the range of their variation. It can be seen in this example that the features are of ii~portance only within certain intervals.
The idea of the application of camputer learning methods to searching for regularities of formation o f
binary intermetallic com po~~n the first time by my colleague Institute: E. M. Savitskii and V. ( ~ a v i t s ~et i i aE., 1968). They great number of problems of predicting various types of binary systems (for example, those solubility or with simple eutectics); th formation of thousands of binary c , A,B, etc. and evaluated some of their properties (type of crystal s t ~ ~ c t u rme1 e, point, critical temperat e of transiti~nto the su conducting state, etc.) avitskii and Gribulya, 1985;
tion about the ~ist~ibution af electro~sin the ener levels of isolated atoms of the chemical elements. such a simple description of binary s y s ~ e allowe ~~s properties with an average reliability of more than 90%. Thus the properties of binary s y s t ~ ~ depend s strongly on the properties of the const~tuentche~i cal e~e~ents. Tlie further development of this approach has followed two interrelated di r~~t i ons: intro
I
B
A
c
2
Map illustrating that features of importance are valid only over discrete intervals of p ~ ~ a ~ e t e r s
824
~ i s c e l l a ~ e Topics o~~s
the complication of the composition of the physicalchemicalsystem, and the development of new predicting iselyova et id., 1977, 1989, 1998a, 1998b; Kiselyova
avitskii et al., 1977, 1978, 1979, 1980, 1981, 1982, a, 1978, 1979, 1983, 1984, 1990; Savitskii and 1985; Talanov and Thousands of new compounds and their properties in ternary, quaternary and more complicated systems approach. The systems were properties of the chemical electrons in the energy levels of isolated atoms of the chemical elements, ionization ~otentials, t h e ~ a land thermodynamic properties, ionic or atomic radii, etc.) and properties of binary compounds.,and also various functions of these properties (ratios of radii, electronic concentrations, etc.). The tions depended on the algorithm of classification of the analyzed ative they were, and also on a s for the description of certain classes of ~hysic~l~~hemical systems. e search for, and development of., effective cting systems were aimed at the creation of more powerful programs capable of analyzing, on the one hand, very large banks of ex~erimentalinformation, and, on the other hand, of allowing construction of multidi~ensionalclassification rules under the condition of small sets. ~ m pro ve~ ents in electronics allowed the development of systems with a user-friendly interrking in real time (Chen et al., 1999; Gladun,
iselyova, 1987, 1993a, 1993b, 989, 1998a, 1998b; d Savitskii, 1977,1979, nov et al., 1987; Pao,
~ o we ver,the most important result of research in this field is that artificial intelligence methods have
become an opera tin^ tool for searching for regularities in experimental data, and the use of these regularities for predicting new intermetallic and other inorganic substances has been achieved, The approach, which had been developed in the A. A, Metallurgy and Materials Science, in different countries. The most powerful groups work
Chinese Academy Let us consider in greater detail the investigations of the Baikov Institute on predicting ternary i n t e ~e t a l l i c compounds. The problem of predicting new substances with desired properties can be divided into four ~onsecutivepro~lems: - prediction of coinpound formation or non-forma-
tion for ternary systems;
- prediction of ternary compou~dsof desired composition; - prediction of phases with a specific crystal structure
type; - estimation o f quantitative properties of the phase
(critical temperature of transition to the superstate, homoge~eityregion etc.). These problems can be solved sequentially, or any one of these tasks can be solved separately, as examples of the compleme~taryclasses are added to the learning set. For example, in the case of predicting the crystal structure at room temperat~reand atmospheric pressure for compounds of a desired composition, it is necessary to include in the learning set examples of both the formation and n o n " f o ~~a t i o n of the compounds in such systems under these conditions. In most cases prediction is carried out for systems at normal conditions, for example, the prediction of a phase at ambient pressure and temperature. In order to predict phases that exist under other conditions (eg. high pressure), it is necessary to enter examples of known compounds that exist at high pressures into the learning set and add pressure as a parameter. ~nfortunately,a majority of the exp~rimentalmeasurements of compound properties are carried out under conditions of incomplete e ~ u i ~ i b r iIn u ~addi. tion, the d e t e ~ i n a t ~ oofn the crystal structure of a given compound is not often u n d e r ~ a ~ einn conjunction with its phase-diagram investigation. Therefore, it is not always clear under what conditions a specific crystal polymorp~is stable. ~tandardizationof the
Computer Design of ~ ~ t e rwith i ~ AI ~ s
presentation of data for compound properties is a task for the future; meanwhile, in extracting examples for computer learning, we have to run a risk - which is not into account the dependence n the conditions of synthesis, it is possible to enter processing parameters for the production of the substance into the learning set, and further to predict a method of synthesis which will t properties to be achieved most effic~ently . Various program versions of this concept formation method (Gladun 1995, Gladun and Vashcheiiko 1995) were used in the calculations to be mentioned below.
new semiconducting, electro-optical, acousto-optical, and B indicate any d in the comp~ter memory as a set of especially (Gladun and Vashchenko, 1995) coded values of properties of the components A and B, whose class (formation or non-formation of a compound with composition ABXz in various systems) was chosen as a target feature. Searches for regularities and predictions were ut separately for the systems A-B-S, Te. Based on physical and chemical of the nature of substances of this kind, three chosen for descrip1. The distribution of electrons in the energy levels of the isolated atoms of the chemical elements A and and their formal valences in ABX, compounds. 2. The types of incoinplete electronic shells, the ~ a u l i n g electrone , the covalent radii ov, the formal valences these compounds, and the enthalpies of formation of appropriate simple chalcogenides. 3. The covalent radii by Bokii and Belov, the standard entropies and enthalpies o f formation of appro~riatesimple chalcogenides. gularity classifications and the predictions of the ation of ~ n ~ n o wcompounds n with composition were obtained separately for each of the three sets of properties of the constituent components. From such a dichotomy, the method of Gladun (1995) and
Gladun and Vashchenko by the sign (2) noncomposition AB& in a event, that the unknown the prediction is inde properties of the che periodically on their a sets of properties of the elements and of simple chalcogenides coincide. Failures (empty squares in tables of the predictions) arose from errors in the learning sets, unsucce~sfulcoding of the initial pro~erties of the components (Gladun, 199 chenko, 1995), or unsuccessful classification in the corresponding space of component properties. Failures can be explained by the n a t ~ rof~ the conc ‘chemical compound’ or may also be due to the metastability of compounds under nor To improve the r e ~ a b i ~ tof y future Shown in Table 1 are some of c o ~p o u n d swith the composition A
a search for new s e ~ c o n d ~ substan~s,new compounds
5*2
ds wi
The chalcogeiiides of this composition belong to a class of compounds that is
826
Miscellaneous Topics
Table 1 Part of a table illustrating the prediction of compounds with the composition ABX,
X
A
S
Li Na
Se
K
Cu Rb Ag Cs T1
O
O
Li Na
Te
K
Cu Rb Ag Cs T1 Li Na
+
Q
K Cu Rb Ag Cs T1
B B A1
sc Ti
Q
O
@
+
+
+
-
@
O @ + @ + @ + @ @ @ + @ + $ Q Q + @ + O + + + + @ + @ + + @
Q
Q
Q
@
+
(0
@
0
+
@ Q + + + + + + + + + @ @ Q @ 63 @ + @ + @ + + + + + + 3. + + + + + + @ FD @ @ @ @ CB + + Q
v
@
@
+ + + + + + + + + @ + + + + Q @ @ @ + @ + + + + + + 0 @ 0 @ 0 @ + + + + + + + + + + +
Cr Mn Fe C O + + + + + + + + + N i + Q + @ + + + + + o + + + Q + + + + + G a @ O O @ o @ o @ + + @ @ + @ o @ @ @ A s + @ o @ + @ + @ @ @ @ @ @ @ @ @ Y Q @ O @ + O + @ Q Q + @ + @ + Q R h + + + + + + + + + + + + + + + In O @ @ @ @ @ @ @ O O @ @ + @ + @ O Sb @ @ F D @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ * t - ) @ L a Q @ $ @ @ - @ @ 83 n + C e O @ @ @ @ @ O @ + @ +-++a + Pr @ @ @ @ @ + + N d @ @ @ @ @ + t t + @ + @ + w + a + t-)
@
+
+
+
@
+
@
+
t
-
)
+
Q
@
@
+
@
+ +
+ + @ + + + o +
@
@ tt
0
0
@ @
@ @
8)
@
+
+
@ O @ + @ @ 63
@
@ @
63
+
@
+
+ + + + + + a3 +
t-)
(E3
@
83
+ + + + @ @ + + @ + €9 63 + @ 63 D y @ @ @ @ e @ + @ o @ + @ + @ + o+ 0 @ €0 H o @ @ Q @ Q @ + @ Q @ + @ + @ + O + 0 @ @ Er @ @ @ @ O @ + @ O @ + @ + @ + O 4@ 8) 6% T m @ @ @ @ O @ + @ + + + @ + @ + Q + @ @ @ Y b @ @ @ @ O @ + @ + + + f B + @ + % + 0 + CD L U @ @ @ @ Q O + @ + + + @ + @ + O + 0 8) @ T1 + + Q @ @ 0 @+ + @ E l @ + + + @ $ + B i $ @ @ $ @ @ @ @ @ @ @ @ @ @ @ $ 8 ) @ @ @ @ @ @ A c + + + + + + + + + + + + + + + + + T h + + + + + + + + + + + + + + + + + + P i t + + + + + + + + + + + + + + + + + U + + O f + + + + + + + + + + + + Designations: +, predicted formation of a compound with composition ABX,; -, prediction of no formation of a compound
P
m
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+ +
S m @ @ @ @ @ @ + @ + @ + @ + + - + @ E u @ @ @ @ @ @ + @ + @ + @ + + + + ( E 3 a d @ @ $ @ @ @ + @ o @ + @ + o + @ Tb @ @ @ @ O @ + @ O @ + @ + % + O
I
with composition ABX,; @, compound ABX, is known to be formed and this fact is used in the computer learning process; w, compound ABX, is not known to be formed and this fact is used in the computer learmng process; 0, predicted formation of a
compound with composition ABX, which is confirmed by experiment; @, predicted formation of a compound with composition ABX, which is not confirmed by experiment; empty square, indeterminate result.
conducting, electro-optical, and other electronic materials. Each A-B-X system was input to computer memory as a set of coded values (Gladun and Vashchenko, 1995) of the properties of components A and B, whose class (a compound with composition AB,X4, formation or non-formation in the system) is indicated as the target feature. As in the previous case (section 5.1) the search for regularities and the prediction were carried out separately for systems A-B-S, A-B-Se and A-B-Te.
Based on physical and chemical information on the nature of compounds of this class, three sets of component properties were chosen to describe the chalcogenide systems as listed in section 5.1. The classifying regularities and predictions of formation of unknown compounds with composition AB,X4 were obtained separately for each of the three sets of component properties. In Table 2 are listed the predictions of compounds with composition AB2S4 (Kiselyova and Savitskii, 1979) and in Table 3 are
@
Part of a table il~ustratingthe prediction of compounds with the composition AB$, A Mg Ca
Ti
V
Cr Mn
Fe
CO Psi
Cu
Zn
Ge
Sr
Cd
Sn
Ba
Sin Eu
Yb
Wg
Pb
Ra
B
~ ‘ ~+ , predicted j g ~formation ~ ~ of a ~ compound ~ ~ with ~ composition : AB& -, prediction of no formation of a cornpound with composition AB$,; @, compound AB,S, is known to be formed and this fact i s used in the computer learning process; f-f, compound AB,S, is known not to be formed and this fact is used in the computer learning process; 0, predicted f o ~ a ~ i of o na compound with composition AB$, which is confirmed by experiment; @, predicted formation of a compound with composition AB$, which is not confirmed by experiment; empty square, indetermnate result. ~
some of the predictions o f c o ~ p o u n d swith composiet al., 1978). In the last two ,Te4 (~avi~skii decades 43 predictions of sulfide compounds and 39 predictions of telluride compounds were tested experimentally. Only five predictions of complicated sulfides (predictions of compounds with composition CdAs2S4, FeNd2S4, FeGdzS4, CuT12S4, and PbT12S4) and six predictions of complicated tellurides (prediction of coi~poundswith compositions CaR2Te4(R = La-Nd, Sm) and Yb,Te4) were in error. More recently it was decided to reteach the computer system using new
experimental data. More exact results are presented in the book by Savitslcii et al. (1990). ~redictionsunder normal conditions o f compounds with composition AB2X4 and the crystal structure types Th3P4,CaFe,04, NiCr,S4, or spinel are given also. Kiselyova (1995) reports on the search for new semiconducting and electro-optical substances; we predicted new compounds of this compositi~nwith the structures of chalcopyrite, spinel, olivine, PbGa2Se4,Vb3S4,Th3P4, Yb3Se4, CaFe204, or I?iCr2S4 (at room temperature and atm~sphericpressure).
828 able 3 Part of a table illustrating the prediction of compounds with the composition AB,Te4 A Mg Ca
Cr Mn Fe
CO Ni
Cu Zn
Ge
Sr
Cd
Ba
@
@
@
+
+
Sm Eu
La
Yb
B
AI sc Ti Cr
+
CB
@
63
-
-
-
-
0
0
Mn
+
a g ; t + + @ +
h
Q
+
+ + +
d m
+
@
+
O
+
@ a + @
a e
+ C
+
+
@
+
@ @
+ +
+ +
@
+ +
@ + + + + +
+
+
+ + + - t
+ + + -
H o Er T r Y b L u
+
+
+
E l l + + C d - t - + T b + + D y + o +
Sm
Bi
@
-
@
+
o
+
+
n + +
+ O o
~ + +
+ + +
+ o + + +
-
-
@
t
+ +
+ + + + + + + + + + + + + + + +
+ + + + +
+
+
+
@
@
+ + + + + + +
+ +
CO
G AS Y R In Sb L C PP N P
-
c23-t-
g ; t @
+ +
Fe
-
+
e f
-
t
@
@
+ + + +
c
-
2
3
+
+
+
g
>
+
Q
@
o
+ + + + + + + + + +
@ + - + @ @ a 3
@
+ + + + + o + + + + + + + + + 0 6 3
+
+
+ c + + + + + + + + + + + + +
@
J
+ ( c J + g J +
+ + + + + + + + + + + + + + + 0
+ - t + s + + - I + o - t - C + + + + + + + + + + + + + + + +
- o + + a + + + + o + B + a + o + a + a + Q + + + a @ + + o
63
e ~ z + ,~predicted ~ ~ formation € ~ of~ a compound ~ ~ : with compo~itionAB,Te,; - , prediction of no formation of a compound with composit~onAB,Te,; @, compound AB,Te, is known to be formed and this fact is used in the computer learning process; e-t, compound AB,Te, is known not to be formed and this fact is used in the computer learning process; 0, predicted formation of a compound with composition AB,Te, which i s confirmed by experiment; @, predicted f o ~ a t i o of n a compound with compa$~tion AB,Te, which is not confirmed by experiment; 0,predicted absence of a compound with composition AB,Te, which is confirmed by experiment; empty square, indeterminate result. ~
of elements were chosen for the description of the systerns:
ComPounds structure ThCr2Si2 are promising for new magnetic and suPerconduc~ing materials. -xwas represented in the computer ~ ' ~ as 0 a' set ~ of especially coded values (Clad'' shchenko, ~ 9 9 of 5 ~the properties of elements A whose class (a compound of composition ,X2 with c r y s ~ ~structure l type ThCr2Si, and for~atio' or noii-fo~ationin the system) is indicated as the tar he searches for classify in^ regularities s were carried out separately for systems 3-Ce, Two sets of properties
1. The distribution of electrons in the energy levels of isolated atoms of the chemical elements A and B. 2, The first three ionization potentials, the metal radii by ]Bokii and Belov, the standard entropies of individual substances, the melting points, the number of complete electronic shells, the number of electrons in incomplete s-, p-, d- or f shells for the atoms of elernents A and The classifyingregularities and predictions of formation of un~nown~ o m p o u n ~ofs co~positionAB, with the ~ h C r 2 ~crystal i2 structure were obtained separately for each of the two sets of component pro~ertie$.
829
Comp~terDesign of M ~ t e r i u ~with s A1 Part of a table illustrating the prediction of the crystal structure type ThCr,Si, for compounds with. the composition AB,%, B
Cr
Mn
Fe
CO Ni
Cu
Zn
Ru
Rh
Pd
Ag
0s
Ir
Pt
Au
A
~ +, formation ~ of a cornpound ~ with ~ the crystal ~ structure ~type ThCr,Si, ~ is predicted; -, formation of a compound with the crystal structure type ThCr,Si, 1s not predicted; @, a compound with the crystal structure type ThCr,Si, was synthesized and appropriate i n f o ~ a t i o nwas used in the computer learning process; t+, a compound with the crystal structure type ThCr,Si, does not exist under n o ~ conditions a ~ and this fact was used in the computer learning process; 0, predicted formation of a cornpound with the crystal structure type ThCr,Si, is confirmed by experiment; @, predicted formation of a compound with the crystal structure type ThCr,Si, i s not confirmed by experiment; 0 , predicted absence o f a compound with the crystal structure type ThCr,Si, which is not confirmed by experiment^ empty square, indeterminate result.
~
Shown in Table 4 are some of the predictions of compounds with composition AB2Si, and struc~ure iselyova and Savitskii, 1983), and in Table 5 are predictions of compounds with composiucture type ThCr,Si, (Savitskii and ecause of the great promise of this class of crystal phases, these compounds have recently been studied intensively, An experi~ental check showed that out of 79 predictions o f silicides checked, only six were wrong and of 37 predictions of germanides only five results did not fit our predictions.
~
5.4 ~ p e $ i c t i oof~ New
More than 10 years ago we predicted hu~dredsof new compounds of aluminum, ~ a l l i uand ~ indium with compositions: AB,X, ABX and A In) and with crystal structures Heusler alloys, TiNiSi (E ph and ThCr,Si, (Kiselyova and used the prototype of our i~fo~mati~n-predict in^ system for the first time.
0
830 able 5 Part of a table illustrating the prediction of the crystal structure type ThCr,Si, for compounds with the composition AB,Ge,
A
Cr
Mn
Fe
CO
Ni
Cu
Ru
Rh
Pd
Ag
Ir
OS
Au
Pt
B Li Ma K Ca
v
Rb Sr
Y
Nb Cs Ba La Ce Pr Nd Prn 5111
+ -
+
+
-
-
@
@
@
@
-
+ +
+
-
+
+ @ - - + @ -
-
@ @
-
+
+
+
+
+
@
@
@
+
@
+
@
-I-
+
@
0
0
+ @ @ @ @ @ a @ @ a + @ @ @ @ @ o a + + + + + + + + + + + +
@
@
@
@
O
!
Q
+
+
+
Lu Hf Ta T1 Pb Bi Fr Ra Ac Th. Pa U NP Pu A m
+
ern
@
+
+ + @
+
@
+
+ +
+
+
% @ @ @ + + + + + + + + +
+ @
@ +
@
+
@ +
@
@
+ c +
+ + +
Q
+
@
I -i-
+ + + -I+ + +
@
-
+
+ +
a
+
+ +
+
+
+
+ +
+
+
@
+
+
+
@
+
+
0
@
+
@
0
+ +
+
+
-
cr)
8
+ + @ + + + 8 + @ +
@
+
+
+
83
i
Eu + 0 @ Q 0 8 0 8 Cd + @ @ @ @ @ o @ + + Tb + @ @ + @ @ o o + + D Y + @ @ @ @ 4 3 0 @ + Ho + @ @ @ @ 0 8 0 + Er + @ @ @ @ @ G O + + Tm + + @ @ @ D c c 3 G + +
M
+
@ +
+ + + + @ @ + @ $ + + + + I + + + @ @ + @ @ + + + + + + + + + + + + + + + + + + + + + + + + +
+
+ 5 ++
+
+
+ + + + + + + + + + + + + @ @ + + + @ @ + + + + + + + + + + + + +
~ ~ s ~ g ~see ~ Table t ~ o 4.~ s :
Two sets of element properties that had allowed us to obtain good results in the solution of similar problems (see section 5.3) were chosen for description of the systems, re diction of the re~ularities of formation of phases with a definite composition and crystal structure type were obtained separately for each of the two sets of component properties. The use of these regularities has allowed us to obtain two tables of predictions of new compounds for each composition and each crystal structure type. Analysis
of these predictions was published by Burkhanov (1989). Table 6 shows some of the predictions of cornpounds with co~position 1 and crystal structure that were checked, all type TiNiSi. Of 16 predic agreed with the new experimental data. Table 7 contains past p r ~ d i ~ t i o of n s~ompoundswith composi$n and crystal structure type resembling the Heusler alloys. Out of 22 predictions that were checked only three were wrong.
83 I
Computer Design of Materials with AI Table 6 Part of a table illustrat~ngthe predictlon of the crystal structure type TiNiSi for compounds with the cornposition A ~~
~~
A
La
Ce
Pr
Nd
Pm
Srn
Eu
Gd
Tb
Dy
Er
If0
Tm
Lu
Yb
B
-
~ e s i ~ ~ ~+ ~formation i o ~ s :of a cornpound with the crystal structure type TiNiSi is predicted; - , formation of a compoui~dwith the crystal structure type TiNiSi is not predicted; @, a compound with the crystal structure type TiNiSi was synthesized and appropriate information was used in the computer learning process; 8, the predicted formation of a compound with the crystal structure type TiMiSi is confirmed by experiment. ~
Table 7 Part of a table illustrating the prediction of a crystal structure type resernbling the Heusler alloys for compound^ with the composition AB,In
One further successfulresult of the sug is a prediction of n structure of the type Two sets of proper results for the solution of similar problems of predicting crystal structure types of intermetallic compounds (see sections 5.3 and 5.41, were chosen for description of these systems. The regularities of formation of Heusler alloys with a definite composition were obtained separately for each of the two sets of properties of the components. Use of these ~egularitiesfor prediction has allowed us to obtain two tables of predictions of new compounds with crystal structure type resembling the Heusler alloys. The results of comparison of these predictions for each were published by Kiselyova (1987). pair of re~ularit~es Table 8 shows some of these results for predicting ABGo, compoun~sand Table 9 shows other results for Cu2 comp~unds,which have a crystal structure type resembling the Heusler alloys. Of the four checked predictio~sfor c o ~ p o u n d swith cobalt, all agreed with the new experimental data. Three predictions of Heusler compounds with copper, and three predictions n o n - f o ~ a t i o nof Heusler alloys in the systems U, coincided with the new experimental data. e results that have been shown thus far, in sections 5.1-5.5 do not exhaust the possibilities for pred~ctionof new i n t e ~ e t a l l i ccompou~dswith our approach. The results of prediction of the crystal structure type of new equiatomic ternary cornpounds with composition ABAl (the crystal structure type ZrNiAl was predicted), ABSi (the crystal structure ZrNiAl, PbFCl, or TiNiSi were predicted), e (the crystal structure types ZrNiAl or TiNiSi
Ru
A
Rh
Pd
0s
Ag
IIr
Pt
Au
@
@
B
Ti Sr Y Zr N Tc La Ce Pr N P S Eu G Tb D H Er T Y Lu Hf Ta Re
@
+
+
El30 +
b
+
+
-
k
+
@
+
+
+
+
@ +
+
+
+
@
+
+
+
@
+
+
+
+
+
+ + @ o + + + $ 0
+
+
d n i
+
Q
+ i n
+ + +
+ +
+
d
+
+ +
+
+ + +
+
+ +
+
+ O
+
Q
+
+
C
~
(
Q
+ +
+
Q
f
)
+
+
+
+
+ +
+ +
+ + +
+
+ + @
+
@
+
+ @
y
+
+
G
O
+
+
+
$
o
+
+
+
O
+
+
@
@
+ m b
+
+
-
+ + +
( + +
+
+
Q O +
@
O
~
O +
+ +
+
+ + +
+
+
+ + +
+
+
+
@
+ +
@ Q @
@
+
a
@ s ~+, gformation ~ ~ of~ a compound ~ ~ ~ with ~ a: crystal structure type resembling the Heusler alloys is predicted; -, formation of a compound with a crystal structure type resembling the Heusler alloys is not predict~d;@, a compound with a crystal structure type resembling the Heusler alloys was synthesized and appropriate information was used in the cornputer learning process; 0, predicted formation of a cornpound with a crystal structure type resembling the Weusler alloys which is c o n f i ~ e d by e x ~ e r i ~ e n t@,; predicted formation of a compound with a crystal structure type resembling the Heusler alloys which is not confirmed by experiment; empty square, i n d e t e ~ n a result. ~e ~
+
832
Miscellaneous Tupics
A A1
Si
Ga
Ce
In
Sn
Sb
Tl
Pb
@
+ +
gj
+
+
+ +
+
+
+
+ +
-
+
-
@
@
WaSPrediCted), AB$,, A structure types CaA1,Si2 were published in a paper b ~ u r ~ h a n o(1987). v The ~redictionsof hundreds of new compounds with various compositions and crystal structures in chalcoge~iidesystems are presented in the book by Savitskii -et al. (1990). Apart from intermetallic compounds, we have also predicted the formation of thousands of new compounds in the oxide and halogenide systems A Hal, and A-B-D-Hal. The res checking of the predictions are presented in Table 10. Comparison of the predictions with exp~rimentaldata, obtained recently, shows (Table 10) that the av reliability of the prediction of met inorganic compounds exceeds 80%. accuracy for a priori predi~tionsof new in~rganic compounds has not been attained by any other known theoretical method.
B Li
Be
+
+
Mg
sc lii v Cr Fe Ni Y Zr N
@
Q
@
+ + + +
@
+
+
@
b o u
N R Rh Pd Ag Lu Hf Ta A u
+
@ + +
+ + +
-
-
@
+
-
-
-
-
+
+
+
0
-
@
+
@
-
i
-
+
-
t
-
-
+
@
+
+
a
-
-
-
-
63 @ + +
@
+ +
_.
+
+
+
+
-
+ + + + + + + + + + + + @ + a + + @ + @
-
+
+
+
-
-
+
+
+
+
+
+
+
-
-
+
Designations: +-+, a compound with a crystal structure type rese~blingthe Heusler alloys does not exist under normal conditions and this fact was used in the computer learning process; see Table 7 for other symbols.
The search for an optimal technology for the production of a material having extreme values of target parameters is an integral part of the design of new substances. As already noted (section 5), it is possible to predict, not only the formation of compounds with a certain co~positionand to estimate their properti~s, but also to pre~ictthe best method of their production. Just such a sequence: the information system the predi~tingsystem --+ the design of m ~ l t i f a ~ t o r i ~ l
(the crystal structure types predicted), ABPd (the crystal structure ZrNiAl was predicted) are presented in a avitskii and Kiselyova (1985). The predictions of the crystal structure type at normal conditions complicated piiictides with compositions: ABP and As (the crystal stru~turestypes ZrNiAl, PbFC1, or
--+
Table 9 Part of a table iilustrati~gthe prediction of a crystal structure type res~mb~ing the Heusler al~oysfor compounds with the composition ABCu, A Li
Be
AI
K
Sc
V
Cr
Fe
CO
Mi
Ga
Ge
Y
Nb
MO Ru
Rh
Pd
Ag
In
B
Designutio~s:0, predicted absence of a compound with a crystal structure type resembling the Heusler alloys which is confirmed by experi~ent;see Tables 7 and 8 for other symbols.
833
~ o m ~ ~Design t e r of ~ u t e r i u l swith AI T a ~ 10 l ~ Co~parisonof predictions with new experimental data Gompounds/Systems
Characteristics to be predicted
Compound formation ABX (X = Se,Te) Compound formation ABX, (X = O,S,Se,Te) Compound formation ABX, (X=O,F,S, Cl,Se,Br,Te,I) Compound formation ABX, (X = O,F,Cl,Br,I) Compound formation A,BX2 (X = S,Se) Compound formation AB2X4(X = O,F,S, Cl,Se,Br,Te,I) ~ o ~ p o formation u ~ d A2BzX7 (X = O,S,Se) Systems w/ compounds A(Hal), - B(Ha1) Structure type AB,X, (X = O,S,Se,Te) Structure type ABX (X =al,Si,P,G.a,Ge,As,Pcl,In,Sb,Bi) Perovskite st~ucture ABO, Pyrochlore structure A,B@7 Structure type AB,X2 (X = Al,Si,P,Ce,As,Sb) MnCu,Al structure ABX, (X = Co,Ni,Cu,Pd) MnCu,Al structure AB,X (X = Al,Ga,In) A~(so4)y-B~(so~)~ and ~ ~ ~ 0 ~ ) ~ - 3 ( ~Compound 0 ~ ) yformation Compound formation ABDO,
Experimental tests as of January 2000
Error of prediction
100 337 420 393 24 76 1 97 108 38 1 78 186 74 200 28 24 130 28
44 10 11 5 9 14 26 10 7 35 13 15 8 14 13 4 4 Average = 14%
(%)
Why are statistical methods for de experiments - was used by Savitskii et al. (1982) in a experiments attractive for chemists simplifiedvariant for predicting superconducting Chevscientists? In the first place, the number of experiments re1phases of composition A,Mo,S, and for optimization for a search for the extremum of a desired property is of the technology of their synthesis. Any mathematical reduced sharply, owing to the s opt~mizingmethod can be used for the solution of the of all the independent variables. latter problem. As a rule we use statistical methods of the design of multifactorial experiments in our work. derives an analytical model that can be used for the First we predicted new compounds of this kind with a u t o ~ a t i o n of the process. Third, a qua~titative critical temperatures for the transition to the superknowledge of the influence o f the technological conducting state (T') greater than the boiling point of conditions on the parameters of optimization can be helium, using for computer learning the information acquired. Furthermore, a physical-c~emicalmodel of from the bibliographic database on Chevrel phases. the process can be developed on the base of this We predicted a new phase with composition A g , ~ o ~ ~ knowledge, , And, finally, the series of planned experiwith a T', above 4 . 2 K Also, optimal conditions of ments can be carried out by unso~hi~ticated stag. synt~esis of this hase have been sought usin We have considered a sequential procedure for the statistical inethods of experimental design. At that design of new compounds with selected properties, time there was but one production procedure: a single namely: a database-a predicting system based on sintering of the powder elements in an evacuated artificial intelligence methods ---$ an optimization of the quartz ampoule. '2 was the parameter of optimization. technology of synthesis of the predicted compounds. We supposed that the phase A g ~ ~ o had ~ S ,a SUE- The predicting of compounds, the forecasting of the ciently large ho~ogeneityrange for cation A, (silver) desired property, using i n f o ~ a t i o nfrom the database, as do the majority of Chevrel phases. The silver and the prediction of the best type for technology of content, the annealing t~mperatureand the annealin synthesis can be considered col1 time were chosen as independent variables. A very for the search for s ~ b s t a n c with e~ those three factors at two levels each The intrinsic properties of the the catalogue of plans for the design and of simple compounds are used to describe of ~xperime~ts. The gradient method of Box and multicomponent physical~chemical s y s t e ~ s , The Wilson was selected for reachin the highest T,. A experimental data for substances (which are similar maximum T', equal to 7.8K, was reached after two to the predicted ones) and for the technology steps consisting of five experiments. synthesis are analyzed using the computer. D
834
~ i s c e l ~ u n e oTopics ~s
experiments to search for optimal conditions of synthesis of substances, predicted in stage one, can be considered as the tactics of the search for new materials. The researcher uses the technological parameters during this stage. The first stage is a theoretical procedure, but the second one is an active experiment using a formal plan. fn the future these experiments may be carried out by a robot programmed for the implei~entationof the procedures of the design of multifactorial experiments. We considered an automation of the search for new substances on the base of new information technologies. Besides the intellectualization of the scientific work, this approach allows ~romotionof the search for new substances with specified properties.
What problems confront the computer design of metallic and other inorganic substances by artificial intelligence methods? The most important problem is the quality of ex~erimentaldata for computer learning. The trouble is that the proposed approach eventually assumes a search of phy$ical-chemical systems for learning sets and sets of predictions, which have similar features. If any physical-chemical system from the learning set has an erroneous character and if the set is small, then it is quite possible that it will yield erro~eous predictions. Our experience is that the number of erroneous predictions varies proportionally with the number of errors in the experimental data processed, and the reliability of the predict~ongrows with an increase of the initial volume of data. (However, reliability approaches a limit with ail n t the representaincrease in size and i ~ p r o v e ~ e in e learning set.) We use databases containing extensive volumes of qualitative information for overcomin~these di~culties.With this aim in mind, we have developed DBs containing data assessed by qualified experts. This allows both an increase in quality and in the volume of the learning sets. However, it should be noted that an infinity of k~owledgenever leads to 100% reliability of prediction. The use of our in~ormation-predictingsystem will allow the enlistment of a user-expert for an assessment of data for computer learning, Usually he can solve the problems of the computer design of substances via analogs which are well known to the expert.
One of the problems of any computer classi~cation in inorganic materials science is the search for those properties of the elements and simple compounds, that are the most i m ~ ~ r t a for n t separation of physicaichemical systems into certain classes. This procedure can hardly be completely formalized, but the system we use, CONFOR (Gladun, 1995; Gladun and Vashchenko, 1995), automatically rejects those properties for the classification process. that have no in~porta~ce The initial set of properties for compute~-aided analysis is prepared by the material scientists, and it is desirable that the artificial intelligence system extrapolates information from this representative set of initial features. We have achieved good predictions of the qualitative properties of physical~chemicalsystems: formation of ~ o m p o u n d ~their , cry§tal structure type, etc. However, the problem becomes even more complicated if it is necessary to predict some ~ ~ ~ n t i t u t i v ~ property (e.g., the melting point, homogeneity range, etc.). The hypothesis of class compactness, based on methods of computer learning, presupposes that the different classes are located compactly in the multidimensional feature space and that there are no intersections between these classes. But we found some sets of properties whose space occupancy contradicts this hypothesis. The ap~lication of cluster analysis to the exemplar learning set, in combination with the grouping of features according to a statistical correlat~on,allows us to decrease the intersections of classes, but only slightly, owing to the selection of the natural threshold values (for a certain learning set) of the predicted quant~tativeproperties. Note that these natural threshold values are less a consequence of the nature of the phases and more a consequence of the set of examples used for the computer learning method. These observations are based upon the examples of learning sets that we have thus far obtained and i~vestigated. Therefore, as a consequence of the above int~raction problem, the attempt to predict certain threshold values that are important for technological applications, e.g. boiling-point temperatures of helium and nitrogen for su~ercond~cting compounds, is justified only from a practical standpoint. The error of this prediction will be high, but it will be possible to predict (with high reliability) those objects which are widely spaced in the features space. A priori identi~cationof these objects by a researcher seems to be a problem. One possibility to solve this problem is to visualize a two-dimensional projection of points, which correspond to the objects of the learning set, in
~ o ~ i ~Design ~ t e of’ r Materials with AI combination with the cluster analysis of objects and grouping of features according to their statistical corre~atioiis,The algorithms for this system involve cluster analysis based on the method of potential functions (Arkad’ev and Bravernian, 1971; Lzerrnan et al., 1970) and the extreme grouping of parameters As stated above, the prediction accuracy of quantitative properties depends strongly on the volume and representativeness of the learning set. Our experience shows that the number of examples in the learning sets must be in the hundreds or cven in the thousands in order to have an acceptable estimation of the e property. The future of this approach, and AI methods, is connected with the development of info~iiation-predictingsystems. That is a very expensive and time-consuming procedure. However, such systems allow us to cut down the time and expense of a search for and the development of new materials with specified properties. This kind of simulation requires D s containing only ‘good’ iiiformation. Let’s imagine the laboratory of the future. A ~ a t e r i a l sscientist, who must solve the problem of searching for new materials with desired properties, makes a request to the cornputer to find the necessary substances, If the set of substances that the researcher receives from a database does not satisfy the request, he asks for a prediction of new substances having the property sought. He chooses the best prediction, from his point of view, and asks the computer to develop an optiinal plan for the synthesis of the substance to meet an extreme target property. Such a ‘virtual’ laboratory is a tool of automation for searching for new substances on the basis of the use of new information technologies. The proposed approach will allow us to speed up considerably the search for new substances with desired properties.
835
system, is indicative of the advent of a new type of inodeling of cognitive activity, namely knowledge engineering. Such modeling will play an important role in those fields of science and te~hnologywhere nmthernatical simulation and ~ ~ ~ ~ u ~ e rexperi- a i d e ~ mcntation have proven to be inadequate (for example, in physical metallurgy, c ~ e ~ i s t,rsyc i ~ ~ of c ematerials, and the like).
Partial ~ n a i i ~ isupport al from rant N.99-0~90040) is gratefully acknowled~ed.1: should like to thank my colleagues Prof. Victor P. Glad Neonila P).Vashchenko of the institute of of the National Academy of Sciences of Ukraine for their help and support.
~ r ~ ~ cintel~~ence ial - an artificial system, usual~ycon~tructed on the basis of computer technology, which simulates a human solution of complicated tasks. It is intended for perception, processing and storage of i i i f o ~ a t ~ o and n, also for forming solutions of problems in an expedient manner. class - a set of objects chosen ~ccordingto some property (properties). classification - a separation of objects according to some essential property (properties). c ~ a s s ~ c a ~scheme o n - a set of rules d e t e ~ i n i n ga certain classification. computer learning (a inethod of artificial intelligence) - a process of the modificatio~ of the parameters of a classifying system on the basis of the use of experi~ental data with the purpose of improving the quality of the classification. concept - a generalized model of some class of objects that provides for recognizing and generating models of specific elements of this class. (CONcept FORmation) - a set of software tools intended for the logical analysis of large volumes of In the process of a ~ t o ~ a t i scientific ng research ranging experimental data (Gladun, 1995; Gladun and Vasheheafrom the developmelit of databases to the building of ko, 1995) with the purpose of searching for regularities. systems of artificial intelligence, the historical process “data - scientific or technical n~easurei~~nts, values calcuof cognition is repeated: from collection and proceslated therefrom, observations, or facts that can be sing of the empirical source data to the generalization represented by numbers, tables, graphs, models, text, or of the experimental facts. The latter is used as a basis symbols and which are used as a basis for reasoning or for constructing scientific theories that reflect the basic fixther calculation. Note: ‘data’ is a plural form; ‘datum’ relations and the correlations between the processes is the singular. and phenomena studied. Development of an a r t i f i d feature - a property of a constituent component of the i n t ~ l l i ~ esystem, n ~ such as an infor~at~on-predicting physical-c~emicalsystem.
*i
on - a collection of data and facts, so selected, arranged and interrelated that they give relevance, coherence and utility within a defined sphere of interest
in
- a system intended for data ounds, prediction of inorganic compounds not yet synthesized, and the forecasting of their properties. This system employs a database of properties of inorganic compounds, a database of element properties, the system CONFOR, a knowledge base, a tional processor and a monitor (Figure 8). * acquainta~iceor awareness of factual informata together with u~iders~anding of their relationimplications for utilization. * as%- (I) a collection of ~nterrelatedinformation, facts, or state~ents (IEEE 610.12); (2) in artifical intelligence, a representatio~¶of information about human experience in a particular field of knowledge and data resulting from solution of problems that have been sly encountered (IS0 per ANSI X3 (modified)). - a multidimensional array of feature values and a vector of the desired property. Each row corresponds to some physical-che~c~ system already known, whose class is indicated by the row position of the column vector. ata - data about data. Consists of descriptors of data in a database to ematic iiiformation for users, application pro database management software. Netadata tnanipulated and searched. cal system which is described as a set of property (feature) values of the constituent elements. - a system (e.g. compound or solid solution) which is formed from chemical elements. on - an ~dentification(classification) of a new object belonging to a certain class in compliance with a fixed which can litative concept (e.g. a ~u1ti"element system with compound formation or non-formation of a crystal structure type, possibility of forming compounds ty
- an object or element property
- an object or element property which alue taken from some continuum (or ~uasi-continuum) set of numbers (e.g. melting point, genw, index of refraction, and so on). iction - ~1multidimensional array of feature values. Each row corresponds to some unk~iown physicalchemical system, whose class it is necessary to predict.
ote: Terns with an (199 1 b).
*
are from Westbrook and Grattidge
C&S
Andersson, J.-0.. Jansson, B., and Sundman, T ~ ~ ~ ~ O - C A a data L C bank : for e diag~dmcalculations, CODATA Bull.
Ansara, I. (1991). ~ ~ n k r a t i oetn application des bases de donnkes t h e ~ o c h i n ~ q L ~Entropie es, Arkad'ev, A. G. and Braverman, E. Learning and Object ~ l a s s ~ c a ~ Nauka. i o ~ . Moscow (Russ.), Bale, C. W. and Eriksson, G. (1990). ~ e t a l l u r ~ ~ ~ a l thermochei~icaldatabases, a review, Can. Met. Quart., 9, 105. Chelikowsky, J. R., Binggeli, N., and Keskar, N. R. (1993) First principles methods for structural trends in oxides: applications to crystalline silica, J. Alloys (IZ C o ~ ~ o ~ n ~ s 197, 137. Chen, M.Y. (1988). Chemi ecognitiaii research in China, Anal. Chini. Act Chen, N. U, Lu, W, C., Chen, R., and Qin, P. (1999). Software package 'Materials Designer' and its ~pplication in materials research, Proc. Second h t . ~ r o c e s s ~ ndtg~ u n u ~ a c t o~f r~ ~u nt e~r ~ ~ l s , 1999. Honolulu, Hawaii. p. 1417. Cohen, M. L. (1986). new solids and superconductors, Science, Darken, L, S., Gurry, Physical C h e ~ ~ iof~ ~ et ~t ay l~~c ~G,r a w - ~ i lNew l , York. Degtyaryov, Yu. I., Podbel'skii, V. V., Kiselyova, N. N., el al. (1999). Database on the properties of acousto-optical, electro-optical, and n o n l i n ~ ~ r ~ o p tproper~ies ica~ of crystals with Tnternet access, Izvestiya VUZov. Materiuly elekt~on.t e k h n i ~3, ~ ~35, (Russ.). Devingtal, Yu. V. (1968). About optimal coding of objects and their classification by pattern recognition methods, Izvestiya Akad. Nauk SSSR, Tekhn. kibesnetika, 1, 162 (Russ.). Devingtal, Yu.V. (1971). Coding of objects with the application of a dividing hyperplane in their classification, Izvestiya Akad. Nauk SSSR, Tekhn. k ~ b e ~ n e t i ~3,a , 1 (Russ.). Dorofeyuk, A. A. (1971). ~lgorithimsof pattern recognition without a teacher, based on the potential function method, ~ v t ~ ~ a tii~ke al e ~ e k l ~ a n i1k a , Drago, V. J. and K a u f ~ a nJ., C.(1993). T ~ h n i c a features l of the cheniical and materials property data network services on STN International, J. Chem. Inj: and Comput. Sci., 33, 46. Enguchi, K. and S~imura,K. (1990). Fa~tualdatabases for materials design and ~ ~ n u f a c t L ~ ~ ~luat., n g ~30, 409, ISIJ Girgis, K. (1983). Structure of i n t e ~ e t ~ ~ lcompounds lic in ~ ~ y s i c Metc~ll~rgy,Part al I, 3rd Edition (eds R. W. Cahn, and P. ~ a a s e n )North"~o~land, . Amsterdam, p. 219. Gladun, V. P. (1995). Processes of F o ~ r ~ a tof~ uNew ~ nowl ledge. SD 'Pedagog 6'.Sofia. (Russ.). Gladun, V. P, and Vashchenko, N.D. (1995). Local statistical methods of knowledge formation, Cyhernetics and Systems Ana~?s~s, 31, 207. Golikova, M. S., Burkhanov, G. S., Kiselyova, N. N., et al. (1989). Databank on acousto-optical properties of inorganic Coinpounds, l z ~ e u r g aic~eslzie ~i Mu terialy,
837
~ o r n p Design ~ t ~ ~of Muterials with A I
systems AO-B,O,, ~ o k l ~ Akad. d i ~ a z &SSSR, k 30 Gulyev, B. B., and Pavlenko, L, F. (1973). Simulation of the (Russ.). search for components of alloys, A v t o ~ a t i k ai TelevlzeKiselyova, N. N.,Pokrovskii, B. I., Kornissarova, L.N., and k ~ i ~ n i k1, a , 131 (Russ.). Vashchenko, N. D. (1977). Application of a cybernetic Ho, C. Y., and LI, €3. H. (1993). Numerical databases on concept formation system to predicting the f o ~ a t i o nof materials property data at CINDAS/Purdue University, J. ~uIti-compo~entchemical compounds, Zh. Neoqars. Cheipz. In$ and ~ i ~ ~ Sci., ~ p 33, ~ ~36.t . ~ h i ~ i22, i , 883 (Russ.). thery, W., and Raynor, G. V. (1962). The S t r u c ~ ~ r e Kiselyova, N. N., Vashchenko, N. D., Gladun, V. P., et al. tals a& Alloys. The Institute of Metals, London, (1998b). Prediction of inorganic compounds grornlslng for the search for new electro-optical materials, ~ ~ r s ~ e ~ ~ raverinan, E. M., and Rozonoer, L. 1. ~ a ~ e r z a l3,y ,28 (Russ.). (1970). ~ e ~ h oof d ~otential ~ u ~ c t ~ oinn sC a ~ p ~ t e r K~selyova,N. N., and Savitskii, E. M. (1977). Prediction of ear^^^^ Theory, Nauka, Moscow (Russ.). inorganic compounds with the composition A Jackson, A. G., Pawlak, Z., and LeClair, S. R. (1998). Rough computer learning, D o k ~ ~Akad. di SSSR, sets applied to the d of material knowledge. J. (Russ.). Alloys Ce ~ o ~ p o ~ n d s ? Kiselyova, N. N., and Savitskii, E. NI. (1979). Prediction of Jurs, P. C., and Isenhour 975). C h e n ~ i c ~Appli~cition~~ l chalco-spinels with the composition AB,X, by usmg a of ~ a ~ t Recognition, e r ~ John Wiley & Sons, New York. coinputer learni~gmethod, Zh. ~eurgara.~ h z ~ ~ ~ , Kiselyov, N. N. (1991). System for relational database man(Russ .). 10, ~ p r a v ~ y a ~ u s h Sistemy c~ie i Kiselyova, N. N., and Savitskii, E, M. (1981). Cybernetic of Chevrel phases with selenium, 2%. ~ e o r g a ~ . diction of ~e~isler-phases with , 1364 (Russ.). on ABD, (D =CO, Ni, Cu, Pd), h e s t i y a Afzad. N., and Savitskii, E. M. (1982), Prediction of Kis Nuuk SSSR, ~ e t a l l i2, , 213 (Russ.). c o ~ p o u ~ i dwith s the ca~positionABHal, in systems of Kiselyova, N. N. (1 993a). ~nformatiQn-predictingsystem for halides, 2%. N ~ o r ~ ~hi^ n. the design of new ma~eria~s, J. Aliays 159. Kiselyova, N. N. and Savitskii, Lion of the Kiselyova, N. N. (1993b). Prediction formation of new silicides -structure, pounds: expe~encesand perspectives, I~vestiyaAkad. ~ ~ SSSR, ~ k~ eko r g a ~ ~ c ~~ae~~e r~i za el y , Kiselyova, N. N. (1995). The Design @In 19, 489 (Russ.). S e ~ r ~ fho r~ New n ~ Elecfro-optical, ~ e r r a - e ~ e c ~Superr~c~ Kiselyova, N. M,and Savitskii, E. M. (1984). Prediction of conduct~~g, kmd S e ~ ~ c o ~ d ~u ~c ~ e~ rni (~a~ insa repart). l ms AHa1,-BHal, Zh. ~ e o r ~ a ~ . EOARD S~C-94-4097. Kiselyova, N. N. (1997). Application of artificial intelligence Kravchenko, N. V., Bui-khanov, G. S., Kiselyova, N. N., et al. compounds design, Perspektivnyc (1991). Databank on the properties of crystals for the ~kkSSSR, Neorgakhanov, 6.S. (1987). Prediction of crystal phases in ternary systems with elements of 5th Kutolin, S. A., and Kotyukov, V. I. (1978). Chenlical allFnity group using computer ethods, Izvestiya Akad. function and computer prediction of binary ~ompositions Nauk SSSR?N e o r g a ~ ~ c erialy, 32,2006 (Russ.). and properties of rare earth c o ~ p o u n dZh. ~ , Ph-vs. C h e ~ . , Kiselyova, N. PIT., and Burkhanov, 6. S. (1989). The search 52, 918 (Russ.), for new ternary phases with Al, Ca and In using an Kutolin, S. A., and Kotyukov, V. I. (1979a). Prediction of the , Izve~~tiyaA ~ a d . Nauk ~ ro p e rt ~ of e s binary compounds of rare earth e l e ~ e ~on ts the base of their simplified electron confi~irations, Eselyova, N. N., Gladun, V. P., and Vashchenko, N. D. Iz~estiyaAkad. Nauk SSSR, ~ ~ o r ~ a ~ ~ c ~ ~ ue tselr ~i ~i le) ~ , (1998a). Computational materials design usin lntelligence methods, J. ~ l l ~ arad y sC a ~ ~ a ~ n d s Kutolin, S. A., and ~ o t ~ k o V. v , 1. ( ~ 9 7 ~ b Computer ). ~ i s e ~ y o vN. a , N., and Kravchenko, N. V. (1992) D prediction of the compositions of compounds in ternary ternary inorganic conipounds’ properties as a basis for systems and their properties as a function of electron n of new substances, 2%. ~ ~ a r g a~~ z i. ~ ~ ~configu~~tions ~ i , of their components. ~ z v e s t ~ yAkad. a ~ u ~ k SSSR, N e o r g a ~ i c h e s ~~~ e~ ~ ~ 15, ~ I389 i ~(Russ.). ~ l y , avchenko, N. V., and Petu Kutolin, S. A., Vashukov, I. A., and Kotyukov, V. I. (1978). the properties of ternary Prediction of binaiy c o ~ p o u of ~ ~rare s earth e ~ e ~ e n ~ s C version), ~ ~ u r g Matea~~c and their properties using a c o ~ p u t ~ r , ~ a SSSR, ~ k~ e o r ~ a n ~ c~~ f e fs ~~ ie ~~ ~ a ~ ~ } , Kiselyova, N. N., Lratsyk, V. I., Voi-ob’yova, V. P., et al. Laves, F. (1956). Crystal structure and atomic size; in Theory (1989). ~yberneticprediction of new compounds in the of Alloy ~hases.ASM. Cleve~and.p. 124. ~~~~
Manzanov, Ye. E., Lutsyk, V I., and Mokhosoev, M. V (1 987). Influence of the features of system selection on the prediction of compound formation in the systems ~ 2 M ~ 0 4 - B 2 ( Mand 0 0 ~A~MoO~"CMoO~, )~ ~ o k l a dAkad. i 1 relation between superconductivity and the number of valence electrons per
Pao, Y. H., Dman, B. F., Zhao, Y. L., and LeClair, S. R. (1999). Analysis and visualization of category membership distribution from multivariant data, Proc. Second Int. Conf. Intelligent Proces,s~ng& Man~facturin~ of Materi~ls, vol. 2,July 10-15, 1999. Honolulu, Hawaii. p. 1361. Pettifor, D. G. (1983). Electron theory of metals in Ph~vszcal ~ e ~ ~ l l u rPart g y , I, 3rd Edition (eds R. W. Cahn, and P. Haasen). North-Holland, Amsterdam, p. 73 . Popov, E. Q. (1987). Expert System. Nauka. Moscow (Russ.). Pospelov, 6.S, (1988). A r t ~ c i a Intelligence l as a Base of New Injbrmation Technology. Nauka. Moscow (Russ,). Pospelov, 6. S., and Pospelov, D. A. (1985). ArtiJicial ~ ~ t ~ ~ l i-~Applied e n c e Systems. Znanie. Moscow (Russ.). Rabe, K. M., Phillips, J. C., Villars, P., and Brown, T. D. (1992). Global multinary structural chemistry of stable quasicrystals, high-?;, s, and high-T, superconductors, PhyLT.Rev Savitskii, E. M., Deving and Gribulya, V. B. (1948). About recognition of binary phase diagrams of ~ e t a l ~systems ic using the computer, Dokladi Akad. Nauk Savitskii, E. M., Gladun, V. P., and Kiselyova, N.N. (1977). Prediction of compounds with composition A2B20, and the pyrochlore-structure, ~ o k l aAkad. ~ i Naulc SSSR, 233, 657 (Russ.). Savitskii, E. M,, and Gribulya, V. B. (1985). A ~ p l i c a ~ ~ofo n ~omputer ~ e ~ h n i ~ uine sthe ~ r e d i ~ t i o onf Inorgun~c Com~ounds,Oxonian Press, New Delhi. Savitskii, E. M., Gribulya, V. B., and Kiselyova, N. N.(1979). Cybernetic n of superconducting compounds, C A L P ~ A ~ Savitskii, E. M. ,V. B., and Kiselyova, N. N. (1980). On the application of cybernetic prediction systems in the search for new magnetic materials, J. L e s s - C o ~ ~ ~Met., on Savitskii, E. M., Gribulya, V. B., and Kiselyova, N.N. (1981). Forecasti erconducting compounds, Phys. Slat. Sol. (a)., ibulya, V. B., and Kiselyova, N.N. Savitskii, E. (1 982a). Cybernetic prediction of inorganic compounds and its correlation with experiment, Crystal Res. dz ~ e c ~ n o17, ~ .3., Savitskii, E. M., Gribulya, V. B., Kiselyova, N. N., et al. (1990). Prediction in ~ a t e r i a l s Scieme Using tlze C o ~ p u ~ eNauka, r. Moscow, (Russ.). Savitskii, E. M., and Kiselyova, N. N.(1978). Prediction of 1 phases, DoklucEi Akad. Nauk
Savitskii, E. M., and Kiselyova, N. N. (1979). Cybernetic prediction of the foriulation of phases with the composition ABX,, Izvestiya Alirad. Nauk SSSR, ~eorganicl~eskie
, N. N. (1983). Search criteria for chemical compound formation in salt systems using cybernetic methods, D o k l a ~ Akad. i Nauk SSSR, 27 (Russ.). Savitskii, E. M., and Kiselyova, N.N. (1984). Cybernetic prediction of new AB2Ce2 gemianides with crystal structure type ThCr2Si2, Izvestiya Akad. Nauk SSSR, ~ e t a l l i1, , 191 (Russ.). Savitskii, E. M., and Kiselyova, N. N. (1985). Prediction of the crystal structure type of some equiatomic ternary compounds, Izvestiya Akad. Nauk SSSR, ~ e t a l l i , (Russ.). Savitskii, E. NI., Kiselyova, N. N., Pishik, B. N., et al. (1984). Development of an automatized data bank on the properties of ternary inorganic compounds, ~ o k l aAkad. ~i auk SSSR, 279, 627 (Russ.). Savitskii, E. M., Kiselyova, N, N., and Shkatova, T. M. (1982h). Cybernetic-statistica~approach to the search for substances with predefined properties, Izvestiyn Akad. Nauk SSSR, ~ e o r g a ~ i i c ~ e ~Mutericily, skie 18, 235 (Russ.). Savitskii, E. M., Kiselyova, N. N., and Vashchenko, N.D. (1978). Application o f computer learniiig to the search for new compounds with composition AB,Te4, Dokladi Akad. Nauk SSSR, 239, 1154 (Russ.). Shah, M., and Pettifor, D. 6. (1993). Structural stability, local topology and electron couiit in small s-valent clusters, J . Allo-ys d Compounds, 197, 145. STN (1993). 10 years of STN Int~rna~~onal. abases in Science 62 Technology, FIZ Karlsruhe. STN Service Center Europe. Talanov, V. M., and Frolova, L. A. (1981). Investigation of chalco-spine1 formation using the method of potential functions, Izvestiya VUZ. ~ h i m i y a i ~ h i ~ i c ~ e ~ ~ k a y a Villars, P. (1995). Factors converning crystal structures. In ~nter~etallic C o ~ ~ o uPnr ~~ n. ~ ~and l e sPractice, Vol. 1. (eds J. H. Westbrook, and R. L. Fleischer). John VViley and Sons, Chchester, p. 227. Villars, P., Brandenburg, K., Berndt, M. et al. (2001). Binary, ternary, and q u a t e r ~ ~ rcyo ~ p o u n dfo~mer/non~former prediction ma Mendeleev number. J. Alloy,r & Co?npounds, 317131$, 26. Vozdvizhenskii, Q. M. (1975). Predic~ion of Binary Phase Diagrams. Metallurgiya. Moscow (Russ.) Vozdvizhenskii, V. M., and Falevich, V. Ya. (1973). Application of a computer pattern-~ecognition method to the identification of phase diagram type of binary metal systems, in Basic Regularities in ,he Constitution of Binmy Metallic Systems, Phase ~ i a g r a ~ sMoscow. . Nauka. p. 119 (Russ.). Westbrook, J. H., and Grattidge, W. (1991a), The role of metadata in the design and operation of a inaterids database, in Com~uterizatio~ and N e t ~ o ~ koi j~~ ga t e r i a l ~ s
Computer Design qf ~ ~ t e r ~with a l sA I D a t a ~ u s eVol. ~ ~ , 2, ASTM STP 1106 (eds J. G. Kaufman, and J. S . Glazman). American Society for Testing and Materials, Philadelphia, PA (1991) 84. Westbrook, J. M., and Grattidge, W. (1991b). A glossary of terms relating to data, data capture, data manipulation, and databases, CODATA Bull., 23, Nos. 1 and 2, p. 196. White, H. J. (1985). The~odynamicdatabases in the US national standard reference data system, CODATA Bull., 58, 45. Yan, L. M., Zhaii, Q. B,, Qin, P., and Chen, N. Y. (1994). Study of the properties of intermetallic compounds of the rare earth metals by artificial neural networks, J. Rare Earths, 12, 102. Yudina, N. V., Petukhov, V. V., C ~ ~ e r e i n u s h E. ~ nA., , et al. (1996). Data bank on the acousto-opti~al,electro-optical, and noiilinear optical properties of materials, Crystal-
839
Zagoruiko, N. G. (1999). A p p l i e ~~ e t h o ~of s Data and Knowledge Analysis. Institute of Mathematics Publishing (Russ.). Zemskov, V. S., Kiselyova, N. N., Kiselyov, N. N., et al. (1995). DIAGRAMMA a database on the phase d i a ~ ~ d m s of semiconductmg systems, Inorgumc. Muterzals, 31, 1096. Zemskov, V. S., Kiselyova, N. N., Petukhov, V. V., et al. (1998a). Database on phase diagrams of se~iconducting
(Russ.).
Zhou, B., Jin S. M., Shao J., and Chen, N. Y. (1989). IMEC - an expert system for retrieval and prediction of binary intermetallic compounds, Acta ~ ~ t a ~ l uSin., rg. 428.
This Page Intentionally Left Blank
Traditional alloy development has largely followed a ‘mix and measure’ approach, improved only modestly in the second half of the twentieth century by what has been called ‘enlightened empiricism’. In the latter case, the improved ~~nderstandin~ of materials behavior, derived from reductionist analysis, has enabled alloy developers to better know hat to mix, what to measure, and Izow processing might be used to control structure and hence propertics, but has not basically altered the nature of the approach. Limits to the traditional approach are now seen to be imposed a) by the need to work with multinary alloys with four to 10 or more components, not simple binaries or ternaries; and b) by the need to achieve new generations of materials for a given application on a 1% to 2 year cycle, rather than 1%to 2 decades (Olson, 1989, 1997, 2000). In contrast to past approaches, modern alloy synthesis, including IMC-based materials (see also iselyova’s chapter in this volume) follows several separate but co~plementarytracks: 1. Modeling of structure over a dimensio~alran of 1 nm to 1Opm, facilitating definition optimal structure at many different levels and i n c o r ~ o r a t i na~probabilistic approach wherein a distribution of structural p a r a ~ e t e r scan be mapped into a property distri~ution. 2. A combinatorial approach - both by applying micro-scale analytical and structural probes as well as nano~indent~tion assessment of mechans in multi-component diffusion ao, 2001); and by multi-variate ~haracteri~ationof the syn~hesis strategy
(Knowledge Foundation Conferences, 1999, 2000, 2001). 3. Particularly for I ~ C - b a s e dmaterials, considerable effort has recently been applied to prediction of stability of particular crystal structures using a so-called ‘ m ~ p ap p ~ ~ ~ ~ binaries (Pettifor, 1992; Warada et al., 1997) and for ternaries (Chen et al., 1999a,b,c,d). This approach enables prediction of most ~rystal structures or, for a given crystal structure, displays the range of compositions that may
~ i o d o w n i k ,1988), a kind o f ‘ t h e ~ i ~ o d y n a i ~ i c modeling’. This offers the poss~blityof extrapolating incomplete binary data to a full system, o f extending diagrams to higher order systems, or (with appropriate software) of obtaining twodimensional sections of multi-component systems (see chapter by de Fontaine in this volume). 4.. It is now possible, based on thermodyna~icsand alloy fundamentals, to prescribe novel p~ocessing paths to produce an optimal structur al., 1998; Allibert and Pastor, 1998; 1997). 5 . Computational quantum mechanics can be applied to the compositionai and structural dependence of energy, e tions - grain boundaries, faces (e.g. Lill et al., realization of property 19’72; Cohen, 1986; Freeman, 1992; Freeman et al., 1992; Miller 1998; Saito, 1999). The balance of this chapter reproduces (with minor alterations) a previously published article ( ~ u of ~
~ n ~ e r ~ eC~ a~ l ~~ i ~~Vol. o 3,~Principles n ~ :and Puactice. Edited by J. H. Westbrook and R. L. Fleischer 0 2 0 0 2 John Wiley & Sons, Ltd.
~
~
(6), 1997, 635-649) by the authors instance of the application of modern alloy design to IMC-based systems. The current approach for the development of intermetallic materials is essentially based on identifying binary intermetallic compounds, such as Ni and Ti alu~inidesas well as certain silicides, exhibiting some promising basic properties, relatively high melting nd inherent oxidation resistance. ever, the number of potentially interesting coinpounds for structural applications is limited. Each of these coinpounds has a given crystal sym~netry,and if it is complex, the compound is hardly d e f o ~ a b l e ,and one does not expect much macroscopic ductility even at fairly high temperatures. In other words, the compound has a ceramic-like behavior. On the other hand, if the crystal structure is simple or less complex, there are significant ductilization and making the material merous attempts have therefore been made in this direction, both through grain-size refinement or grain-boundary suppression and by micro- or macroalloying. In spate of some successful or even o c~~s io iia~ly spectacular examples result in^ from such efforts, the materials science coniinuiiity has yet to provide a new material that possesses a’ti acceptable balance of mechanical and other essential properties. Thus, if studies are restricted only to a binary compound base, there is an obvious danger of the existing possibilities. In order to uge potential offered by numerous coiiibinatioiis of metallic elements in intermetallics, the authors have been interested over the past few years in exploring ternary, quaternary, or more complex alloy
only may such alloying stabilize the desired simple crystal structure, but other benefits inay derive through lower diKusion ratcs and better solid-solutioii stren~thenin~ at high temperatures (Westbrook, 1996). Thus our aim is to identify new intermetallic phases (absent in binary couples) having a simplc crystal s t r u ~ ~ u and r e then to ~valuatetheir field of existence both in composition and in temperature. It should be emphasized that such efforts cannot be successful without support from the most recent progress in alloy theory, for e the so-called crystal structure maps propose ttifor (1992). The aim of this chapter is to show how materials scientists and engine~rsmay take advantage of a knowledge of modern alloy theory for the development of real structural materials and to indicate the types of theoretical information that can be useful.
Because the lack of both ductility and tougliness often represents the major drawback in intermetallics, the successful development of these new materials essentially depends on the improvei~entof these properties. However, it is also important to increase the mechanical strength of iiiany intermetallic compounds, a considerati~n often o v e r l ~ o ~ eby d many research workers. For example, stoichiometric or Ni-rich NiAl has a yield strength o f 100 to 200MPa, which is very low from the engineering viewpoint, although this strength (corresponding roughly to lOW3pwhere p is the shear modulus) is fairly high when viewed by plasticity theory. ~ a r d e n i nof~ intrinsically strong crystals that have a high lattice-fricti~nstress is not very easy. There is, however, ample experimental evidence available today to show that the mechanical strength of multiphase materials is higher than that of single-phase internietallics (Khan et al., 1990b). The authors have attempted for several years to generate two-phase microstructures in various alloy systems, ~ i t y the assuming that a good phase c o ~ ~ p a t i b ~betw~en second phase and the matrix phase is a major factor for obtaining useful high-t~~perature, high-strength materials. This chapter describes the approach, the experimental results, and the degree of success accomplished during this endeavor.
~ m o nmultipha$e ~ alloys with high m e c ~ a n i ~ a l strength, the most prominent example is the so-called Ni-base y-y’ superalloys (Westbrook, 1996). Some of the Ni-base y-y’ superalloys now being used in modern aeroengines have been pushed to an operating temperature of about 0.8 T, (T, is the melting point). These alloys contain a very high volume fraction (up to 70%) of y’ phase in the y-matrix; the y (LI,) phase is crystai~ograpli~cally a superstructure of the y (AI) phase, These two phases not only have a cube-cube orientation relationship but also a very small lattice misfit. In other words, the com~atibilityof the two lattices is very high. This compatibility, combined with the high volume fraction of y’ phase, is certainly the key to the outstanding performance (creep strength and ductility of these alloys. In these alloys, the y‘ phase is hard enough to resist dislocation cutting, the yly’ interfaces constitute barriers to dislocation propagation, and thus dislocations are restricted to the relatively confined space that is the y
843
Alloy Design
ch a situation, resulting in high work is responsible for the high creep performance. On the other hand, the y’ phase can be sheared; dislocations may propagate through this phase, if any stress concentration occurs at y/y‘ interfaces. This is probably an important reason for the good ductility of these alloys. In spite of considerable research devoted to Ni,Al (y’), ever since the discovery of a ry strong ductilizing effect of trace-level additions of by Aoki and Izumi (1979) and followed by extensive investigations undertaken, in particular, by Liu et al. (1985), the relatively poor high-temperatu~emechanical strength of singlephase y’ alloys seems to seriously restrict their field of application. It is worth recalling here recent experiments (Khan et al., 1990a) that illustrate the advantages of Ni-base y-y’ superalloys over singlephase y’ alloys. They compared the creep behavior of an advanced y-y’ superalloy, designated ‘GMSX-2’ to that of a single-phase y’ alloy whose composition corresponds to that of the constituent y’ phase of this superalloy. Both these materials were directionally solidified as [0011-oriented single crystals and then tested in creep at various temperatures. The creep curves (Figure 1) showed that the two-pliase y-y‘ superalloy was far stronger than the single-phase y’ alloy at 760 “C (creep life: about 1150h at a stress of 750MPa for the former and less than l00h at a stress of 650MPa for the latter), in spite of the fact that the tensile strengths of the two materials determined at the same temperature were almost identical. These results practically rule out the possibility of rcplacing highstrength y-y’ superalloys by y’-based alloys in many high-temperature/high-stress applications.
Creep strain (Yo) 30
Assuming that a two-phase microstructure with good lattice compatibility and a high volume fraction of the second phase are the keys to obtaining useful hightemperature materials, the authors attempted to create a y-y’ type microstructure in several alloy systems. Siniilnr efforts have been made on Pt-group metal systems by Yamabe-~itaraiet al. (1997) and others; see WolfPs chapter in this volume.
2.2.1 ~ e - ~ i ~ A l ~ ~ The first case deals with the pseudobinary Fe-Ni,AlTi system and its derivatives. The initial idea was to incorporate the NiAl phase (B2) in a bcc (A2) matrix. Among the ternary systems X-Ni-A1 (X=bcc transition metals such as Fe, Cr, V, MO, VV, Nb, and Ta), three systems Fe-Ni-Al, Gr-Ni-Al, and V-Ni-A1 possess a two-phase (A2 + B2) field in a certain co~~positional range. It is iiiterestiiig here to compare the lattice parameter (Table 1) of each of the above metals with that of NiAl (aNiA,=2.886x 10-lom). have a paranieter close to that of NiAl wliile V lias a slightly larger parameter ((av-aNtAl)/av rn 0.04). All the others (MO, W, Nb, and Ta) have a much larger parameter, especially for Nb and Ta (e.g. (aNb-oNIN)/ (aNbrn0.13). It is important to note the difference between the pseudobinary Fe-NiAl phase diagram on the one hand and Cr-NiAl and V-NiA1 diagrams on the other hand (Figure 2). For the Fe-NiAl system, two single-phase solid-solution fields, either disordered (A2) or ordered (B2), occupy the whole composition range at high temperatures, and these solid solutions decompose into two phases (A2+B2) at low temperatures; a schematic phase diagram is shown in Figure 2(a). The other two systems show a eutectic reaction, like that depicted in Figure 2(b). In general, a system having a phase diagram of the type shown in Figure 2(a) offers great flcxibility for controlling microstruc-
20
Table 1 Lattice parameters of some bcc transition metals Element
Lattice parameter (a), x 10-lOm
10
Ta Nb 0
w
200
400
600
600
1000
1200
Time (h)
Figure 1 Creep curves at “C of [OOl] CMSX-2 and y CMSX2 single crystals
MO
V Cr Fe
3.30 3.30 3.16 3.15 3.03 2.89 2.87
A
Schematic of two types of phase diagrams
tural parameters such as size and distribution of the second phase and its voluine fraction through a suitable choice of composition and heat treatments, ban, 1991) showed that the microstructure observed in alloys of the Fe-NiAl system after solidification was indeed characterized by a very fine two-phase morpho~ogyprobably due to a sp~nodal-li~e decomposition, but that this microstructure is very stable daring lieat treatment. These alloys were therefore always very hard and brittle. ~ h e i ipart of the A1 is replaced by Ti (the composition of the alloys studied lies on the Fe~ i 2 A l T tie-line), i the ~icrostructureo~servedafter unid~rectiona~ solidification has a morphology very similar to that of nickel-base y-11 ' alloys, characterized by a very regular distribution of cuboidal particles (size x 0.2 pm) of L2, phase in the bcc matrix with a volume fraction of about 50?h (Figure 3). The L2, phase is a s~iperstructureof the bcc (A2) lattice, and these two phases have a cube-cube oriei~tation relationship, but their lattice coherency is not very high (mismatcli GZ 1%). Interestingly, these alloys show a recrystallized grain structure with a grain size of about 100pm after u~id~rectional solidification. The recrystal~ization is presumably induced by phase decomposition (solid solution-aA2 + 152,). ~ l t h o u g h these alloys exhibit at room te~peraturea tensile elongation% 1 to 2% with a yield stress of 1000MPA Pa at 600 "C,their strength shows a sharp decrease above this temperature, probably because of their poly~rystallinecharacter. 2,2.d
~
~
Alloys
"
~
~
S
The second example is Nb-based alloys where an atternnt was made to create a Y-v'-like microstructure.
~ i o b i ualloys ~ are c ~ n ~ i d ~for t e s~igh-temp~rature applications, because they are refractory and have reasonably low density (Tm=2468"C and d=8.G g/cm3 owever, they have poor oxidation resistance and their strength is low at low and intermediate temperatures (25 to 900 "C). The purpose of this ~ ~ v ~ s t i g was a t i other~fore ~ to develop Mb alloys with substantial strength in this temperature range. ecause Nb has a bcc: (A2) lattice a suitable second phase might be of the ordered among about 300 binary B2 compou~dsinventoried in the literature (Pearson, 1967), no binary B2 compoiind is shown formed with Nb. The experiments were therefore extended both to ternary Nb-XY and to quaternary Nb-X2YZ systems. The choice of BZtype
~
~
Figure 3 Two-phase microstructure (A2 + L2,) obtained in the Fe-Ni,AlTi system
845
Alloy Design
Nb-XY or Mb-&YZ compounds was made by taking into account the difference in the lattice parameters between Nb (a,,=3.30 x 10-lOm) and these 8 2 compounds. If the difference is too large, the system cannot permit coexistence of two coherent phases A2 and B2. Furthermore, even if the system accepts such a coexistence, it will tend to take the shape of a phase diagram with a eutectic reaction (Figure 2b) rather than that with a solid-state decomposition (Figure 2a). The compatibility of two lattices may also be very low.
aNb 0.03), the as-cast N ~ ~ * C O ~ alloy * Z ~ (one , ~ of the compositions of the pseudobinary system NbCoZr) was found to be two-phased bxit showed a very coarse dendritic two-phase structure: Nb (A2) in dendrite cores and GoZr (B2) in interdendritic regions. icroanalysis showed that the Nb phase contains only small amount of CO and the solubility of Nb in CoZr is also low. ~ubsequentmicrostructural examinations after heat treatment suggested that this pseudobinary system has a eutectic reaction similar to that schematically shown in Figure 2(b). Although this alloy shows ductile behavior at room temperature, its strength is quite low, probably because of the coarse microstructure. A survey of the literature (Pearson, 196 1983) indicates that there are some c o ~ p o u n d sof the Ti2AlX (X=Mo, Fe, Cr, Nb) type. Their lattice para~etersare not well measured but seem to be 3.10 to 3.15 x 10-lOim. The field o f existence of these phases, both in terns of concentration and tem~erature,is also not well known. Because preliminary experiments indicated that Ti2AIMo was the most stable among the ove compounds, alloys of the quaternary Nb-Ti-A1 system along the NbTi,AlMo tie-line wer nvestigated. Niobium, /3-Ti and MO are totally miscible in binary couples, while the solubility of A1 in Nb, in Ti, and in MOis limited to about 10 at.%. In the Nb-A1 phase diagram, there are three intermetallic phases: Nb,Al (cubic A 1.9, N b,Al (tetragonal DSb)and NbA13(tetragonal in Ti-A1, Ti3A1 (hexagonal DOl9), TiA1 (tetragonal Llo), TiAl, (DO,,), and so forth; in MO-Al, Mo,Al ( A15). Mo,Al, (monoclinic), MoAl,, and so forth (Naka et al., 1992). icrostructural examinations of three alloys (see Figure 4) of the Nb( t- Cr)-Ti2A1Mosystem, both in the as-cast state and after heat treatment, indicated that they were totally single-phase, but electron diffraction studies showed that, at room temperature, they were either disordered (A2) or ordered (B2), depending on the composition. Small quantities of Cr were totally
soluble and were not supposed to modify the nature of the phases present. The size of the antiphase doinains observed in the as-cast ordered alloys also depends on compositi~n: it decreases with incr~asin content. The results of these i~~icrostructural observations suggest that the pseudobinary system NbTi2AlMo has a second order (or continuous) orderdisorder transition (Figure 4). In this schematic phase diagram, both A2 and B2 phases have a wide solidsolution range, but there is no t ~ o - p ~ aA2 s e + B2 field. Although the above three alloys showed a surprisingly high yield strength between 25 and 800"C, comparable to that o f the c ~ r r e ~ t l ~ - u sIN e d 100 superalloy, the single-phase ~ i c r o s t ~ c t ~did r e not provide sufficiently high creep resistance. Work has therefore been conducted to identify suitable additional alloying elements that would make the Nb( + Cr) - Ti,AlMo alloy system show the first-order order-disorder transition, encouraged by recent theoretical u~derstandings.The results of the work of Ackerrnann et crl. (1989), usin the so-called cluster variation method (CVM) show that both first-order and second-order transitions are possible between A2 and B2 phases, depending on the nature of the interaction between atoms of both the first and the second neighbors. 2.2.3 Ta-( T i , ~ ~ ) ~ A l ~ ~ o , ~ b ~
To illustrate such efforts, consider a successful result obtained in the Ta-base system (Ta-TiTantalum is also a bcc (A2) metal, parameter is almost identical to that
Nb (+Cr)
1
Schematic phase diagram of the pse Nb( + Cr) - Ti,AlMo, suggested from microstructur~l observations
846
ure 5 Two-phase 8 2 + A2 microstructu~eobserved in an alloy of the T ~ - ( ~ ~ , Z r j , A ~ (system ~o,~b)
starting from a quaternary Ta-Ti-Al-~o system presenting a second-order orderdisorder transition, a partial substitution of Zr for Ti as well as that of Nb o prov~dedTa-rich alloys with a two-phase B2 + A2 niicrostructure, as illustrated in Figure 5. Although totally different crystallographically, this microstructure is morphologically similar to that of the y-y' Ni-base superalloys. The two phases (matrix and precipitates) of the observed microstructure show a cube-cube orientation relationship, and the volume fraction of regularly distributed cuboidal particles is at least 50%. The examination of the interface by TEM indicates that the lattice misfit i s quite small. There is, however, an important difYerence between this twophase microstr~ctu~e and that of the y-y' Ni-base
superalloys. ~ h i l ethe y-y ' Ni-base superalloys are constituted by ordered y' particles and disordered y matrix, it is the second-phase cuboidal particles in the Ta-rich alloys that are disordered, bcc (A2), and the matrix is ordered B2. Various subsequent attempts made by preparing numerous alloys of the same alloy system, in order to reverse the nature of the matrix and second phase (namely, two-phase A2 + 232 microstructure), have been unsuccessful. These results can be understood by using the schematic representation of pseudobinary phase diagrams shown in Figure 6. Microstructural analyses of alloys of the Ta-Ti2A1Mo system after various heat treatments suggest that this pseudobinary system has a second-order order-disorder trmsition (Figure 63);tlie alloys of this system are therefore always single-phase. The alloys, in which the above-mentioned two-phase 5 2 + A2 niicrostructure were observed, belong approximately to the pseudobinary Ta-(Ti,Zr),Al ~ ~ o ,system ~ b (Figure ) 6b). ere, the system shows a first-order order-disorder transition. In this case, if an alloy has the cornpositiori indicated by the dotted line, the phase transition sequence observed during cooling after solidification corresponds to A2+52-+B2 + A2; hence, the matrix of tlie final state is ordered B2 and the second phase is disordered A2. If the composition is located between .xi and x2 in the diagram, the sequence should be A2+A2+B2, arid the fiwd microstructure would be reversed: A2 matrix and 5 2 precipitates. However, the absence of such a microstructure in the authors' experiments suggests x1% x2; in other words, the phase separation line between A2 and A2 + 5 2 has a very steep slope, as illustrated by the broken vertical line in the diagram.
1
1
Ta ure 6 Schernatlc of p~eudoblnaryphase diagrams. (a) Ta-Ti,AlMo. (b) Ta-(Ti,Zrj A l ( M ~ , ~ b )
847
Alloy Design The origin of the change in order of the orderdisorder transition through the partial substitution is not SO clear, but it is interesting to note that there IS probably some atomic size effect. Indeed, in the TaTi,AlMo system, the size of Ti and Ta atoms i s alniost the same, and the size difference between these atoms and MO atom is about 5%. These three elements are therefore completely miscible. In the presence of Al, which fwors ordering, the transition from disorder to order can occur continuously (second-order transition). In the ~a-(Ti,Zr),Al(~b,Mo) system, the size of Nb atom is similar to that o f Ti or Ta, but the Zr atom is about 10% larger. As previously stated, Zr and Ta tend to demix in the binary couple. The size difference betwecn Zr and MO is even more important (about 15%); the miscibility between these atoms is low. The presence of Zr may therefore render the order-disorder transition less continuous, hence become a first-order transition.
x A
Figure 7 Field of existence of coniplex B2 aluminides in the pseudoternary diagram. Starting from the composition around X,AlY, such a B2 field is often extended both toward the X-and Y-rich corners, as indicated by arrows in the diagram. Moreover, the B2 phase seems to exist only tn the AI-lean side, as delimited by a broken line (X and Y are defined at left).
Indeed, three elements of group X show a total mutual solubility; this is true not only in their bcc form but also in their cpli form. Titanium of group X exhibits, at high temperatures, total miscibility in its bcc form with all the elements of the second group Y, and there is no intermetallic phase between Ti and 3.1 ~ ~ e r~ n ~ i I~ ~~ s~ t ~ t ~ ~ these ~ ~ elements ? ~ except Cr; in this case, the large difference in lattice parameters leads to the f o r ~ a t i o n These complex B2 aluminides can be formed by three of Laves-type phases, e.g. TiCr2. Titanium also shows groups of metallic elements, namely X=(Ti, Zr, Hf), A1 a miscibility gap, in the low-temperature range, with and Y=(V, Nb, Ta, Cr, MO, W) (Figure 7). Thus, the and with W (decomposition of bcc-+bcG,-,+bccMo alloy system is of multiconstituent type but may be because the difference in lattice parameters is considered as pseudoternary X-Al- Y. Here, both X and relatively large. On the contrary, Zr and Hf show Y may correspond either to an isolated element or to a total miscibility only with Ta and Nb among the co~binationof the elements of each group. It is also elements of group Y , but this miscibility is also important to note that in binary couples (X-Y, X-A1, accompanied by a tendency toward decomposition and Y-Al), there is no B2 compound. (bcc-+bcc, or bccTaor Nb). The miscibility between Each element of the group X (transition metals of Zr or Wf and the other elements of group f-' is quite column IV of the periodic table) has two crystal limited, and hence the formation of the Laves-type structures: cph (A3) at low temperatures and bcc (A2) Zr Y, or Hf Y,. at high temperatures. The elements of the group Y (the Similarly, the miscibility ~etweentwo elements of first three belong to the column V, and the last three to group Y is total, the difference in lattice parameter is the column VI) are always of bcc type. It is interesting not so large; this is the situation for most of the binary to examine the mutual solubility between two elements couples of this group. If the difference is larger, there is of these groups X and Y in the bcc phase by comparing a miscibility gap at low temperature ( the lattice parameters (Table 1). When the difference in Cr). A large difference in lattice parameter results in lattice parameter is small, the mutual solubility is total the formation of Laves phases (TaV,, TaCr,, and or very high; if not, it is restricted.
The existence of a complex B2 phase was identified during the autliors' various investigations, for instance in T&AI or Ti,AlNb base alloys, in TiAl base alloys, and in Nb or Ta base alloys.
~ b ~ r , Finally, ). it should be mentioned here that a umn VII) or Fe (column VIII) in some alloys during the present investigation. Rhenium exhibits a large soluf the elements of the groups X Hf. Fe shows little solubility in Xand Y? but it forms a binary B2 compound with Ti.
3.
&2 phase. In the case of the ternary Ti-Al~Tasystem, the B2 phase was observed around the composition Ti,AITa, for example, TiSOA125Ta25 and Ti58A117Ta2s. Finally, in the ternary Ticould be observed in Tib6Al2,Fel2and Ti70.5Al,35Fe6; these compositions, especially the second one, are fairly different from the compo~itionTi2A1Fe. According to Seibold (1981), Ti,AIFe can be considered as a 2 TiFe phase. In the ternary systems Zr
Various ~ x ~ e r i i ~carried ~ i i t s out on many alloys of the using TEM suggest that there i s a X-AI-2' SY more or le field of existence for the I32 phase in the above ternary system, sometimes extending deeply into both A'= and Y-rich regions (Figure 7). In what follows, the field of existence of these complex aluniinides is illustrated by numerous examples. Previously, the existence of such alurninides has been literature only with certain ternary B2 conipounds, such as Ti2A1Mo and Ti,AlCr (Kornarek, 1983). More recently, it has been recognized (Pere) that the B2 phase exists in the system around the composition TjS,A1,,Nb,,(=Ti2AINb). These ternary B2 compounds have often been neglected by the phasediag~am-assessment cornmunity, for example, the ~ A L group. ~ ~It isAinteresting ~ to note that the
It is interesting to mention that the B2 phase could be observed in t alloy of the Nb-rich composition N b ~ ~ A l ~whi ~~r3(~~ is fairly distant from Zr,AlNb. Finally, because Zr and Hf are chemically very similar, it is probable that the B2 phase exists around In quarternary or more complex systems, it is obvious that a thoroughly a tremendous task (but see numerous ~ossibilities for combining elements of groups X and Y. Many alloys examined up to now belong only to some selected syst~ms,but the examples mentioned below may suggest a large field of existence of the B2 phase. For Ti-rich systems, the
A1-X3A1; therefore, Ti,AlX can be
For Nb- or Ta-base alloys, this phase was found in compositions such as T ~ 3 ~ ~ A l ~ ~ Hf2,A1,, Nb~0~o~~ 0, ~Ti20A1 ,oNb60Felo, TiJ4Al1 7Ta32Mo and N b ~ O l
Nb $ystem, the B2 phase was o~servedin a variety of *, alloy compositions, for example, Tibl,sA120.5Nb Tis,A12sNb2,, Ti30AlIsNbs5, and Ti,0AllsNb7s. By n e g ~ e ~ t ifor n ~a moment two questions, namely, if e is stable and whether it coexists not, one realizes that its field of tion can be extremely extended. oAl,,Nb75is located deep in the Nb-rich corner. Further experiments indicate that this t ~ o ~$?bSs c o ~ ~ O ~ i tas i owell n as the c o ~ p ~ s i Ti3&ll are two-phase in e ~ ~ i l i at b low ~ u te~peratures, ~ as suggested by the phase diagram; but the two phases
In the above sections, the field of existence of the B2 phase is discussed only in terms of the alloy composition. A more detailed exaniination by TEM of the phase before and after various heat t r ~ a t ~ e nshowed ts that a large variety of situations can be found. F i ~ u r e8 shows three typical examples of < 001 electron diffraction patterns. The presence of superlattice (100) spots indicates that the crystal structure is B2. These diffraction patterns, however, also contain other important information. In particular, super(a) and (c), while they are ~=Ti,A~Mo), TissA125M020, diffuse streaks observed in streaks are weak in (b) and absent in (c).
Alloy Design
Examples of (001) 132 diffraction. patterns. fa) In the as-cast Ti,,AI,,Nb,, treatment of 2 h at 990 “C. (c) In the as-cast TiS,A1,,Mo,, alloy
alloy. (b) In. the same alloy after beat
3.3.1 ~ e ~ r oef e~ r d e r
3.3.2 ~ t o i ~ ~ i o ~of e the t r y~ b s e ~ v e d
The variety of intensities of superlattice spots suggest that the ‘degree’ of order may vary strongly. variation exists not only in different alloy c tions, but also in a given alloy due to some change in the composition of the B2 phase after heat treatments that precipitate a second phase. This is precisely the case with two diffraction patterns (Figure 8a and b) obtained in the T i s ~ A l z s ~alloy; b Z ~ a heat treatment o f 2 h at 990 “C after ca resulted in the decomposition of the initial met le 8 2 phase giving rise to a cipitation of the so-called 0 (orthorhombic) phase nerjee et al., 1988). It is now necessary to discuss the m e a ~ i nof~ the lerni ‘degree’ of order. The degree of order is generally expressed by the l o ~ ~ - r ~order n g ~eter. er. For a stoichiometri~binary alloy, the order parameter is 1 if perfectly ordercd, and 0 if completely disordered (above the critical teni~erature~. When the composition deviates from the stoichiometric value, the order parameter is necessarily smaller than 1, even if the alloy is fully ordered. In the present cases of phases formed in ternary or pseudoternary alloys, assessment of the degree of order requires, strictly speaking, a definition of two order parameters, because there are two distinct sites in ut the discussion here is qualitatively valid and useful for understanding the authors’ e x ~ e r i ~ e n tresults. al
correspond to the stoichiometry of the B2 phase in, Ti-A1-A? If the chemical siniilarity between Ti and X is taken into account, Ti2A1Xis considered as ( T i , ~ ~ AInl .this case, however, the sites Q and b of the 232 s ~ r ~ c t u r e cannot be occupied separately by (Ti-t-m on the one hand and A1 on the other hand. It is thus at suppose that the s t o i c ~ i o ~ e t royf the corresponds to (Ti,X)Al, although there is no such example reported in the literature. Et is interesting here to note that some compounds forme composition TiA1,X (for example, reported (Mabrichi and Nakayama, 1991) to take the cubic but very complex D8,c over, some compo~ndssuch as (Komarek, 1983). Accord results already mentioned Nb, it i s inferred that the B2 phase can be observed in the alloy of the composition TiAIMbz ( = T i ~ ~ A l Z 5 ~ b ~ ~ On the contrary, it is unlikely to observe the B2 phase at the c o ~ ~ o s i t i oTiAlzNb n (TiZ5 situation may be tentatively underst approaching the AI-rich corner of the ternary system, an increasingly ~ r ~ ~ ochemical ~ i nc ao ~n ~t ~ of~ ~ ~
this element tends to stabilize other compounds to the detriment of the 8 2 phase. Thus, the observed B2 phase field in the ternary system may be considered as formed by the extension in ordered solid solution starting from the hypothetical stoichiometry TiA1,Nb. In this case, the degree of order decreases when the composition of the B2 phase moves away from TiA1,Nb to either the Ti-rich side or the Nb-rich side. The arguments just developed are, however, difficult to prove, based on the current state of knowledge. Indeed, previous work (Banerjee et al., 1987), on site occupancy in the B2 phase of ternary Ti-Al-Nb alloys evaluated through A ~ ~ ~ E (atom M location I channeling enhanced microanalysis), indicates that Ti atoms tend to occ~ipyone of the two distinct sites, while Al+Nb occupy the other. If this is unambiguously verified by cross-checking with other techniques such as extended X-ray absorption fine structure (EXAFS), the stoi~hiometriccomposition may correspond to T i ~ ~ l Nrather b, than TiA1,Nb as mentioned above. Furthermore, some of the alloys of the ternary Ti-Al-X system examined in the present study skowed the existence of the B2 phase at very high temperatures around the cornposition Ti,AlX, in particular in the case of Ti,AIMo. This was suggested by examining the existence or absence of antiphase boundaries (APBs) in various alloys.
3.3.3 ~ n t ~ p ~Ba #s ue n ~ u ~ i e ~
In many of the alloys examined, APBs such as those illus~ratedin Figure 9 were observed in the as-cast state. This implies that there is a disordered bcc phase field at high temperatures. When these alloys pass through the critical temperature during cooling from a high temperature, the nucleation and subsequent growth of doiiiains of the 8 2 phase takes place, s; and, if the coarsening of 8 2 d o ~ a i n si s not complete, these APBs are quenched to room temperature. The Fact that no APB was observed in the Ti,AlMo alloy in the as-cast state therefore suggests that the 132 phase exists at rather high temperatures; if this phase is stable even up to the melting point, the composition Ti,AlMo may correspond to the stoichio~etry. In conclusion, further work is necessary to understand the large variety of ordered states observed in various B2 phases, and the key issue is certainly the nature of their stoichiornetry; in the absence of this information, the degree of order cannot be quantified by order parameters. For a more precise comparison
~~~u~~ 9 Example of APBs observed in the as-cast state in the Ti-AI-Nb system
of the degree of order, some effort inust be directed to ~uantitativeanalysis of electron d i ~ r a ~ t i ointensities. n
As previously mentioned, the field of existence of the 8 2 phase often covers n large ~ o ~ p o s i t i o nrange al of the 1-Al- Y system. Consequently, there are various routes of decomposition of the B2 phase. Iln Ti-rich alloys, for example, the decomposition is probably initiated by ‘displacive’ shearings. It is necessary to remember liere that in some as-cast alloys diffraction patterns of B2 showed strong diRuse streaks, as shown in Figure 8(a). Such a streaking in diffraction patterns results p r e s u ~ a b ~from y elastic distortions of the crystal lattice, originating from some coinpositional fluctuations typical of a pre-precipitaLion stage. Moreover, these diffuse streaks were often a c c o ~ p a ~ i ebyd a ‘ t ~ e e ~ - l i kcontrast e in dark field TEM images formed using B2 superlattice spots (Naka et al., 1993). Both diffuse streaks in diffraction patterns and the tweed-like image contrast are a manifestation of a lattice instability, which may induce displacive shearings during the decomposition on heat treatment. These displacive shearings, followed by ‘replacive’ chemical rearrangements, lead to the formation of a second phase of complex crystal structure such as 0 and w-type phases. Detailed information on these phases is available in the literature ( 1988; Bendersky et al., 1990). As for the Nb-rich alloys, in particular those of the Ti-AI-Nb system and
85 1
Alloy &sign of its derivative Ti-Al-Y, the precipitation of the 6 phase possessing the cubic A15 crystal structure was observed after heat treatments in some alloys such as Ti3,AlI$Jb,,, initially single-phase 8 2 in the as-cast state. It is worth noting that this precip~tationtakes place in a heterogeneous manner, probably due to the fact that the nucleation barrier is so high that the grain boundaries become preferential precipitation sites.
Our discussion of the deformation behavior of the B2 phase is restricted to a comp~rativestudy of glide dislocations, operative in two alloys: one is is osingle-phase-ordered B2 ~ i 3 ~ A l ~ ~ N bwhich ~~M ~~, and the other is T i ~ , A l ~ , ~ b ~ , M osingle-phase-dis~,, ordered A2 fbcc). Table 2 summarizes the pertinent deformation microstructures. In the disordered alloy, the slip system is { lTo)(lll), both at 25 and at 800 "C. This is usual for a bcc alloy. The straight screw dislocations (Burgers vector =& (111)) observed in the alloy deformed at 25 "C indicate that they exhibit a high lattice friction stress. The most important infor~ationis that (111) superdislocations are active in the ordered B2 alloy, and such dislocations after deformation at 25°C are shown in Figure 10; the absence of dissociation suggests a fairly high APB energy. Although more detailed and careful investigations are necessary, the observed activity of (111) s~perdislocationsis encouraging. indeed, if this is proved to be true in most complex B2 aluminides of this category, some macroscopic ductility can be expected, because the propagation of these dislocations in 8 2 crystals is generally believed to enhance their plasticity.
Now we examine the influence of the variation of ordered state on the mechanical behavior of 8 2 alumi~ides, because after extrusion some of them
Figure 10 (111) superdislocations observed in the B2 T i ~ ~ A l alloy ~ ~after ~ db ~ ~~ o~r ~~ ~atot 25 i ~o"C n~
show a room-temperature tensile ductility unusually large for i n t e ~ e t a l ~materials. ic
4.1.1 Formation o j VGS
Optical microscopy of the cross-se~tions of bars obtained by extrusion of alloy ingots revealed a wavy contrast bearing a strong resemblance to 'Van Gogh's the crosssky' (VCS) (Figure 11). ~acroscop~cally, section of the bars exhibited a surprisingly irregular shape (Figure 12a) instead of the usual circular one. Moreover, in their peripheral region, waves of 'v closely followed the irregular contour of the crosssection (Figure 12b). These observations suggest that during extrusio~the alloy ingots e~peri en~e plastic instabilit~esarising from a region-to-region variation in deformability of the ingot; in other words, apparently there are soft and hard zones within the material. ~xam i nat ~onof the deformation behavior during extrusion, conducted by interrupting the extrusion of an ingot, showed progressive alignment of the preexisting dendrites parallel to the e ~ t r ~ saxis. io~
Table 2 S u ~ ~ a of r ythe observations of ~ e ~ o r ~ a tmicrostructures ion 25 "C Slip systems
Disordered alloy [I To]( 1 I I ) Ordered alloy [l21](111) and [IiO] (001)
Characteristics af dislocatioiis
800 "C Slip systems
Straight screw b=% (111) pia](I 1i l Straight screw b=(lI 1) [lTO](lII ) and [110](001)
Characteristics of dislocations b = x (111)
Dissociation into two partials k ( I 11)
852
~ ~ a p hsp~cimen ic preparation. uch a local v a r i ~ t i ~inn composition was indeed demonstrated, for example, by electron backscattering imaging in a SEM; in Figure 13, bright contrast co~respondsto the heavy-elem~nt-rich zones and dark contrast to the li~ht-elei~ent-rich zones. To a lesser extent, the occurrence of VGS metallographic contrast may correspond to local variations in the density of crystal defects such as dislocations; such variations in dislocation density were perceptible during TEM observations.
A direct correlation between VGS and preexisting 1 Example of VGS contrast in Nbs4Ti30AI,sZrt, typical of the ~icrostructuresobserved in the cross-section of the bars obtained by extrusion of various alloy ingots
these observations, we postulate that the abovementioned variation in deformability originates from dendritic segregation.
4.I .2 at^^^ of’ the VGS Contrasi e now exam in^ the nature of the ~etallographicVGS contrast. It should be remembered that most of the asbars were sirigle~phase-orderedB2 according examination; hence, the VGS metallogra~hic contrast cannot be related to precipitation of a second phase. Under these circumstances, the most plausible exp~anationis based on the local variation of chemical com~ositionexemplified by etching during metallo-
dendritic segregation has been established. We also suggest that variations of chemical composition associated with the dendritic segregation lead to local variations in defo~iabilityor hardness (hard and soft zones), which are required to induce plastic instabilowever, it should be emphasized that the VGS on accompanied by the creation of an irre lar-shaped cross~se~tion of ingots after extrusion is a very unusual and surprising situation, while the dendritic segregation is a well-known, conimon phenomenon in metallic materials. It is therefore im~ortantto discuss the reasons why there is, locally, a strong difference in hardness rn the alloys of the present investigation. As already mentioned, within a wide B2 phase field of the pseudoternary X-AI-Y system, there is a large variation in the degree of order d e ~ ~ n d i nboth g on alloy composition and on temperature, and the orderdisorder transition (Bc-tA2) is second order. A l ~ h o u ~ h
(a) A very irregular cross-section observed on the extruded bars (right) of the Ti,oZr~Nb,,AI,, alloy, instead of a standard circular one (left). (b) Metallographic image observed in the peripheral region of the cross-section ( T i ~ ~ Z r ~ ~ b ~ ~ A l
853
Alloy Design
ulre 13 SEM electron bac~scatteringimage of the cross section
detailed examination of the ordered states has not been made on the B2 matrix of the alloys studied in the present work, it is probable that some variation in degree of order, which may be associated with the change of chemical composition due to dendritic segregation, plays a significant role in the variation of hardness; a low degree of order corresponds to a highly d e f o ~ a b l esoft zone, and a high degree of order to a hard zone, difficult to deform.
The room-te~perature mechanical behavior, and particularly the tensile ductility of the above-mentioned alloys prepared through various processing routes, has also been e ~ a ~ i n efor d the presence or absence of the VGS structure. Table 3 summarizes the
most significant results. These ordered intermetallics, usually exhibit an exceptionally large tensile elongation (10 to 28%). In the follo~ingparagra~hs,an a t t e m ~ist made to analyze critically the results in this table. For Nb-rich Ti,,Zr,Nb,,All, ingots, the ductility was 24% after extrusion at 1100°C and the corresponding ~icrostructure was B2 si exhibited a typical VCS aspect. comparison can be made with the as-cast state of the same alloy which proved to be brittl~and free from VGS; its B2 single-phase microstructure is of the ordinary dendritic type. Thus comparing these two results, a very strong ductility-~nhancingeffect of V ~ S is implied. Note here that both the dendritic structure and the VGS structure (which originates during extrusion from the dendritic segregation) are associated with a local variation of chemical c o ~ p ~ s i t i o n . The difference between them is that VGS is a morphologically 'well-organ~zed' structure formed by an alignme~tof the preexisting dendrites, compared to the dendritic structure, which is more 'disorganized'. ~xaminationafter rupture of the VGS-bearing tensile specimen provided further useful information. The fracture surface was irregular and extremely uneven; although numerous 'humps' and 'troughs' of the fracture surface co~tainedsmall dimples, typical of a ductile fracture, regions between humps and troughs had brittle features. Moreover, the tensile specimen, initially cylindrical, exhibited quite an irregular contour. All these observations strongly suggest that during the room-temperature tensile test the deformation behavior is composite~li~e because of soft and hard zones, as was observed during high-te~perature extrusion. Now consider the influence of var ments on the ductility of extruded Tij heat treatment of 50 1.1 at 1550 "C prior to extrusion at
Table 3 Results of the r o o ~ - t e ~ p e r a t u tensile re tests ~
Alloy
Processing condition
T i ~ * ~ r , N b ~Extrusion ~ A l ~ ~at 1100 "C Heat treatment at 1550"CjSO h + extrusion at 1100 "C Extrusion at 1100 "C+- heat treatment at 900 "C/8 11 As cast Ti,,~,Nb,5Al,, Extrusion at 1100"C + heat treatment at 900 "C/8 h T i ~ * N b ~ , ~ l , ,Extrusion at 930 "C + heat treatment at 900 'C/24 h Isothermal forging at 980 "C + beat treatment at 900 "C/24 11 T i ~ ~ ~ r ~ ~ TExtrustion a , * A l ~at~ 1000 "C
VGS
Strong Weak Slightly weakened NO Strong Strong No Strong
-~
Ultimate Yield tensile Elongati~nstrength strength (Yo) (MPa) (MPa) 24.0 13.0 10.4 0 19.0 27.6 16.2 0
991 990 1002 ? 856 842 733 7
1007 1004 __
1068 965 872 -
854
Miscellaneous Topics
1100 "C weakened the VGS microstructure by reducing segregation. Room-t empesature tensile ductility iiieasured after this processing sequence decreased significantly to 13'/0, in comparison with the case of extrusion without prior heat treatment (24%). In the same way, a post-heat treatment of 811 at 900°C applied to the alloy ingot extruded at 1100°C S structure, tends to reduce roomteiiiperature tensile ductility (10.4% instead of 24%). In various Ti-rich alloys, the correlation between room-temperature tensile ductility and VGS is not so direct. For the extruded Tis5.sNb25A.l19,s ingot, which shows a very clear VGS structure, the ductility is high (19.0°/~).ln spite of the fact that the alloy is two-phase (B2 + 0) because of the 0 phase precipitation during a beat treatment of 8 h at 900 "C, there is apparently the usual correlation between ductility and VGS. For a similar Ti-rich alloy (TiS8Nb21A121),however, good ductility was obtained, not only after extrusion at 930 "C + heat t r e a t ~ e n of t 24 h at 900 "C (27.6%) but also after isothermal forging at 980 "C + heat treatment of 2411 at 900*C). Moreover, despite the VGS structure found in the former case, examination of the tensile specimen after rupture showed no clear evidence of the composite-like deformation behavior, such as that suggested for the extruded Ti3QZrlNbs4Al15 ingot. Because the alloy was two-phase (B2t-0) in both cases (after extrusion and after forging) and because there was no evidence of VGS after forging, roo~~temperature tensile ductility seems to be more directly related to high deformability of a two-phase structure than to VGS per se. It is worth mentioning that the 0 phase is generally recognized as quite ductile anerjee et U / . , 1988); Consequently both phases (B2 and 0 )are ductile, and the alloy does not seem to show soft and hard zones. In all the alloys the B2 phase was considered as more or less deformable at room temperature; especially, this was a necessary condition for explaining the ductility obtained for single-phase B2 Ti,,ZrlNb54A1,5 after extrusion. A counter-example is provided by another Ti (more precisely X)-rich alloy Ti45Zr3sTaloAll,. In this case, the alloy exhibiting a very clear VGS structure after extrusion at 1000°C was B2 single-phase, but no roo~-temperaturetensile ductility was found. This result implies that VGS formed due to the existence of soft and hard zones during hightemperature extrusion does not show any soft zones during room"temperature tensile tests. More detailed study by TEM on the room-temperature dislocation slip behavior in the B2 phase of the above-mentioned alloys is now under way to understand
their ~acroscopicbehavior, in particular, a possible correlation between tensile ductility and degree of order. Finally, from an engineering viewpoint, it is worth noting that Ti-rich alloys of Ti2Al~b-type(Ti55.5Nb,,AlI9,, and Ti5,Nb21A121)exhibit 'better' tensile behavior than that of the Nb-rich Ti,,Zr,NbS4Al,, alloy, because the former alloys show significant strain hardening whereas the latter does not show strain hardening and in some cases exhibits a tensile softening indicative of a plastic instability (Table 3).
This chapter has described ways in which materials scientists and engineers can use knowledge of modern alloy theory to develop useful structural inter~etallics. First, the background and approach for developing the two-phase 7-y' type ~icrostructurewas reported by illustrating some successful examples such as FeNi2A1Ti and Ta-(Ti,Zr)2A.l(Mo,Nb).The second part dealt with complex B2 aluminides of refractory metals. The stability field of the ~omplexB2 phase in the pseudo-ternary X-A1-Y system was found to be very large. A large variety of ordered states exist for these aluminides; in particular, their degree of order varies from one alloy to another. In the last section, the meclianical behavior of sevcral alloys of the X-AI- Y system, characterized by an unusually large roomtemperature tensile ductility (10 to 28'/0), was reported. Plastic instabilities encountered during high-temperature extrusion, resulting in the formation of a 'van Gogh's sky'-like microstructure, were tentatively related to the variation of the degree of order associated with a change of the local chemistry due to preexisting dendritic segregation.
This work was conducted with financial support of the . authors are French ini is try of Defence ( ~ R E T )The grateful to Dr F. Ducastelle for his very useful advice. We thank Dr M. Thomas and Mr M. Marty for their important contrib~itionto the present work. Thanks are also addressed to Mr P. ThCveniii, Mrs A. Bachelier-Locq, Mr J. L. RaEestin, Miss C. Grisot, and Mr Y, Lebreton for their helpful e x p e r i ~ e ~ ~ a i contribution, A. new introdu~tionwas provided by J. H. Westbrook in preparing this paper for inclusion in the present treatise, for which the authors would like to express their gratitude.
Alloy Design
A c k e ~ a n n ,H., Inden, G., and Kikuchi, R. (1989). Acta Met., 37, 1. Allibert, C. H., and Pastor, H, (1998). In Therniodjmanzic ~ o ~ ~ e and l ~ nMater~als g Data Engineering, (eds J .-P Caliste, A. Tryol, and J. H. Westbrook). Springer, p. 263. Ansara, I,, Durand-Charre, M., Wright, C. S., Wronski, A. S., Mascarenhas, J., Oliveira, M., Lernoisson, E., and Bienvenu, Y., ibid., p. 255. Aoki, IS.,and Izum, 0. (1979). J. Jap. Inst. Met., 1196. Banerjee, D., Nandy, T. K., and Gogia, A. K. (1987). Scr. ~ e ~ a l l96, . , 597-602. Banerjee, D., Gogia, A. K., Nandy, T. K., and Joshi, V. A. (1988). Acta Met., 36, 871-882. Bendersky, L. A., Boettinger, W. J., Burton, B. P., Biancaniello, F S., and Shoemaker, C. B. (1990). Acra Metall. Mater., 38, 93 1-943. Chen, N.-Y., Lu, W.-C., Chen, R.-L., Qin, P., and Villars, P. (1999a). J. Alloys & Compounds, 289, 120-125. Chen, N.-Y., Lu, W.-C., Qin, P., Chen, R.-L., aiid Villars, P. (1999b). J. Alloys & C o ~ p o u n289, ~ , 1236-130. Chen, N.-Y., Lu, W.-C., Li, C.-H., Qin, P., Chen, R.-L., Yao, L.-X., and Tao, L. (1999~).J. AIIouvs & Compounds, 289, 131-134. Chen, N.-Y., Chen, R.-L., Lu, W.-C.-, Li, C.-H., and Villars, P. (1999d). J. Alloys Csi Cohen, M. (1986). Sczence, Freeman, A. J. (1992). Ber. ~ u ~ s e n - GPhys. ~ s . Chem., 96, (1 I), 1512-1518. Freeman, A. J., Xu, J.-H., Hong, T., and Lin, W. (1992). In Ordered I n t e r ~ t a l l i cs Physical MetalIurgy and Mechanical ~ e h a v i o r ,Vol. 213 (eds C. T. Liu, R. W. Cabn, and 6. S a u t h o ~ NATO . ASI Series E: Applied Sciences, Kluwer Academic, Dordrecht, p. 1. Harada, Y., Morinaga, M., Saito, J.-I., and Tagaki, Y. (1997). J . Phys. Cond. ~ u ~ ~9,e8011-8030. r , Khan, T., Caron, P., and Naka, S . (1990a). In High Temperature Aluinindes and Iriterr.P-letullics,Proc. Joint ASM/TMS Symp. (eds S. H. Whang, C, T. Liu, D, P. Pope, and J. 0. Stiegler). 219-241. Khan, T., Naka, S., Veyssitire, and P. Costa (1990b). In High Temperature ~ a t e r i ~for ~ l sPower Engineering. COST 501 and COST SOS ConE. Proc., Kluwer Academic, Dordrecht. Knowledge Foundation Conferen~s: 1999: Rational Approaches to New Materials Design and Synthesis; 2000 ‘Combinatorial Approaches and High Throughput Screening for New Materials Discovery; 2001, ‘Cornbinatonal Approaches for New Matenals Discoveiy’.
855
Komarek, K. L. (ed.) (1983). Titanium: Physico-C~emi~al Properties of its Compounds and Alloys, Intl. Atomze Energy Agency, Vienna. Lill, J. V., Skinner, A. J., and Broughton, J. Q. (1997). J. Phase Equil., 18, 495. L ~ uC. , T., White, C. L., and Horton, J. A. ( I 985). Acta MPt., 33, 213- 229. Mabuchi, H., and Nakayama, Y. (1991). Bull. Jap. Inst. Met., 30, 224-230. Miller, G. J. (1998). Eur. J . Inorg. Chem., 5, 523-536. Naka, S . , and Khan, T. (1991). In Intermet~llicC o i n p o u n ~Structure and Mechanical Properties, Proc. Intl Symp. JIMS-6 (ed, 0. Izuini). Japan Inst. of Metals, Sendai, Japan, 165-171. Naka, S., Thornas, M., and Khan, T. (1992). ~ a ~ e Scz, r . Tech., 8, 291-298. Naka, S., Thomas, M., Marty, M., Lapasset, G., and Khan, T. (1993). Proc. First Intl Symp. olt Structural Inte~ine~ullics (ISSI), TMS, ~arrendale,PA, 647-656. Naka, S., Thomas, M., Sanchez, C., and Khan, T. (1997). Proc. 2nd Intl Synqp. on S t r ~ ~ c ~In~ermetullics ur~l (eds M. V. Nathal, R. Darolia, C. T. Liu, P. L. Martin, D. B. Miracle, B. Wagner, and M. ~ a ~ a g u c h i313-322. ). Olsoti, G. B. (1989). J. Materials Educ., 11, S1.5-528. Olson, G. B. (1997). Science, 277, 1237. Olson, G . B. (20~0).Science, Pearson, W. B. (1967). A Ha Structures of Metals and Alloys, Vol. 2, Pergamon, Oxford. Perepezko, J. H., Chang, Y. A., Seitzman. L. E., Lin, J. C., Bonda, N. R., Jewett, T. J., and Mishurda, J. C. (1990). In High Te~peratureAlumindes & I n t e r ~ e ~ ~ l l iProc. c s , Joint ASM/TMS Symp. (eds S. E. Whang~C. T. Liu, D. P. Pope, and J. 0. Stiegler). Warrendale, PA, 19-47. Pettifor, D. G. (1992). Mater. Sri. Tech., 4, 2480-2490. Saito, T. (ed.) (1999). C~mputational Materials Design, Springer-Verlag, Saunders, N., and Miodownik, A. P. (1998). CALPP-TAD, Calcula~ionof Phase Diagrams - A C~~prehen.si~ie ~ui~e, Elsevier Science/Per~amon. Seibold, A. (1981). 2. ~ e t a l l k ~ c , Shah, M., and Pettifor, D. G. (19 197, 145. Waugh, J. L. T. (1972). The Con~stitut~~n of ~norgaiqi~~ Coinpoundv: Quantum Mechanzcs: Metals and Inter~etallicCompoun~.Wiley-I~terscience,New York. Westbrook, J. H. (1996). In D ~ ~ ~ l o c ~in~ iSolids, o n s Vol. (F. R. N. Nabarro, and M. S. Duesbery). 1-26. Yamabe-Mitarai, Y., Koizuini, Y., Murakaini, H., Ro, Y., Maruko, T., and Rardda, H. (1997). MRS Symp. Proc., 460, 701-705. Zhao, J.-C. (2001). Adv. Eng. Mutis, 3, 143.
This Page Intentionally Left Blank
The amount of inforination available on the Internetl is incredibly large, over three billion pages of text and graphics, equivalent to a stack of paper more than 100 miles high and growing faster every day (a doubling time of about 10 months), And all of this is searchable within a few seconds! One might suppose that such an arcane, esoteric, and narrow topic as ‘intermetallic compounds’ (IMCs) would be poorly represented in this vast collection. This is not so, as even the simplest searches will quickly turn up 1000s of hits on this topic as we shall see. The purpose of this chapter is not to catalog or even s u ~ m a r i z ethe informat~onon intermetallics that is available on-line, but rather to provide: it) a brief introduction to the means for threading one’s way through the enormous mass of information and b) some examples of the kinds of information available, I11 this way the reader may be encoura~edto familiarize himself or herself with this powerful resource and use it to find needed information in the reader’s own sub-specialty in intermetallics, be it crystallography, quantum niechanics, or engineering properties. A final caveat: the whole field of computer access to information is moving so rapidly that by the time this chapter is available in print much of its content will be obsolete, incorrect, or incomplete. Nonetheless, Z hope that it will prove a useful introduction for the novice. A still useful overview of materials science and en~ineeringon the Internet is provided by ~ e l t s ~ (1995), er although the reader is cautioned that in this fast-moving field passage of six years renders it somewhat out of date. ‘This and other itaEicized key terms are defined in an appended Glossary. An excellent brief history of the lnternet ttp://www.isoc.or~/intern~t/
To begin, the prospective user needs: a computer, a physical means (a modern) for linking that computer telephonically to an access point (an ~ n ~ e r nService e~ Provider - monthly fee usually charged, but some are t.com), and either knowledge ired source (its URL) or a search engine (mostly fr ince more than a billion URLs exist, a search engine is a virtual necessity for browsing or ~ a t f fmining, yet there are hundreds of them differing in scope, speed, strategy, and scoring, For these reasons it has been found (Lawrence and Giles, 1999) that no one search engine covers more than 16% of the contents of the Web. This in turn has led to the concept and implementatio~of ‘ ~ e t a - , ~ e a r ~ ~ engines or ‘super browsers’ that automatically search using several individual search engines s i ~ u l t a n e o ~ s.l y Some of the best of these2 in the area of science and technology are:
custom designed, letting you create a persoiial inetasearch engine by choosing several individual engines from among 100 available.
2Thoselisted are for PGs but may work for Macintoshes as well; a corresponding one, especially far Macintosh computers, is Sherlock (www.sherloc~ ’Here and elsewhere in this chapter shown in bold.
~ ~ t ~ ~ r ~ ~~ Vol. ~ 3,aPrinciples ~ ~ ~ and ~ ~ Practice. c Edited ~ by J.~ IT.Westbrook ~ and R. ~ L. Fleischer. s 0 2 0 0 2 John Wiley & Sons, Ltd.
:
individual engines in order to take advantage of specific ‘power searching? tools of each that are tailored to their own content and strategy. Another factor to be aware of: conventional searching facilities miss much that is really out there. The firm Bright as estimated that a virtually invisible reposielectronic i n f o ~ a t i o nexists, the so-called eb’, that is perhaps 500 times greater than that accessible by conventional search engines! More about
.More on search strategies may be found in a short e quantity of int~rmetallicinformation available on the Internet, consider the number of hits obtained with each of the meta-search engines noted above, searching on ‘intermetallic’: Raging Search IQ758
Copernic 1791 and for a few of the individual search engines:
s are not recorded because these engines rctrieved relatively few hits on ‘intermetallic’. Apparently the sources these engines cover contain little on science and t~chnologybut concentrate on other fields 4For completeness we might, for example in the ChemWeb oolean logic and ORed ‘in~~rmetallic’ with such terns as ‘silicide’ (7658 hits), ’selenide’ (625 1 hits), ‘telluride’ (5794 hits), or ‘antimonide’ (4059 hits), thereby picking up hundreds o f additional hits that might otherwise have been missed.
such as news, sports, entertainment, consumer products, etc. Considering once more the numbers of bits obtained with different engines, there are, of course, duplications between the different searches, but not within any one search, be it with a single search engine or a meta-engine combining several subsidiary engines. The differences in the n u ~ b e of r counts in the different searches are a result, as alluded to above, of differences in the scope of original sources assessed, the time period sampled, and the search strategy, among other factors. (See Thornas, 1998,for a discussion of these factors and quantitative coinparisoils of search engines.) The numbers listed above are reassuring that there is a substantial body of relevant information on intermetallics accessible on-line, but the mass is clearly too vast even to simply scroll through the titles on a computer screen. What to do? There are two major options which we will illustrate: some engines are set up to provide counts by sub-ccttegories their builders have chosen; for others it is necessary to narrow the search with Boolean logic. As an example of the first type, consider NorthernLight with its 9306 intermetallic hits derived from two sectors: science and mathematics, and technology. These are broken down by the eiigine builders as follows: special collections 235 1 non-ferrous 745 solid state physics 432 physical chemistry 399 metallurgy 483 NTIS 1956 composite ~ a t e r i a l s187 iron industry 233 corrosion 188 clay, ceramics, and refractories 90 synthetic chemistry 38 all other 6571 All in all, this is not very helpful. The categories are not what we might have chosen; nearly all are still too large; and furthermore there is a certain illconsistency or lack of equivalency in the categorical groups designated. Any topic of particular interest might well be found in several different categories, and furthermore the three largest roupings (special collections, NTIS, and all other) are the~selvesvery broad categories. By clicking the mouse on the appro~riate topic, one can automatically get a further breakdown of any one group, again into sub-cat~gorieschosen by the system builders. Take ‘all other’ for example. These 6571 hits are then seeis to be broken down further as follows:
I n t e ~ ~ e t a lan l i ~the ~ Internet
special collections 325 alloying 504 physics 1712 carbides 129 sintering 125 superconductors 22 1 hydrides 69 aerospace materials 35 solidification 95 sur~~ce-mount technology 41 machining 53 all other 3460 This exper~mentreveals not only the arbitrariness and illogic of the assigned categorization but still leaves us with a huge, niiscellaneous, ‘all other’ group. Following further successive breakdowns of ‘all other’ through four more generations, 156 items remained in an ‘all other’ category, but the search ultimately showed a modest number of hits in unexpected and quite surprising sub-categories, e.g. astronomy (4), birds or ornithology (2) and biomedical engineering (4). And these are included under ‘intermetallic’? To illustrate the other approach using Boolean logic, let’s examine the 3802 hits obtained on CheinWeb using only ‘intemetallic’. Although many choices could be made, a few common sub-categories were studied by ‘AND’ing Yiitermetallic’ with some selected terms, yielding the following results: structure 1432 properties 1 159 order 78 1 application 34 1 bonding 319 synthesis 114 microstructure 110 refractory 30 defect 29 dislocation 28 grain boundary 25 Most of the categories are still impra~ticablylarge (admittedly we could have chosen narrower search terms), but we are now on track. All we must do is refine the search with further application of Boolean operators. Suppose we were concerned with the crystallography of certain structures of interest, we D ‘stiucture’ with some particular types, say ‘BE32’ (32 hits) or ‘E-phase’ (18 hits). These numbers are small enough that we might then scroll through these hits to find whether or not we have turned up s~methingof real interest. Extending the
859
illustration, consider the Properties category. This could be broken down by ANDjng ‘properties’ with such terms as ‘electrical’ (163 hits), ‘niechanical’ (354), ‘structural’ (415), ‘magnetic’ (385), ‘chemical’ (532), and ‘catalytic’ (136). (Note that perhaps we should have truncated some of these terms, e.g. ‘eelectr”’ or ‘cataly”’, so as to obtain additional hits on electrochemical properties of IMCs or catalysis with IMCs.) This gives us a better idea of the coverage of different kinds of IMC properties included, but further narrowing of the search is still required. Examples of the use of other Boolean operators can be given: ‘intermetallic’ AND ‘defect’ AND ‘vacancy’ (23 hits); ‘intermetallic’ AND ‘defect’ NOT ‘vacancy’ (6 hits). For a truly comprehensive search, on ChemWeb or any engine, to be sure that there is a real association of two or more terms of interest, it is sometimes desirable to use NEAR in place of AND. NEAR requires that the linked target terms niust occur within say 1020 words of each other in the particular document, not just anywhere in the document. One hazard with the use of Boolean searches: it is not good to get too specific too soon. For example, if we were to search for ‘intermetallic + superconductor + Nb,Ga’ we might find no hits at all; yet there are 74 hits on ‘intermetallicfsuperconductor’. Practice, while not making perfect, can be a big help in searches. A useful way to augment one’s searching by terms or categories is via links. Some search engines offer the following option: Having found one or more sites on the Net of particular interest, one can click on ‘more like this’. The engine, with its own programming strategy, then brings up to view similar sites, Other engines, e.g. AltaVista, have a link command. Bring up the AltaVista engine, type link, and then the address of a particular site known to be relevant and important. At that point click on ‘Search’, and AltaVista will present a list of Web sites that link to the one you entered. Another factor to be aware of in searching is terminology. Even in science, terminology is not always precise, unique, consistent, and unam~iguous; for any term or concept has a wide variety of synonyms, near synonyms, and related super-categories or sub-categories. As an example, let’s use the Google search engine to look for icosahedral alloys, an important class of intermetallics. Only 22 hits are found. But look at the results of the use of some alternative, virtually synonymous terms: pentagonal intermetallics 55 hits penrose icosahedral metals 6 1
860
Penrose alloy 330 quasi-crystal 350 ~ u a s i c r y ~ ~alloy ~ l ~ 459 ne ver 95% of the potentially relevant information would have been missed by using only the original search term! See further on this particular topic below. In using search engines it is important to realize that most offer, or have already adopted, various options, both in mode of search and in display of results. For example, you may be able to choose what elements of the pages to search on (title, full text, meta-tags (keywords, notes, etc.), author(s), or date range). When the search is c o ~ ~ ~ land e t eyou are ready to review the results, there are again options for display: all hits or some maxim^^ number of hits; for a given hit, title only, abstract, or the first few lines. Not all engines even tell you the total number of hits; some simply present some number, say 20 or 50, of what they consider the most i m p o ~ a n t , automatically ranked by a paraineter they have chosen, e.g. most hits contained within the page, most links to other sites, pages most visited by other searchers, most recent, etc. Another approach in searching might consider the we are seeking, not the particular ing located a substantial body of relevant intermetallic hits, we could search within it for various kinds of inforination. For example, we could AND with ‘i~termetallic’specific terms within categories such as ‘people or person’; ‘society or organization’, ‘publication or paper’; ‘book’; ‘conference, s y ~ p o s i uor~~ e e t i n ~and ’ ; ‘univ~rsity , college or institute’ to get more quickly to what we are seeking. Again, it takes practice to get good at this, and the o p t i ~ ustrate~y ~ may difl’er considerably from one search engine to another and from one kind of information to another. The people category is a very mixed bag; searching on name alone yields hits ranging from biographical sketches of faculty members, to individuals active in the field, to graduate students looking for a position (see Figure 1). Thus it may well be more fruitful to take a different approach. For information on faculty members known to be active in the intermetallics page of the field, it is better to go to the individual’s ii~stitutionand extend the search from in Figure 2 for Ian Jones at which details his address, means for contacting him, his research interests, publications, current projects, planned projects, published books, etc. To illustrate a search for work by, or ~~~~
1am graduate sbdent
of Physics of metals Chair of Phys~calPacuiQ of Ivan F r d o L‘vlv National V&ea&y
‘Nng of a ~ n o ~ h ohlms u ~ of intermetallic compounds by thermalevaporation g m h n sputkring methods Inwfihgation of electrical propertiesof thin films Sc-Ch and Fffid compounds Structural mnveshgationsof melallic amorphoushlms by rleczconical micmscope investigationof magnehc prnperttes of films
~blicat~o~: 1. Mycotaychuk O.G.,Dutsyak IS., Lutsyk N.Yu. Frywazhnyuk V.I. ShcIural clrangLT in films of system GaSb-Sn / / Mater. of 4 Int Con€.“Phys~csand Technologyof thin Mms“ Ivano-Franluvsk, 1997., P.31.(Ukr.) 2. PrysyazhnyukV.L, Derkach V.O., Margolych LL The strucNre and e Y ~ ~ n d u c hof~disorder ty films %a4 // Mater of 4 Ifit. Conf “Physicsand
Technology of thin films" I v a n ~ ~ r a n 1997., k ~ v P~.4~1.(Ukr.) 3. My~~Iaychuk O,G, Lutsyk N.Yu,, hfiva!zImyuk V.L TIE kinetics o€ZorMation and f h x @f melastable solid soluhon m Films of system Ge-GaSb. // Mater. XI Int. Conf. t’Coastru&vityand functionahty matertals” 4 Dutsyak I d , ~ ~ y V.i ,Kodovsky a z ~ ~ ~ obtiuning on s m c r u and ~ ~l~trophysical properttm o~amorphousthin films &Cu4 // Mater. ~ t e ~ o~ ni ~. n h f i c - p ~Conf. ~ c a“Phyacs l @fconde~ matgnals”.Uzhorod, 1998.P.91. (Ukr r$clcical properties of thin films 5. Prwyazhnyuk V I. S&ntcturiil
(PDe f Visnyk of L%v Wniversity
~ i 1 ~The h~o ~ e p~ a g eof e a graduate student, Viktor Prysyazhnyuk, at L’viv Natiotial University, seeking a position, This continues for 12 pages listing his p u b l i ~ ~ ~ ~ o n s , areas of research interest, coinputer programs he is knowledgeable about, lists of journals publishing in his field, universities and institutions he knows to be active there, and relevant databases - an obvious effort to promote as many hits searches by others. ( ~ ~ w . reproduced with permission
ita at ions to a particular worker in the intermetallics field, 1 offer a self-centered example: ~intermetal~ic +Westbrook’ using the Google engine. Although the hits were d i s ~ ~ p o i n t i n ~few l y in number (152), some interesting and unexpected results were obtained: use of our book in seine graduate course outlines, references to us included in papers still ‘in press’, some review papers in unfamiliar journals, and even data on the current book that were unknown to the editors (publication date, price, and ISBN number!). Some useful directory sites for people are ~ ~ ~ o S p a c e
name, phone number, or e-mail address. Planning a trip to Japan and want to know which universities might be worth a visit to check up on
86 I
I n t e ~ ~ e ton~ the ll~ Intemet ~~
2000, could be 1 from h y ~ ~ r l i non . ~ sthe TM home site ~ w w w ); 45 pages of abstracts were instan tly available. sites are accidentally Some o f the most interest~n~ found simply by su~$ng through some of the larger bodies of data isolated by a preliminary crude sort. Thus in the preparation of this chapter we learned of
a tally maintained of all the Japanese ~ a t i o n a lResear over the past 10 years, their country of ori the subject of each research program; a Japanese-speaking Russian who is an expert on. planar defects in ordered alloys and now works in Japan; the micro-rover for the Mars Pat contained an Alpha Proton Xwhose Russian-built sensor head used ~ u r i u m ~ ~ ~ silicides. Pt-group inter~etallics with ~urium isotopes were also tried. The comb~nationof alpha back-scattering, proton emission, and X-ray emission enabled ~etermi~ation. of all elements
(la) P h n c Defomt~onBehawour LI@Iu&snc Defo~ation BehaQour
Figure 2 The hoinepage of Prof. Iaii Joiies, a faculty member at the University of Birmingham, active in i n t e ~ e ~ a l l i c
sfBAIPST Caart&
current research in our field? CO directly to the proposed institution, e.g. Osaka University, the
, which include millions of books offered by thousands of booksellers all over the world. From 20 to 100 hits will be found for inte~ietallicsat any one of these sites. For details on recent or forthcoming conferences or symposia it would be best to go to the home site of the sponsoring society or other organi~ationand use the h ~ ~ found there to get to the technical program, abstracts, or even full papers on-line. For example, the 5th International Conference on Structural and Functional Intermetallics, held in Vancouver, British Columbia in
&.QPhaae T ~ ~ f o m B~watBls n ~ m of Fe-based Allovs hc Compounds
e
~
~
~
~
~
s
Figure 3 The research program in the iiiaterials engineering dept. at Osaka University in Japan. (ht osaka-u.ac.~p/mse~/Al
862
Miscellaneous Topics
(except H) encountered on the ~ a r t i a nsurface at concentrations typically above a fraction of 1%; the staff member at Florida State ~niversitywho i s an expert on Kondo insulators as a class of in t ermetallics. None of these facts could have been easily foreseen so as to have been targets of a directed search. Furthermore, we are now equipped to search further on any of these hits which spark a particular interest. For example, on the third hit above, the Research Institute for Atomic Reactors in Russia was identified as the location for the intermetallic work. Searching for this on the Net brings up the home page of the i r nwith hyperlinks to Institute ( w w w . ~ ~ ~ ~ ~ ~ s.su) location, organizational structure, publications, achievements, etc. Thus we learn that the Institute is located in Dmitrovgrad, ll00km SE of Moscow; employs 700 scientists, 2500 engineers, and 3300 other workers; and offers a catalog of available nucleide sources and their specifications! Color photos of the Pathfinder on the surface of Mars may be viewed, as well as the tri~modalAPXS spectrometer. A reference to the paper describing the successful application of the radio-nucleide intermetallic sources is also given r et al., 1997) and one to the Russian work on sources led by V. M. Ryadchenko (Ryadchenko et al., 1997), together with an e-mail address One of the latest trends is to develop search procedures that rely neither on predetermined directories and hierarchical classifications nor on keyword searching alone, but instead use approaches such as statistical techniques to measure the co-occurrence frequency of pairs of terms 01-co-citation analyses and hence identify sources that have a semantic relationship (Nadis, 1996; Alper, 1998). Another program in effect reads (niore than ZOO0 times Faster than a human) electronic text from specified archives using a proprietary semantic searching algorithm to identify the actual subject matter of the document. Abstracts generated are then stored in an electronic file folder and/or provided with a hypertext link to the full text of the document (Gavacs, 2000). At the University of California a program called Cha-Cha (Chen et al., 1999) is being developed that determines the homepage of every item retrieved from a search, records the shortest path to get from there to the retrieved page and then groups together information sources that share pathways.
proprietary ranking algo~thnithat considers both the relative importance o f a given page (intrinsic value) and the relative relevancy of that page to the query posed as d e t e ~ i n e d by its external hyperlinkage (extrinsic value). In ranking hits, it also makes an assessment of inbound links in terms of both their number and the authoritativeness of the source. Weighting of these two parameters to determine ranking of hits is a function of the particular query. Hearst (1997) has described recent research at Xerox Palo Alto Research Center and at Stanford ~niversity on novel user interfaces employing color, pattern recognition, and animated 3-D displays to facilitate the search process. Clearly, despite the growing volume of information, getting to exactly that which is needed will become ever faster, ever cheaper, and more reliable. Fayyad et al. (19961, Hearst (2000) and Han and Kamber (2000) review the current situation. A final caution on Net searching: Josh Dubeman, at a recent symposium on artificial intelligence
em/^
1,
gave some good advice on ‘practicing safe surf, i.e. how to protect your identity and interests so as not to be subjected to targeted a~vertising,voyeur studies of your Net behavior by unknown persons, etc.
I
Assists
These are Web sites that list, describe, and provide links to hundreds or thousands of Internet-based media. Among those useful for science and technology are:
above as a search engine, it is really more like a directory inasmuch as it is a humongous hierarchical index, created by humans to search the Web, and hence is more logical, less errorone and more efficient. net An electronic compendium of directories usenet, communication directory, information Telnet: archie.
ca This tells you where a given
Gopher A directory service, available on most public access sites, will scan the Internet for files and programs re1 Japanese Science
I n t e r ~ e t ~OH l ~the i ~Internet ~~
Engineering Electronic Library Sweden (EELS)
863
Information Retriev www.mac~donia.
Engineering Resources Online ( w ~ w . e r - o n ~ i i n e . ~ ~ r . e d Associated ~). with the libraries of the University of California, provides access to scholarly Internet resource collections using its own sophisticated search engine. e m ~ ~ n ~ ~ ~Tlie .org) he American Chemical Society with links to many of its searchable databases. Chemical Abstracts Service, an ACS Division,
directory offering guides to information on the periodic table, atomic bonding, fundamental properties, and links to > 5000 chemical sources. CheniDex. A well-organized directory of chemistry Web sites, accessible via a hyperlink on ChemWeb (q.v.)
Materials Science Resources on the Internet. This meta-site was originally developed by Cathy D. Stewart of National Steel and has since been augmented by Antonio Gorni of ~ ~ S I P ~ ~ I M E .
Mats Ericsson at the Royal Institute of Technology, Sweden. 1 1 parts + 3 Appendices.
hysies
Sites mainly providing links to particular sites in their subfields.
Physics Web (http:/ American Institute overnment
Government information ov) NTTS ( w ~ w . n t i § . ~ oIncl ~) a1 reports on completed research, 60 000 summaries of US and foreign government-sponsored R& and engineering, and ‘Published Searches’, completed computer searches on 3000 topics. Covers > 20 000 NSF-sponsored NSF (www.n§~.~ov) projects as well as information on NSF’s internal affairs. For a typical hit relative to TMCs, see Figure 4. GrayLIT Network ( ~ w text reports from OSTf ( w w w . ~ ~ ~ i ~ gResearch ov) award summaries for DOE, NSF, and NIH. *
World Wide Web Hub for (documents, archives, on-line journals, downloadable software, etc.).
allows searches by name, formula, and CAS Registry No.; it can also supply some physical properties data and
and tables of contents are free in all these; for 45 entries, everything is free; others require a
I The Spider’s Apprentice ( spidap.htm1). A guide to search engines; provides useful tips on the use of many cominon search engines.
~ i s c e l l a n e o ~Topics s
864 Title
8
Type N8F Org ~ Latest n ~ Dats File
. f
n I
2.
Third rntarnacional Workshop on Ordered Intennatmllic Allaye and Compesitse; Hmgzhou, Chinat April 5-10, 1998 award DMR t @ebruLcy
2,
ic
1998
a9714853
Award Numher: 9714853 Award 1nsEx.i Standard Grant Prgm Manageri B r u m A. Mamonald DXR DIVXSION OF MATERIAtS RESeRRCX MPS DlRECT EVR MATHEWTICAL 4 PHYSICAL SCSEN Star0 Dats February 1, 1998 January 31, 1999 IEetimatedt Wphea Expacted Total Amt. I 96,000 (Estimated) Investiqatori Stephen b Sess [email protected] Sponsor Cornell VnAvaxaraity-Endowad 123 Day Hall Ithaca, NY 148532801 607/255-2000 I
NSI Program 1771 Fld Applictn: 0106005 Abstract I
A list of search options and links to
METALS Materiala Research
8714863 Bass The objective of this international workshop is a provide a forum f o r material scientists, and engineers from various countrim to partieip#te in an in-depth diScusai.cn of recent advances and critical iaaues f o r the stmctural use ot ordered Inte*metal,lic alloys and eompositsn. Th4 structural use8 of intemtallic alloys suffer a major dranbackt m e t are brittle at ambient ~ ~ @ r % ~ u r This e ~ .poor ductility and eoughners r~etficta the U88 cf inte~e~ellics a8 critical compcnents Ln induotrral myatmu. For the past 10 years, a ~ a t a n t ~effcrtn ~l have been devoted CO this class of structural materials, and, as a zeuult, significant progress had b4en mad5 in improving the tensile ductility and fracture re~i=tanceof nickel, iron, and titetmum aluminidbis. The objeotivs o f this worksbop i s a provide I form fc snparta from variou# Countries to discuss I l l fundment41 variables controlling defQrraation and fracturo of ordered 1ntennataLlice at axlbL@nt anc slavatsd t ~ ~ ~ ~ a t (21U %he K ~do4%W 5 ~ Of ducCfle IntCSSmt6lliC alLQy5 for structural ~ 5 ~ I31 5 ; further ~ ~ r c v ~ n i n tIOWa e h i g h - t e ~ ~ r a t uatrongth ~e by & cmposite approach; and ( 4 1 the procaa#ing of i n ~ e ~ t a l l i a11QyS c end COnQorritee by both cenventionel and fancvative nuathodS. This international workahop 1s co-organised by Prof. Dcnglihng Lin Ibsse) from shanghsl Jiao Tong Univar5ity. Dr. C. T. Liu from Oak Ridgs National Laboratory, and Prof. sfephen 1. Sa.8 from Cornell Unlv%rElty. IL IS held in Hangahou, China, frcarapril 5-10, 1998. Dertlcigsticn in %his of th&n wrkahop is workahop ia by invitation only. Tbe proce4&ngn fantatively planned to be publiahsd in the Journal of Kuteriale Science and Teohnology. Lt Le expected tu have 25-30 6~i(lntistafrom outside Chip% and 2s-30 aaientis~~ from China. %%% Ordsred intemtallics based on aldnLdas end silioides possess many attractive propertisa tot high-tamparature a t g ~ ~ t use. ~ a l In gensral. these intawetallLc 6 show excellent strength a t e1evEt.d temperacurea. Many intecmtailir alloys mhibit an inosease Ln r e than a decrease aa in conventional szrengrh with t a ~ ~ x ~ t urqther materials. Aluarlnides and eilicidcri are rrtzsmaly resistant to oxidatfen anc corrosion at elevated twuperatures in hostiiot snvironments. In addition? theoe ~ n t e ~ ~ aalloys ~ l ~havs a a density lawer than that of rteala and c ~ for supenlloys by 16 mueh a6 406, thus making the ~ n f m m t e ~ l iideal fsbrinnting moving parts i n advancacl heat engine@, gas turbines and a:her wecgy canversLon ayatw. ***
dictionary definition, encyclo~ediaentry, and list of Web links.
vides a history, u ~ ~ e r l sy t~r un ~~ ~ ~current re,
Mtp NWWW a a P g o v / a w ~ ~ a ~ ~ ~ ~ ~ 7 / a w d ~ txt 1 ~ 7 ~ 1 ~ ~ 7 1 4 ~o 5M 3z5mOo
An example o f a hit from a search on llic+NSF” showing NSF’s support of an American scientist to pa~tici~ate in an I ational Workshop on i n t e r ~ ~ ~ ~ l ~i i c s ://www.~~sf,~ov/awards/
Yahoo Internet How-TO Large Science Databases
Pri~~
home l a n ~ ~ a g or e s find English language engines for various countries nd compares engines.
Ackerrnann, E. and Hartrnan, K. (2000). The I n f o r ~ ~ a t i oSpecialist’s n Guide to S e ~ r c ~ i nand g ~ e s e a r c ~OiMnthe ~ Internet and the World Wide OR, 438 pp. Web, ABF Content, W~lso~ville, Basch, Et. (1998). ~ e s e ~ r c OnEine ~ i n ~f o r ~ u ~ ~ IDG Books, 334 pp. B e r k ~ a n R. , L. (1994). Find it Onl~ne,W i n ~ c r ~ s t ~ ~cGraw-~ill. Clement, G. P., ed. (1996). Science and Technology on the rnt~rnet:An Inst~uctionalGuide (Internet W o r ~ s h o pSeries, No. 4), Library S o l ~ t i ~ n s Institute, 376 pp. Detcher, A. L., ed. (2000). ~ a t e ~ ~ e lGuide d ‘ s to ~ o ~ p u t Se or ~ t ~ ~Waterfiel re, Gaffin, A. (1996). ~ v ~ ~ y b Go ~ y ’ ~ ~ MIT Press, 21 1 pp. Gilster, P. (1995) ~ i n ~ i it n gon the ~ n t e ~ n (1995). et J. Wiley and Sons, 376 pp. Grossbrenner, A. and Gross Little Web Book, Peach
~
~ n t e r ~ e t ~ l lon i c the s Internet Hu, Y., Xiau, Y., and Zhang H ~ ~ ~ d ~ojScience ?ook and (Chinese and English) People’s Post & Telecom~~inicationsublishing House, 329 pp. ., Young, M. L., and Baroudi, C. (2000). f o r ~ u ~ ~ i7th e ed., ~ s IDG, , 384pp. (1998). How to Access the ~ e d e r a l G ~ v e r n ~ on ~ nthe t Internet, i999; ~ a . s h i n ~ t o n QnEine. Now to Access Federal G o v e r ~ ~ e n t ? ~ j o r ~ ~ t i4th o n ,ed. Congressional Quarterly, Inc., 300pp. ., eds (1990). The Internet, DK P~blishing,128 pp. Newquist, H. P., ed. (2000). YGhoo! The ~ Z t i ~ a t e Desk Reference to the Web, Saxper Collins, 504 pp. ay, E. J., Seltzer, R., and Ray, D. (1997). AZta~ista Seapch RevoLution, O s b o r n e / ~ c C r a w - ~ i l l ~ 395 pp. Renehan, E. (2000). S c ~ ~ n~t ~~ ec r i Guide c a ~ to Science on the Internet, ibooks, 470pp. J. (1997). The Internet jor Scientists and rs, SPEE Press, 3rd ed., 497pp. o ~ W e~b f o r d ~ Thomas, B. J. (1998). The ~ Scientists and Eifgiizeers, SPIE SAE Int’l, and IEE, 1 The keptic’s Guide to I ~ t e r n e tResearch, Bibliodata, Needham Heights, MA ) An 8pp. monthly news(w letter.
A library directory listing libraries with
but categorized directories are available. New York Public Library (w catalogs of holdings are as guides to doing research, and some substantive information. H a r v a r ~On Line Library ~nformationSystem
links to the Web, lin ublications, exhibitions, and searching guides, as starting points for research. Minuteman Library (htt Useful Internet resources,
86.5
2.6 F e e - c ~ ~ The Scientific World ( ~ w w based in Boynton Beach, F engine from LION bioscience in H offers a suite of four p r o g r a ~ ssc : bibliographic database covering journals and > 12 x 106 articles since 1993); worldMEET (an electronic link to conference databases, meeting alerts, and links to event homepages); scienceand services of WAREHOUSE (a link to the Fisher Scientific, Inc.); and I H (a means of ~on-line ~ publishing e of proceedings of technical conferences). Although hea d toward biology and file nonetheles~showed medicine, a test of the 2492 hits on ‘inteme ’tles and bibliograp~ic citations are available on-line, but full text costs $12 per article. On the other hand, the I-PU conference organizers the possibility to put on-line, without cost to orgmizers or authors, the abstracts, extended abstracts, or full text of all or selected papers from their meeting. All that is asked in return is promotion of ScientificWorld by conference organizers in their notices and m nLXCHTn (www~n~i searches not only Internet sites but also other coniinercial databases, newsletters, reference sources, etc. It attempts to provide a single place for the user to search all of the world’s electronic information. s ~ organization * ~ CliemSW.com (www ~ ~ ~ e r n) This offers a wide variety of individual software programs and databases in Boppy disk or CD format. They are designed to provide n comiiion interface for all their products and to offer bases compatible across all platforms. Among the products of interest CISProWeba (for chemical inventory); TA (theri~o-chemical and physical properties); Cheniand site@Pro (a comp~ter-aidedchemistry m od~l ~i ng drawing program with an interactive crystal builder); and Physical Properties!Pro@ (1 1 programs for calculating selected physical pro~erties).
866
Misce llalte ous Topics
US Government Patent Office (ww~.~spato.gov).
(Elsevier, Pergamon, North Holland, and Exaccessible through their ogram (free). First choose an rtkahaslu, 8 , Aoki, K ,Masumota, T ,pp 207science), then journal (Inter11 ~ e t a l ~ i cand s J o u r ~ o~ f lAlloys and ~ ~ ~ p o u n ~ s o Mitta, R ,Rami~RW,V V ,VC~lugopalRao, A ,pp 213-232, PDF &89L8U are most relevant to our subject), then table of http h w w hbz-nnv du/~IsevierlO9~9795/v~O7i02/ 09/14/1999 contents, then year and issue (see Figure 51, and finally, if desired, author or keyword. Full text of several journals is available, but you must be a Figure 5 The table of contents of one issue of the journal Intemzetullics reproduced with permission from Elsevier subscriber to a print publication in the Elsevier Science. (http://www.hbz-nrw.de/elsevier~09669795~v0007i02[) family or your institution must have an existing contractual agreement. TMS (The Minerals, Metals, and Materials Society TULIP (The University Licensing Program) s,org) Tables of contents, and Elsevier’s collaborative project with 9 U.S. abstracts from four TMS journals and from TMS-pLiblished volumes. MRS (Materials Research Society) ~ w w w Provides hyperlinks to tables of cont journals, conference proceedings, published tables of contents for the complete backlist of books, etc.; some full te journals published by the Institute (for our ASM International (www.a$ subject, see especially condensed matter and Provides hyperlinks to t applied physics). I~formationis provided on how journals, conference proceedings, published to arrange for full text access to particular books, etc.; journals. Annual Reviews Physical Review On-Line re~iewsor^^ our subject include materials science and physical chemistry. Full text is available (for a fee) from 1996 on. Tables of contents (free) are available Fulltext Sources Online ( w ~ ~ . i n f o t o ~ a y . c o m ~ f s ~ / online from 1985 on. tm). A directory, print (1144pp., Jan, Angew. Chernie Intl. (in English) (w 2001) or electronic (F O/e)* of 709 periodicals conte~ts/)Tables of contents on-line. that cover science or technology and can be Journal of Applied Crystallography accessed through an aggregator or content Tables of contents on-line. provider (paid subscription required).
Intermetallics on the Internet
2.9 Translations AltaVista Translations (http://babelfish.altavista.com/). Babylon (www.babylon.com). Both the above sites provide translations to or from English for a variety of other languages. AltaVista offers French, German, Italian, Spanish, Portuguese, Chinese, and Japanese; and Babylon offers these and many more including Hebrew, Russian, and Dutch. AltaVista goes beyond translating simply words or phrases; by typing in the URL of a site of interest, the full site will be recovered but with all text translated into the language of your choice, perhaps not smoothly or completely correctly, but adequate for general intelligibility. An example, using Babelfish, is presented in Figure 6 showing its translation of ‘intermetallic compound’ and a common term, ‘large table’ into several languages. The consulting firm, Global Reach, (http://glreach.com) found that the languages most heavily represented on-line, following English, are Japanese, Chinese, German, and Spanish in that order. As of 30 Dec. 2000, there were 39 million Japanese on-line users compared to 192 million English-speaking users. gmde table
French
intermetalliccompose’
gross Tabelle
German
halbleitende Verbindung *
grande tabella
Italian
intermetallicresiduo *
grande tabela
Portuguese mtermetallic compost0
gmde vector ?
Spanish
867
deleted to make room for new files. Some sites are moderated by a Webmaster to ensure quality and some standardization of format; most are not. Perhaps the most attractive feature of the system is the ability to pose questions that have previously been impossible to find answers to. The welcome responses can come from people who know the answer, know people who might know, or can at least identify others with common concerns. Unfortunately, there appears to be no newsgroup focused solely on intermetallics. There are, however, discussions and questions about interinetallics floating about the Usenet. To find them we might try Copemic 2001, using as search term ‘newsgroup+intermetallic’ and obtain 17 hits on topics ranging from the AuAI, purple plague to announcement of the availability of most of Vols. 1 and 2 of this treatise as a set of four soft-cover volumes. Alternatively, using the Google search engine for ‘newsgroup + science’ and within that ‘science.materials’, we find 2140 hits, many of which relate to intermetallics, e.g. shape memory alloys, compound semiconductors, silicides, etc. The other mode of access to discussion groups is via discussion lists or list servs. Among these are: http://paml.net (7344 lists) www.liszt.com www.tile.net www.clearinghouse.net
Again we find no lists specifically devoted to intermetallics, but many to science or to materials. It would seem that there is opportunity for someone to build and upload a list of individuals and organizations with strong intermetallic interests for the future benefit of all concerned.
intermethim compuesto
3. Some Useful Sites and Interesting Results Figure 6 Translations of the terms ‘large table’ and ‘intermetallic compound’ from Enghsh into several languages using AltaVista’s program ‘Babelfish‘. Note that Babelfish makes several errors (*) w t h the technical term, but does rather better with the ordinary phrase
2.10 Newsgroups
There are >40 000 electronically accessible newsgroups that constitute another valuable means of acquiring information. Usenet (www.usenet.com) is a global bulletin-board system provided by your ISP for a fee that allows individuals with a common interest to share information. The information posted is accessible only for a finite time (days) following which it is
So far we have described the search process and some of the assists that may be used to get to what we are seeking about intermetallics on the Intemet. Now we tabulate some particularly useful sites and present some unusual findings. Admittedly, most of the hits resulting from any on-line search will be to print publications of one sort or another. This mode of access to printed papers may be cheaper or more convenient than conventional means - chemical abstracts, metals abstracts, or encyclopedic volumes like the present work - but the results are familiar and need no further discussion. Here we present examples of uncommon information, unlikely to be encountered
~ i s c e l l a n e o ~Topics s
868
in a conventional literature search, but accessible on the Internet with a few key strokes.
Access to the periodic table may be of assistance in understan~ingwhat p ar t i c u~ ~properties r of elei~ents are relevant to the formation of intermetallics and what their properties might be. Several sites present the table on screen and let you look up various properties of any element you click on. Some sites of this kind include:
hnology w w w . c s r r i . i ~ ~ . ~ ~ u /
I standard representations of the table
This subject is certainly the sine qua ROM of any study of intermetallic co ~p o u n ds .With 100 individual elements, there are 4950 binary systems and 161 700 ternary systems. Of these the diagranis for about 4000 binaries and 7380 ternaries are known, but not all of these have been gathered together in printed reference works and fewer still are directly available electronically. A search on ‘(phase diagram) + metal’ gave 63 800 hits, a number so great that most of these must be to the original literature. Fortunately, Coogle ranked these by their idea of order of importance. Thus the electronic all^ accessible collections appear near the top of the list. We found the following sites:
diagrams, evaluations, and references, not for all own binary systems. ganrath, Germany ted the thousands of systems M’s binary and ternary diagram rican Ceramic Society’s of these are sold as
sub-collections on or single systems are offered on-line STGE This cooperative European group sumniarizes their work on calc found at w w w . m ~ t ~ Red Books (www.~kos It is, of course, possible to use an engine to search for a single system. For example, a search with Coogle for ‘(phase diagram) + Ni-Ti-Si’, a system this author worked on many years ago, led not directly to a diagram but to a reference to the literature; the same report by DuPont et al. was keyed to a ~ e ~ a r t ~ofe n t Energy report, a Sandia Lab. Report, and an article in the Welding Journal. ~nfortunately, none of these could be called up on-line.
As DeVries showed us in ~ h a p t e r25 of Vol. 1 of this work, while elemental metals are relatively rare in nature, a large number of intemetallics are encouiitered among minerals. Here are two sites that offer images, descriptions, and data on minerals, including intermetallics, that may be searched in various ways:
to >20 sites. A color image of dycrasite, Ag,Sb, from this site is shown in Figure
To the best of my knowledge, there are no sites on the Internet devoted exclusively to the properties of i nt e~et al l i c com~ounds. onet the less some useful information in this category can be found. There is a . ~ o tabulates m) site called MatWeb ( w ~ ~ . m a t w ~ ~ that the properties of a large number of c o m ~ ~ r c i a l l y available materials - metals, ceramics, polymers and composites. To find the desired data you must know the commercial name of the material, or that given it by the producer or marketer, or an ASTM specification munber, industry standard number, etc. Materials in each broad category, say metals, are further broken down into sub-categories, ferrous, nonferrous, superalloys, and these latter further sub-divided into Nibase, Co-base, and Fe-base. Once at this point in the index, say Ni-base superalloys, one can find some inte~etallics,e.g. Tribaloy, Nitinol, etc. Under Cubase alloys one may find Muntz metal (p brass or CuZn); under ceramics, pnictides, chalcogenides,
869
Figure 7 A colored interinetallic mineral, dyscrasite (Ag,Sb), as shown by the Mineral Gallery (http://webmineral,c~m/ yc~asite.shtml).Individ~alprismatic crystals up to 8 mm in length. From Mine 21, Pribram, Czech Republic, reproduced with permission. See also Figure 7 (colour plate section) between pages 870 and 871
Figure 8 Some crystal structures from the Naval Research Laboratory's crystal structures page, organized by Pearson ~ -w w . n r l ~ ~ a v ~ . ~ ~ / l a symbol. ( h ~ p : / / c sw reproduced with permission. See also Figure 8 (co'lour plate section) between pages 870 and 871
Figure. 10 2D-drawing of the structure? approximately the (1 10) plane, of an inter~etalliccompound, NiAl, with I32 (cP2) structure showing four different types of point defects. ~http://~~fects.ph~sics.wsu.e-lattice-pic.html), reproduced with permission. See also Figure 10 (colour plate section) between pages 870 and 87 1
F i g ~ e9 Quasi-crysta~ research by Miraga at Tohoku University, Japan. A Kikuchi pattern showing five-fold rotational symmetry in an Al-Fe-Cu quasi-crystal (left figure) and a dodecahedra1 500-atom cluster of a structura~unit (the right figure) for an AI-Pd-Mn#icosahedral~quasi-crystal, ~ttp:/~hirag~b.i~r.toho~u.ac.jp/q~asi~rystal. See also Figure 9 (colour plate section) between pages 870 and 871
870
~ i s c e l l u ~ Topics e~~s
borides, and silicides. Another site for properties structure^ are shown in color at a site built by the (US) data, again with only a scattering of intermetallics is Eagle International Software ( w w w . r n ~ t a ~ ~ ~ 0 . c ~ ~ ) with easy access to metal standards, property data, son symbol. An example is shown in Figure 8, noth her a glossary, posting forum for technical questions, conipany profiles, etc.
Some inte~etallicsmay be purchased directly from their ~ a n ~ ~ f a ~ort ~a rc~emical er supply house and need not be synthesized in the laboratory. Some sites of this type include:
e.g. aluminides, phosphides, carbides, and e.g. AlAs, As,Zn,, Bi,Te,,
e.g. ZnAs2, PbAs, Ag,Se, TlSe, a,Te, GaSb, Cd3Sb,
devoted to structural data of elements and intermetallic phases. These comprise: Vol. XXVIII, Group HI,% (1971), and its supple~entsXI1/14, sub vols. a bl (1986), and b2 (1986). They may be found at See also the c ~ ~ t ~ l o g r a ~ ~ section of the www Virt
One of the fascinating and totally unexpected events in the intermetallic field was the discovery by Shechtman et al. (1984) that certain IMCs exhibit a new phase of condensed matter with non-crystallographic symmetry and quasi-periodicity (see also Chapter 20 by Kelton in Vol. I of this work). An excellent site providing an introductory tutorial to quasi-crystals, links to reviews, software, and current research groups may be found at
crystal images OEthe Net, see Figure 9.
Cr2Te3,DySi,, LaNi5, Mn2Sb, Mo,Al, NbSi,, e.g. alurninides, borides, silicides and > 60 intermetall~cs .g. BaAl,, Cd3As2,CeSi,, Cs,Bi, rAl,, LaSb, NiAl, Rb3Sb, TmAs, information on and products for joining difficult-to-bond materials reparation of alkali (or pyrophoric, poisonous, or other dangerous) mation of an IMC that is subnulated and encapsulated
More than 90% of all known binary IMCs adopt less than 100 crystal structure types (see Villars, Chapter 11 of Vol. 1 of this work). The great majority of these
The intermetallic work at such universities as Carnbridge, MIT, and Genoa is well known. We list here some not-so-well-known, but active in IMCs, stumbled upon while surfing the Internet:
I n t e ~ ~ i e ~ a lon l i ethe ~ Internet
871
U. of Texas at Austin, Dept. of Chemistry U. of the ~ a l e a r i cIslands Washington State University ~~
3.
es
In our own surfing of the Net we have run across many striking images of intermetal~cs,some of which we reproduce here to show the variety that is available. Some are drawings, some are still photos and some are even animated. Fig 10 is a drawing of a plane of atoms in In-doped MA1 showing different defect structure types: three Ni vacancies, one A1 vacancy, one substitutional defect - a Ni atom in an anti-site position, and one site where an In atom occupies an AI position. Figure 11 is a view of a grain boundary in a boron-doped Ni,Al sample as seen in the ORNL field ion microscope. Figure 12 is a HREM image of a superlattice dislocation in CoTi (€32, cP2). The image on the right has been improved by noise filtering. The result of a computer simulation of the martensitic transformation of Ni-rich from a recent doctoral thesis at U. of Duisburg. may be
Figure 11 Field ion micrograph of boron-doped Ni,Al. The bright dots are iiidividual boron atoms that have segregated to a grain boundary (arrows). Photo from work funded by the US Dept of Energy at the Oak Ridge National Laboratory; see the OliML Review, 28 (4)’ (1995). (~~p://www.ornl.gov/ ~ ~ t e x t / a ~ o ~ s . h tAI1 m ) . rights reserved, reproduced with permission
Figure 12 Lattice image (left) of a superlattice edge dislocation by HREM in a BZtype inter~etallic(CoTi) and the same image (right) as asme7.ia~p~toho~u,a~.jp~E ~i6-~ho~.~tml)
High Temperature ~ a t e r i a l sInfor~at i onAnalysis
As promised at the outset, what has been presented is but a sampling of the myriad of riches on i ~ t e r m e t ~ l l i ~ s available on anyone’s computer with a few clicks of the mouse. We hope that it is sufficiently intriguin~to encourage each reader to explore for him- or herself. We would be interested to learn of any errors or of findings of new sites that should become more widely known. One last caveat: While all URLs cited in the chapter were accessible via the address shown at the time of writing, a~dressesfrequen~lchan~e. ~ Some addresses require the http:// prefix; others do not, or the search engine automatically supplies it. In case of d i ~ c u l t yin accessing a site, try t r u ~ c a t from i ~ ~ the
872
~ i s c e l l a n e o uTopics ~~
right the address as shown until a response is obtained; then follow hyperlinks to the desired item.
home page a Web document, intended as the introductory ‘page’ for a subject individual or organization, contain in^ hyperlinks to related information about the subject html HyperText Markup Language http HyperText Transmission Protocol IA4C Intermetallic compound. The ~ n c ~ c l o p ~ e d i a for a guide to 675 on-line ~ritannica( w w w . e ~ . c o ~provides ) the following dictionaries. Definitions of terms in ‘netspeak’ landefinition: any o f a class of substances ~ o ~ ~ ~ofo ds ~e ~d n i t e ~ u b / g n udirectory; the gz file proportions of two or more metals, rather than . Other glossaries are located at continuously variable proportions (as in solid r e sthe propers o l ~ t i o n .s The ~ crystal ~ t r ~ c t ~and 0 . ~ are 0 ~ pa~icularlygood for ties of intermetallic c o ~ p o u n d soften d$fep. ~ a r l ~ e dfrom l y those of their constituents. In a d d ~ t ~ oton the ~ o r ~ valences a l o j their components, the relativ~sizes o f t h e atoms and the ratio Roget ~ h e s a u may r ~ be found at w w w . t ~ ~ s ~ ~ ~ . c ~ ~ . of the total number of valence electrons to the total The definitions which follow are those for the number of atoms have ~ i ~ p o r t efjCects a ~ t on the italicized words in this chapter. com~ositionof interinetallic ~ o ~ p o u n d s Internet (often just ‘the Net’) a worldwide system b o o l ~ m a ra~means of noting a Pwored site for later linking (via wire, fiber optic cable, or wireless instant recall with a single click, rather than the transmission) smaller computer networks tomore complex sequential means by which one gether with the aid of a communication standard first found it (also known as ‘Favorite Places’ in TCP/IP (Transmission Control Protocol/Internet some systems) this means a client (requesting ~ o o l e a nlogic a system of operations performed on onnected to a serv~r(a computer binary variables, called Boolean variables, after program making information available to other George Book (18 1 5-1 864). These variables programs or computers) which then transmits the , OR, and NOT, together with NAND, requested i nfor~at i onback to the client , and XOR, may adopt only two values, ISP Internet service provider: an agent providing true and false, that can be represented by the computer access to the Internet for a fee of binary digits 1 and 0. $5-$20 per month. A compre~~nsive listing of brows~nguse of special softw~re(a browser) to search ISPs around the US may be found at: the Internet for information of interest w ~ wthelist.com . click a procedure in which you place the mouse ineta-search engine an engine that simultaneo~sly pointer on an item of interest and click the mouse multiple browsers, each button once (left, if two buttons) cope, etc. Some limit the data mirzing originally used in the broad sense of ne or the total number of searching computerized files for desired inforunique hits mation; now used more narrowly to mean inetasite a web site that includes all (most?, many?) seeking unknown relationships or patterns in a important URLs within its subject field body of data m o ~ e ma modulator/demodulator between the digital FAQ fre~uentlyasked questions (with answers) language of the cornputer and the analog languge ,ftp file transfer protocol. A public access Internet of the telephone system site, essentially a standard, making it possible to mouse a palm-sized control unit contain in^ a ball on download information from a remote computer its lower surface and buttons on the upper. via the Internet to your own computer Motion of the ball positions the cursor on the Gopher a public access system for browsing menus of computer screen and pressing the bu ~~ierarchi~ally organized in~ormation vokes various actions such as runnin hit successful location of a site containing the word, or opening files words, or phrases (with or without Boolean NTIS National Technical I~formationSystem (US) operators) specified by the target search terms
I n t ~ r ~ e t a l l i cons the Internet on-line when your coinputer is connected to the Net via an on-line service, bulletin board system, or public access site on-line database a searchable collection of related information search engine a software program that searches the Internet for keywords in files and documents. Depending on the particular engine, the search may cover titles of documents, URLs, headers, or the full text surfing negotiating the Interiiet via known URLs or with the aid of a browser or ‘meta’ search engine, often with little or no predetermined plan of acti 011 URL Uniform Resource Locator or Web address W A I S Wide Area Informatio~Service WWWWorld Wide Web or simply the Web, the universe of network-accessible information.
Alper, J. (1998). Science, 2 Chen, M., Hearst, M. A., Hong, J., and Lin, J. (1999). Proc. 2nd USENIX Symp. on Internet Technologies and Systems,
873
Oct. 1999 (in press) (see: http://citeseer.nj.ilec.com/ chen99chacha.html). Fayyad, U, M,, Piatetsky~Shapiro, G., Smyth, P., and Uthurusamy, R., eds (1996). Advances in ~ o ~ l e d g e Discovery and Data Mining, MIT Press, 625 pp. Gavacs, J. (2000). R&D ~ a g a z ~ ? zApril, e , E17. Han, J-W., and Karnber, M. (2~00).Data ~ i n i ~ gConcepts : and Techniques, Kaufmann, 500 pp. Hearst, M. A. (1997). Scientijic American, March issue. Available online ( ~ ~ ~ . s c ineering ~ u l l e t i iSept., ~, Hearst, NE. A. (2000). IEEE special issue on Next Geiieratioii Web Search. Lawrence, S., and Giles, C. L. (1999). ~ a ~ u395. ~ e , Meltsner, I(.J. (1995). J. o ~ ~ e t a l s , Nadis, S. (1996). Science, 27
Ryabinin, M. A., and Ekonomou, 7‘. (1997). RIAR Proc. Dmitrovgrad, Issue 3, 93-99. Shechtman, D., Blech, L., D., and Cahn, J. W. (1984). Phys. Rev. Lett., Thomas, B. J. (1998). The ~ o r ~ d ~ Webfor i d e ~cienri,stsand Efzgzneers,SPIE Press, IEEE Press, ASME, SAE I d . , and IEEE, pp. 141-165, 327-334.
Knowledge is of two kinds. e know a subject ourselves, where we can find i n f o r ~ a t i o nupon it, s a rule . . . he who has the most i n f ~ r m a t i owill ~ have eli the greatest success in life, ~nformation is knowledge. To work for business, government, education, or industry, it has to be o control informati is to store it, retrieve it, or display it. nd that requires the creation of functional innovations. innovations interact, a system is created that makes information work. An information system. - from a Be adv~rtisement,1980
This Page Intentionally Left Blank
Aaronson, H.I. 1: 762, 765; see also Hehemann, R.F et al.; Lorirner, G.W. Aarts, J. see Steglich, F. et al. Abbaschian, R. see Schlesinger, et al.; Xiao, L. and Abbaschian, R.; Zliu, N.X. Abbundi, R. 2: 382, 393, 394 Abdel-Hamid, A.A. 1: 643 Abdellaoui, M. et al. 3 730 Abe, E. et al. 3: 411 Abe, M. see Koon, N. C. et al. Abe, M.see Yamakoshi, S. et al., Yamaoka, T. et al.; Yokoyama, N. et al. Abe, M. et al. 2: 335 Abe, 0. 1: 651 Abe, T. see Hashimoto, H. et al. Abe, Y. see Nagasawa, A. ef al. Abegg, R. 1: 12 Abell, J.S. see Bi, Y.J. et al. Abell, J.S. et al. 2: 398 Abelson, J.R. see Park, D.G. et al. Abraham, F. 1: 612 Abrahain, T. see Rittner, M.N. and Abraharn, T. Abrahams, E. see Palmer, R. et al. Abrikosov, N.Kh. see Morokhovets, M.A. et al. Abromeit, C. 1: 559 Abstreiter, G. see Brunner, K. et al. Abuniya, T. see Kdwabata, T. et al. Achard. J.C. see Percheron-Gukgan, A. et al. Achete, C. see Niehus, H. and Achete, C. Achiwa, N. 3: 803 Acker, F see Giorgi, A. L. et al. Ackermann, H. 2 388; see also fnden, G. et al. Ackland, G.J. 1: 88, 598, 600, 601 3: 201; see also Vitek, V. et al. Wallace, W.E. and Acklaiid, G.J. Ackland, G.J. and Vitek, V 3: 447, 766 Adachi, G.Y. et al. 3: 799 Adachi, H. see Morinaga, M. et al. Adachi, K. 3 803 Adani, C.M. 2 175, 176, 178; see also Brown, A.M. et al., Das, S.K. et al.: Kim, N.J. et al. Adam, E. see Sanger, P. et al. Adam, E. et al. 2: 360 Adams, D. see Rozgonyi, G.A. et al. Adams, J. see Flukiger, R. et al. Adams, N. see Hill, P.J. et al. Adams, R.A. see Fowler, P.H. et al. Adamson, G.M. 2: 523 Adamson, R.B. see Yang, W.J.S. et al.
Adamson, R.J. 2: 515 Adaptive Control Systems 2: 403 Adcock, F. 1: 98, 99 Addison, R.C. Jr. see Cox, B.N. et al. Adelerhof, D.J. see de Reus, R. et al. Adey, J.M. 2: 583, 585, 586; see also Mahler, D.B. tit al. Adibi, F et al. 3: 666 Adkins, N.J.E. see Zuttel, A. et al. Adrianovskii, B.P. et al. 3: 422 Adroja, D.T. 1: 217 Aebi, P see Pilo, Th. et al. Aebi, P et al. 3: 139 Aebischer, H. 2: 477 Aebli, E. see Blau, B. et ul. Aeppli, G. see Goldman. A.I. et al.; Lacerda, A. et al.; Mason, T.E. et al. AfIolter, K. see Kattelus, H.P. et al. Afyouni, M. see Venndgues, P. et al. Agarwal, A. see Rockett, A. et al. Agarwal, B.K. 3: 142 Agarwal, K.L. see Baldi, R.W. et al. Ageev, N.V. 1: 712 Ageeva, C.N. see Khalim, A.A.R. et al. Agrawal, R.C. and Gupta, R.K. 3: 250 Agyernan, K. et al. 1: 735 Ahlbehrendt, D. see B l u , R.P. et al. Ahlman, R. see Carnpisi, I.E. et al. Ahniadieh, A. 1: 917 Ahmed, A. see Sellmyer, D.J. et al. Ahmed, M. 1: 649 Ahrens, L.H. 1: 243 Ahrens, T. et al. 2: 186 Ahrens, T.J. see Schwarz, R.B. et ul. Ahzi, S. 3: 371 Ahzi, S. see Schoenfeld, S.E. et al. Aidelberg, J. 1: 797 Aiken, R. 1: 977, 1000, 1001, 1003, 1008 Aiken, R.M. 3: 486 Aikin, B.J.M. et al. 3 657 Aindow, M. see Kazantzis, A.V. et al.; Shmg, P. et al.; Shyue, J. et al.; Smith, L.S. et al. Ainger, F.W. see Li, J. et al. Ainsworth, P.A. 2 654 Aitchison, R. 1: 3 Aitken, E.A. 1: 912, 927, 997; 2: 17, 507 Aitken, E.A. and Smith, J.F. 3: 44 Aitkin, R.M. Jr. 3: 574 Aitov, R.G. 2: 502 Aizawa, T. see Thadhani, N.N. and Aizawa T.; Yen, B.K. et al. Aizawa, T. and Tokumitu, K. 3: 728 Aizawa, T. et al. 3 728 Ajioka, T. see Tachikawa, K.Y. et al. Ajisawa, A. see Kornatsu, IS.et al. Akaishi, M. see Kagamida, M. et al.
Akayama, M. see Fujii, H. et al. Akhtar, A. and Teghtsoonian, E. 3: 74 Akhtdr, D. 1: 743 Akimoto, K. et al. 2: 326 Akinc, M. see Meyer, M. K. and Akinc M.; Meyer, M. K. et nZ. Akinc, M. et a/. 3: 486, 490 Akita, S. see Tsuji, H. et al. Akselrud, L.G. see Kalychak, Ya.M. et al. Akuezue, H.C. 1: 923 Alamgir, F.M. see Jin, 0. et al. Alamgir, F.M. et al. 3: 689 Alamo, A. see Doyama, M. et al. Alamo, A. et al. 1: 560, 567, 568, 573, 574, 578, 794, 795, 796 Alan, R.P. see Williams, D.B. et al. Alatalo, M. et al. 3: 287, 288 Alben, R.S. see Sundaram, V.S. et al. Albers, R.C. see Alouaiii, M. et al.; Asta, M.D. et al.; Chen, S.P. et al.; Voter, A.F et al.; Wang, C.S. et al. Albers, W. 2: 654 Albert, B. see Posgay, G. et al. Albert, D.E. 2 115 Albert, D.E. and Gray, G.T. 3: 368, 411, 419 Albertini, F. see Ibarra, M.R. et al., Morellon, L. et al. Albin, D.S. see Mooaey, G.D. et al.; Tuttle, J.R. et ah. Albin, D.S. et al. 2: 423 Albrecht, J. sec Lutjering, G. et al. Albright, S.P, et al. 2: 423 Alcock, C.B. 1: 95; 2: 642 Alcock, C.B. et al. 1: 681 Alcock, C.B. see Kubaschewski, 0. et al. AlcouEe, G. see COUJOU, A. et al. Alder, B.J. 1: 128 131 Aldinger, F. 2 584 Aldinger, F and Petzow, G. 3: 49 Alefeld, G. 2: 476, 484 Alekeseev, V.1. 1: 69 Aleksandrov, B.N. 1: 955 Aleksandrov, B.N. et al. 1: 956 Alekseeva, M.A. see Shcherbakov, A.S. et al. Alexander, D.J. see Maziasz, P,J. et al. Alexander, D.J. et al. 3: 655 Alexander, H. 1: 917; see Cox, G. et al. Alexandrov, 1.V see Valiev, R.Z. et ul. Alexiades, V. 2: 642 Alferness, R. see Shani, Y et al. Alford, T.L. see Ranianath, G. et al. Algarabel, P.A. see Ibarra, M.R. et al.; Morellon, L. et al. Alheim, U. see Fraas, IS.et al.
876 Ali, N. et al. 1: 484 Alisova, S.P. 1: 724 Alisova, S.P. et al. 1: 721, 722, 723, 724 Allaverdova, N.V. 1: 983, 987 Allemand, J. et al. 3: 100 Allen, C.E. 3 805 Allen, C.W. et al. 1: 815; 3: 671, 674 Allen, C.W see Birtcher, R.C. et al. Allen, J.L. see Lee, M.C. et al. Allen, J.M. and Whitlow, G.A. 3: 304 Allen, J.W 1: 218; see also Seaman, C.L. et al. Allen, J.W. et al. 1: 218 Allen, L.H. see Bergstrom, D.B. et al., ~ d m a n a t h G. , et al.; Wang, S.Q. et al, Alien, R.E. and de Wette, F.W 3: 220 Allen, S.J. Jr. see Palmstrom, C.J. et al. Allen, S.M. 1: 35, 850, 852, 853, 865, 866; 2: 199, 201; see also Liu, Y. et al. Allen, W.P. et al. 1: 860 Alleno, E. see Tominez, E. et al. Alley, G.D. 2: 335; see also Vojak, B.A. et al. Alley, G.D. et al. 2: 335 Alley, P. 1: 944 Allia, P. see Caciuffo, R. et al. Allibert, C.H. and Pastor, H. 3 841 Allison, J. see Dowling, W. et al. Allison, J.E. see Dowling, W.E. Jr. c’r al.; Hartfield-Wunsch, S.E. et al. Allman, D.J. see Allen, C.W. et al.; Ramanath, G. et al. Alman, D.E. 2 298; see also Korinko, P.S. et aE. Alman, D.E. et al. 1: 391 Almazouzi, A. see Nonaka, K. et al. Alonso, J.A. 1: 244, 247, 682, 683. 684; 2 610; see also Gallego, L.J. et al. Alonso, T. et al. 3 756 Alouani, M. et al. 1: 60, 874, 876 Alpas, A.T. see Ding, Y et al. Alper, J. 3: 864 Alterovitz, S.A. see Oh, J.E, et al.; Park, D.G. et al. Altounian, Z. see Liao, L.X. et al.; Strom-Olsen, J.O. et al. Altstetter, C.J. see Chang, H. et al. Alvdrez, J. see Castro, C.R. et al. Alven, D.A. and Stoloff, N.S. 3: 338, 339 Aly, S.H. see Cadieu, F.J. et al. Al-Yasiri, L.H. see Nicholls, J.R. et al. Amador, C. et al. 1: 41 Aniako, Y . see Courtois, D. ~t al. Aman, Y. see Sato, K. et al. Amat di San Filippo, P. 1: 7 Amaya, K. see Kobajashi, T. et al. Amazigo, J.C. see Budiansky, B. et al. Amelinckx, S. 1: 521; see ul,so Delavignette, P. et al. Arnemiyd, N. 2: 382 Ames, 1. et al. 2: 654 Anieyanam, K. et al. 3: 762 Amiotti, M. et al. 3: 235 An, S.U. see Seo, D. et al., Seo, D.Y. et al. Aiiantha Swamy, N.K. 2: 309 Anantharaman, T.R. 1: 453
Author Index Ananyn, V.M. et al. I: 594 Anderko, K. 1: 101; 2: 309, 578, 589; see also Schubert, K. et al. Andersen, H.C. 1: 479 Anderson, I. see Zimm, C.B. et al. Anderson, I.M. et al. 3 288 Anderson, K. see Sikka, V.K. et ol. Anderson, M.U. see Thadhani, N.N. et al. Anderson, M.U. et al. 3 738 Andersen, N.H. et al. 3 250, 251 Andersen, O K . 1: 58, 60, 83, 114, 133, 198, 960; see also Meth~essel,M. et al. Anderson, C.D. et al. 2 84, 85 Anderson, E.H. see Erdogan, T. et al. Anderson, C.see Titran, R.H. et al. Anderson, K. see Sikka, V.K. et al. Anderson, O.L. 1: 883, 885; see also Schreiber, E. et al. Anderson, P.W. 1: 213; see also Palrner, R. et al. Anderson W.T. Jr. see ~apanicolaou, N.A. et al. Anderson, W.T. Jr. et al. 2: 626 Anderson, S. et al. J: 275 Anderson, S A . and Lang, C.I. 3: 55 Anderson, T. see Kryliouk, 0. et al. Anderson, Y.R. see Bao, Z. et al. Anderson, J.-0. et al. 3: 813, 817 Andler, S. see Trautmann, C. et al. Ando, K. 2: 330 Ando, S. see Inui, H. et al. Ando, T.et al. 2: 376 Andoh, El. see Minemura, T. et al. Andre, J.P. see Bellon, P et al. Andreatch, P. see McSk¶min, H.J. and Andreatch, P Andreenko, A.S. see Nikitin, S.A. et al. Andreoni, W. et al. 1: 242, 243 Andreoni, W. see KohanoR, J. et al., Wengert, S. et al. Andres, K. 2: 654; see also Bucher, E. et al. Andres, K. 3: 522 Andres, K. and Darack, S. 3 522 Andresen, A.F. et al. 2 309 Andrews, J. see Hu, Y.Z. et al. Andrews, M.R. 1: 10 Andrews, P V. et al. 1: 956 Andreyenko, A.S. see Nikitin, S.A. et al. Andreyeva, L.P. see Balina, Ye.A. et al. Andreyshikov, B.M. see Ryadchenko, V.M. et al. Andrieux, J.L. 2: 506 Anfiteatro, D.D. ‘seed’Heurle, F.M. et al., Finstad, T.G. et al. Angelini, P. see Liu, C. et al.; Liu, C.T. et al. Angelo, J.E. see Baskes, M.I. et al.; Mediin, D.L. et al.; Mills, M.J. et al. Angelo, J.E. et al. 3: 766 Angers, L.M. see Ayer, R. et al.; Hawk, 5.14. et al.; Wilsdorf, H.G.E. et al. Angers, L.M. et al. 2: 185, 187, 188 Angilello, J. see Heiney, P,A. et al.; Olowolafe, J.O. et al., Thompson, R.D. et al. Angles, R.M. see Parkinson, N. et al. Anisimov, V.I. see Greenberg, B.A. et al.
Anisimov, V.I. et al. 1: 540 Anlage, S.M. et al. 3: 78 Annaorazov, M.P. et al. 3: 527 Annaorazov, P.P. see Nikitin, S.A. et al. Annapurna, J. 1: 994; 2: 167, 168 Anongba, P.N.B. see Steinemann, S.G. e f al. Anongba, P.N.B. and Steinemann, S.G. 3 241 Anoshkin, N. see Bondarev, B et al. Anosov, V.Ya. et al. 1: 712 Ansara, 1. 3: 813, 917 Ansara, I. et al. 3 841 Anselmi~Tamburini,U. 1: 646; see also Bertolino, N. et al. Munir, Z.A. and Anselmi-Tamburiai U. Anstead, R.J. see Hunter, W.R. et al. Anstis, G.R. see Chou, C.T. et al. AntclifFe, G.A. 2: 328 Anthony, L. 1: 856 Anthony, L. et al. 3: 199, 201 Anthony, T.R. 1: 191, 760, 764, 765 see De Bussac, A. et al. Anton, D.L. 2: 10,225,241,244,245,246, 247, 249, 294, 295, 298, 650; see also Giama, A.F et al.; Miracle, D.B. et al., Shah, D.M. et al., 3: 373, 440, 492 Anton, D.L. see Giamei, A,F et al.; Lee, T.-S. et al.; Lee, T.S. et al.; Shah. D.M. et al. Anton, D.L. and Shah, D.M. 3: 486, 487 Anton, D.L. et al. 3: 501, 552 Antonopo~lous,J.G. et al. 1: 590 Antonova, O.V. see Greenberg, B.A. et al. Antreasyan, A. see Temkin, H. et al. Antula, J. 2: 510 Aoki, A. 1: 591 Aoki, K. 1: 85, 564, 896, 899, 924; 2 29, 30, 38, 39 Aoki, K. and Izumi, 0. 3: 492, 604, 843 Aoki, K. et al. 1: 695; 2: 479 Aoki, M. 3 : 434 Aoki, M. see Pettifor, D.G. and Aoki, M. Aoki, T. see Sawada, H. et al. Aoki, Y. see Takabatake, T. et al. Aoyagi, M. see Takai, M. et al. Appalonia, D. see Meier, G.H. ef al. Appel, F see Cliatterjee, A. et al., Imayev, R.M. et al.; Oehring, M. et al.; Paul, J.D.H. et al.; Wagner, R. et al. Appel, F. and Wagner, R. 3: 297, 361, 405,413,419,420,618,621, 622 Appel, F. et al. 3 275, 368, 413, 618, 619, 621,624,626,627,628,430,634,635 Appelbaum, A, 2: 618 Appelbaum, A. et al. 2: 617, 618, 624 Aptecar, I.Z. et al. 3: 155, 160 Arai, K. see Osaka, T. et al. Araj, S. see Jaccarino, V. et al. Arqs, S. et al, 1: 056 Arayashiki, T. see Nonaka, K. et al. Arbman, G.O. 1: 199
Arch, D. see Stassis, C. et al. Archuleta, J.J. see He, Y. et al. Ardakani, M.G. see Basoalto, N.C. et al.
877 Ardell, A.J. 1: 817; 2: 258, 259, 260, 261, 262,263,265,271,272,274,279,281, 282; see also Chellman, D.J. et al., Chen, F.C. et al.; Cheng, J. et al.; Li, H. et al. Ardell, A.J. et al. 2 268, 271, 273, 274, 280, 281, 283, 284 Ardell, J. 1: 809 Argon, A.S. 1: 743, 917 Argon, A.S. see Kocks, U.F. et al. Argon, A.S. and Haasen, P. 3 362 Argon, A.S. and Kuo, H.Y 3: 693 Argoud, R. see Obbade, S. et al. Argyres, P.N. 2: 439, 440 Arita, M. see Koiwa, M. et al. Arjona, F. see Fatas, E. et al. Arkad’ev, A.G. and Braverman, E.M. 3: 822, 835 Arkharov, V.I. et al. 1: 955 Arklarov, A.M. see Nikitin, S.A. et al. Arko, A.J. see Campuzano, J.C. et al., List, R.S. et al.; Olson, C.G. et al. Arlt, G. 1: 174, 178, 180, 183 Armstrong, R.D. et al. 2: 506, 508 Aimstrong, R.W. 3: 408 Arnaud d’Avitaya, F. Arnberg, L. see Inoue, A. et al. Arnhold, V. 1: 575 Arnold, R.see Lugscheider, E. et al. Arnold, 21. see Morellon, L. et al. Aronin, L.R. 1: 803 Aronson, M.C. see Demczyk, B. et al. Aronson, M.C. and Coles, B.R. 3: 37 Arrot, A. see Toth, R.S. et al. Arrott, A. 1: 440 Arsenault, R.J. 2 290, 298 Artaki, 1. see Ray, U. et al. Arthey, R.P. see Hill, A.D. et al. Artz, E. and Grahle, P 3: 494 Artz, E. et al. 3: 647, 656 Arunachalam, V.S. 1: 845 Arunachalam, V.S. see Shi, X. et al. Arunachalam, V.S. and Sargent, C.M. 3: 405 Arunachalam, V.S. and Sundaresan, R. 3: 644 Arzt, E. 1: 918, 925; see also Whittenberger, J.D. et al. Arzt, E. see Schaefer, H.-E. et al., Schropf, H. et al. Arzt, E, et al. 1: 918 Asachi and Sasao 3: 805 Asada, T. et al. 3 56 Asahi, H. see Tanaka, H. et al.; Wakita, K. et al. Asahi, M. et al. 2: 336 Asai, T. see Mabuchi, H. et al. Asanabe, S. et al. 2: 329 Asano, H. see Hirabayashi, M. et al. Asano, T. see Tachikawa, K. er al., Tachikawa, K.Y. et al. Asanum, N. see Taniguchi, S. et al. Asaro, R.J. see Barnett, D.M. et al.; Dao, M. et al.; Kad, B.K. and Asaro, R.J., Kad, B.K. et al.; Schoenfeld, S.E. et al. Asatryan, ISA. see Annaorazov, M.P. et al.; Nikitin, S.A. et al. Asaumi, K. et al. 3: 161 Asayama, K. see ISyogaku, M. et al.; Nakamura~H. et al.
Ascencio, M.C. see Castro, G.R. et al. Ashary, A. et al. 1: 999 Ashby, M.F. 1: 913, 916, 917, 918, 919, 925; 2 241, 268, 269, 608; 3: 647; see also Artz, E. et al.; Helle, A.S. et al.; Kocks, U.F. et al. Ashby, M.F. et al. 1: 918 Ashcroft, N.W. 1: 200, 207, 663, 1018 Ashcroft, N.W and Langreth, D.C+3: 245, 247 Ashenford, D.E. see Duddles, N.J. et al. Asher, H.H. 2: 562 Ashida, A. et al. 3: 672 Ashida, K. see Nakano, T. et al. Ashida, Y. see Futjisuna, N. et al. Ashok, S. see Kuruvilla, A.K. et al. Ashraff, J.A. et al. 1: 161 Askenazy, P. et al. 1: 701 Aslanidis, I. see Vedula, K. et al. Aslaiiov, L.A. 1: 366, 381 Asner, A. see Wenger, S. et al. Assmus, W see Cornelius, A.L. et al.; Sievering, A. et al. Ast, D.G. see Lilienfeld, D.A. et al. Asta, M. et al. 1: 41; see Wolverton, C. et al. Asta, M.D. see Althoff, J.D. et al., Tepesch, P.D. et al.; Wolverton, C. er al. Asta, M.D. and Foiles, S.M. 3: 201 Asta, M.D. et al. 3: 193, 205 Astbury, I: 309 Astle, M.J. see Weast, R.C. et al. Asundi, M.K. see Bhanumurthy, K. et al. Aswath, P.B. see Soboyejo, W.O. et al. Aswath, P.B. and Suresh, S. 3: 337 Aswath, P.B. et al. 3: 337 Atalla, M. and Kahng, D. 3: 787 Atkinson, R. 2: 448 Atobe, Y. see Sawada, H. et al. Atou, T. see Yamasaki, T. et al. Atrei, A. et al. 3: 219 Attaran, E. 2: 449 Atzmon, M. 3 758 Atzmon, M. et al. 1: 734 Atzmon, M.J. 2 599; 3: 727 Atnnony, U. 2 394; see also Klimker, H. et al.; Rosen, M. et al. Atzmony, U. et al. 2 393, 394 Au, P. see Beddoes, J. et al. Au, Y.K. 2 54 Aubauer, H. 1: 782; see also Veith, G. et al. Audier, M. 1: 454,461,465,471,472,475, 482, 483; 2 180, 186; see also Dong, C. et al.; Guyot, P. et al.; Launois, P. et al. Audouard, A. 1: 745 Audouard, A. see Defour, C. et al. Audouard, A. et al. I: 821; 3: 267 Auerbach, A. 1: 214 Augarde, E. see Bellon, P. et al. Augis, J.A. 2: 507 Augustine, M. see Campisi, I.E. et al. Auld, J.H. 2 282 Auleytner, J. see Sobczak, E. and Auleytner J. Auran, L. see Westengen, H. et al. Aust, K.T. 1: 955 Aust, K.T. and Westbrook, J.H. 3: 211 Austin, C.M. 2: 83, 87, 88
Austin, C.M. and Kelly, T.J. 3: 592, 601, 648 Austin, C.M. et al. 3: 483, 598, 601 Aver, W see Eggeler, G. et al. Averback, R.S. 1: 705; see also Barbu, A. et al.; Caro, A. et al.; Hoshino, K. et al.; Nastasi, M. et al. Averback, R.S. see Barbu, A. et al.; Chang, H. et al. Averill, F.W. 1: 907 Avrami, M. 1: 773 Axe, J.D. 1: 162, 163; see Shirane, G. and Axe, J.D. Ayache, C. see Thomas, F et al. Ayada, M. see Taguchi, K. et al. Aydin, M. see Pctersen, J.F. et al. Aydinli, A. see Compaan, A. et al. Ayer, R. et al. 2: 178, 179 Aylesworth, K.D. see Oh, J.E. er al. Ayres de Campos, J. et al. 3: 99 Ayres de Carnpos, N. see Ayres de Campos, J. et al. Ayushina, G.D. et al. AZIZ,M. 1: 772, 773; see also West, J. et al.
Baba, K. see Flanagan, T.B. et al. Babbitt, I. 2: 592 Baburaj, E.G. see Khina, B.B. et al. Bacalis, N.C. see Sigalas, M. et al. Bacanov, S. 3: 160 Baccino, R. and Moret, F. 3: 648 Baccino, R. et al. 3: 648 Bachelet, G.B. see Manghi, F. et al. BBchler, M. see Ziittel, A. et al. Bachorczyk, R. ,Tee Danielewski, M. et al.; Datta, P.K. et al. Backman, D.G. 1: 137 Backofen, W.A. see Paton, N.E. and Backofen, W.A. Bacmann, M. see Ayres de Campos, J. er al. Bacon, D.D. see Sherwood, R.C. et al.; Testardi, L.R. et al. Bacon, D.J. see Barnett, D.M. et al. Bacon, D.J. et al. 1: 524; 2: 268; 3: 392, 394 Bacon, G.E. 1: 441, 445, 935 Badcock, C.C. 2: 509, 511 Badding, J.V. 3: 724 Bader, S. see Orent, T. et al. Badoz, P.A. see Rosencher, E. et al. Badrinarayanan, K. et al. 3 344, 346, 347 Badura, K. see Brossmann, U. et al.; Kumrnerle, E. et al. Badura, K. and Schaefer, H.-E. 3: 281 Badura-Gergen, I(.3: 280, 28 1, 282; see alAsoWurschum, R. ct al. Badura-Gergen, K. and Schaefer, H.-E. 3: 279, 281 Baenziger, N.C. 1: 413 Baer, Y, see Grioni, M. et al. Baeri, P et al. 3: 664 Baeslack, W.A. I11 1: 656; 2 128; see also Cieslak, M.J. et al. Baeslack, W.A. 111 et al. 2 128 Bafi, F. see Ferro, R. er al. Baggio-Saitovitch, E. see Xia, S.M. et al. Baginski, W.A. see Regan, R.E. et al.
878 Bagley, B.G. 1: 465 Baglin, J.E.E. 2: 608 Baglin, J.E.E. et al. 2 616 Bagno, P et al. 1: 130, 132 Bahadur, D. see Dunlap, R.A. et al. Bahn, S. and Schubert, K. 3: 236 Bai, B. see Collins, G S. et al. Bailey, P.J. see Paunovic, M. et al. Bailey, R.E. see Baldi, R.W. et al. Baillargeon, J.N. et al. 2 424 Bain, E.C. 1: 10, 11, 827, 828, 835; 2: 562 Bam, K. 2: 65; see also Wright, P.R. et al. Bain, K, see Wright, P. et al. Bain, K.R. 2: 292 Bain, K.R. et al. 3: 320 Bains, G.S. et al. 2: 443, 449 Baird, H.W 2: 576 Baird, J.D. 1: 768 Baird, R. et al. 1: 615 Bak, P. 1: 459, 480 Baker, A. see Fang, J. et al. Baker, A.D. see Brundle, C.R. and Baker A.D. Baker, C. 3: 600 Baker, D.R. see Frasrer, F.R. et al.; Kestner-Weykamp, H.T. et al. Baker, H. 3: 801 Baker, H. see Massa~ski,T.B. et al. Baker, H.H. see Verhoeven, J.D. et al. Baker, I. 1: 522, 535, 536, 539, 591, 594, 655,864,9Q2,903,913,920.921,924; 2: 157,202,205,206,237; 3: 309, 31 1, 312, 368, 373, 772; see also Nagpal, P et al., Schmidt, B. et al.; Schulson, E.M. et al. Baker, 1. and George, E.P 3: 613, 614 Baker, I. and Nagpal, P. 3: 361, 368, 372 Baker, 1. et al. 1: 522, 529, 586, 587, 902, Baker, J.M. see ~ u r a k ~M.i ,et al. Baker, M.C. see Kirk, M.A. et al. Baker, N. see He, X.-M. et al. Baker, T.J. 1: 15 Bakker, H. 1: 572,574,578,700,777,849; 2: 610,611; 3: 761,805; see also Beke, D.L. et al.; Stolwijk, N.A. et al., Van O m e n , A. et al., Van Ornmen, A.H. et al., Zhou, G.F. and Bakker H. Bakker, H. et al. 1: 575, 849; 3: 759, 760, 76 1 Balagurov, A.Y. et al. 2: 412 Balanzat, E. 1: 782; see aiso Audouird, A. et al. B a ~ a ~ z aE.t , et al. 1: 743 Balart, S. see Birchenall, C.E. et al. Balbach, W ,see Somrner, F et al. Baidan, A. 2 296 Baldereschi, A. 1: 132; see also Andreoni, W et al. Baldi, R.W. et al. 2 373 Baldo, P see Sclieuer, U. et al. Balducci, P. see Rozgonyi, G.A. et al. Baldwin, R.H. see Sikka, V.K. and Baldwin, R.H.; Sikka, V.K. et al. Baldwin, R.H.N. see Sikka, V.K. et al. Bale, C.W. and Eriksson, G. 3: 813 Bales, J.W see Goodhue, W.D. et al. Baligidad, R.G. see Sundar, R.S. et al. Baligidad, R.G. et al. 3 612 Balina, Ye.A. et al. 1: 948 Balk, P see Gruter, K. et al.
Author Index Balkanski, M. 3: 805 Balkanski, M. see Kunc, K. et al. Ball, A. 1: 504, 542, 574; 2: 58, 61 Ball, J. et al. 1: 786, 787, 788 Ballal, N.B. see Baligidad, R.G. et al. Ballestracci, R. 1: 249 Ballingall, J.M. see Chao, P.C. et al. Ballone, P. see Rubini, S. and Ballone, P. BalluB, R.W 1: 561, 598; 2: 604 BalluE, R.W et al. 1: 588 Balogh, J. et al. 3: 731 Balsone, S.J. 2: 112, 118, 292; see nlso Larsen, J.M. et al., Smith, P.R. et al.; Worth, B.D. et al. Balsone, S.J. et al. 3: 337 Balsone, S.N. 1: 993 Baluc, N. 1: 505, 527, 529; see also Bonneville, J. et al.; Mills, M.J. et al. Baluc, N. see Yu, D.P. et al. Baluc, N. and Schublin, R. 3: 443, 463, 467 Baluc, N. et al. 1: 521, 529, 549; 2: 271, 283 Balzan, M.L. see Geissberger, A.E. et al. Bampfylde, J.W 2: 203 Bampton, C.C. 1: 656 Ban, Z. 1: 997 Bancel, P.A. 1: 465; see also Heiney, P.A. et al.; Litbensky, T.C. et al. Bancel, P.A. et al. 1: 453, 460, 463 Banda, E. see Rockett, A. et al. Banerjee, D. 1: 657; 2: 103, 104, 106, 107, 110, 115, 116, 117, 118, 119, 120, 121, 127; 3: 361,446,494; see also Gogm, A.K. et al.: Koss, D.A. et al.; Mura~eedharan,K. et al.; Nandy, T.K. et al.; Nandy, T.K. and Banerjee. D., Rowe, R.G. et 111.; Sagar, P.K. et al.; Singh, A.K. et al. Banerjee, D. et al. 1: 538, 539, 857, 858; 2 95, 98, 102, 103, 104, 105, 107, 120, 243; 3: 494, 495, 639, 849, 850, 854 Banerjee, S. 1: 785,786, 800,802, 803,857 Banerjee, S, et al. 1: 802, 803 Banhart, F see Schaefer, €€.-E. et a/. Bankstahl, H. 1: 102, 103, 104 Banovic, S.W et cd. 3: 572 Bao, G. see Davis, J.B. et al. Bao, G. et al. 2: 290 Bao, 2;. et al. 3 161 Bar, L. see Hoenig, H.E. et al. Baranov, N.V. and Barabanova, E.A. 3: 56 Baranski, A.S. 2: 510, 511 Barbara, B. see Penney, T. et al. Barbee, T.W et al. 2 481 Barbieri, A. see Chen, W et al. Barbour, J.C. see de Reus, R. et al.; Denier van der Gon, A.W et al.; Mirkarimi, P.B. et al.; Nastasi, M. et al. Barbour, J.C. et al. 2 611 Barbu, A. see Dammak, H. et al.; Dunlop, A. et al. Barbu, A. et al. 1: 821; 3: 263, 265, 266, 267, 268 Barclay, J.A. see Rowe, J.R. et al.; Zimm, C.B. et a[. Bardi, U. 1: 616; see also Atrei, A. et al. Bardos, D.M. 1: 394 Baricco, L.H. et al. 3: 683
Barin, I. 1: 109, 110, I l l , 112, 121 Barin, I. et al. 1: 109, 110, 111, 112, 121 Barinov, S.M. et al. 1: 986, 987, 989, 993 Barinov, V.A. see Elsukov, E.P. et al.; Ermakov, A.E. et al. Baris, J.M. 2: 8 Bark, P 3: 389 Barkalov, 0.1. 1: 702, 703 Barkalow, R.H. see Corey, R.G. et al. Barker, C.S. see Fretague, W.J. et al. Barker, D.A. see Prasad, Y V.R.K. et al.; Seniiatin, S.L. et al. Barker, D.R. 2: 58, 60, 61 Barker, J.A. 3: 193 Barker, R.E. see Malhotra, M.L. et al. Bdrkow, A.G. see Ryge, G. et al. Barlow, M. and Planting, P.J. 3: 31 Barlow, W. 1: 10 Barna, P.B. see Csanady, A. ef al. Barnardini, J. see Jennane, A. et al. Biirner, K. see Kuhrt, Ch. et al. Barnes, A.C. 1: 670, 671 Barnes, A.C. et al. 3: 252 Barnes, C.B. see Heer, C.V et al. Barnes, J.W and Bailay, E.H. 3: 21 Barnes, R.G. 2: 476 Barnett, D.M. see Bacon, D.J. et al. Barnett, D.M. and Lothe, J. 3: 394 Barnett, D.M. et al. 2: 264, 265, 266 Barnett, S.A. see Hultnian, L. et al. Baro, G. see Herrnanti, G. et al. Baro, M. see Yavari, A. et al. Baro, RI. et al. 1: 788 Baro, M.D. et al. 3: 761 Baron, M. et al. 1: 705 Baroni, S. 1: 196, 208; see also Gianozzi, P et al. Baroiir, S. et al. 1: 157 Barr, T.L. 3: 140 Barrac~o~igh, K.G. 2: 217 Barradi, T. see Abdellaoui, M. et al. Barratt, S. 1: 681 Barret, D.L. see Hobgood, W.et al. Barrett, C.A. 1: 987, 988, 991, 995, 1007; 2: 57; 3: 501; see also Doychak, J. et al.; Hehsur, M.G. et al.; Khan, A.S. et al.; Lowell, C.E. et al.; Nesbitt, J.A. et al. Barrett, C.A. see Nesbitt, J. A. et al. Nesbitt, J.A. and Barrett C.A. Barrett, C.A. et al. 1: 1008 Barrett, C.S. 1: 105, 106 Barrett, C.S., see Batterman, B.W and Barrett, C.S. Barrett, J.R. see Ram, S.V. and Barrett, J.R. Barron, T.H.K. 1: 1025 Barron, T.H.R. et al. 1: 1022, 1025 Barsch, G.R. 1: 875. 890, 891 Bartels, A. see Chatterjee, A. et al.; Clemens, H. et al.; Kestler, H. et al.; Koeppe, C. et al.; Schillinger, W. et al.
Bartels, A. et al. 3: 634, 635 Barth, E.P. see Tien, J.K. et al. Bartho~in,H. see R o ~ s a t - ~ i g n oJ.d ,et al. Bartinger, M. see von Schnering, H.G. et al. Bartko, J. see Mentzer, M.A. et al.
Bartlett, R.J. see Campuzano, J.C. et al.; LlSt, R.S. et al. Bartlett, R.W 1: 998, 999, 1003 Bartlett, R.W. et al. 3: 574 Bartolotta, P. and Krause, D.L. 3: 638 Bartolotta, P.A. see Brindley, P.K. et al. Bartrarn, S.F. 1: 1025 Bartsch, M. see Feuerbacher, M. et al.; Guder, S. et al., Messerschi~idt,U. et al., Urban, K. et al. Bartur, M. 2: 605 Bartynski, R.A. see Palmstrom, C.J. et al. Barun, N.A. see Panteleimonov. L.A. et al. Basavaiah, S. see Huang, H.C.W et al. Bashara, N.M. 2: 436 Basili, N. see Magini, M. et al. Basinski, M.B. ef al. 1: 960 Basinski, S. and Basinski, Z. 3: 361 Basiiiski, Z.S. see Nabarro, F.R.N. et al. Basinsky, Z.S. et al. 3: 439 Baskes, M. 1: 61 1; see also Foiles, S. et al. Baskes, M. et al. 1: 611 Baskes, M.I. 1: 78, 79, 155, 523: 3: 765, 766, 767, 773, 774; see also Foiles, S.M. et al., Yoo, M.H. et al.; Angelo, J.E. et al., Daw, M.S. and Baskes, M.I.; D ~ wM.S. , et al.; Gdll, K. et al.; Mitchell, T.E. et al. Baskes, M.I. and Johnson, R.A. 3: 766, 773 Baskes, M.I. et al. 3: 765, 766, 767, 769, 770, 772 Basoalto, H.C. et al. 3: 305 Basol, B.M. 2: 330, 423; 3: 663, 672 Bass, J. 1: 944 Bassani, J.L. see Vitek, V et al. Bassas-Alsina, J. see Pannier, J. et al. Bassett, D. see Matteazzi, P. et al. Bassett, W.A. see Bird, J.M. et al. Bassi, C. 1: 655; see nl,so Wittenauer, J. et al.
Bassi, C. et al. 1: 655; 2: 128 Bastien, P. 1: 98 Bastin, J. see Van Loo, F.J.J. et al. Basu, A. see Gaibala, R. et al. Batalla, E. see Altounian, Z , et crl, Bateman, T.B. see Testardi, L.R. et al. Bates, J.F 2 590, 649 Batlogg, B. see Allen, J.W. et al.; Cava, R.J. et al. Batsanov, S.S. 1: 232, 233, 243, 260, 261, 424 Batsanov, S.S. et al. 3: 741 Batsch, M. see Feuerbacher, M. et al. Battaglin, G. et al. 2: 610 Battelle Pacific Northwest 2: 371 Batterman, B.W. and Barrett, C.S. 3: 258 Battezzati, L. see Baricco, L.H. et al. Baublitz, M. Jr. 1: 184 Baudin, K. et al. 3: 267 Baudoing, R. 1: 611, 615, 616, 617 Baudoing-Savois see Gauthier, Y. et al. Baudry, A. et al. 2: 484 Bauer, C.L. see Basile, D.P et al. Bauer, E. see Hauser, R. et al. Bauer, E. et al. 1: 1028 Bauer, R. 2: 227, 228 Bauer, R. et al. 2: 490
I
Baufeld, B. see Feuerbacher, M. et al.; Rosenfeld, R. et al. Baugh, D.A. see Kim, Y.K. et al. Baume, L.J. 2: 649 Bauminger, E. see Atzmony, U. et al. Baur, H. see Chatterjee, A. et al.; Kestler, H. et al., Knippscheer, S. et al. Bauer, W.H. 1: 354, 356 Bavarian, B. et al. 1: 988, 993, 995 Baxter, D.V 1: 743 Baxter, W.F. see Baldi, R.W et al. Baylits, S.C. see Brazhkin, V.V. et al. Bayuzick, R.J. see Anderson, C.D. et al. Bean, C.P. et al. 3 266 Bean, J.C. see Temkin, ET. et al.; Tuiig, R.T. et al. Beard, D.S. 2: 374 Beardmore, P. see Warlimont-Meie~~ B. et al. Beardmore, P. et al. 2: 13 Beasley, M.R. 2: 384, 653 Beattie, H.J. Jr. 1: 282, 407 Beauchamp, P see Dirras, G. et al.; Douin, J. ef al.; Lasalmonie, A. et al.; Tounsi, B. et al., VeyssiGre, P. et al. Beauchamp, P. et al. 1: 496,497,499, 500, 501, 544; 3: 461 Beaudry, B.J. see Han, S.H. et al. Beaufort, M.F. et al. 1: 562, 563, 567 Beaulaigue, L. see Campuzano, J.C. et al. Beaven, P.A. see Appel, F. et al. Beaver, W.W. see Paine, R.M., Paine, R.M. et al. Beccard, R. see Gruter, K. et al. Bechet, D. 1: 737, 739 Beck, A. 1: 7 Beck, P.A. 1: 282, 292, 394, 440; 2: 9 Becke, A.D. 1: 130 Becker, B.F 2: 510 Becker, G. 1: 758, 759 Becker, J.D. et al. 1: 41, 67 Becker, J.J. 2: 312, 314 Becker, K. 1: 10 Becker, R.S. see Kopf, R.F. et al.; Kortan, A.R. et al. Becker, S . et al. 1: 987, 993 Beckman, G.W see Libsch, J.F. et al. Becla, P. et al. 2: 419 Becquart, C.S. see Rifkin, J.A. et al. Bedair, S.M. see Hussien, S.A. et al. Beddoe, R.E. et al. 2: 273 Beddoes, J. see Chen, W. R. et al., Dudzinski, D. et al. Beddoes, J. et al. 3: 297, 317, 318 Bednorz, J.G. 2: 352 Bedwell, K.H. see Brown, J.D. et al. Beeler, J.R. 1: 496, 762, 947 Beeli, C. see Nissen, H.U. et al. Beeli, C. et al. 1: 467 Beer, A.C. 2: 327 Beer, N. and Pettifor, D.G. 3: 236 Beert, W.B. 1: 925 Beers, D.S. see Dismukes, J.P et al. Beevers, C.J. 2: 217; see also Chave, R.A. et al. Begurn, R.J. see Satya Murthy, N.S. et al. Behgozin, A. see Yasuda, H.Y et al. Behgozin, A. et al. 3 328 Behr, R. see Clemens, H. et al. Behrendt, M. see Larsen, S.E. et al. Beicher, P see Adarn, E. et al.
Beiler, T.R. see Seo, D. et al. Beke, D.L. see Tokei, Zs. et al. Beke, D.L. et al. 1: 809 Belakhovsky, M. see Durr, H.A. et al. Btlanger, A. 2: 504, 51 1 BClanger, @. see also Vijh, A.K. et al. Belash, I.T. see Aptecar. I.Z. et al., Degtyareva V.F. et al.; Ponyatovsky, E.G. and Belash. T.T. Belin, C. see Tillard-~harbonnel,M. et al. Belin, C. and Ling, R.G. 3: 123 Belin, C. and Tillard-~harbonnel,M. 3: 113, 123 Belin, C. et al. 3: 1 1 5 Bclin, E. 1: 485 Belin, E. see Sadoc, A. et al., Trambly de Laissardiere, G. et al. Belin, E. et al. 3: 142, 143, 147 Belin-Ferre, E. see Traverse, A. et al. Belin-Ferre, E. and Dubois, J.M. 3: 147 Belin-Ferre, E. et al. 3t 144 Bell, T.M. see Migliori, A. et al. Bellisent, R. see Funnel-Bellisent, M.G. et al. Bellissent, R. see Goldman, A.E. et al., Sadoc, A. et al. Bellissent, R. et al. 1: 484 Bellon, P. 1: 517. 821; see also Martin, G. et al. Bellon, P, et al. 1: 792 Belousov, O.K. see Kornilov, 1.1. et al. Belov, N.V. 1: 268,403; 3: 10, 12; see also Smirnow, N.L. et al. Belozerov, Ye.V. see Ivanova, G.V. et al. Belskiy, V.K. see Bodak, 0.1. et al. Belson, H.S. 2: 317; see also Clark, A.E. et al.; Beiici, S. et al. 3: 290 Benci, S. et al. 1: 565 Benck, R.F see Niiler, A. et al. Bender, €3. see de Potter, M. et al. Bender, 0. 1: 567, 785 Bendersky, L. 1: 453, 465,467; 2: 185; see also Schaefer, R.J. et al. Bendersky, L. et al. 1: 925; Bendersky, L.A. 1: 454, 740, 859; see also McAlister, A.J. et al.; Mozer, B. et al.; Robertson, J.L. et al.; Schaefer, R.J. et al.; Waterstrat, R.M. er al. Bendersky, L.A. et ul. 1: 857, 858, 860; 2: 95, 98, 101, 176; 3: 850 BenC, R.W 2: 606, 608, 609 Benedek, R. 1: 804 Benedek, R. et al. 1: 208 Beneking, H. 2: 335 Beneking, H. see Su, L.M. et al. Benesh, G.A. see Ellis, D.E. et al. Benesovsky, I3 see Wowotny, H. et al. Beneteau, A. see COUJOU, A. et al. Bengtzelius, U. et al. 1: 679 Benhaddane, K. 1: 540 Benjamin, J.S. 1: 700; 3: 645 Benjamin, J.S. and Volin, T.E. 3: 645 Benkaddour, A. see Dirnitrov, C. et al. Bennemann, K, 1: 80, 618 Bennernann, K. see Mortin-L6pez, J. et al.; Mukherjee, S. et al. Bennett, H.S. 2: 439, 440, 441 Bennett, J.E. 3: 31; 507
880 Bennett, L.H. 1: 241, 242, 243, 420; see also Carter, G.C. et al.; Goodman, D.A. et al.; Massahki, T.B. et al.; Rubinstein, M. et al. Bennett, L.H. see McMichael, R.D. et al., Massalski, T.B. et al. Bennett, L.H. et al. 3: 691 Bennett, M.R. 2: 637 Bentley, J. see Anderson, I.M. et al. Benyagoub, A. 1: 821; see also Garrido, F et al. Benz, M.G. 2: 353, 360, 378 Berczik, D. see Shah, D.M. et al. Berczik, D.M. 3: 487, 488, 491 Berera, A. see Dreysst, H. et al. Beretz, D. 1: 802 Beretz, D. et al. 1: 577 Berg, 0. see ~nippscheer,S . et al. Berg, S. see Ostling, M. et al. Berger, C . see Belin-Ferre, E et al., Klein, T. et al. Berger, C. et aE. 1: 480, 484 Berger, H. see Pilo, Th. et al. Berger, P. see Tominez, E. et al. Berger, S.B. see Harbeke, G. et al. Bergerhoff, G. et al. 3: 10 B e r p a n , G. et al. 1: 475 Bergmann, G. 1: 689 Bergniann, H.W. 1: 733 Bergmann, 5. see Poschmann, 1. et al. Bergmark, T. see Seigbah~,K. et al. Bergstrom, D.B. see Creene, J.E. et al. Bergstrom, D.B. et al. 3 666, 674, 675 Berko, S. I: 41; see also Singh, D. et al. Berkowitz, A.E. et al. 3: 691 Berkowitz-Mattuck, J.B. 1: 998, 1000 Berkowitz-Mattuck, J.B. et al. 1: 1003, 1004; 3: 574 Berman, H. 2 576; see also Palache, C. et al. Bernian, R, 1: 1025, 1027 Bernal, J.D. 1: 479, 679, 740 Bernard, F. see Charlot, E. et al.; Charlot, F. et al.; Gauthier, V. et al., Gras, 6. et al. Bernardi, J. see Fidler, J. et al. Bernardini, J. see Tokei, Zs. et al, Bernas, H. see Jaouen, C. et al. Bernauer, 0. 2 485 Berndt, M. see Bergerhoff, G. et al. Beriier, D. 3: 280 Berner, D. see Epperson, J.E. et al. Berner, D. et al. 1: 565 Bernstem, H. 1: 100 Berry, B. 11: 778 Berry, G. see Rockett, A. et al. Berry, R.L. 1: 107 Bertaut, E.F. 1: 249; see also JoubertBettan, C.A. et al.
Bertocci, U. et al. 1: 967, 970 Bertolino, N. et al. 3: 736, 737 Bertram, M. see Quyen, N.H. et al. Berztiss, D. et al. 1: 937, 998, 999, 1003, 1004, 1005 Berztiss, D.A. et al. 3: 574
A u t h o ~Index Berzon, E.M. see Smirnova, N.L. et al. Besag, F.M.C. see Hutchinson, W.B. et al. Besenbacher, F. 2: 479 Besenbacher, F. see Sprunger, P.T. et al. Besenhard, J.O. 2: 510 Besenhard, J.O. see Winter, M. et al. Beshers, D.N. 1: 760 Besmann, T.M. see Stinton, D.P. ei at. Besnus, M.J. et al. 3: 46 Besocke, K. see Niehus, H. et al. BessiZre, M. 1: 40 Bessoud, A. see Colinet, C . et al. Bethe, H.A. 1: 440 Bethoux, 0. see Zougmort, F. et al. Betterton, J.O. 2 517 Betterton, J.O. see Hume-Rothery, W. et al. Betts, K. 2: 383 Betz, J. see Fulap, G. et al. Betz, U. et al. 3: 542 Beuneu, F. see Defour, C. et al. Beuth, J.L. see Knaul, D.A. et al. Bever, M.B. 1: 97, 109, 786, 832; 2: 502, 640, 642; see also Darken, L.S. et al.; Jena, A. et al. Bever, M.B. et al. 1: 960; 2: 654 Bevis, M. and Crocker, A.G. 3 405, 416 Bevk, J. et al. 2: 63 Bewlay, B.P. see Henshall, G.A. et al.; Jackson, M. R.et al.; Jackson, M.R. and Bewlay B.P. Bewlay, B.P. and Sutliff, J.A. 3: 547, 548 Bewlay, B.P. et al. 3: 346, 489, 490, 492, 541, 545, 547, 550, 552,553,554, 555, 556, 559, 658 Bewlay, P.A. et al. 3 312 Beyer, W.B. see Weast, R.C. et al. Beyermann, W.P. see Canfield, P.C. et al. Beyss, M. see Jia, C.L. et al.; Rosenfeld, R. et al.; Wollgarten. M. et al. Bezinge, A. see G~~eramian, M. et al. Bhadra, R. see Grirnsditch, M. et al.; Okamoto, P.R. et at.; Rehn, L.E. Pb
al.
Bhaduri, S . and Bhaduri, S.B. 3 725, 732, 733 Bhadun, S.B. 3: 734 Bhaduri, S.B. see Bhaduri, S. and Bhaduri S.B. Bhalla, A S . see Li, J. et al. Bhalla, A.S. et al. 1: 172, 181 Bhandari, C.M. 2 453 Bhangu, J.K. see Broomfield, R.W. et al. Bhanumurthy, K. et al. 1: 646 Bhargava, R.N. 2 326 Bhat, A. see Compaan, A. et al. Bhatia, A B . 1: 663, 665, 671, 673, 674, 676, 677 Bhatt, K.B. see Kulshreshtha, S.K. et al. Bhattacharya~P.K. see Ojirna, M. et al. Bhattacharya, R.N. 2: 511 Bhatta~~iarya, R.N. et al. 2 511 Bhattacharyya, S.K. 1: 866 Bhowal, P. see Seo, D.Y et al. Bhowal, P.R. et al. 3: 415 Bi, Y.J. et al. 2: 395 Biancaniello, F.S. 1: 454; see also Bendersky, L. et al., Bendersky, L.A.
et al.; McAlister, A.J. et al.; Schaefer, R.J. et al. Bianchessi, A. see Lupinc, V et al. Bibring, H. 3: 554 Bickmann, K. see Sajovec, F. et aE. Bieber, A. 1: 28, 29, 34, 36, 849, 850 Biedermann, A. see Hebenstreit, W. et al. Bieger, H. see Klaumiinzer, S. et al. Bieler, T. see Seo, 23. et al. Bieler, T.R. see Cheong, S.W. et al.; Jin, Z. and Bieler, T.R.; Jin, Z. et al.; McQuay, P.A. et al.; Seo, D.Y. et al. Biemont, A. see Andreoni, W. et al. Bienenstock, A. 1: 889 Bienvenu, Y. see Ansara, I. et al. Bierlein, J.D. 2: 414 Biery, N. see De Graef, M. et al. Biggs, B.D. et al. 1: 484 Biggs, T. see Hill, P.J. et al. Bigot, J. see Yu-Zhang, K. et al. Bigot, J.Y. et al. 2: 412 Biham, 0. et al. 1: 480 Bijkerk, K.R. see van der Kolk, G.J. et al. Bilbrey, A.R. see Chang, Y.A. ct al. Bilby, B.A. 1: 844 Bilby, BA. see Cottrell, A.H. and Bilby, B.A. Bilby, B.A. and Crocker, A.G. 3: 405 Billard, L. 1: 482; see also Lnncon, F et al. Billebaud, A. see Baudin, K. et al. Billman, F.R. see Paris, H.G. et al. Billy, J. see Macko, D. et al. Bilonizhko, N.S. 2: 312; see also Kuzma, J.B. et al. Bilz, H. 1: 150 Bimberg, D. 3: 806 Binder, K. 1: 38; see also Helbing, W. et al.; Schweika, W. et al. Binder, I(.et al. 1: 39 Binggeli, N. see Chelikowsky, J.R. et al. Binnig, G. and Rohrer, H, 3: 212 Binnig, G. et al. 3 215 Birch, F. 1: 199 Birchenall, C.E. 1: 766; 2 637, 639, 642, 643, 644 Birchenall, C.E. et nl. 2: 637, 640 Bird, J.E. gee Mukherjee, A.K. et al. Bird, J.M. et al. 1: 627, 631 Birgeneau, R.J. and Horn, P.M. 3 212 Birinpccio, V. 1: 3 Birkmne, R.W see Shafaman, W.N. et al. Birnbaum, H.K. see Bond, G.M. et al. Biro, D. see Manaila, R. et al. Birringer, R, see Bohn, R. et at., Karch, J. et al. Birtcher, R.C. see Allen, C.W. et al.; Brown, B.S. et al.; Rest, J. et al. Birtcher, R.C. et al. 1: 791, 8'15, 816, 875; 2: 648 Biscondi, M. 1: 955 Biscondi, M. see Fraczkiewicz, A. et al. Bishop, H.E. sec DeVan, J.H. et al. Bishop, R.R. see Bewlay, B.P. et al. Bittner, H.F 2 509, 511 Black, P.J. 2 177, 178 Black, S.A. 2: 646; see also Sergev, S.S. et al. Black, T.J. et al. 1: 807, 808
Author Index Blackburn, M.J. 1: 534; 2: 59, 60, 64, 75, 80, 82, 83, 91, 93, 107, 294; 3: 618; see also Lipsitt, H A . et al., Russell, S.M. et al.; Shechttnan, D. et al.; Sheetman, R. et al., Williams, J.C. and Blackburn, M.J. Blackburn, M.J. &weShcchtnian, D. et al.; Blackburn, M.J. and Smith, M.P 3: 648 Blackburn, P.E. see Berkowitz-Ma~tuck, J. et al. Blackford, J.R. et al. 3: 572 Blaha, P. et al. 1: 133 Blair, H.D. see Ray, U. et al. Blakely, J. 1: 618 Blakely, J.M. 1: 585, 586; 2: 604: 3 225; see also Potter, H.C. and Blakely, J.M. Blanco, J.A. et al. 1: 949; 3 177, 179 Blander, M. see Saboungi, M.L. et al. Blander, boungi, M.L. et al. Blank, D de Reus, R. et al. Blank-Be , M. see Koster, U. et al. Blaiike, H. 1: 747 Morellon, L. et al. Blatt, F.J. et al. 2: 469 Blatter, A. 2: 610 Blau, B. et al. 2: 374 Blau, P.J. 2 598, 599 Blau, W see Muller, Ch. et al. Blavette, D. et al. 3: 307 Blech, I. 1: 453, 482; 2: 185 Blech, I. see Shechtman~D. et al. 5lenkinsop, P. see Davey, S. et al. BlCtry, J. see Lamparter, P et al. Blewitt, T.H. 1: 804, 805, 808; see also Brown, B.S. et al., Kirk, M.A. et al. Bloch, A.N. 1: 242, 243, 319,419,420 Bloch, J. 1: 694 Block, G. 2: 314 Block, H. 3: 542 Blom, H.-0. see Ostling, M. et al. Bloosuberg, D. 2: 435, 441, 450 Blouin, M. et aZ. 3: 78 Bloyer, D.R. et al. 3 342 Blugel, S. see Takizdwa, S. et al. Blum, A.N. et al. 1: 16 Blum, M. et al. 3: 593, 601 Blum, R.P. et al. 3: 219 Blum, V see Hammer, E. et al. Bobev, S. see Xu, L. et al. Wobev, S. and Sevov, S.C. 3: 128, 131 Bobrov, E.S. see Zhao, Z.P et al.; Zhukovsky, A.Y et al. Bocelli, G. see Sanchez, J.L. et al. Bochu, B. 1: 773 Bocquet, A.E. see Ogawa, S. et al. Bocquet, J.L. et al. 1: 576 Boda, G.D. 1: 843 Bodak, 0 . 1 . 1: 378, 406; 3 100; see also Gladyshevskii, E.I. et aE.; Kalychak, Ya.M. et al.; Levin, E.M. et al.; Marusin, E.P. et al. Bodak, 0.1. et al. 1: 412 Bodyrev, G. see Ivanov, E. et al.
Boehler, C. J. et al. 3: 303 Boerhof, W. 1: 191 Boettinger, W.J. 1: 772,857; 2: 95; see also Bendersky, L. et al.; Bendersky, L.A. et al.; Mozer, B. et al., Schaefer, R.J. et al. Boettiiiger, W.J. et al. 2 221, 225 Boettner, R.C. et al. 3: 325, 332, 333 Bogdanov, E.1. see Larikov, L.N, et al. Bohtn, G. see Brunner, K. et al. Bohm, H. 2: 95 Bohm, M. 3: 805 Bohm, V. see Scholl, R. et al. Bohn, M. see Tominez, E. et al. Bohn, R. et al. 1: 652 Bohn, R.G. see Compaan, A. et al. Bohr, J. see Majkrzak, C.F. et al. Bohsung, J. and Trebin, H.R. 3: 379 Boily, S. et al. 2: 411 Bojarski, Z. 1: 394 Bok, L.D.C. 1: 352 Bokhonov, B.B. et al. 3: 758 Bokii, G.B. 1: 713; 3: 3, 4 Boldyrev, V V see Bokhonov, B.B. et al. Bolle, U. see voii Schneriiig, H.G. et al. Bolling, G.F. 1: 845; 2: 150 Bolling, G.F. and Richinan, R.H. 416 Bollmann, W. 1: 590, 598; see also Grimmer, H. et al. Bollmann, W. et al. 1: 598 Bolt, P.J. et al. 1: 653 Boltaks, B.I. 1: 764, 766 Bommel, F see Hoenig, H.E. et al. Boiia, M. see ten Kate, H.H.J. et al. Bonafede, S. see McGahan, W.A. et al. Bond, A.M. 2: 646 Bond, G.M. et al. 1: 927; 2: 30 Bonda, N.R. 2: 599; see alLwPerepezko, J.H. et al. Bonda, N.R. et al. 2: 25 Bondarev, B. et al. 2: 88 Bondarev, V.N. 2 630 Bonefaac, A. 2: 178 Bonhomme, E;. et al. 1: 393 Boni, B. see Majkrzak, C.F. et al. Boniface, T.D. see Horner, I.J. et al. Bonito, A. see Bruzzone, P. et al. Bonjour, E. see Daudin, B. and Bonjour, E. Bonnell, D.A. 3: 215 Bonnelle, C. 3: 142 Bonnelle, C . see Fargues, D. et al. Bonnet, R. see Loubradou, M. et al. Bonnett, J.D. see Howe, L.M. et al. Bonneville, J. 1: 545, 547; see also Baluc, N. et al.; Rruml. T. P i al.; Viguier, B. et al. Bonneville, J, et al. 1: 547; 2: 25 Bonsach, W. 2: 577 Bontemps, C . 1: 543, 548 Bontemps-Neveu, C. 1: 545, 550, 551, 552 Boodey, J.B. see Gao, M. et al. Booker, J. et al. 3 37 Boom, R. see de Boer, F.R. et al., Miedeina, A.R. et al. Boone, D.H. 1: 987, 996; 2 57, 490,492, 493; see also Dust, M. et al.: Goward, G.W and Boone, D.H. Boone, D.H. and Sullivan, C.P. 3: 298 Boone, D.H. et al. 2: 499 Boone, D.J. see Demaray, R.E. et al.
88 1 Boonk, L. see Evers, C.B.H. et al. Booth, J.G. 3 803 Borchardt, G. see Jedlinski, J. et al. Bordeau, R.G. 3: 572 Bordeaux, F. and Yavari, A.R. Bordenet, M. see Goken, M. et al. Borelius, G. 1: 758, 759 Boren, B. 1: 409 Borghesi, A. see Amiotti, M. et al. Borgman, H. 2: 402, 405 Borgstedt, H.U. et al. 3: 33, 34 Boring, A.M. see Chen, S.P. et al., Eriksson, 0. et al.; Voter, A.F. rzl al.; Wang, C.S. et al.; Wills, J.M. et al. Boriskiiia, N.G. 1: 714 Borisov, B.S. see Arkharov, V.I. et al. Bonsov, V.T. et al. 1: 768 Bormann, R. see Oehring, M. et al., Schultz, L. et al.; Yan, Z. et al. Born, M. 1: 319 Bornstem, N.S. 1: 998, 1003 Borshchevsky, A. see Caillat, T. et al., Chen, B. et al. Bortel, G. see Stephens, P.W et al. Borstel, G. 1: 132 Borusevich, L.R. 2: 226 Borzillo, A.R. see Z;occola, J.C. et al. Bose, A. see Gernian, R.M. et al.; Sims, D.M. et al. Bose, A. et al. 1: 646, 765; 2: 44; 3: 656, 725, 734 Bose, S. 2: 221 Bose, T.K. see Foldeaki, M. et al. Boss, D.E. and Yang, J.M. 3: 584 Both, E. see Libsch, J.F. et al. Battger, G. see Schobinger-Papa~ant~llos, P. et al. Battger, R. 2: 5 Bsttiger, J. see Greer, A.L. et al. Bsttiger, J. see Samwer, K. et al. Bouchard, M. et at. 3: 416 Boudreau, R.A. 2 511 Bouffard, S. see Audouard, A. et al.; Dunlop, A. et al.; Toulemonde, M. et al. Boukamp, B.A. see Wen, C.J. et al. Bsuldin, E.E. see Stern, E.A. et al. Boulet, R. see Reinders, P.H.P. et al. Boulogne, B. see Vedula, K. et al. Bououdina, M. see Ayres de Campos, J. et al.; Obbade, S. ct al.; Vert, R.et al. Bourdeau, R.G. 2: 178 Bourgedt"Lam1, E. see Courtols, D. et al. Bourgoin, J.C, et al. 1: 580 Boursier, D. see Fruchart, R. et al. Bouten, P.C.P. see Buschow, K.H.J. et al. Bovenkerk, H.P. et al. 1: 180 Bowen, P. 2 83; see also Chave, R.A. et al. Bowen, P. see Peiiton, R.J.T. et al. Bowen, P. and Jatnes, A.W 3: 337 Bower, E.N. 2: 277, 282 Bower, J.E. see Kwo, J. et d. Bowers, J.E. 2: 342, 428 Bowers, J.E. et al. 3: 28 Bowers, L A . see Ren, J. et al. Bowles. J.S. 1: 828, 829; see also Stevens, G.T. et al. Bowman, K.J. 3: 574 Bowman, R.C. see Richter, D. et al.
882 Bowman. R.C. Jr. 1: 305 Bowman, R.R. 1: 926, 1007, 1008; 2: 290; 3: 317, 320; see also Doychak, J. et al., Locci, I.E. et al., Noebe, R. et al.; Noebe, R.D. et al.; Raj, S.V et al. Bowman, R.R. et a/. 2: 61; 3: 656 Boyce, J.B. see Bridges, F. et al. Boyce, J.B. et al. 1: 479 Boyer, L.L. 1: 207; see also Chen, J. et al., Feldwm, J.L. et al. Boyer, L.L. et al. 1: 195, 207 Boyer, P. see Baudry, A. et al. Bozek, B. see Danielewski, M. et al. Bozler, C.O. 2: 335: see also Alley, G.D. et al.; Vojak, BA. et al. Bozorth, R.M. 2: 306, 307, 309; see also Clark, A.E. et al. Bozzolo, G. 1: 926 Brack, K. 1: 172 Bradley, A.J. 1: 10, 11, 309, 565; 2: 18, 54, 55, 199,200 Bradley, A.J. and Taylor, A. 3: 275, 769 Bradt, H.L. see Frier, P.S. et al. Brady, M.P. .vee Hanrahan R.J. Jr. et al,; Lm, C.T. et al. Brady, M.P. and Doychak, J. 3: 44 Brady, M.P et al. 1: 987, 992,995; 3: 577, 580 Bragg, W.L. 1: 502, 509, 775, 799 Brag~nski,A. see Sinharoy, S . et al. Braisaz, T. et al. 3: 410 Braithwaite, D. see Thomasson, Y et al. Brainpton, C.G. see Rhodes, C.G. er al. Branagan, D.J. and McCallum, R.W. 3: 97 Brandenburg, see Bergerhoff, G. et al. Briindle, H. 2 448 Brandle, H. et al. 2: 444 Brandon, D.G. 1: 598 Brandon, D.G. et al. 1: 598 Brandt, D. see Tomaschko, Ch. et al. Brandt, N. 1: 211 Brandt, W. 1: 136 Branover, H. and Lesin, S. 3: 31 Brauer, S . see Koster, U. et al. Braun, D.J. see Evers, C.B.H. et al. Braun, H.F. see Dumas, J. et al. Braun, P.B. 2: 309 Braunovik, M. 1: 955 Braunstein, J. 2: 637 Braverman, E.M. see Izerrnan, S.A. et al. Bray, J.W 2 165 Brazhinikov, M.A. see Batsanov, S.S. et al. Brazhkin, V.V. et al. 3: 161 Brebec, G. see Bocquet, J.L. et al. Brebrick, R.F see Vanyukov, A.V. et al. Brkhet, Y. see Kubin, L.P. et al. Bstchet, Y. et al. 1: 792 Bredl, C.D. see Steglich, F. et al. Brehm, H. and Glatzel, U. 3: 305 Brehm, W see Bruzzone. P et al. Breig, P.G. 1: 640 Breinan, E.M. et al. 1: 733; 2: 653 Brekkeler, K.H. et al. 2: 610 Brener, R. see Appelbaum, A. et al.; Eizeiiberg, M. et al. Brennan, P.C. et al. 3: 317 Brenner, A. e f al. 1: 733, 792; 3: 681
Brenner, S.S. et al. 1: 590, 591, 901; see also Sieloff, D.D. et al. Brentnall, W.D. et al. 2: 228 Bresson, L. see Caillard, D. et al. Bresson, L. and Gratias, D. 3: 397, 400 Bresson, L. et al. 1: 743 Breusov, O.N. see Dremin, A N . and Breusov O.N. Brewer, L, 1: 63, 100, 116; 2: 518 Brewer, L. see Nosrnan, J.H. et al. Brewer, W.D. and IJnnam, J. 3 582 Briane, M. 1: 684 Briant, C.L. 1: 591, 593, 594, 604, 896, 898, 899, 900, 902, 904, 907, 908; 2: 27, 214 Briant, C.L. see Bewlay, B.P et al., Fleischer, R.L. et al.; Huang, S.C. et al., Taub, A.Z. et al. Briant, C.L. et al. 1: 897, 899 Bridges, F.B. see Boyce, J.B. et al. Bridges, F.B, et al. 1: 483 Bridges, P.J. see Gilman, PS.et al. Bridgman, P W 3: 542 Briggs, D. and Seah, M.P. 3: 214 Briggs, J.K. see Hawk, J.A. et al.; Wilsdorf, H.G.E. et aE. Briggs, L.J. see Schulson, E.M. et al. Brik, K.A. see Shemyakin, V.S. et al. Brik, V.B. see Larikov, L.N. et al. Brill, U. 1: 989 Bsimhall, J.L. see Bruemmer, S.M. et al., Charlot, L.A. et al. Brimhall, J.L. and Brueminer, S.M. 3: 42, 46,47 Brimhall, J.L. et al. I: 809, 818 Brindley, P.K. 3 584 Brindley, P.K. see MacKay, R.A. et al., Smialek, J.L. et al. Brindley, P.K. et al. 2: 168, 292 Brmdley, W.J. see Brady, M.P. et al. Brindley, W.J. et al. 3: 577 Brison, J.-P. see Thomasson, V et al. Brite/Euram Programme 2: 499 Britt, J. see Chu, T.L. et al. Brittain, J.O. 1: 574; see crlso Hughes, T. et al., Lautenschlager, E.P. et al. Brittan, S.C. 2 506, 51 1; see nlvo Parkinson, N. et al. Broddin, D. see Leroux, C. et al. Broder, J.D. 1: 17111, 178, 180, 181, 192; 2: 623 Broderick, T.F. see Kestner-Weykamp, H.T. et al. Brodsky, M. see Orent, T. et al. Broholm, C. see Mason, T.E. et al. Bronimer, P.E. see Buis, N. et a/. Bronevoi, I.L. 1: 960 Bronfin, M.B. see Kovalev, A.I. et al. Bronoel, G. et al. 2: 506, 508 Bronsveld, P. see Schurer, P.J. et al. Bronsveld, P.M. see Elswij, H.B. et al., Van Der Wegen, G.J.L. et al. Brook, G.B. 3: 77 Brook, G.B. and Iles, R.F 3: 56, 77 Brooks, C. 1: 788; see also Choudhury, A. et al. Brooks, C.R. et al. 2 18 Brooks, J.W. see Gilman, P.S. et aZ.; Millett, J.C.F et al. Brooks, J.W et al. 3: 635 Broomfield, R.W. et al. 3: 297, 307
Broser, 1. 3: 806 Broska, A. see Wolff, J. et a/. niann, U. see Appel, F et nl. Brossmann, U. et al. 3: 277. 279 Brotzen, F.R. 1: 960 Brouer, G. see Zintl, E. and Brouer, G. Brouers, F. 1: 944 Broughton, J.Q. 3: 841 Broughton, J.Q. see Boyer, L.L. et al.; Lill, J.V. et al. Brown, A. 1: 639, 657; 2 506 Brown, A.C. 1: 479 Brown, A.M. 2: 608; see also Bye, R.L. et al.; Kim, N.J. et al., Skinner, D.J. et al. Brown, A.M. and Ashby, M.F. 3: 300 Brown, A.M. et al. 2: 178, 194 Brown, B.S. 1: 808 Brown, B.S. et al. 1: 792, 803 Brown, C.O. see Breinan, E.M. et al. Brown, D. et al. 3: 137 Brown, G.G. 2 278, 279, 282 Brown, I.D. see Rabe, K.M, et al. Brown, J.R. 1: 82 Brown, K.M. see Wakugawa, J.M. ci al. Brown, L.D. see Fine, M.E. et al. Brown, L.E. see Chatterjee, A. et al. Brown, L.M. 1: 512, 514; 2: 7, 258, 259, 261, 262, 265, 267 Brown, N. 1: 496,497, 500, 502, 503, 509, 51 1, 544, 574, 960 Brown, N. see Marcinkowski, M.J. et al. Brown, R.A. 1: 955, 956, 960 Brown, R.H. 1: 81 Brown, S. see Whittenberger, J.D. et al. Brown, S.A. 1: 924; 2: 357, 358, 160, 161, 166, 167; see also Kumar, K.S. et al.; Wbitteiiberger, J.D. et al. Brown, S.A. et al. 2: 158, 161, 164, 165, 166 Browning, R. 2: 168 Bruce, R.A. 1: 645 Bruch, C.A. see Menzies, R.G. et al. Bruchle, W see Trautmann, C. et al. Bruck, E. see Tang, N. et al. Bruck, H.A. et al. 3: 695 Bruemmer, S.M. see Charlot, L.A. et al.; Henager, C.H. et al.; Song, S.G. et al. Bruemmer, S.M. et al. 3: 44, 67 Bruesch, P. 1: 15111 Brunim, M.W. 1: 987, 989, 999, 100.5; see also Grabke, W.J. et al. Brun, T.O. see Said, M.R. et al. Brun, T.O. et al. 1: 566; 3: 252 Brundle, C . 1: 412 Brundle, C.R. and Baker, A.D. 3: 140 Brunelle, A. see Baudin, K. et al.; Dammak, H. et al. Brunet, Y. see Tixador, P et al. Bruning, R. see Strom-Olsen, J.O. et al. Brunner, G.O. 1: 237, 238, 250, 364 Brunner, K. et al. 2 425 Bruns, S. see Xnden, G. et al. Bsuskov, V.A. see Il’nitskaya, O.N. et at. Bsuzzone, G. et al. 1: 366 Bruzzone, P. et al. 2 374 Bryant, A.W. 1: 64 Brynnt, J.D. 1: 644; see also Semiatin, S.L. et al. Bryant, J.D. et al. 3: 691
883 Bryant, M.D. see Hiller, M.W et al. Buchanan, R.A. 2: 208 Buchanan, R.A. 1: 967, 970, 973 Bucl~enauer,G. 2: 572 Bucher, E. see Mason, T.E. et al. Bucher, E. et al. 1: 218 Biichler, H. 2: 572 Buchner, H. 2: 476 Buchta, R. see Nygren, S. et al.; Ostling, M. et al. Buck, T.M. see McRae, E.G. et al. Buck, T.M. et al. 1: 613, 615, 616, 618, 619; 3: 218 Biickel, W 1: 733, 792 Buckley, D.H. 2: 597 Buckley, RA. 1: 921; see also Manaf, A. et al., Midson, S.P. et al., Otani, Y. et al., Prakash, U. et al. Buckley, R.A. see Blackford, J.R. et al. Buckman, R.W 3: 544, 545 Buckremer, H.P. see Ishiyania, S. et al. Budai, J.D. 1: 454 Budberg, P.B. see Alisova, S.P. et al. Budiansky, B. et al. 1: 927 Bud’ko, S.L. see Canfield, P.C. et al. Buehler, E, see Kunzler, J.E. et al., Menth, A. et al.; Nesbitt, E.A. et al. Buehler, W.J. 2: 307 Buehler, W.J. et al. 1: 832; 2: 529 Buerger, J.M. 3: 12 Buffered, A.S. 3: 646 Buhrer, W. see Reinhard, L.et al. Buhrer, W. et al. 1: 157, 158 Buhrman, R.A. see Granqvist, C.G. and Buhsrnan, R.A. Buis, A. et al. 1: 588, 589, 598 Buis, N. et al. 1: 71 BuJdoso, L. see Balogh, J. et al. Bukman, S.P. see Shernyakin, V.S. et al. Bulat, T.J. 2: 390 Bulatov, V see ~ a g h m a r eU.V. , Bull, M. see Palmer, I.G. et al. Bullough, R. 3: 408 Bullough, R. and Gilman, J.J. 3: 270 Bulycheva, Z.N. et al. 1: 717 Burnm, HI. 2: 564, 565 Bundy, F.P. see Bovenkerk, H.P. et al. Bungardt, W. 2: 647 Bunshah,, .%. 4933,k.6 7 Bunshah, R.F. see Nimrnagadda, R. et al. Burchard, W.-G. see Dahl, W. et al. Burdett, J.K. 1: 242, 428 Burdett, J.K. et al. 1: 243 Burger, J.P. et al. 2 483 Burgers, M.J. 3: 403 Burgers, W.G. see Six, W. et al., Veer, F.A. et al. urgeson, J.E. see Baldi, R.W. et al. Burgio, N. see Magini, M. et al. Burke, M.G. see Brenner, S.S. et al.; SielofT, D.D. et al. Burke, N.R. 1: 32 Burke, P.M. 1: 916 Burkhanov, G.S. see Golikova, M.S. et al.; Kravchenko, N.V. et al. Burkhardt, U. see Grin, J. et al. B ~ r k ~ a rM.W. t , 1: 834 Burkholder, P.S. see Broomfield, R.W. et al. Burkhov, S.E. see Chen, H. et al.
Burlet, P see Morin, P. et al., Rossat-Mignod, J. et al. Burnashova, V.V. see Markiv, V.Ya. et al. Burnell-Gray, J.S. see Chan, W.Y. et al.; Datta, P.K. et al.; Du, H.L. et al.; Fisher, G. e f al.; Xiang, Z.D. et al. Burns, G. 2 438, 439 Burns, P.C. and Hawthorne, F.C. 3: 10 Burov, I.V. see Kripyakevich, P.I. e f al. Burrest, C.A. see Miller, D.A.B. et al. Burrus, C.A. 2: 342 Bursill, L. 1: 453 Burton, B. 1: 918 Burton, B. et al. 1: 773 Burton, B.P. see Bendersky, L.A. et al. Burton, P.B. and Cohen, R.E. 3: 205 Burton, B.P et al. 1: 41, 69, 71 Burzlaff, H. see Fischer, W. et al. Burzo, E. 3 803 Burzo, E. and Kirchrnayr, H.R. 3: 804 Busby, P.E. et al. 1: 594 Busch, G. et al. 1: 682, 683 Busch, R. see Kirn, Y.J. et al. Buschov, K.H. see Lord, J.S. er al. Buschow, K.H.J. 1: 419, 434, 735; 2: 308, 310,314. 315,316, 394,396,443,448, 449,476, 509,511,610,611,612,613; 3: 804; see also Colinet, C. et al., Grossinger, R. et al.; Srnit, H.H.A. et al., van Engelen, P.P.J. et al., van Engen, P.G. et al.; Hautot, D. et al., Kuncser, V et al., Liu, J.P. et at.; Msze, 0. and Buschow, M.H.J.; Palstra, T.T.M. et al., SchobingerPapamantellos, P et al.; Tang, N. et al.; Velge, W.A.J.J. and Buschow, K.H.J., Zhou, R.J. et al.; van der Groot, A S . and Buschow, K.H.J. 2: 394 Buschow, K.H.J. and Wohlfarth, E.P. 3: 804 Buschow, K.H.J. et al. 2: 315, 443, 444, 445, 446, 447, 448, 476; 3: 97 Bush, D.H. 1: 109 Bushin, I N . et al. 1: 768 Busse, P. see Blum, M. et at. Bussiere, J.F 2: 380 Bussiere, J.F. et al. 2: 380 Busso, E.P. and McClintock, F.A. 3: 368 Butkevitch, V.G. see Khryapov, V.T. et al. Butler, B. et al. 1: 875 Butler, D. see Abell, J.S. et al. Butler, E.P. 1: 784, 785, 786, 800 Butler, J.L. 2 400, 403 Butler. S.R. see Wasilewski, R.J. et al. Butler, W.H. 1: 133; see a00 Stocks, C.M. et al.; Turchi, P.E.A. et al. Butt, D.P. see Hanrahan, R.J. Jr. et al. Butterfield, M.T. see Brown, J.D. et al. Buttger, G. see Schobinger-PapamantellOS, P et d. Buyanov, Yu.1. see Eremenko, V.N. et al. BLIYCO, E.H. 1: 1019, 1020, 1022 Buyers, W.J.L. 1: 562, 564 Bye, R.L. see Brown, A.M. et al.; Das, S.K. et al.; Kim, N.J. et al.; LaSalle, J.C. et al.; Skinner, D.J. et al. Bye, R.L. el al. 2: 178 Byk, D.S. see Kuz’man, Yu.B. et al.
Bykovskii, Yu.A. see Matveeva, N.M et al. Bylander, D.M. see Zhu, M.J. et al. Byrom, E. see Ellis, D.E. et al. Byun, T.S. 1: 926 Cabanas-Morcno, J.G. see Calderon, W.A. et al. Cabane, G. see Sainfort, G. et al. Cabarat, R. et al. 1: 886 Cable, J.W. 1: 447; see also Majkrzak, C.F. et al. Cable, J.W. et al. 1: 935 Cabral, C.J. see Svilan, V et al. Cabrera, A. see Calderon, W.A. et al. Cabreros, E. see Pui, C.S. et al. Cabn, L.J. I: 627, 628, 629, 630, 631, 635 Cac1uff0, R, et al. 1: 743 Cadek, J. see Fiala, J. et al. Cadevillc, M.C. I: 777, 778; see U ~ S O Kozubski, R. et al.; Leroux, C. et al.; Vennkgues, P. et al. Cadeville, M.C. et al. I: 777, 778, 779, 946, 947 Cadieu, F.J. see Hegde, H. et al., Karnprath, H. et al., Rani, R. et al. Cadoff, I. 2: 453 Cadogan, J.M. see Campbell, S.J. et al.; see Courtois, D. et al.; Li, H.S. et al.; Margarian, A. et al., Ryan, D.H. ef al.; Yang, F.N. et al. Cadogan, J.M. atid Li, H.S. 3: 174, 175 Cadogan, J.M. et al. 3; 169, 171 Cady, C.M. see Jin, 0. et al. Cahn. J.W 1: 35, 496, 499, 502, 511, 774, 8442849,850,851, 852, 853,854, 857, 865, 866, 867; 2: 199; 3: 410; see also Gratias, D. et al., Johnson, W.C. and Cahn, J.W.; Portier, R, et al., Shechtman, D. et al. Cahn, J.W. et al. I: 460, 461, 483, 537 Cahn, R.W. 1: 1411, 51 1, 680, 772, 780, 787, 788, 817, 857, 863, 864, 925; 150, 590; 3 75,275,278, 501; see also Gialanella, S . et d.; Lawley, A. et al., Liu, C.T. et al.; Toloui, B. et al.; Vidoz, A.E. et al., Yang, R. et al. Cahn, R.W. and Coll, J.A. 3: 404, 410, 416 Cahn, R.W. et d. 3: 624 Cai, T. see Shen, 2. et al. Caignaert, V. see Pannier, J. et af. Caillard, D. 1: 547, 549, 915; 2: 27; see also CICrnent, N . et al., Couret, A. et al.; Legros, M. et al.; Paidar, V. et al. Caillard, D. et al. 1: 522. 547, 548, 549; 27; 3: 365, 693 Caillat, T. see Chen, B. et al. Caillat, T. et aZ. 3: 104 Caillet, M. see Kabbaj, M. et al. Calawa, A.R. 3: 781 Calbrick, C.J. and Marcus, R.B. 3: 416 Calderon, H.A. et al. 3: 65, 653 Cale, T.S. et al. 3: 664, 669 Calestani, G. see Sanchez, J.L. et al. Calka, A. et al. 1: 735 Callegari, A. see Murakami, M. et al. Callegari, A. et al. 2: 621
884 Callen, M.B. 2: 392, 393, 394 Caloni, 0. see Conserva, M. et al. Calvayrac, U. 1: 498; see also Uu~Zhang, K. et al. Calvert, F.C. 1: 5, 7 Calvert, L.D. 1: 120, 138, 227, 228, 230, 234,241,248,255,272,277,278, 231, 289,294,298,303,305,306,320,328, 363,3645367,370,385,386,388,391, 393, 394, 396, 397, 398, 40311, 425, 431,433,437,625,632,876,878,881, 885,887; 2: 147; see also Mishima, 0. et al. Calvert, L.D, see Villars, P. and Calvert, L.D. Cam, G. see Kestler, H. et al. Camare, E. see Fatas, E. et al. Cameron, B.P. see Forster, J. et al. Cameron, C.B. 2: 597 Cameron, D.J. 2: 145 Cameron, D.J. et al. 2: 136 Camerson, S. see Chen, W. et al. Camlibel, 1. see Forrest, S.R. rt al. Canimarata, R.C. see Edelstein, A S . and Cammarata, R.C. Cammarota, V.A. 3: 21 Camp, F.E. see Kim, A S . and Camp, F.E. Camp. J.M. 2: 519, 523 Campagna, M. 1: 136 Campbell, I.A. et al. 1: 944 Campbell, J. and Schrobilgen, G, 3: 115 Campbell, J.B. see Davidson, D.A. et 01.; 326, 334, 335, 336
Cam~uzano"J.C. et al. 1: 136; 3: 222 Camus, C,see Matuszyk, W er al. Cainus, G.M. et al. 1: 970, 971, 973 Canadell, E. see Tillard~C~arbonnel, M. et al. Candro Skekar, N.V. see Sahu, P.Ch. et al. Canepa, F see ~anfrinetti,P. et al.; Merlo, F. er al. Canepa, F et al. 3: 95, 527 Canfield, P.C. see Mmdley, M.F. et al., Lacerda, A. et al.; Severing, A. ef al.; Shen, Z. et al. Canfield, P.C. et nE. 1: 218; 3: 103 Canit, J.C. 2 416 Canova, G. see Kubin, L.P. et al. Canova, G.R. 1: 921 Cantella, M.J. see Kosonocky, W.F. et al. Cao, H.C. 2 290; see also Davis, J.B. rt al. Cao, H.C. et al. 1: 862 Cao, L. see MuZler, K.H. et al. Cao, S.-C. see Zhang, D.-L. et al. Cao, Y. see Sun, Y Y. et al. Capelli, R. 1: 64; see also Ferro, R. et al.
Caplin, A.D. 1: 1021 Capone, B.R. see Kosonocky, W.F. et al. CappeiIetti, R.L. see Neumann, D.A. et al. Capper, P 3 32 Car, R. 1: 133 Carden, J.L. see Grenga, H.E. et al. Cardona, M. and Ley, L. 3: 140 Cardozo, F.A. see Farkas, D. and Cardozo, F.A. Carews, J.A. see Forster, J. et al. Carey, M.J. see Berkowitz, A.E. et al. Carey, R. see Bains, G.S, et al. Cargill, G.S. 111 1: 479, 739, 742 Carisey, T. see Gogia, A.R. et al. Carlson, O.N. see Smith, J.F. et al. Carlsson, A.E. 1: 41,56,64, 65,66, 67,70, 71, 81, 499, 530; 2 156, 157, 219; 3: 765; see also Jeong, E.K. et al.; PhilLips, R. et al., Philligs, R.B. et al. Carlsson, A.E. and Meschter, P.J. 3: 185 Carlsson, A.E. et al. 1: 81 Carlsson, J.R.A. see R a ~ a n a t h G. , et al. Carmichael, C.A. see McKarney, C.G. et al.; see Sctineibel, J.W. et al. Caro, A. 1: 791, 798; see also Riaz de la Rubia, T. et al.; Pedraza, D.F. et al. Caro, A. et al. 1: 570, 792, 798 Caro, J.A. 1: 570 Caron, P. see Blavette, D. et al.; Khan, T. et al. Caron, P. et al. 1: 527, 546; 3: 297, 302, 303 Carpenter, G.J.C. 1: 803; 2: 133; see also Gilbert, R.W. et al.; Schulson, E.M. et al. Carpenter, L.G. 1: 110 Carr, M.J. 1: 453; see also Simonsen, I.K. et al. Carrabba, M.M. et al. 2: 415, 416, 417 Carrillo-Cabrera, W. see von Schnering, H.G. et al. Carrillo-Cabrera, W. see Somer, M. et al. Carrillo-Heian, E. see Xue, H. et al. Carry, C. and Strudel, J.L. 3: 313, 315 Carsi, M. see Limmez, J.A. et al. Carstens, D.H.W. and David, W.R. 3: 30 Carter, C.B. see Wong, Q.Z. et al., Medlin, D.L. et al.; Palmstro~,C.J. et al. Carter, D.H. 2: 295; see also IJnal, 0.et ale Carter, D.H. et al. 2: 295 Carter, F. see Koon, N.C. et al. Carter, G.C. et al. 1: 10 Carter, S.see Cava, R.J. et at. Carter, S.A. see Cam, R.J. et al, Carter, W. see Burton, B. et al. Caruso, R. see Jordan, A S . et al. Carver, T.G. 2: 506 Caspary, R. see Fraas, K. et al. Cassada, W.C. et al. 1: 454 Castagna, A. 3: 338, 339, 340 Castagna, A. see Shea, M. et al.; Stoloff, N.S. et al. Castagna, A. and Stoloff, N.S. 3: 332, 338, 339, 340 Castagna, A. et al. 3 339 Castagnt, J.L. 2: 267 Castanet, R. 1: 114, 1 1 8 Castaiio, V. see Yacaman, M.J. et al. Castro, G.R. see Hum, V. et al.
Castro, G.R. et al. 3: 139 Castro, R.G. et al. 1: 642; 3: 47 Cathcart, J.V. 1: 982, 995; see also Worton, J.A. et al. Catteral, J. 1: 117 Caudron, R. see Ricolleau, C. et al, Causey, A.R. et al. 2: 144 Cava, R.J. see Grier, D.H. et al.; Ikushima, K. et al.; Siegrist, T.et al.; Z a ~ d b ~ r g eH.W. n , et al. Cava, R.J. et al. 3: 103, 251 Cavin, O.B. see Koch, C.C. et al. Cayron, R. 3: 186, 187 Cebulak, W.S. see Paris, H.G. et al. Cedell, T. 2: 401, 402 Ceder, G. 3: I94 Ceder, G. see Tepesch, P.R. et al.; Wolverton, C. et al. Ceder, 6. et al. 1: 41 Celis, P.B. 2: 225 Celis, P.B. et al. 2: 225 Celler, G.K. 1: 650, 792; Celotti, G. see Nava, IF Cenzual, K. see ~ladysh~vskii, R.E. et al.; Jorda, J.L, et al., PartZrC, E. et al. Ceperley, D.M. 1: 128, 131; see also Li, X.-P. et al. Cerchiara, R.R. see Bertziss, D. et al. Ceresara, S . see Bruzzone, P. et al. Cerrina, F. see Boily, S. et al. Cetas, T.C. et al. 1: 1022 Cetel, A. and Duhl, D.N. 3 297, 307 Cezairliyan, A. 1: 1019 Cliaara, M.B. .see Cale, T.S. et al. Chabik, St. 1: 564 Chabot, B. 1: 417; see also Fornasini, M.L. et al.; ParthB, E. and Chabot, B., ParthC, E. et al. Chabot, M. see Baudin, K. et al. Chadi, D.J. 1: 132 Chadwick, G.A. 1: 585, 5 Chafe, J.N. see Zimm, C. Chahine, A. see Tillard-Charbonnel, M. et al. Chahme, R. see Foldeaki, M. et al. Chat, B, see Kryliouk, 0. et al. Chat, L. see Yu, S. et al. Chakachery, E.A. 1: 993 Chaker, M. see Boily, S. et al. Chaki, T.K. 1: 655; see Li, H. and Chaki T.R. Chakoumakos, B.C. see h i a d o r , C. et al.; Sales, B.C. et al. C ~ a k r a b a r ~D.J. i , see ~ k a m o t oH. , et al.; Subramaman, P.R. et al. Chakraborty, S.B. 1: 845 Chakraborty, S.B. and Stark, E.A. 3: 404, 405,411 Chakravorty, S. 2 54,55 Chakravorty, S. and West, D.R.F. 3: 60 Chakravorty, S . et al. 3: 66 Challis, G.A. 1: 631 Chalmers, B. 1: 535, 542 Chamberlin, R.R. see Albright, S.P. et al. Chainberod, A. see Audouard, A. et al.; Garrido, F. et al. Charnberod, A. et al. 3: 272 Chambers, S.A. 3: 784 Chambers, S.A. and Loeb, V.A. 3: 783, 785 Champier, G , see George, A. e~ al.
A u t ~ Index o~ Chan, C.M. see Van Hove, M.A. et al. Chan, C.T. see Kong, X.J. et al. Chan, H.M. see Sung, C.M. et al. Chan, J.Y et at, 3: 104 Chan, K.S. 2 33,78, 120, 121, 122; 3: 479 Chan, K.S. and Shih, D.S. 3: 326, 334, 335 Chan, S.S. see Markiewicz, W.D. et al. Chan, W.K. see Harbison, J.P. et al.; Sands, T. et al. Chan, W Y . et al. 3: 567 Chaiidley, G.D. and Flerniiigs, M.C. 3: 593 Chandra, G. see Mazumdar, C. et al. Chandra, N. see Markiewicz, W.D. et al. Chandra Shekar, M. 'see Sahu, P.C. et al. Chandrasekaran, L. see Saunders, N. et al. Chandrasekaran~M, 2: 55 Chandrashekhar~S. see Chatterjee, S. et al. Clmig, C.A. 2: 624 Chang, C.C. see Palmstrom, C,J, et al.; Sands, T. et al. Chang, C.P. 2: 98 Chang, C.S. 2: 224, 231 Chang, H. see Gibala, R. et al. Chang, H. et al. 3: 753, 762 Chtng, K.M. 2:65, 202, 296; see also Darolia, R. et al., Huang, S.C. et al., Taub, A.Z. et al. Chang, K.M. et al. 2: 39, 64, 65; 3: 226 Chang, K.T. see Merer, G.H. et al. Chang, L.C. 1: 832, 834; 2: 529 Chang, L.C. see Gottstein, G. et al. Chang, L.C. and Read, T A . 3: 55 Chaiig, L.L. aiid Giessen, B.C. 3: 783 Chang, L.L. and Ploog, K . 3: 779 Chang, M. see Hu, J. et al. Chang, S.N. see Ferreira, A. et al. Chang, W.C. see Luo, H. et al. Chang, Y.A. 1: 565, 573; see also Kao, C.R. and Chang, Y.A., Chang, Y.A. and Neurnanii, J.P. 3: 276, 358; Kattner, U.R. et al.; Lin, Y.-C. et al.; ~ e u m a n nJ,P , et al., Pike, L.M. et al.; Perepezko, J.H. et al. Chang, Y.A. et al. 1: 875; 3: 275 Chang, Y.C.see Blson, C.G. et al. Chang, Z.P. 1: 875, 890, 891 Chantsell, R.W and Wohlfarth, E.P 3: 693 Chdo, C.C. et d.1: 703, 705 Chao, G.Y. 1: 627, 628, 629, 631 Chao, P.C. et al. 2: 335 Chaplin, M.R. see Bruzzone, P. et al.; Markiewicz, W.D. et al. Chapnian, G.R. see Patten, E A . et al. Chapnik, I.M. 1: 1021 Chapon, L.et al. 3: 107 Char, C. 2: 45 Charai, A. see Harper, J.M.E. et al., Thomas, 0. et al. Charai, A. et al. 2: 623 Chargin, A.K. 2: 376 Charles, J, see Gachon, J.C. et al. Charles, S.W. see Keeling, L, et al.; Windle, P. L. et al. Charlot, F. et al. 3: 723, 744 Charlot, J.P. see David, 5. et al. Charlot, L.A. see Brimhall, J.L. et al.; Bruemrner, 5.M. et al.
Charlot, L A . et al. 3: 44 Charmers, S.A. and Loebs, V.A. 3: 566 Charpenay, S. see Clapp, P.C. et al. Charpy, G. 1: 7 Chart, T.G. 1: 67 Chartouni, D. see Ziittel, A. et al. Chartouni, D. et al. 3: SO7 Chase, M.W.3: 800 Chaston, J.C. 3: 54, 108 Chaston, J.C. and Sloboda, M.H. 3: 76 Chatterjee, A. et al. 3: 495, 651 Chatlerjee, D.K. 2: 204; see alm Mendiratta, M.C. et al. Chatterjee, S. et al. 3: 669 Chatterji, D. see McCarron, R.L. et al. Chatterji, K. see McCarron, R.L. et al. Chattopadhyay, K. 1: 467; see also Dong, C. et al., Mu~hopadhyay,N.K. et al.; Swarny, V.T. et al. Chattopadhyay, K. see Chitralekha, J . et al. Chattopadhyay, K. et al. 1: 456,457,465, 466 Chaturvedi, M.C. 2: 276; see also Wan, Y.F. et al.; Zhang, Y.G. and Chaturvedi, M.C.; Zhang, Y.G. et al. Chaturvedi, M.C. et al. 2: 276, 277, 283 Chaudari, P. see Dimos, D. et al. Chaudhan, P. see Lancon, F, et al. Chaudhan, P et al. 2: 435 Chaudouet, P.see Fruchart, R. et al. Chave, R.A. et al. 2: 122 Chavez-Pirson, A. see Ojima, M. et al. Chavineau, J.P. et al. 1: 956 Chazalviel, J.N. 2: 416 Chaze, A M . 1: 993 Chaze, A.M. and Coddett, C. 3: 575, 576 Cheeks, T.L. see Pahstrom, C.J. et al.; Sands, T. et al.: Tabatabaie, N. et al. Chelikowski, J.R. 1: 133; 1: 229,232, 246, 247, 612; 2: 610 Chelikowsky. J.R. et al. 3: 811 Chelkowski, A. 3:804 Chellman, D.J. 2: 262, 279; see also Ardell, A.J. et al. Chellrnan, D.J. et al. 2: 280, 281 Cliemla, D.S. see Miller, D.A.B. et al. Chemla, D.S. et al. 2 424 Chen, B. and Franzen, H. 3 56 Chen, B. et al. 3: 106 Chen, B.H. see Yi, S.S. et al. Chen, C. see Feldman, J.L. et al. Chen, C.H. 1: 454,465,475 Chen, C.H. et al. 3: 106 Chen, C.Q. see Zhang, Y.G. et al. Chen, C.W. 2 318; see also Shirane, G. et al. Chen, C.Y. 2: 231 Chen, D. 1: 88; see also Lin, D.L. et: al.; Liu, Z. et al. Chen, D. e f al. 3: 772 Chen, D,Y et al. 3: 691 Chen, F, see Mu, J. et al. Chen, F.C. see Li, H. et al. Chen, F.C. et at. 2: 136 Chen, G. see Chu, T.L. et al. Chen, G. et at. 1: 993; 3: 618 Chen, G.H. 2 45; see also Suryanarayana, C. el al. Cheii, G.L. see Sun, Z.Q. et al., Zhang, L.C. et al.
885 Chen, G.L. et al. 3: 472, 478 Chen, H.1: 776, 777; see also Prieskorn, J.N. et al.; Yang, Y. et al. Chen, H. et al. 1: 483, 748 Chen, H.S. 1: 454,465,475,680,735,141; 3: 682, 685, 686, 687; see also Goldman, A.I. et al.; Hauser, J.J. et al.; Hong, M. et al.; Hsieh, Y,F et al., Inoue, A. et al., Kortan, A.R. et al.; ~ o s k e n m a kD.C. i ~ et al.; Saito, Y. et al., Villars, P et al.; Warren, W.W. Jr. et al. Chen, H.S. et al. 1: 743 Chen, J. see Kaviani. K, et al. Chen, J. et al. 1: 195 Cheii, J.F. see Cheii, L.J. et al. Chen, J.S. see Kolawa, E. et al. Chen, K. see Cadieu, F.J. et al.; Hegde, H. et al.; Rani, R. et al. Chen, K.C. see Hanrahan, R.J. Jr. et al. Chen, L. see Larnarchi, A.M. et al. Chen, L. et al. 1: 774 Chen, L.C. see Spaepen, F. et al. Chen, L.C. et al. 1: 740 Chen, L.J. et al. 3: 664 Chen, L.K. and Tu, K.N. 3: 675 Chen, L.Q. 1: 44, 853; see also Khachaturyan, A. et al. Chen, L.Q. et al, 1: 853, 855, 856 Chen, L.Y. see McGahan, W.A, et al. Chen, M. see Luo, H. et al. Chen, M. et al. 3: 864 Chen, N.Y. 1: 247; see also Yan, L.M. et al.; Zhou, B. et al.; 3: 812 Chen, N.Y.et al. 3: 812, 824, 841 Chen, Q.M. see Huang, L.J. et al. Chen, R. see Chen, N. Y. et al. Chen, R.-L. see Chen, N.-Y et al. Chen, S.P 1: 82, 84, 85, 598; 2: 28; see also Che, X.F. et al.; Farkas, D. et al.; Nu, G.X. e f al.; Srinivasan, S. et al.; Vitek, V. et al., Voter, A.F. et al. Chen, S.P see Mitchell, T.E. et al.; Voter, A.F. and Chen, S.P.; Yan, M. et al. Chen, S.P. et al. 1: 82, 83, 84, 85, 86, SO5, 507,508,598,600,60l,603,604,616, 874, 876; 2: 168; 3: 766 Chen, S.R. and Gray, G.T. I11 3: 363 Chen, T.et al. c): 728, 729, 730, 731, 739 Chen, T.T. 1: 629, 630 Chen, W. see Prieskorn, J.N. et al. Chen, W. et al. 3: 219 Chen, W.R. et al. 3: 297, 317, 318 Chen, W.X. see Lau, S.S. et al. Chen, X. see Tang, W. et al. Chen, X.F. see Hu, G,X. et al. Chen, X.L. see Huang, F. et al. Chen, X.Y. see Yang, C.P et d. Chen, Y. see Angers, L. et al., Wuttig, M. et al. Chen, Z. and Wang, Y. 3: 62 Cheiial, B. see Lasalnionie, A. et al. Cheng, B. see Yang, W.J.S. et al. Cheng, C.W. 1: 875, 887 Cheng, C.J. see Ho, C.T. et al.; Sekhar, J.A. et al. Cheng, C.T. 3: 604, 609 Cheng, C.T. see Ho, C.T. et al.
886 Cheng, C.Y. et al. 1: 571 Cheng, G.H. 1: 923 Cheng, G.H. et al. 1: 925 Cheng, H. see Haase, M. er al. Cheng, J. et al. 1: 817; 2: 143 Cheng, K.Y. see Baillargeon, J.N. et ul, Cheng, S.C. et al. 3 651 Cheng, S.F. see Huang, M,Q. et al. Cheng, S.F. et al. 3: 174, 177 Cheng, T.T. 3: 598 Cheng, T.T. see Shang, P et al., Smith, L.S. et al. Cheng, X.R. see Rao, G.H. et al. Cheng, Y.T. see Juhnson, W.L. rt al. Cheng, Y.T. et al. 1: 698; 2: 610 Cheng, Z.H. see Shen, B.G. et al. Cheng, Z.H. et at. 3: 98 Cheong, S.-W. see List, R.S. et al.; Jin, Z. et al. Cherchiara, R.R. see Berztiss, D.A. et al. Cherkashenko, V.M. see Galakkov, V.R. et al. Chernikova, LA. see Nikitin, §.A. et al. Chernov, D.B. see Khachin, V.N. et al. Chesarek, W see Mitchell, K.W. et al. Chesnutt, J.C. 2: 88; see also Austin, C.M. et al.; Huang, S.C. and Chesnutt, J.C.; Huang, S.C. et al.; Marquardt, B.J. et al.; Rowe, R.G. et al.; Shih, D.S. et al., Tien, J.K. er al. Chester, 6.. see Nicholls, J.R. et al. Chester, C.V. see Leung, P.W. et al. Cheung, C. et al. 3: 671 Cheung, N. et 01. 2: 623, 624 Cheung, N.W. see von Seefeld, H. et al. Cheuiig, T.D. see Cadieu, F.J. et al. Chevalier, J.P see Bellon, P. et al.; Bresson, L. et al.; Yu-Zhang, K. et al. Chevary, J.A. see Perdew, J.P. et al. Chew, N.G. see Baesi, P. et al. Cheynet, B. 3: 800 Cher~~ushkxn, E.A. see Yudina, N.V. et al. Chiang, H.C. see Hsut, Y.Y. et al. Chiang, K.H. er: al. 2: 414 Chiang, K.T. see Meier, G.H. et al.; Perkins, R A . er al. Chiba, A. et al. 1: 899, 925; 2: 39; 3: 58, 64, 307, 3 1 1 C h i d a m b a ~ dP. ~ , see Farkas, D. et aE. Chien, C.L. 1: 440; see also Majkrzak, C.F et al. Chieux, P. 1: 663 Chikazumi, S. 2: 306, 390 Childs, K.D. see M u ~ k a ~M. i , et al. Chin, G.Y. see Mahajan, S. and Chin,
et al.; Lee, T.S. et al. Chipenko, C.V see Degtyareva, V.F. et al. Chirba, V.G. see Testardi, L,R. et al. Chitralekha, J. et al. 3: 731 Chittipeddi, S. et al. 3: 674
Cho, A.Y. see Tsang, T.S. and Cho, A.Y. Cho, A.Y. and Dernier, P.D. 3: 783
Author Index Cho, B.K. see Canfield, P.C. et al. Cho, K. 1: 926 Cho, W. et al. 2: 115 Cho, W.D. see Natesan, K. and Cho, W.D. Cho, Y see Ashida, A. et al.; Kanaya, H. et al. Cho, Y.S. and Koch, C.C. 3: 759, 761 Choe, S.J. see Stoloff, N.S. et al. Choe, W. et al. 3: 531, 532 Choi, B.W. et al. 1: 645 Choi, C.-H. et al. 3: 670 Choi, J. see Henry, M.F. et al. Choi, W.-C. see Lee, J.-Y. et al. Chokshi, A.H. see Nieh, T.G. et al. Choo, Y.H. see Stmtt, P.R. et al. Choppin, G.R. see Gschneidner, K.A. Jr. et al. Chopra, K.L. 2: 410 Chou, C.T. 1: 512, 513; see also Song, Z.Y. et al. Chou, C.T. et al. 1: 512, 513, 514 Chou, T. 1: 620 Chou, T.C. 1: 927, 987, 996, 1003, 1004, 1005, 1007; see also Nieh, T.G. et al. Chou, T.C. et al. 1: 1005; 3: 47 Chou, T.S. see Hon, W.P et al. Chou, Y . 1: 620 Choudhury, A. see Blum, M. et al. Choudhusy, A. et al. 1: 591, 593, 594; 3: 505 Choudhury, N.A. et al. 1: 987, 991, 992, 993 Chow, G.-K. 2 263, 279 Chow, T.P see Machlin, E.S. et al. Chowdhary, R.B. see Singh, N.K.P. et al. Choy, T.C. et al. 1: 470 Christensen, N.E. 1: 66, 70; see also Wang, C.S. et al. Christian, J.W. 1: 690, 829, 915; 2 19, 150; 3: 363, 437; see also Sun, Y.Q. et al. Christian, J.W. see Sun, Y.Q. et al. Christian, J.W. and Crocker A.G. 3: 422 Christian, J.W and Laughlin, D.E. 3: 405, 410 Christian, J.W. and Mahajan, S. 3: 403, 405 Christian, T. see Bruck, H.A. et al. Christides, C. et al. 3: 257 Chnstman, T. 1: 647; see also Jam, M. and Christmaii T. Christman, T. et al. 2: 290 Christmann, K.R. 2 482 Christodolou, L. see Larsen, D.E. et al. Christodoidou, J.A. see Bryanl, J.D. et al. Chsistodoulou, L. 1: 864, 865; see also Larsen, D.E. Jr. et al., Patterson, R.A. et at. Christoph, U. see Appel, F et al. Christopher, H.A. see Rao, B.M.L. et al. Christou, A. see Anderson, W.T. Jr. et al.; Papanicolaou, N.A. et al.; Tseng, W.F et al. Chroniik, R.R. see Zsibi, A. et (d. Chrzan, D.C. 1: 548 Chu, C.W. see Marezio, M. et al.; Sun, Y.Y et al. Chu, F. and Pope, D.P. 3: 415 Chu, IF;. et al. 3: 455, 775 Chu, J.P. et al. 3: 505
Chu, S.S. see Chu, T.L. et al. Chu, T.L. et al. 2: 423 Chu, W.K. see Ottaviani, G. ef al. Chu, W.Y 1: 925 Chu, W.Y. et al. 2: 95 Chuang, T.M. 1: 927; see also Pan, Y.C. e€ d . Chuang, Y -C. see Zhang, D. et al. Chubb, S.R. et al. 1: 205 Chudinov, V.G. see Moseev, N.V. et al. Ckudley, C.T. and Elliot, R.J. 3: 251 Chulkov, N. see Nefedov, V. et al. Chumbley, L.S. see Slueld, J.E. er al. Chung, D.W. 2: 276; si.e also Chaturvedi, M.C. et al. Chung, H.F. see Eples, I.E. et al. Cbung, H.H. et al. 3: 645 Chung, P.L. see Wh~tten,W.B. et al. Chung, S.S. see Hirano, T. et al. Chupenko, G.V. and Degtyareva, V.F. 3: 160 Chuprina, V.G. see Arbuzov, M.P. et al. Chuvildeev, V.N. see Perevezentsev, V.N. et L d . Chuyanov, V see Conn, R.W. et al. Cialone, H.J. see Holbrook, J.H. et al. Ciancetta, G.M. see Markiewicz, W.D. et al. Cichy, M.A. see Zhu, Q. et al. Cicognani, G. see Cristofolini, L. et al. Cieslak, M.J. Z: 128; see also Baeslack, W.A. I11 et al. Cieslak, M.J. et al. 1: 655; Ciosek, S.J. 2: 403 Cirafici, S. see Canepa, F, et al. Cisar, A. see Belin, C . et al. Cisar, A. and Corbett, J.D. 3: 114 Claeyssen, F. 2: 403 Claeyssen, F. et al. 2: 403 Claisse, F see Bouchard, M. et al. Clapham, V.M. see Fowler, P.H. et al. Clapp, P.C. 1: 28; see also Cohen, M. et a/,; Moncevicz, A. et al.; Rifkin, J.A. et al., Russell, S.M. et al. Clapp, P.C. et al. 1: 82, 566, 611; 2 58 C~areborough,L.M. 1: 540; see aIso Head, A.K. et al. Clark, A.E. 2: 317, 382, 388, 391, 392, 393, 394, 399; see also Moffett, M.B. et al.; Sato, K. el al.; Savage, H.T. et al. Clark, A.E. et al. 1: 885; 2 390, 391, 394, 398, 399,400 Clark, A.M. 1: 628, 629 Clark, H.M. 1: 830 Clark, R.K. see Wallace, T A . et al.; Wiedemann, K.E. et al. Clark, R.W. 1: 1024; 2: 57, 58 Clark, S.M. see Brazhkin, V.V. et al. Clarke, D.R. see Lipkin, D.M. and Clarke, D.R.; Ma, Q. and Clarke, D.R. Clarke, M. 2: 506, 51 1 Clarke, M. et al. 2: 507, 511 Clarke, R.L. 2: 492 Clarke, R.S. 1: 3 Clarke, R.S. et al. 1: 3 Clarke, R.S. Jr. 1: 631, 633; 2 306 Clasen, R. 3: 806 Claus, H. ,see Campuzano, J.C. et al. Clausen, K.M. see Mason, T.E. et al.
Author Index Clarasen, K.N. see Andersen, N.H. et al. Clauss, A. see Guille, J. et cif, Clavaguem, N. see Baro, M. et al. Clavaguera-Mora, M.T. see Baro, M. et al. Clay, B.D. 1: 918 Clayton, C.R. see Hubler, G.K. et al. Clegg, W.J. 1: 918 Clernens, B.M. 2: 610, 613; see also Nix, W.D. and Clernens, B.M. Clemens, B.M. et al. 1: 734, 735 Clernens, D. see Wang, J.N. et al. Clemens, D.R. see Liu, C.T. et al. Clemens, H. see Appel, F. et al.; Bartels, A. et al.; Chatterjee, A. et al., Kestles, H. et al.; Koeppe, C. et (11.; LeHolm, R. et al.; Yolton, C.F. et d. Clernens. H. and Restler, H. 3: 626, 632, 633, 638 Clemens, H. and Schretter, P 3: 631, 632 Clernens, H. et al. 3: 475, 626, 631, 632, 633,634,635,637,638,650,651,652 ClCment, N. see Caillard, D. ClCment, N.et al. 1: 521, 523 Clementi, E. and Roetti, C. 3: 766 Clements, H. see Markete, W.T. et al. Clemrnens, D. see Larsen, S.E. et al. Clevenger, L.A. see Ma, E. et al., Svilan, V. et al. Cliff, G. see Lorimer, G.W. et al. Cliff, G. et al. 1: 840 Clift, M.W see Mirkarnni, P.B. et al. Cline, H.E. 1: 191; 2: 296 Clurn, J. see Zribi, A. et al. : 176; see also Kieschke, R.R. et al. Coad, J.P. 2 492, 495 Coble, R.L. 11: 918 Cochran, W.T. see Chittipeddi, S. et al. Cochrane, R. see Howson, M.A. et al. Cochrane, R.F. see Vittra, S . et al. Cockayne, D.J.H. 1: 522; see QLW Korner, A. et al.; Ray, 1.L.F et al. Cockayne, D.J.H. and Vitek, V. 3: 463 Cockayne, D.J.H. et al. 1: 497, 508; 3: 462 Cockeram, B.V 3: 556 Cockeram, B.V. and Rapp, R.A. 3: 556, 574, 577 Cockeram, B.V. et al. 3 574 Coddett, C. 1: 993; see also Chaze, A.M. and Coddett, C. Cody, G.D. see Rehwald, W et al. Coetzee, R. see WolE, I.M. et al. Coey, J.M.D. 1: 683; 2: 310, 314, 315; see also Hu, B.P. et al., Leithe-Jasper, A. et al.; Mitchell, Z.V. et al.; Otani, Y. et al.; Sun, H. et al. Coey, J.M.D. and O’Donnell, K. 3: 102 Coey, J.M.D. et at. 2: 314. 315; 3: 100 Cogan, S.F see Schwall, R.E. et at. Cogan, S.F. et al. 2: 361 Cohen, E. 2: 344 Cohen, J. 1: 777, 786 Cohen, J.B. 1: 478; 2: 576; see al;\o Hughes, T. et al.; Mikkola, D.E. and Cohen, J.B.; Terauclii, H. et al. Cohen, J.B. and Weertman, J. 3: 422 Cohen, M. 1: 830, 831, 832, 897; 3: 841; see also Mehrabmi, R. el al.; Ruhl, R.C. et al.
Cohen, M. see Christian, J.W et al.; Green, M.L. and Cohen, M. Cohen, M. et al. 1: 829 Cohen, M.H. 1: 679 Cohen, M.L. 2: 132,252,319,422; 3: 811; see also Barbee, T.W. et al.; Zhang, S.B. et al. Cohen, P.I. see Wowchak, A.M. et al. Cohen, R.E. see Burton, P.B. and Cohen, R.E.; Mehl, M.J.et at., Pickett, W.E. et al.; Singh, D. et al. Cohen, R.W see Rehwald, W. et al. Cohen, U. 1: 652 Cohn, J.L. see Tritt, T.M. et al. Cole, J.L. see Song, S.G. et al. Coleridge, P.T. see Remders, P.H.P. et ul. Coles, B.R. 1: 212, 221, 943, 953; see also Dernczyk, B. et al.; Murani, A.P et al., Roy, S.B. et al. Coles-Hamilton, C, see Lacy, D.E. et al. Colgan, E.G. 2 625, 630; 3: 673, 676; see also Hung, L.-S. et al., Olowolafe, J.O. et al. Colgan, E.G. et al. 2 608, 609 Colinet, C. see Le, D.H. et cd.; Pasturel, A. et al. Colinet, C. and Pasturel, A. 3: 799 Colinet, C. et al. 1: 41; 2: 610 Coll, J.A. 2: 150; see also Cahn, R.W. and Coll, J.A.; Lawley, A. et al. Collen, B. see Anderson, S . et a2. Collings, E.W 1: 669, 670; see also Enderby, J.E. and Collings, E.W. Collings, E.W. et al. 1: 1018, 1028, 1029 Collins, D.A. see Lile, D.L. et al. Collins, G.S. et al. 3: 278 Collins, J.G. see Barron, T.H.K. et al.; White, G.K. et al. Collins, M. 1: 777, 778, 779 Collins, M.F. see Wood, J.H. et al. Collocott, S.J. et al. 1: 1020, 1024 Collomb, A. see Ayres de Campos, J . et al. Collver, M.M. 1: 689; 2: 610, 612 Colornbani, D. see Claeyssen, F. et al. Colter, P. see Hussien, S A . et al. Colvin, G. 3: 593 see Jones, P.E. et al. Compaan, A. et crl. 2: 423, 424 Compans, E. see von Lolineysen, H. et al. Compton, D.N. et al. 3: 62 Compton, V.B. see Mathias, B.T. et al. Comsa, G. see Morgenstern, K. et al., Niehus, H. and Conisa, G; Niehus, H. et al. Comstock, R.L. 2: 448 Condat, M. 1: 574 Condat, M. see Rubin, L.P et al. Conde, F. et al. 3 691 Conn, R.W. et a1 2: 375 Connell, G.A.N. 2 435, 441, 450 Conner, R.D. et al. 3: 684 Connolly, J.W.D. 1: 24, 137 Conradi, M.S. see Jeong, E.K. et al.; Kimmerle, F. et al. Conradson, S.D. see List, R.S. et al. Conservx, M. et al. 2: 188 Cook, B. see Akinc et al. Cook, D. see Rawers, J.C. et al. Cook, H. 1: 775, 781 Cook, J. see Lee, E.W et al. Cook, J. et al. 1: 1007 Cook, J.C. see Skripov, A.V et al.
887 Cook, J.W. Jr. see Ren, J. et al. Cook, N.C. 2 506, 521 Cook, W.R. Jr. see Zare, a. et al. Cooke, C.M. see Eylon, D. et al. Coolidge, W.D. 3: 31 Cooper, A.S. see Bucher, E. et al. Cooper, B.R. see Lim, S.P. et al. Cooper, C.V. see Inoue, I3.R.P et al. Cooper, M. 1: 473; 2: 177, 178, 182, 186, 187 Cooper, M.J. 1: 565 Cooper, R.F. see Allen, W.P et al. Cooper, R.M. 3: 656 Cope, M,T. see Penton, R.J.T. et al.; Postans, P.J. et al. Copley, J.R.D. see Christides, C. et al.; Neurnann, D.A. et al. Copley, S.M. 1: 546, 882, 896; 2: 11, 13, 17, 24 Coqblin, B. see Martinez, G. et al. Corant, J.W 1: 413 Corbel, C. 1: 561; see also Dimrtrov, C. et al., Doyama, M. et al. Corbel, C. er al. 1: 580 Corbctt, J.D. 1: 303; 3: 113, 120, 123, 126 Corbett, J.D. see Belin, C. et al.; Cisar, A. and Corbett, J.D.; Dong, Z.C. and Corbett, J.D.; Edwards, P.A. and Corbett, J.D., Guloy, A.M. and Corbett, J.D.; Hennmg, R.W. and Corbett, J.D.; Huang, D.P. and Corbett, J.D., Huang, D.P. et al.; Kaskel, S. and Corbett, J.D.; Kwon, Y.U. et al.; Leon-Escamilla, EA. and Corbett, J.D.; §aboungl, M.L. et al.; Sevov, S.C. and Corbett, J.D.; Sevov, S.C. et al. Cordi, R.C. 1: 651 Cordier, G. see Eisenmann, B. and Cordier, G. Corey, C. 1: 777 Corey, R.G. et al. 2: 492 Cornelis, I. 1: 836, 837, 838 Cornelius, A.L. et al. 3: 162 Cornies-Quinquandon, M. see Devaud-Rzepski, J. et al. Cornish, G.R. see Seiiiiatin, S.L. et al. Cornish, J.B. see Alcock, C.B. et al. Cornish, L.A. see Compton, D.M. et al. Harte, A S . et al.; Hill, P.J. e f al.; Hohls, J. et al., Horner, Z.J. et al.; Levey, F.C. et al.; WOE, I.M. et al., Wong-Kian, M. et al. Cornwell, L.R. et al. 2: 22 Corti, C.W. see Coupland, D.R. et al. Cork C.W. et al. 3: 66 Cortie, M.B. see Horner, I.J. et al.; Levey, F.C. et al., Wong-Kian, M.et al. Cortie, M.B. et al. 3: 62, 66 Costa, P see Khan, T. et al.; Lasalmonie, A. et al. Coster, D. et al. 1: 172 Cotton, J.D. and Field, R.D. 3: 49 Cotton, J.D. et al. 2: 296; 3: 352 Cottrell, A.13. 1: 437; 504, 512; 3: 308, 364 Cottrell, A.H. and Bilby, BA. 3: 422 Cottrell, §.A. 1: 835 Cotts, E.J. ,see Zribi, A. et al. Cotts, E.J. er al. 1: 699 Cottstem, G. 3: 917 Couch, D.E. see Brenner, A. et at.
888 Coujou, A. see Caillard, D. et al.; Couret, A. et al.; Farenc, S. et al. COUJOU, A. et al. 3: 411 Coulomb, P. see Reynauld, F. et al. Counihan, P.J. et al. 3: 645, 723, 740 Coupland, D.R. see Corti, C.W. et al. Coupland, D.R. et al. 3: 66 Couret, A. 1: 549; 2: 27; see also Caillard, D. et al., ClCment, N. et al.; Farenc, S. et al.; Legros, M. et al.; Paidar, V. et al.; Sun, Y.Q. et al. Couret, A. et al. 1: 548, 549; 3: 407, 408, 409,422 Cour~ngton,S.V see Pharr, G.M. et al. Court, S.A. see Evans, D.J. et al., Lofvander, J.P.A. et al., Vasudevan, V.K. et al., Wiezorek, J.M.K. et al. Court, S.A. et al. 1: 533, 534, 535, 539, 541, 543, 545, 552; 2: 77, 102, 103 Courtney, T.H. 1: 647; see also Maunce, D. and Courtney T.H.; Maurice, D.R, and Courtney, T.H. Courtney, T.H. and Maurice, D. 3: 730 Courtois, D. see Ei, H.S. er al. Courtois, D. et al. 3: 94 Courtright, E.L. see Mueller, A. et al. Coutts, T.J. et al. 3: 896 Coweti, V.C. see Fowler, P.H. et al. Cowlam, N. see Sakata, M. et al. Cowles, B.A. see DeLuca, D.P. et al. Cowley, J.M. 1: 28, 50,498, 504, 507, 781 Cowley, R.A. 3: 258; see ralso Harris, M.J. et al. Cox, A.P.D. see Webster, G.A. et al. Cox, A.R. and van Reuth, E.C. 3: 306 Cox, B.N. et al. 2: 288, 292 COX,D.E. 1: 221, 441, 803; see also Yamada, T. et al. Cox, D.M. see Neumann, D.A. et al.
Cracknell, A.P. 1: 1018 Craft, A. see Nesbit, S. et al. Craft, A.P see Flanagan, T.B. et al. Craig, J.R. 1: 635 Craig, R.S. see Wallace, W.E. et al. Cramb, A. see De Craef, M. et al. Crainb, A.W see Murafeedharan, K. et al.; Rishel, L.L. et al. Crapper, M.D. see Brown. D. et al. Crawford, A. see Counihan, P.J. et al. Crawford, J.W. Jr. 1: 559 Crawford, K. see Campisi, I.E. et al. Crawford, R.C. 1: 504,506,507, 509,521, 537; see also Ray, I.L.F. et al. Crawford, R.C. and Ray, I.L.F. 3: 454 Creegan, K.M. see Neurnann, D.A. et al. Crespo, P. see Yavari, A. et al. Cressoni, J,C. 1: 420 Cretu, C. and Van der Lingen, E. 3 77, 232 Criddle, A.J. 1: 625, 626 Crimp, M.A. 1: 503, 505, 509, 529, 540, 905; 2: 201,202,203,204,205; 3: 443; see crkso Sun, Y.Q.et al. Crimp, M.A. et al. 1: 536; 2: 203, 204 Crisp, V.H.C. see Lock, D.G. et al.
Cristofolini, L. et al. 3: 258 Crites, J.W. see Geissberger, A.E. et 171. Croat, J.J. 2: 314; 3: 102 Croat, J.J. et al. 1: 748; 2: 314 Croce, P. see Chavineau, J.P. et al. Crocker, A.G. 1: 844; 3 416; see also Bew, M. and Crocker, A.G.; Bilby, B.A. and Crocker, A.G.; Christian, J.W. and Crocker A.G. Crombeen, J.E. see Reader, ASH.et al. Cromer, D.T. et al, 1: 413 Cronin, G.R. 2: 417 Croockewit, K. 1: 5 Cros, C. see Rasper, J.S. et al.; Reny, E. et al. Cros, C. et al. 3: 130, 131 Cross, C.R. see Rowe, J.R. et al. Cross, L.E. see Li, J. et al. Crossland, I.G. 1: 918 Crossman, F W. 1: 925 Crowder, C. see Hardman-Rhyne, M. et al. Crowder, C.E. 1: 447,448 Crowder, D.N. 2 394 Crowe, C.R. see Feng, C.R. ei al. Crowell, C.R. and Sze, S.M. 3: 787 Crowell, W. see Wise, E.M. et al. Crozier, E.D. 1: 671 Crucq, A. et al. 1: 610, 612, 613, 618 Crussard, C. 1: 829; 3: 352 Cruzan, P.D. see Sun, R.C. et al. Csanady, A. see Nissen, H.U. et al.; Urban, K. et al. Csanady, A. et al. 1: 454 Cserti, G. see Vitek, V. et al. Cserti, J. see Khantha, M. et al.; Vitek, V, et al. Cserti, J. et al. 3: 439, 450 Cubin, L.P. see Petez, K. et al. Cuevas-Diarte, M.A. see ~ o n d i e i gD. , et al. Culbertson, G. and Kortovlch, C.S. 3: 572 Cullen, G. see Hanak, J. et al. Cullers, C.L. see Noebe, R.D. et al. Cullis, A.G. see Baeri, P. et al. Cunat, C. et al. 3: 699 Cunat, Ch. et al. 1: 743 Cunat, S. see Wotin, M. et al. Cuomo, J.J. .we Chaudhari, P. et al.; Krusm-Elbaum, L. et al. Cupschalk, S.G. 1: 496, 503, 509, 511, 574, Curda, J. see von Schnenng, H.G. et al. Curran, P.M. see Erickson, J.S. et al. Curtin, W.A. see He, M.Y. et al. Cusack, N.E. 1: 677 Cuthill, J.R. see Carter, C.C. et al. Cutler, M. 1: 671 Cyrot, M. 1: 57 Cyrot-Lackmann, F. 1: 57; see also Klein, T. et al.; Pasturel, A. et al.; Singh, D.J. et al. Czakb-Nagy, 1. see Leidheiser, H. et al. Czopnik, A. sec Morin, P. et al. Czybulka, A. et al. 3: 238 Czyzyk, M.T. et al. 3: 149 Daadbin, A. see Griffin, D. et al. Daalmans, G.M. see Hoenig, H.E. et al. Daams, J.L.C. 1: 237, 256, 259, 365,437
Daams, J.L.C. et al. 1: 227, 237, 256, 259, 272, 364, 365, 367, 370, 437; 3: 797 Daane, A.H. see Spedding, F.H. et al. Dabrowski, J. 1: 580 Dahl, K. 1: I5 Dahl, 0. 1: 923; 2: 306 Dahl, W. et al. 2: 282 Dahlgren, S.D. 3: 65 Dahmani, C.E. see Cadeville, M. et al. Dahms, M. see Schneider, D. et al. Dahms, M. et al. 1: 646 Dai, C. see Fang, R.-Y. et al. Dai, D . 4 . see Fang, R.-Y. et al. Dai, C. see Sitaud, B. et al. Dai, M. see Miyazaki, T. et al. Dai, M.X. see Feng, J.L. et al.; Wang, R.H. et al. Dais, S . et al. 3: 290 Dalal, R.P. et al. 3: 305, 306 Dalder, E. see Uang, W.S. et al. Dalgleish, B.J. see Cao, H.C. e f al., Sigl, L.S. et d. Dalichaouch, Y. see Lopez de la Torre, A. et al. Dalton, J. 1: 8 Dam, B, see de Reus, R. et al. Damask, A.C. see Gilbert, J. et al.; Koczak, M.J. et al. Damen, T.C. see Miller, D.A.B. et al. Damjanovic, A. see Yeh, L.S.R. et al. Damjailovic, D. see Li, J. at id. Damkroger, B.M. see Patterson, R.A. et al. Darnmak, H. see Barbu, A. et al.; Dimitrov, C. et al.; Dunlop, A. et al. Dammak, N. and Dunlop, A. 3: 267 Dammak, H. et al. 3: 267, 268 Damsa, H. 1: 392 Dainsa, H. see Ilavinga, E.E. et al. Damson, B. see Schaefer, W.-E. et al. Dana, E.S. 1: 17811, 180, 181 Dana, J.D. 1: 180, 181 Dandekar, D.P see Ho, J.C. et al. Danesi, W.P. 2: 4 Danhazi, Z . 1: 485 Daniel, M.R. see Randall, R. et al. Danielewski, M, see Datta, P.K. et al.; Holly, K. and Danielewski, M.; Mrowec, S. et al. Danielewski, M. and Filipek, R. 3: 568, 569 Danielewski, M. et al. 3: 564, 568 Danielson, G.C. 1: 1027, 1028; see also Whitten, W.B. et al. Dankhazi, Z. see Belin-Ferre, E. et al.; Sadoc, A. et al.; Trambly de Laissardiere, G. et al.; Traverse, A. et al. Dan’kov, S.Yu. see Pecharsky, V.K. et al. Dan’kov, S.Yu. et al. 3: 525, 527, 529 Dann, T. see Kryliouk, 0. et al. D’Anna, E. et al. 1: 641 Dannohl, H.-D. 1: 69; 2 54 de Provincia, 3. 1: 129 Darack, S. see Andres, K. and Darack, S. Dardel, B. see Grioni, M. et al. Dardi, L.E. see Dalal, R.P. et al,
889 Dariel, M.P. et al. 2: 394, 396; see also Atzmony, V. et al.; Klimker, H. et al.; Rosen, M. et al. Darin, A. see Duret, C. et al. Darken, L.S. 1: 245, 246, 247, 667, 668, 712 Darken, L.S. et al. 1: 102; 3: 812 Darling, AS. 3: 53, 78 Darnell, A.J. et al. 1: 707 Darolia, R. 1: 904, 1027; 2: 56, 57, 58, 65, 202; 3: 275, 602, 603; see also Bowman, R. et al.; Chang, K.-M. et al.; Field. R, et al.; Field, R.D. et al.; Garg, A. et al.; Kumar, K.8. et al.; Lahrman, D.F. et al., LOCCI, I.E. et al.; Maloy, S.A. et al., Miracle, D.B. and Darolia, R.; Schneibel, J.H. et al.; Walston, W.S. and Darolia R.; Walston, W.S. et al.; Walston, W.S. and Darolia, R. Darolia, R. and Walston, W.S. 3 297, 305, 307, 316 Darolia, R. et al. 1: 539,924; 59. 60, 61, 62, 66, 67, 68 351, 451, 494, 604 Dartyge, E. see Isnard, 0. et al. Das, B.K. see Schmerliiig, M.A. et al. Das, B.K. and Liebeman, D.S. 3: 56
Das, G. 1: 1007 Das, G. and Cleinens, H. 3: 637, 638 Das, G. and Krtshna Das Gupta, A. et al Das, I. see Sarnpath Das, M.B, 2: 335 Das, S.B. 1: 944 Das, S.K. 2: 175, 178, 180, 185, 188, 190, 194; see also Brown, A M . et al.; Kim, N.J. et al.; LaSalle, J.C. et al.; Zedalis, M.S. et al. Das, S.K. et al. 2: 178, 180, 194 Dasbach, D. et al. 2: 377 da Silva, C.G. see Kumar, V. et al. da Silva, X.A. see von Ranke, P.J. et al. Datta, A. et al. 1: 748 Datta, P.K. see Chan, W.Y. et al.; Danielewski, M. et al.; Du, H.L. et al.; Fisher, G. et al., Griffin, D. et al.; Strafford, K.N. and Datta P.K.; Xiang, Z.D. et at, Datta. P.K. et al. 3: 563, 564, 566, 567, 569, 707, 716 Dliubler, M.A. see Winkler, P.J. et al. Daudey, J.P. see Malneu, J.P. et al. Daudin, B. and Boiijour, E. 3: 527 Dauebler, D. see Lutjering, G. et al. Daulton, T.L. 1: 457, 460, 465, 467, 470, 477 Daulton, T.L. et al. 1: 466, 467,469,470, 471,483 Daum, C. see teii Kate, H.H.J. et al. Daunt, J.G. see Heer, C.V. et al. Dautovich, D.P. 1: 716 Dautreppe, D. see Ntel, L. et al.; Paulevt., 3. et al. Davenport, E.S. 1: 835
Davenport, J.W. 1: 60, 908, 921; see also Fernando, G.W. et al.; Lui, S.-C. et al.; Watson, R.E. et al. Davenport, J.W. et al. 1: 60, 67, 68, 133 Davenport, R.M. see Dickson, C. et al.; Oglesby. P.L. et al. Davey, J.E. see Anderson, W.T. Jr. et al. Davey, J.E. and Pankey, T. 3: 779 Davey, S. et al. 3: 623, 624 Davey, S.J. see Gouma, P.I. et al, Davey, T.R.A. 2: 516 David, J. et al. 1: 359 David, S.A. 2: 202; see also Joslin, D.L. et al.; Santella, M.L. et al. David, S.A. et al. 2: 45 David, W.I.F. et al. 1: 278 Davidson, D. see Larsen, S.E. et al. Davidson, D.A. et al. 2: 122 Davidson, D.E. 3: 601; see also Frecke, T. and Davidson D.E. Davidson, D.L. and Anton, D. 3: 342 Dawdson, D.L. et al. 3: 337 Davidson, R.D. 2: 135 Davies, C.K.L. et al. 2: 10 Davies, C.M. Jr. 2: 390, 396 Davies, D.K. 2: 432 Davies, E.A. 1: 671 Davies, G. see Smith, G.D.W. et al. Davies, G.J. and Willianis, D. 3: 779 Davies, H.A. 1: 735, 745; see also Li, Y et al.; Harland, C.L. and Davies, H.A.; Manaf, A. et al.; Otani, Y . et al.; Sakata, M. et al. Davies, P.W. et al. 1: 917 Davies, R.A. see Pettifor, D.G. et al. Davies, R.G. see 1: 521, 534, 546, 896, 913,914,915,920,921,925; 2: 18,24, 203, 213; see also Beardmore, P. et al.; Boettner, R.C. et al.; Stoloff, N.S. and Davies, R.G.; Thornton, P.H. et al. Davila, M.E. see Castro, G.R. et al. Davis, G.L. 2: 653 Davis, H. 1: 616 Davis, H.L. see Noonan, J.R. and Davis, H.L. Davis, N.L. and Naonan, J.R. 3: 219,225 Davis, J.B. et al. 1: 913 Davis, L.A. 2: 175, 178, 180, 185, 188, 194; see also Datta, A. et al. Davis, L.A. and Kavesh, S. 3 694 Davis, R.F. 1: 579; 3: 663 Davis, R.L. see Li, H.S. et al.; Tibballs, J.E. et al. Dams, R.M. et al. 1: 647; 3: 756 Davis, S. et al. 2: 610 d’Avitaya, FA. see Rosencher, E. et al. Daw, M.S. 1: 78, 79, 82, 155, 505, 507, 523, 564, 570,575,611,797,798, 874, 875; see also Baskes, M. et al.; Foiles, S.M. and Daw, M.S.; Foiles, S. et al.; Foiles, S M . et al.; Mills, M.J. et al., Phillips, R. et al.; Phillips, R.B. et al.; Srinivasan, R. et al.; Yoo, M.H. et al. Daw, M.S. and Baskes, M.I. 3: 215, 765, 766 Daw, M S . et al. 3: 766 Dayananda, M.A. 1: 923 Deadmore, D.L. I: 980: see also Santoro, G.J. et al. Deal, B.E. 1; 997
Dean, T.A. see Brooks, J.W. et al., Davey, S. et al. Deb, P. see Dust, M. et al. Debek, V. see Dimitrov, C. et al. de Bergevm, F, see Bertaut, E.F. et al. de Bithre, P. see Sajovec, F. et al. de Boer, F.R. see Liu, J.P. ct al.; Miedema, A.R. et al.; SchobingerPap~mantel~os, P. et al.; Tang, N. et al,; Yang, F.M. et al. de Boer, F.R. et al. 1: 99, 100, 231, 232, 698, 799 de Boer, J.L. see Meetsma, A. et al. DeBoissieu, M. see Dong, C. et al. de Bondt, M. 1: 836, 842 de Bruijn, N.C. 1: 457 de Bussac, A. et al. 1: 547 Debye, P. 1: 10; 3 519, 536 Dkamps, B. see Toniinez, E. et al. de Chltel, P.F. see Liu, J.P. et al. Decker, R.F. 2: 5, 9, 283; 3: 308 DeCristofaro, N. 3: 698 DeCristofaro, N,J. see Datta, A. et al. Dedek, U. see Dimitrov, C. et al. Dederichs, P.H. see Stefanou, N, et al. Dedericks, P.H. 1: 1019 Deevi, S.C. 3: 344, 345, 655, 733; see also Hajaligol, M.R. et al.; Jake, C.E. et al.; Lilly, A.C. et al., Sadananda, K. et al.; Sikka, V.K. and Deevi, S.C.; Sikka, V.K. et al., Stoloff, N.S. C? al. Deevi, S.C. and Sikka, V.K. 3: 275, 502, 604, 607, 645, 656, 657 Deevi, S.C. and Zhang, W. 3: 503, 506 Deevi, S.C. et al. 3: 504, 506, 507, 639, 654, 656 Defay, R. 1: 675, 678 Deffeyes, J.E. see Soboyejo, W.0, et al. de Fontaine, D. 1: 26, 523, 773, 775, 781, 855, 857; 3: 307; see also Asta, M. et al.; Ceder, G. et al.; Dreysst, H. et al.; Kikuchi, R. et al.; Mohri, T. et al.; Sluiter, M. et al.; Turchi, P. et al.; Wolvertoii, C. et al. de Fontaine, D. see Althoff, J.D. et al.; Amador, C. et al.; Asta, M.D. et al.; Sluiter, M.H.F et al.; Vul, D. and de Fontaine, D., Wille, L.T. and de Fontaine, 13.; Wolverton, C. et al. Defour, C. et al. 3: 268 Degawa, T. see Kamata, K. et al. Degawa, T. et al. 1: 638, 639 DeGelormo, J. see Basile, D.P. et al. Degen, T . see ~ e r g e r h o G. ~ , et al. de Gironcoli, S . see Cianozzi, P. et al. Degols, L. see Crucq, A. et al. De Graef, M. see Muraleedharan, K. et al. De Graef, RI. et al. 3: 594, 595, 597, 598 de Groot, C.H. see SchobingerPapamantellos, P. et al.; Hautot, D. et al. de Groot, D.G. see Salomons, E. et al. de Groot, R.A. see Sarma, D.D. et al.; Weger, M. et al.; Xu, R. et al. Degtyareva, V.F. see Chupenko, G.V. and Degtyareva V.F.; Tissen, V.G. et al.
890 Degtyareva, V.F. and Ponyatovsky, D.G. 3: 159, 161 Degtyarcva, V.F et al. 3: 159, 160, 161 Degtyaryov, Yu, 1. et al. 3: 823, 814, 816 Dehghan, K. see Dubois, J.M. et al. Dehlinger, U. 1: 758 De Hosson, J.T.M. sec Van Der Wegen, G.J.L. et al. deHosson, J.Th.M. see Elswij, H.B. et al.; Kruisman, J.J. et a/.; Pestman, B.J. et al.; Vitek, V. et al. Delirli, M.L. see Shoemaker, J.R. et al. Deis, D.W. see Randall, R. et al. Deitch, R.H. see Parthk, E. et al. De Jong, W.F. 3: 4 DeJonghc, L.C. see Murugesh, L. et al. Dejus, R. see Brun, T.O. et al. Dekhtiar, I.J. see Hertzriken, S.D. et al. de Kort, K. see Buschow, K.H.J. et al. Dela Torre, S.D. et al. Delaey, L, I: 837 Delafond, J. see Jaouen, C. et al. Delage, S. see Rosencher, E. et al.; see Thomas, 0. et al.; Thompson, R.D. et al. Delagi, R.G. see Jha, S.C. et al. Deia~ey,D. see Shen, 2;. et al. Delaney, J.A. 1: 762 Delapalme, A. see Bertaut, E.F et al. Delavignette, P. et al. 1: 566 Deline, V.R. see d’Heurle, F.M. et al.; Finstad, T.G. et al. DellaCorte, A. see Blau, B. et al. Dclla-Negra, A. see Dammak, H. et al. Della-Negra, S. see Baudin, K.. et al.;
et al. DeLuca, D.P. et al. 2: 93, 122, 123 DeLuca, M.S. 2: 523 DeLuca, P.M. see Wiezorek, J.M.K. et al. Demania, D.A. see Walston, W.S. et al.
~ e i ~ c ~B.y ket ,al. 1: 215, 221 Demczyk, B.G. see Cheng, S.F. et al. Deniianczuk, D.W. 1: 955 de Miranda, J. 1: 577, 775, 777 de Mooij, B. 1: 419, 434 de Mooij, D.B. see Buschow, K.H.J. et al., van Engelen, P.P.J. et al. Dempsey, N.M. see Mller, K.H. et al. Demukai, N. see Yamada, J, and Demukai, N. Dench, W.A. see Kubaschewski, 0. et al. de Neufville, J.P 1: 732 Deng, D.-C. 1: 74 Deng, D.P. see Widom, M. et al. Deng, H. see Phillips, R. et al. Deng, W.F. see Wang, Z.G. et al. Deng, Y.G. see Choi, B.W. et al. Denier van der Gon, A.W. see Barbour, J.C. et al. Denier van der Gon, A.W. et d.2: 610, 612, 614 Denike. K.K. see Fleischer, R.L. et aE. Denissen, C.J.M. see Buschow, K.H.J.
Author Index Dennis, T.J.S. see David, W.1.F. et al. Dennison, D.H. see Spedding, F.H. et a/. de Novlon, C.H. 1: 295, 304, 579; see also Alamo, A. et al. den Ouden, A. see ten Kale, H.H.J. et al. Denoyer, F see Launois, P. et al. Dentenerr, P.J.H. 1: 208 de Oliveira, I.G. see von Ranke, P.J. et al. Depauw, J. see Baudin, K. et al. DCportes, C. see Fabry, P. et al. Deportes, J. et al. 1: 449 de Potter, M. et al. 2: 621 DePristo, A.E. 1: 79 DePuydt, J.K. see Haase, M. et al. DePuydt, J.M. see Cheng, H. et al.; Haase, M.A. et al.; Neumark, G.F. et al.; Park, R.M. et al. de Raedt, W. see de Potter, M. et al. Derby, B. 1: 917 Derdau, D. see KBppers, M. et al. De Renzi, R. see Cnstofolini, L. et al. de Reus, R. 2: 610, 61 1, 619, 625; see also Barbour, J.C. et al., Denier van der Gon, A.W et al.; Pretorius, R. et al. de Reus, R. et al. 2: 610, 613, 618, 619, 620, 626, 627, 629 Dergacheva, M.B. see Kozin, L.F. Dernier, P.D. see Cho, A.Y. and Dernier, P.D.; Marezio, M. et al. Derouwaux, P 2 572 Derrien, J. 2: 231 Deruyttere, A. 1: 836, 842 Dew, P. see Hultgren, R. et al. Desai, P.D. 1: 69, 70, 117; ,see also Hultgren, R. et al.: Touloukian, Y.S. et al. Desamot, G. see Alamo, A. et al., RiviZre, J.P. et al. DeSavage, B. see Clark, A.E. et al. Desborough, G.A. et al. I: 629, 630 Desch, C.H. 1: 8, 9, 11, 14 Desch, P.B. see Schwarz, R.B. et al.; Sriiiivasan, S. et al. Desch, P.B. et al. 3: 731 Deschanvres, A. see Rebbah, A. et al. Deschler, M. see Gruter, K. et al. Deshpandey, C.V. and Bunshah, R.F 3: 666 Desjonqueres, M.C. see Papon, A.M. et al. Desoyer, J.C. see Junqua, N. et al. Desport, J.A. see DeVaii, J.N. et al. DesrC, P.J. 3: 701 Desk, P see Molinan, C. et al. DesrC, P.J. see Funnel-Be~li~nt, M.C. et al. Desu, S.B. see Aning, A.O. et al. Deutsch, M. see DiMasi, E. et al. DeVan, J.H. 1: 927, 988, 995; 2: 33, 208; see also Hippsley, C.A. et al., McKainey, C.G. et al., Tortorelli, P.F. and De Van, J.H.; Tortorelli, P.F. et al. DeVan, J.H. et al. 1: 986, 989 Devanathan, R. 1: 792,797,798,817,818, 819; see also Lam, N.Q. et al., Zhu, H. et al. Devanathan, R. et al. 1: 792, 827, 818, 819, 820 Devant, G. see Chavineau, J.P et al. Devaud, G. and Turnbull, D. 3: 683
Devaud-~epski,J. 1: 463 Devaud-Rzepski, J. et al. 1: 537 Deve, H.E. see Cao, H.C. et al. Deve, H.E. et al. 2: 78 Devenish, R.W. see Mullan, C.A. et al. Devenyi, A. see Manaila, R. et al. Devincre, B. sipe Kubin, L.P et al. Devingtal, Yu. V. 3: 823 Devingtal, Yu. V. see Savitskii, E.M. et al. DeVita, J. see Zribi, A. et al. Devore, C.E. 2: 598, 599 de Vos, K.J. see Koch, A de Vnes, J. et al. 1: 743 DeVries, R.C. 3: 271 de Wette, F W. see Alleii, R.E. and dc Wette, F.W. de Wijs, G.A. see Verkerk, P et al.; Xu, R. et al. de Wit, G. 2: 263 de Wit, J.H. 1: 352 DeWit, J.H.W 1: 989, 990; see also Young, E.W.A. et 01, de Wollf, P.M. 1: 453 Dey, A N . 2: 510, 511 Dey, G.K. and Sekhar, J.A. 3: 723, 737 Dhar, S.K. see Nagarajan, R. et al. Dhere, A. see Porter, W.D. et al. Dhere, N.G. 2: 421, 422, 423 Dhesi, S . S . see Durr, H.A. et al. d’Heurle, F.M. 1: 768; 2: 606,608; see also Ames, 1. et a[., Baglin, 3.E.E. et al.; Finstad, T.G. et al., Gas, P. et al.; Thonias, 0. et al. d’Heurle, F.M. et al. 2: 616, 617 d’Heurle, F.M. and Ho, P.S. 3: 666 Di, G.Q. et al. 2 443, 444 Diamand, Y.-S. see Lopatin, S. et al. Dianoux, A.J. see Cristofolini, L. et al.; Zabel, H. et al. Diaz de La Rubia, T. et al. 1: 808 Diaz, S. see Saiichez, J.L. et al. Dibble, D.C. see Mirkarimi, P.B. et al. DiCerbo, R.K. see Westbrook, J.W. et al. DiCioccio, L. see Hewat, E.A. et a2. Dickerson, R.M. see Aikin, B.J.M. et al.; Doychak, J. et al., Locci, I.E. et al. Dickman, 3: 286 Dickson, R.W. 1: 875 Diefenbacher, J. see Ramachandran, G.K. et al. Diehl, J. see Schaefer, H.-E. et al.; Seeger, A. et al. Dienes, G. 1: 775, 778, 785 Dienes, G.J. see Welch, D.O. et al. Dikiy, 1.1. see Zarechnyuk, O.S. et al. Diko, P. see Macko, D. et al. Diller, D. see Parthasarathy, T.A. et al. DiMasi, E. et al. 3: 216 Dimiduk, D. 3: 352; see also Rao, S.1. et al.; Yoo, M.H. et al. Dimiduk, D.M. 1: 419,498,505,539,540, 545, 916, 927, 928; 2: 18, 19, 20, 22, 23, 35, 37, 83, 86, 271, 283, 295; 3: 352,442,477,478,481,503,591,617, 644, 648; see also Fleischer, R.L. et al.; Kim, Y.-W and Dimiduk D.M., Kim, Y.-W. et al.; Lipsitt, H.A. et al., McQuay, P.A. et al.;
Author Index Mazdiyasni, S. et al.; Mendiratta, M.G. and Dimxduk D.M., Mendiratta, M.G. et al.; Mendiratta, M.G. et al.; Nekkanti, R.M. and Dimiduk D.M.; Parthasarathy, T A . et al.; Parthsarthy, T.A. et al.; Rao, S. et al.; Rao, S.I. et al.; Rigney, J.D. et al.; Simmons, J.P et al.; Sriram, S. et al.; Stucke, M A . et al.; Subramanian, P.R. et al.; Yoo, M.H. et al. Dimiduk, D.M. and Vasudevan, V.K. 3: 618 Dimiduk, D.M. et al. 1: 523, 540, 859; 2: 547, 597, 619, 626, 630, 658 Dimmo, G.M. see Bridges, F. et al. Dimitrov, C. see also Sitaud, B. et al. Dimitrov, C. et al. 1: 573, 577, 777, 778, 785, 792, 794, 821 Dimitrov, D.V et al. 3: 691 Dimitrov, 0. 1: 782, 787 see Dimitrov, C. et al.; Njah, N. et al.; Sitaud, B. et al. Dimos, D. et al. 1: 960 Ding, B. see LOLL, T. et al. Ding, B.Z. see Lu, L. et al. Ding, D.H. see Hu, C.Z. et al., Qin, Y.L. et al.; Yang, W.G. et al.; Yao, D.Z. et al. Ding, D.H. et al. 3: 379, 384, 389, 392, 396 Ding, J. see Jeon, H. et at, Ding, J. et at. 2: 428, 430 Ding, Y . et al. 3: 565 Ding, Y .-F. see Yang, Y.-C. et al. Dingle, R. et al. 2: 334 Dinhut, J.F. see Rivikre, J.P et al. Dim, K. 1: 454, 465 Dionne, S. and Lo, J. 3: 572 Diorio, M.S. 2: 348 Dip. A. see Hussien, S.A. et al. DiPasquale, J. see Soboyejo, W.O. et al., Srivatsan, T.S. et al. DiPasquale, J, et al. 3: 342, 343 DiPietro, M.S. see Whittenberger, J.D. et al. DiPietro, M.S. et al. 2: 161, 166, 170; 3: 495 Dippel, Th. see Lechner, R.E. et al. Dirac, P.A.M. 1: 127 Dirkmat, A.J. see Palstra, T.M.M. et al. Dirks, A.G. 1: 733 Dirks, A.G. et al. 2: 622, 627 Dirras, G. 1: 536; see also Beauchamp, P. et al. Dirras, G. et al. 1: 506, 536, 540, 542 Di Russo, E. see Conserva, M. et al. DiSalvo, F.J. see Kwo, J. et al. DiSalvo, F.T. see Bucher, E. et al. Dismukes, J.P. see Moustakas, T.D. et al. Dismukes, J.P and Moustakas, T.D. 3: 786 Dismukes, J.P. et al. 2: 467 Ditchek, B.M. 3: 235 DiVincenzo, D.P. 1: 491; see also Horn, P.M. et al.; Onoda, G.Y. et al. Divinski, S.V. et al. 3: 289, 290 Djuanda, F. see Cheung, C. et al. D.juric, Z. 2: 419 Dlaugatch, L. 1: 7 Do, H. see McGahan, W.A. et al.
Dobbins, T.A. see Luzzi, D.E. et al. Dobbs, J.R. see Darolia, R. et al. Dobbyn, R.C. see Carter, G.C. et al. Dodd, R.A. 1: 565; 2: 63,64,200; see also Bevk, J. et al.; Hocking, L.A. et al. Dodson, B.W 1: 88 Dogan, B. see Wagner, R. et a/. Doggett, A.G. see Jacobs, M.H. et al. Doherty, J.E. et al. 3: 325, 329 Dohler, G.H. 2: 425 Doi, H. see Hashimoto, K. et al.; Kasahara, K. et al. Doi, Y. see Ochiai, S. et al. Doig, P. 1: 836, 840 Doldon, R. 2: 524 Dollar, M, see Nash, P. et al. Dolle, P. see Gauthier, Y et al. Domian, H.A. 1: 762, 765 Domrachev, G.A. et al. 3: 670 Don, J. see Heilmann, P. et al. Donald, I.W. 1: 735 Donaldson, A.T. 1: 565 Donati, C. et al. 3: 694 Doiichev, T. see Norstrom, H. et d. Dong, C. see Launois, P. et al. Dong, C. et al. 1: 467, 475, 477 Dong, J. see Ramachandran, G.K. et al. Dong, S. see Yang, J. et al. Dong, Y .D. see Gregan, G.P.J. et al. Dong, Z. et al. 3: 368 Dong, Z.C. see Huang, D.P. et al. Dong, Z.C. and Corbett, J.D. 3: 119, 120, 121, 123, 254 Doniach, S. 1: 215 Donkersloot, H.C. and Van Vucht, J.H.N. 3: 56 Donlon, W. see Dowling, W. et al. Donlon, W.T. see Dowling, W.E. Jr. et al. Donnay, J.D.H. 1: 167, 171; see also Fischer, W. et al. Donnay, J.D.H. et al. 1: 309, 313, 317 Donnelly, N. et al. 3: 663 Donohue, J. 1: 203 Donovan, P.E. 3: 694, 695 Donovan, P.E. and Stobbs, W.M. 3: 694 Dons, A.L. 2: 175, 176, 186 Doraivelu, S.M. see Prasad, Y.V.R.K. et al. Doremus, R.H. see Klug, F.J. et al. Darn, J.E. see Cheng, C.Y. et al., Mukherjee, A.K. et al.; Vandervoort, R.R. et al.; Webster, G.A. et at. Dorner, B. 1: 152, 153, 157 Dorner, W. et al. 1: 734, 735 Dorofeyuk, A.A. 3: 822 Dorsi, D. see Wernick, J.H. et at. Dosch, H. see Voges, D. et al. Dosch, H. et al. 1: 614 Doty, M. see Fulap, G. et al. Dou, S.X. see Li, H.S. et al. Douglas, A.F. see Armstrong, R.D. et al. Douglas, A.M.B. 2: 177 Douglas, J.F. see Donati, C. et al. Douglass, D.L. 1: 982, 983, 986; see also Kafstad, P. et al. Doum, J. 1: 525, 527, 528, 545; 2: 133, 139; see also Beauchamp, P. et al.; Hug, G. et al.; Veyssidre, P et al. Douin, J. and Veyssidre, P. 3: 443, 461
89 1 Douin, J, et al. 1: 498. 505, 524, 528, 538, 540, 548, 551; 2: 19, 2, 22, 23, 270, 217: 3: 370, 443 Dove, D.B. see Jaswon, M.A. and Dove, D.B. Dove, M.T. see Harm, M.J. et al. Dowa Mining Co. and Tokyo Metro Dowling, W.E. 1: 983, 986, 996; see also Hartfield-Wunsch, S.E. et al. Dowling, W.E. Jr. et al. 2: 83; 3: 442, 601 Downey, J.W. see Nevitt, M.V. et al. Downie, D.B. 1: I17 Downing, J.M. 2: 523 Dowson, A. see Datta, P.K. et al. Dowson, A.L. et al. 3 592 Doyama, M. 1: 693; see also Hashimoto, M. et al.; Kuczynski, G.C. et al.; Wakayama, S. et al., Wang, T.M. et al. Doyama, M. and Yabe, M. 3: 809 Doyama, M. et al. 1: 564, 565, 566, 573, 574, 794 Doychdk, J. 1: 982, 986, 987, 989, 990, et al.; Raj, S.V. et al.; Smialek, J.L. et at,
1003, 1004, 1007, 1008; 2: 57 Doyle, B.L. et al. 2: 621, 626 Doyle, N.J. 1: 566; 2: 54, 55, 58 Drago, V.J. and Kaufman, J.G. 3: 813 Dragsdorf, R.D. 2: 276 Dran, J.C. see Garndo, F. et al. Draper, S.L. see Gaydosh, D.J. et al. Drehman, A.J. 1: 690, 733; see also Poon, S.J. et al. Drehman, A.J. et al. 3: 682, 683, 685 Dreizler, R.M. 1: 129 Dremin, A.N. and Breusov, 0.". 3: 740 Dresner, L. et al. 2: 361 Drews, J. 3: 238 Drews, J. et al. 3: 74, 238 Dreyss, H. see Asta, M.D. et al.; Wolverton, C. et al. DreyssC, H. see Ceder, G. et al.; Wolverton, C. et al. Dreysse, €3. et al. 1: 24, 32 Drigo, A.V. see D'Anna, E. et al. Drobyshevskiy, see Segal, V.M. et al. Droher, J.I. see Hogman, N.J. et al. Drury, M. see Campisi, I.E. et al. Du, H.L, see Datta, P.K. et al.; LJn, C. et al.; Zhang, D. et al. Du, H.L. et al. 3: 575, 576, 716 Du, J. see Yang, C.P. et al. Duan, B. E;. see Pao, Y . H. et al. Dub, A. see Bouchard, M. et al. D~bbeldam,R. see ten Kate, H.H.J. et al. Dubey, K.S. 1: 737 Dubinin, G.N. see Borisov, V.T. et al. Dubols, J.M. 1: 735, 736; see also Belin-Ferre, E. and Dubols J.M.; Cunat, C. et al.; Dong, C. et al.; Launois, P. et al. Dubois, J.M. et al. 1: 486. 741, 742 Dubrovsky, V.A. see Kablov, E.N. et al. Due, N.H. 3: 97
892 Ducastelle, F. 1: 24, 25,26, 28, 34, 35, 36, 38, 43, 277, 284, 295, 305, 849; see also Landesman, J.P. et al.; Ricol~eau,C. et al.; Sanchez, J.M. et al.; Trkglia, 6. et al.; Turchi, P, et al. Duchateau, J.P W.B, see Reader, A.H. et al., van der Kolk, G.J. et al. Duchateau, J.P.W.B. et al. 2: 618, 620, 629 Ducros, B. see Claeyssen, F. et al. Duda, A. see Albin, D.S. et al. Duddek, G. see Somrner, F. et al. Duddles, N.J. et al. 2: 426 Dudley, R.D. and Desai, P.D. 3 37, 38, 44, 808 Dudnik, S.F. see Glushko, P.1. et al. Dudzik, E. see Durr, H.A. et al. Dudzinski, D. see Beddoes, J. et al. Dudzinski, D. ef al. 3: 297, 299, 302 Duetias, A.R. see Summers, L.T. et al. Duesig, T. see Coo, E. et al. Duerig, T.W see Moberly, W.J. et al. Duesbery, M.S. 3: 437, 438, 451, 463; see also Basinsky, Z.S. et al.; Kaxiras, E. and Duesbery, M.S.; Waghmare, U.V Duesbery, M.S. and Vitek, V. 3: 437 Duff, W.R. see Swami, P.R. et al. Dufour, C. see Toulemonde, M. et al. Dufour, C. et al. 1: 821 Dufour, Ch. see Wang, Z.G. et al. Dugdale, J.S. 3: 808; see also Basinski, M.B. et al. Dugdale, R.A. 2: 492,495 Duh, K.-H.G. see Chao, P.C. et al. Duhl, D.M. see Anton, D.L. et al. Duhl, D.N. 2: 25,41,42; 3: 298, 313, 314; see also Anton, D.L. et al.; Cell, M. et al. Duhl, D.N. and Cetel, A.D. 3: 66 Dullenkopf, W. 1: 319; see also Zintl, E, and Du~~enkopf, W. Dumas, J. et al. 3: 806 Dumitrescu, C. see Dahl, W. et al. Duiiand, A. et al. 1: 579 Dunbar, E. see! Thadhani, N.N. et ul. D u ~ b a rE. , et al. 3: 744 Duncan, A.J. see Anderson, I.M. et al.; Viswanathan, S. et al. Duneau, M. 1: 457, 483 Dunlap, R.A. 1: 454, 465; see also Mitra Ghemawat, A, et al.; Stroink, G. et al. Dunlap, R A . et nl. 1: 484 Dunlop, A. 1: 693, 733, 750; see also Audouard, A. et al.; Barbu, A. et al.; Dammak, H. et al., Dimitrov, C. et al., Garrido, F. et al. Dunlop, A. and Lesueur, D. 3: 267 Dunlop, A. et al. 1: 821, 822; 3: 265, 267, 268, 271 Dunlop, G.L. see Liu, P. e f al. Dunlop, J.B. see Li, H.S. et ai.; Margarian, A. et al. Dunmead, S.D. et al. 3: 736 Dunne, D.P 1: 830 Dunning, F. see Jamson, K. et al. Dunweg, B. see Helbing, W. et al. Dunxu, Z. see Dong, Z. et al.
Duo, R. see Fatas, E. er al. Dupasquier, A. 1: 136 DuPoat, J.N. see Banovic, S.W. et al. DuPreez, G. and Wolff, I.M. 3: 62 Dupuy, M. see Hewat, E.A. et al. Duquette, D. see Fish, J. and Duquette, D. Duquette, D.J. 1: 992,993, 1007, 1008;see also Camus, G.M. et al.; Korinko, P.S. et al.; Krishna~oosthy,A. et at., Matuszyk, W. et aL; Scott, J. et al. Duraj, M. et al. 3: 179 Duraj, R. see Duraj, M. et al. aural, J. see Defour, C. et al.; Riviere, J.P. et al. Durand-Charre, M. see Ansara, I. et al. Duret, C. see Lernoine, V et al., Merrel, R. et al. Duret, C. et al. 2 491, 495 Durham, P.J. 1: 136 Durlu, N. 1: 924 Durovic, D. et al. 3: 724 Durr, W.A. et al. 3 663 Durrani, S. and Bull, R.K. 3: 263 Durvall, G. 3 726, 743 Dust, M. et al. 1: 988, 996 Duthie, J.C. 1: 431 Dutta, A. 1: 657; 2: 127; see also Ranganath, S. et al. Dutton, R.E. et al. 3: 447, 648 Dutzi, J. see G e ~ a nA, ~ et , al. Duvall, D.S. 2: 494; see also Sullivan, C.P. et al. Duwez, P.E. 1: 688, 735, 792; see also Bever, M.B. et al.; Chao, C.C. et al.; Johnson, W.L. et al.; ~ l e m ~ nW.J. t, et al. Duwez, P.E. et al. 1: 104, 703 Dwight, A.E. 1: 107, 411; 2: 317 Dworschak, F. see Dimitrov, C. et al. Dwyer, D.J. see Chen, W. et al. Dyberg, J. 2: 403 Dyson, B.F 1: 927 Dyunusov, A.K. see Mi~hai~enko, S.D. et al. Dziuba, C.M. see Chittipeddi, S , et al. Eager, T.W. see FNkiger, R. et al. Eash, J.T. 2: 564; see also Wise, E.M. et al. Eason, P.D. see Jayashankar, J.S. et al. Easterling, K.E. see Artz, E. et al.; Helle, A.S. et al.; Li, W.B. et ul.; Porter, D.A. and Easterling, K.E. Eastman, L.F. 2: 621 Easton, D.S. see Joslin, D.L. et al.; Liu, C.T. et al.; Schneibel, J.H. et al. Ebalard, S. 1: 457, 463, 537; see also Spaepen, F. et al. Ebe, H. see Sh~nohara,K. et al. Ebel, T. see Evers, C.B.H. et al. Ebeling, R. 2: 269 Eberhardt, J.E. et al. 2: 346 Eberhardt, N. see Clemens, H. et al. Clemens, H. et al.; Knippscheer, S . et al. Eberhart, M.E. 1: 604, 605 Eberhart, M.E. et al. 1: 907; 2: 157 Eberspacher, C. see Mitcliell, K.W. et al. Ebert, F, 1: 10 Evert, L.J. 2: 63, 64
Eberz, U. 3: 74, 238; see also Drews, J. et al. Eberz, U. et ul. 3: 239 Eckerlin, P. and Hellwege, M.-H. 3: 797 Eckert, J. see Ledig, L. et al. Eckert, J. et al. 1: 454, 701, 706 Eckert, W see Schaefer, H.E. et al. Eckold, G. see Herget, G. et al. Ecob, R.G. see Ricks, R.A. et al. Ecom, C.B. see Shen, Z.X. et al. Econornou, N.P. see Vojak, B.A. et al. Edagawa, K. et al. 1: 477, 485 Eddingtoi~,A.S. 1: 441 Edelin, G. 1: 571 Edelstein, A.S. and Cammarata, R.C. 3: 749 Eden, G.T. 2: 580 Edington, J.W. 1: 534 Edmonds, A.R. 3: 172 Edwards, A.R. 1: 250, 252 Edwards, D. 3 80 Edwards, D.R. see Jackson, R.O. et al. Edwards, P.A. and Gorbett, J.D. 3 115 Eer, W. see Grossmann, J. et al. Effenberg, G. 2: 717,718, 719,726; 3: 801 Egarni, T. 1: 743,745,820; see also Boyce, J.B. et al.; Koflat, D.D. et al.; Kulp, B.T. et al.; Vitek, V. et al. Egelstaff, P.A. see Enderby, J.E. et ul. Eggeler, G. et al. 2: 519 Eggersmann, M. 3: 287; see also Mehrer, H. et al.; R u m e l , G. et al. Eggersmann, M. and Mehrer, H. 3: 287, 290 Eggert, S . see Appel, F. et al. Eguchi, T. et al. 1: 782 Ehlers, S.K. see Mendiratta, M.G. et al. Ehmann, J. and Fiihnle, M. 3: 439 Ehrenreich, H. see Gelatt, C.D. Js. et al.; Velick9, B. et al. Ehrhart, P. 1: 567, 573, 785 Ehrlich, A.C. see Tntt, T.M. et al. Eibler, R. 1: 137 Eibner, J.E. et al. 1: 574 Eichelber~er,P.M. see Jang, H. et al. Eicher, S, 1: 645 Eichinger, 6. 2: 510 Eisaki, H. see Cava, R.J. et al. Eisen, F.H. 1: 766 Eisenmann, B. 1: 360; see also Schafer, €3. and Eisenmann, B. Eisenniann, B. and Cordier, G. 3: 113, 128 Eisenmann, B. and Klein, J. 3: 128 Eizenberg, M. 2: 618, 619, 621; see also Appelbaum, A. et al.; Lahav, A, et al. Eizenberg, M. et al. 2: 617, 622, 623 Ekin, J.W. 2: 354; see also Gregory, E. et al. Ekman, W. 3 239 Ekonomou, T. see Ryadchenko, V.M. et al. Ekstrom, L. see Dismukes, J.P et al. Elabd, H. 2: 232; see also Kosonocky, W.F. et al. Elagin, D. see Bondarev, B. et al. Elagina, E.I. see Morokhovet§, M.A. et al. Ham, R.C. see Talboom, F.T. et al. EI-Bahraoui, J . see Xu, L. et al. El Boragy, M. et al. 1: 725
893 Elbourne, R.G.P. see Clarke, M. et al. Elcock, E.W 1: 575, 762, 763; 2: 134 Elcock, E.W. and McCoinbie, C.W. 3: 77 1 Elder~Randall,S.P see Brady, M.P, et al. Eldndge, G.W see Hobgood, H. et al. Eldridge, J. 1: 94; see also Noebe, R.D. et al. Eldridge, J.I. see Gaydosh, D.J. et al. Elford, L. et al. 1: 108 Elgueta, J. et al. 1: 777, 778 El-Hanany, U. 1: 684 Elliot, R.J. see Chudley, C,T. and Elliot, R.J. Elliot, R.O. 1: 690, 703, 705 Elliot, R.P. 3: 786 Elliott, C. see Cao, H.C. et al. Elliott, K.E. see Castro, R.G. et al. Elliott, R.J. 1: 1028 Ellis, D.E. et al. 1: 137 Ellis, D.L. 1: 987, 989, 990 Ellis, P see Hill, P.J. et al. Ellner, M. 1: 109, 112, 115; see also Grin, J. et al.; Stadelmaier, H.H. et al. Ellner, M. et al. 1: 112, 114, 115, 116, 117, 122 Ellwanger, R. see Hu, J. et al. El-Masry, N.A. 2: 310, 31 1 El-Masry, N.A. et al. 2: 310, 311, 312 Elmen, G.W. 2: 307 Elser, V. 1: 457, 460, 473, 476, 477, 482, 483; 2: 180 El-Soudani, S. see Semiatin, S.L. et al. Elsukov, E.P. see Ermakov, A.E. et al. Elsukov. E.P. et al. 3: 759 Elswij, H.B. et al. 1: 453 Eltoukhy, A.H. and Greene, J.E. 3: 666 Elvira, R. see Madurga, V. et al. Elwenspoek, M. 1: 191 Embry, P.G. 1: 632 Embury, J.D. 1: 655; see also Cornwell, L.R. et al.; Mukhopadhyay, N.K. et al.; Srinivasan, S.R. et al. Emery, J. 1: 611 Emes-Misenko, E.I. see Zarechnyuk, O.S. et al. Emmenegger, F.P. 2: 332; see also Harbeke, G. et al. Emrick, R.M. 3: 296 Enami, I(.2: 22, 55; see also Nagasawa, A. et al. Enanii, K. et al. 2 54, 55; 3: 56 Ence, E. 2: 75 Enderby, J.E. 1: 669, 670, 671: see also Collings, E.W. et al.; Fortner, J. et aE. Enderby, J.E. et al. 1: 682 Endo, T. 1: 959; see also Kino, T, et al.; Takizawa, H. et al. Endoh, H. see Song, Z.Y et al. Eng, S. see Sanchez, J.M. et al. Eng, T. see Prasad, A. et al. Engbretsen, J,E. see Andresen, A.F et al. Engdahl, G. and Mayergoyz, I.D. 3: 804 Engel, N.N. 1: 116 Engell, H.J. 1: 566, 568; see also Eibner, J.E. et al. Engell, H.J. et al. 1: 911, 913 English, A.T. 1: 509, 510 English, C.A. 1: 807 English, C.M. see Black, T.J. et al. Ennas, G. see Magini, M. et al.
Ennis, P.J. see Appel, F. et al.; Oehring, M. et al., Wagner, R. et al. Ensselin, F. see Hartmann, H. et al. Entwisle, A.R. see Paxton, A.T. and Entwisle, A.R. Eow, K. et al. 3: 58 Epelbom, I. et al. 2: 510, 51 1 Epicier, T. I: 579 Epler, J.E. et al. 2: 424 Epperson, J.E. 1: 574 Epperson, J.E. et al. 1: 574; 2 58 Erb, U. 3: 754; see also Cheung, C. et al. Ercker, L. 1: 4n Ercolessi, F. 1: 82 Ercolessi, F. et al. 1: 82 Erdogan, M.U. see Hussieii, S.A. ei al. Erdogan, T. et al. 2: 431 Eremenko, V.N. et al. 1: 720 Erhardt, H.G. see Kosonocky, W.F. et al. Erickson, G.L. 2: 3; 3: 307; see also Broomfield, R.W et al.; Harris, K. et al. Erickson, J.S. et al. 3: 542 Eriksson, 0. 1: 138; see also Wills, J.M. et al. Ermakov, A.E. see Elsukov, E.P. et al. Ermakov, A.E. et al. 1: 700, 701; 3: 759 Ermakov, A.I. see Vanyukov, A.V. et al. Erman, M. see van Engen, P.G. pf al. Ermer, J.H. see Mitchell, K.W. et al. Ernazarova, L.A. see Mukimov, K.M. et al. Ernst, F. et al. 3: 624 Erskine, J.L. et al. 440, 441 Erwin, S.C. et al. 1: 133 Esaki, L. 2: 424, 425 Esaki, L. and Tsu, R. 3 779 Escaig, B. 1: 547; 2: 25, 150; see also Vanderschaeve, G. and Escaig, B.; Vanderschaeve, G. et al. Escaravage, C. see George, A. et al. Escher, C. et al. 3: 639 Eshelby, J.D. 3: 418 Eshelby, J.D. et al. 3: 392, 393 Espe, W. et al. 2: 646 Espevik, S. 1: 988, 997; 2: 590 Estrade, L. see Ulvensoen, J.H. et al. Estrin, Y. 1: 917 Etienne, J. see G h h a r d , G. et al. Ettenberg, M. et al. 2: 54 Evans, A.G. 1: 925, 927; 2: 217, 290; see also Budiaiisky, B. et al.; Cao, H.C. et al.; Davis, J.B. et al.; Deve, H.E. and Evans, A.G.; Deve, H.E. et al.; He, M.Y et al.; Heredia, F.E. et al.; Rosler, J. et al.; Sigl, L.S. ei al. Evans, B.S. 2: 647 Evans, D.J. et al. 3: 455, 775 Evans, D.S. see Prince, A.A. et al. Evans, E.LI. 1: 115 Evans, R.W. see Davey, S. et al. Evans, R.W. and Wilshire, B. 3 300 Evans, W.J. see Davies. P.W. et al. Everett, R.K. 2 298; see also Skowronek, C.J. et al. Evers, C.B.H. et al. 3: 106 Evetts, J.E. 1: 646, 745; see also Highmore, F.J. et al.; Wangts, C.H. et al. Ewald, P.P. and Hermann, C. 1: 322; 3: 11 Ewing, F.J. see Shoemaker, D.P. et al.
Ewing, W see Kosonocky, W.F. et al. Exner. H.E. and Artz, E. 3: 643 Eylon, D. 2: 87; see ufso Clemens, €3. et al.; Jones, P.E. et al.; Semiatin, S.L. er al. Eylon, D. et al. 3: 650 Eyring, L. see Gschneidiier, K.A. Jr. ~ 7 1 wl. : Ezekwenna, P.C. see Luo, H. et al. Ezz, S.S. 1: 498, 521, 528, 547, 915; 2: 18, 25, 162; see also Pope, D.P. and Ezz, S.S. Ezz, S.S. and Hirsch, P.B. 3: 368 Ezz, S.S. et al. 1: 547; 2: 24, 25; 3: 368 Faber, K.T. 2: 290 Faber, T.E. 1: 663 Fabry, P et al. 1: 95 Facci, P. see Cristofolini, L. et al. Fadeev, A.V. see Klakov, M.P. ef al. Fahlman, A. see Siegbahn, K. et al. Fahnle, M. see Ehmann, J. and Fahnle, M. Fahnle, M. 3 278 Fiihnle, M. et al. 3: 276, 277, 281, 282, 287, 288 Fahrenheit, D.B. 1: 479 Fahy, S. 1: 59 Faigel, G. see Stephens, P.W. et (d. Fairhank, G.B. et al. 3 58 Fairbanks, J.W see Demaray, R.E. et al. Fairhurst, C.W. 2: 576 Fairhurst, G.W. 2: 576 Fairhnrst, W. 1: 534 Faiz, M. see Carnpuzano, J.C. et al. Faktor, M.M. see Hanks, R. and Faktor, M.M. Fal’chenko, V.N. see Larikov, L.N. et al. Fal’chenko, V.M. et al. 1: 758 Falicov, L.M. 1: 24, 32, 610, 613, 614 Fallavier, M. see Baudin, K. et al. Fallot, M. 2: 307 Fan, G. see Lou, T. et al. Fan, J.W see Collins, G.S. et al. Fan, Y.D. see Huang, L.J. et al. Fan, Z. see Park, D.G. et al. Fang, J. 1: 925; 2 137, 143 Fang, J. et al. 3: 443 Fang, L.H. see Chen, X.F. et al. Fang, R.-Y. et al. 2: 444 Fantozzi, G. see Liang. K.N. et al. Fanzen, H.F. see Yi, S.S. et al. Faraday, N. 1: 5; 3: 250 Fardig, D.A. see Sh~~arrnan, W.N. et al. Farenc, S. see Caillard, D. et al.; Couret, A. et al. Farenc, S. et al. 3 422 Farges, J.P see Martin, G.M. ei al. Farkas, D. 1: 88, 598, 600, 614, 615, 616, 617; 2: 27, 637, 642, 643, 644; 3: 220, 223, 772, 773, 774; see also Caillard, D. et al.; Jones, C. and Farkas, D,; Mishin, Y, and Farkas, D.; Mutasa, B. and Farkas, D.; Pasianot, R. et al., Pedraza, D.F. et al.; Ternes, J.K. et al.; Vailhe, C. and Farkas, R.; Valh, C. and Farkas, D.; Ye, F. et al. Farkas, D. and Cardozo, F.A. 3: 771 Farkas, D. and Jones, C . 3: 774 Farkas, D. and Ternes, K. 3: 771 Farkas, D. and Vailhe, C. 3: 452
894 Farkas, D. et al. 1: 536, 611, 616, 617, 618; 3: 439, 452, 768, 769, 770 Farkas, Z.D. 2: 377 Farmer, S. see Aikin, B.J.M. et al. Farmer, S.C. see Whittenberger, J.D. e f al. Farne, G, see Matteazzi, P. and Farne, G. Farnsworth, H.E. 1: 183 Farques, D. et al. 3 142 Farrell, B. see Sundaram, V.S. et al. Farrow, L.H. see Inoue, H.R.P. et al. Fasel, R. see Aebi, P. et al. Fasman, A.B. see Mikhailenko, S.D, et al. Fasman, V. see Ivanov, E. et al. Fatas, E. et al. 3: 670, 671 Fat-Halla, N.K. see Takasugi, T. et al. Fattah, A. see Sajovec, F et al. Faulkner, J.S. 1: 25 Faust, J.W. Jr. see Marshall, R.C. et al., Suri, S.K. ea al. Fauth, M.I. 2: 647 Fawcett, E. see Collocott, S.J. et al.; Joss, W et al. Fawcett, W.R. 2: 510, 511 Fayard, M. 1: 498, 574; see also Bresson, L. et al. Fayyad, U.M. et al. 3: 864 Featherstone, F.J. 1: 204, 206 Fecht, H.J. see Hellstern, E. et al., Lee, M.C. et al. Fecht, H.J. et al. 1: 701 Fecke, T. and Davidson, D.E. 3 483 Fedders, P.A. see Carlsson, A.E. et al.; Jeong, E.K. et al. Feder, R. et al. 2: 18; 3: 222 Fehling, D.T. see Dresner, L. et al. Feijen, F.W. see Buschow, K.H.J. et al. Feil, H. see Otto, M.S. et al. Felder, R.J. see Cava, R.J. et al. F e l d e ~ ~ a n H. n , see Ronning, C. et al. Feldman, J.L. see Boyer, L.L. et al. Feldinan, J.L. et ul. 1: 195 Feldman, L.C. see Hong, M. et al. Feldtkeller, E. 2: 444 Fella, R. 2: 410 Feller-Kni~p~eier, M. 1: 528 Felten, E.J. 1: 988, 996; see also Berkowitz-Mattuck, J. et al. Felter, T. et al. 1: 616; 3: 225 Feng, A. and Munir, Z.A. 3: 734, 735 Feng, C.R. see Sadananda, K. and Fcng C.R., Sadananda, K. et al. Feng, C.R. and Sadananda. K. 3 485, 486 Feng, C.R. et al. 2: 75; 3: 407 Feng, H.J. see Moore, J.J. and Feng H.J. Feng, J. see Sadananda, K. et al. Feng, J.L. see Wang, R.H. et al. Feng, J.L. and Wang, R.H. 3: 382, 383, 384 Feng, J.L. et al. 3: 383, 398 Feng, Q. see Whang, S.H. et al. Feng, W F . see ~ a n gZ.G. , et al. Fenga, A. and Munir, Z.A. Ferekides, C. see Chu, T.L. et al. Ferguson, P.E. 2: 446 Fernandes, M. see Xia, W. et al. Fernandez, E M . see Gallego, L.J. et al. Fernando, G.W see Lui, S.-C. et al. Fernando, G.W. et al. 1: 64, 68, 133 Ferracane, J.L. see Tsutsumi, S. et al.
Author Iridex Ferrante, J. 1: 926; see also Rose, J.H. et al. Ferrari, E. see Baricco, L.H. et al. Ferreira, A. et al. 1: 649 Ferreira, I.C. see Ayres de Campos, J. et al. Ferreira, L.G. see Lu, Z.W. et al.; Wei, S.-H. et al.; Zunger, A. et al. Ferreira, L.G. et al. 1: 31 Ferreira, L.P. see Ayres de Campos, J. et al.; Baudry, A. et al. Fern, M. see Thome, R.J. et al. Ferrier, A. 1: 114, 118 Ferriss, D.P. 2: 597 Ferro, R. 1: 64 Ferro, R. et al. 1: 103 Ferry, D.K. 2: 417, 432 Fert, A. see Campbell, I.A. et al. Fertig, W.A. see Moodenbaugh, A.R. et al. Fetcenko, M.A. et al. 2: 486 Feucht, K. 2 485 Feuerbacher, M. see Rosenfeld, R. et al., Urban, K. et al.; Wang, R,H. et al.; Yang, W.G. et al. Feuerbacher, M. et al. 3: 380, 384, 385, 397, 399. 400,401, 693 Feygenson, A. and Zemel, J.N. 3: 673 Fiala, J. e f al. 1: 918 Fialkov, Yu.Ya see Anosov, V Ya. et al. Fichet, R. see Rivet, J. et al. Fick, Z. see Movshovich, R. et al. Fidler, J. 2: 313; see also Urban, K. et al. Fidler, J. et al. 2: 31 1 Fidleris, V see Causey, A.R. et al. Field, R.D. 1: 453, 883; 2: 63, 64, 241, 242, 248; see also Darolia, R. et al., Fleischer, R.L. et al., Krueger, D.D. et al., Lahman, D.F. et al. Field, R.D. et al. 1: 536, 860, 913, 921; 2: 58, 59, 60, 61; 3: 356, 371, 772 Field, R.J. and Ghandi, S.K. 3: 670 Field, T.T. see Sullivan, C.P. et al. Figueiredo, M.O. 1: 282; 3: 3, 5, 10, 12 Fihey, J.L. 2: 360 Filipek, R. see Danielewski, M. and Filipek, R.; Danielewski, M. et al., Datta, P.K. et al. Filippi, A. 1: 994 Fillion, G. see Yavari, A. et al. Filoti, G. SCP Kuncser, V et al. Filyand, M.A. 2: 518, 523, 524 Fimland, R.-0. see Palmstrom, C.J. et al. Fine, M.E. 2: 158, 178; see also Angers, L. et al.; Calderon, H.A. et al. Fine, M.E. et al. 1: 195, 196, 206, 207; 2 24 1 Finel, A. 1: 35, 39, 43 Finel, A. et al. 1: 35 Finetti, M. see Maenpaa, M. et al. Finetti, M. et al. 2: 66, 624, 626 Fink, J. see vom Felde, A. et al. Fink, J.L. see Bertocci. U. et al.; Ricker, R.E. et al. Finley, D.W see Ray, U. et al. Finnegan, J.J. 2: 623 Finney, J.J. see Desborough, G.A. ef al, Finney, J.L. 1: 479 Finnis, M. 1: 611
Finnis, M.W. 1: 80, 523; 2: 153; see also Ernst, F et al., Hagen, M. and Finnis, M. W.; Scliroll, R. et al. Finnis, M.W. and H e m , V. 3 220 Finnis, M.W. and Sinclair, J.E. 3: 215 , 447,450, 766 Fiiistzd, T.G. 2: 615, 616; .see ulso d’Heurle. F.M. et al. Finstad, T.G. et al. 2: 616 Fiolhais, C. see Perdew, J.P. et al. Fires, S. 3: 800 Fischer, J.E. see Neumaiiii, D.A. et al.; Setton, M. et al., Zhu, Q. et al. Fischer, J.F. 1: 912 Fischer, M. see D a m , S. et al. Fischer, M. et al. 1: 735 Fischer, P 1: 393, 394; 2: 478; see also ~ o n h o ~F~ eef al. , Fischer, R. see Knecht, J. et al.; Schmutzler, R.W. et al. Fischer, R.J. 2: 431 Fischer, W. 1: 317, 327; see also Hellner, E. et al. Fischer, W. et al. 1: 309, 315 Fish, J.S. 1: 992, 993, 1007, 1008 Fish, J. and Duquette, D. 3: 575 Fisher, F.D. see Markete, W.T. et al.; Schogl, S.M. and Fisher, F.D. Fisher, G. see Chan, W.Y. et al.; Datta, P.K. et al. Fisher, 6. et al. 3: 566 Fisher, 1.R. see Shen, Z. et al. Fisher, M. see Jacob, 1. et al. Fisher, R.M. 1: 498, 509; see also Mar~inkowski,M.J. and Fisher, R.M; Marcinkowski, M.J. et al.; Swann, P.R. et al. Fishman, S.G. et al. 1: 575 Fisk, Z. 1: 211,215,216; see also Canfield, P.C. et al.; Guntherodt, C. et al.; Hundley, M.F et al., Lacerda, A. et al., List, R.S. er al., Migliori, A. et al., Ott, H.R. et al.; Pintschovius, L. et al.; Severing, A. et al.; Stewart, G.R. et al.; Watako, Y. et al. Fisk, Z. et al. 3 46 Fitzer, E. 1: 16, 1003; 2: 295, 490; 3: 574 Fitzer, E. et al. 2: 228; 3 574 FitzGerald, J.D. see Choy, T.C. et al. FitzGerald, J.D. et al. 1: 466 Fitzgerald, T.J. and Singer, R.F 3: 543 Fjellvag, H. see trauback, B. C. ef al. Flack, H.D. see Dunand, A. et al. Flahaut, J, see Ghkinard, G, et al.; 011itrault-Fichet, R.et al.; Rivet, J. et al. Flanagan, T.B. et al. 1: 306 Flanders, D.C. see Alley, G.D. et al.; Vojak, B.A. et al. Flank, A.M. see Sadoc, A. et al. Flannery, B.P. see Press, W.H. et al. Fleet, M.E. 1: 352, 628 Fleischauer, GS. see Deevi, S.C. et al. Fleischer, M. 1: 626, 627, 628, 629, 630, 631 Fleischer, R.L. 1: 116, 138, 144, 195, 196, 204, 205,248,266,267,391, 539,552, 560, 561, 567, 580,883,885, 886,887, 891,920,921,922,923,928,988,996, 999, 1001; 2: 237,238, 239, 240, 241, 242,243,244,245,246,247,248,249, 251, 252, 253, 254, 259; 3: 42, 55, 56,
Author Index 59, 60, 78, 263, 270, 288, 351, 352, 353,354, 355, 356,357, 358,485,486: see Bein, C.P et al.; Westhrook, J.H. and Fleischer, R.L Fleischer, R.L. and Hibbard, W.R. 3: 353 Fleischer, R.L. and McKee, D.W. 3: 55 Fleischer, R.L. and Taub, A.I. 3: 501 Flerscher, R.L. and Zabala, R.J. 3: 43 Fleischer, R.L. et al. 1: 137, 206, 595, 849, 850, 883, 885, 906,913, 928 2: 242, 247, 248,250, 251; 3: 42, 56, 78, 263, 264,265,268,269,270,271,272,501 Fleischer, T. see Clemens, H. et al. Fleitman, A.H. and Weeks, J.R. 3: 31 Fleming, R.M. see Cava, R.J. et aE., Majkrzak, C.F. ei' al. Fleinings, M.C. see Chandley, G.D. and Flemings, M.C. Fleurial, J.P. see Caillat, T. et al.; Chen, B. et al. Flewitt, P.E.J. 1: 836, 838, 840, 841 Flick, W. see Kolawa, E. et al. Flinn, P.A. 1: 496,497,498, 500, 501, 546, : 19, 22; 3: 311, 364, 365, 372; see also Shirane, G. et al. Flokstra, J. see de Reus, R. et aE. Florence, D. see Rossat-Mignod, J. et al. Flores, K.M. and Dauskardt, R.H. 3: 341 Florez, L.T. see Harbison, J.P et al., Sands, T. et al.; Tabatabaie, N. et al. Florio, J.V. et ail. 1: 396; 2: 31 1 Flower, H.M. 2: 188, 190; see also Cheng, G.H. et al.; Khatee, A. et al. Flower, H.N. et al. 2 282 Fliikiger, R, 2: 353 Fliikiger, R. et at. 2 360 Flukiger, R, and Klose, W. 3: 806 Flynn, C.P. 1: 682 Fogg, C.T. and Cornellisson, J.L. 3: 54 Foiles, C.L. see Blatt, F.J. et al. Foiles, S.M. 1: 82, 83, 88, 155, 505, 507, 564,570,575,598,600,797,798, 874, 875; 3: 225; see also AlthoiT, J.D. et al.; Asta, M.D. and Foiles, S.M.; Baskes, M. et al.; Daw, M.S. et al.; Mills, M.J. et al. Foiles, S.M. and Daw, M.S. 3: 769, 770 Foiles, S.M. et al. 1: 88, 569, 611 Foldeaki, M. see Giguere, A. et al.; Mitra ~ h e m a w a tA. , et al. Foldeaki, M. et al. 3: 527 Foley, J.C. see Men, W.P. et al. Foley, R. see Nesbit, S. et al. Foley, W.T. 1: 669 Foley, W.T. and Reid, L.E. 3: 26 Follstaedt, D.M. 1: 454; 3 681 Folzer, A. see Ziittel, A, et al. Fomenko, V.S. 3: 806 Fomine, I.A. see Thomas, F et al. Fomitcheva, L.N. 1: 707 Fonda, R.W et nl. 3: 56 Foner, S. see Fliikiger, R. et al. Fonstad, C.G. 2: 328 Fontana, M.P. see Cristofolini, L, et al. Foo, C. see Yoo, M. and Foo, C. Forbes, K.R. see Hemker, K.J. et al. Ford, D.A. see Broomfield, R.W et al.; Hill, A.D. et al. Ford, W.E. 1: 178n Foreman, A.J.E. 2: 259, 262, 281 Foreman, A.J.E. et al. 2 19
Foreman, S. see Fliikiger, R. et al. Fornasini, M.L. see Bruzzone, G. et al.; Manfrinetti, P. et al.; Merlo, F. and Fornasini, M.L.; Merlo, F. et al. Fornasmi, M.L. et al. 1: 325, 417 Fornwalt, D.E. see Kear, B.H. et al. Forrat, F see Bertaut, E.F. et al. Forrest, S.R. et al. 2 339 Forro, L. see Stephens, P W. et al. Forster, J. et at. 2: 492 Forster, J.A. see Jha, S.C. et al. Forsythe, E.B. 2: 380 Fort, D. see Zhu, W. et al. Fortner, J. see Saboungr, M.L. et al. Fortner, J. et al. 3: 254, 256, 257 Forto, V.E. see Altshuler, L.B. et al.; Kanela, C.L. et al. Forwood, C.T. see Head, A.K. et al. Foster, K. et at. 3: 253 Foster, M.S. see Carnpuzano, J.C. et al. Foster, N.F. 1: 17811 Fougere, G.E. see Siegel, R.W and Fougere, G.E. Fourdeux, A. see Weber, D. et al. Fournee, V. see Shen, Z. et al. Fournier, J.M. and Gratz, E. 3: 806 Fournier, P.R. 2: 624; see also Garceau, W.J. et al. Fowler, C.M. et al. 1: 136 Fowler, H.A. et al. 1: 483 Fowler, P.H. et al. 3: 263 Fox, A.G. 1: 541; 2: 54, 58 Fox, A.J. see Radmilovic, V. et al. Fox, D.S. see Hesus, M.G. et al. Fox, T. et al. 1: 973 Foxall, R.A. see Mitchell, T.E. et al. Foxon, C.T. and Harm, J.J. 3: 779, 782 Foxon, C.T. and Joyce, B.A. 3: 779 Foy, P W see Hayashi, I. et al. Fraas, K. et al. 1: 213 Fraczkiewicz, A. et al. 3: 614 Fraga, G.L.F see Besnus, M.J. et al. Franchet, J.M. see Gogia, A.K. et al. Francliuk, V.J. see Larikov, L.N. et al. Franchy, R. see Wuttig, M. et al. Francis, C.B. 2: 519, 523 Francis, R.W. see Rao, B.M.L. et al. Frangois, A. 1: 530, 536, 540, 543 Frangois, A. et ul. 1: 531, 536; 3: 446, 447 Francois, M. see Venturini, G. et al. Francois, M. et al. 3: 91 Frank, F.C. 1: 167,409,453,473,474,479 Frank, F.C. and Stroh, A.N. 3: 408,418 Frank, G.R. Jr. see Hollingsworth, E.H. et al. Frank, St. see Divinski, S.V. et al. Franke, R. see Clemens, H. et al. Franklin, R.J. see Libsch, J.F. et al. Frankowski, I. 1: 217 Frankwicz, P.S. 2 219, 221; see also Boettmger, W.J. et al. Franse, J.J.M. 1: 441; 3: 804; see also BUIS,N. et al., Sinnema, S. et al. Franse, J.J.M. and Radwansh, R.J. 3: 171 Franz, M. see WOE, J. et al. Franz, M. et al. 3: 284, 286
895 Franz, W. see Steglich, F et al. Franzen, H. 1: 109 Fraser, H. see Shyue, J. et al. Fraser, H.L. 1: 453,990; also Court, S.A. et al., Evans, D.J. et al.; Hou, D.H. e~ al.; Kaufman, M.J. et nl.; Konitzer, D.G. et al.; Lofvander, J.P.A. et al.; Marquardt, B.J. et al.; Vasudevan, V.K. et al.; Wiezorek, J.M.K. et al.; Wheeler, R. et al.; Youngquist, S.E. et al. Fraser, H.L. et al. 2: 59 Frasier, D.J. see ~ r o o ~ ~ eR.W l d , et al. Frasier, F.R. et al. 2 157 Fratilom, D. see Manaila, R. et al. Frawley, J.J. see WolE, G.A. et al. Frear, D.R. 2: 654 Frear, D.R. et al. 2: 523 Freei~an,A.J. 1: 65, 66, 69, 70, 71, 72, 137, 507, 508; 2: 58, 156, 157, 278; 3: 841; see also Asta, M. et al.; Field, R.D. et al., Gonis, A. et al., Guo, X.-Q. et al.: Hong, T. er. nE.; Jansen, H.J.F. and Freeman, A.J.; Krakauer, €3. et al.; Lee, J. et al.; Lin, W. et al.; Lin, X. et al.; Min, B.I. et al.; Mryasov, O N . and Freeman, A.J.; Mryasov, O.N. er al.; Sluiter, M. et al.; Sluiter, M.H.F et al.; Wirnmer, E. et al.; Xu, J.-H. et al. Freeman, A J . et al. 1: 65, 66, 69, 70, 72, 137, 506,924; 2: 156; 3: 841 Freeman, B.L. see Fowler, C.M. et al. Freeman, C.F 2: 417, 418 Frefer, A., see Suryanarayana, C. et al. Freiburg, Ch. see Grushko, B. et al. Freidinan, S. see Welsch, G. et al. Freise, E.J. 2: 216 Freitag, K. see Rumniel, G. et al. French, J. D. et al. 3: 486 Frenkel, J. 1: 760, 763 Frekel, J. and Kontorova, T. 3: 408 Frenner, K. see Schaefer, H.-E. et al. Frennet, A. see Crucq, A. et al. Fretague, W.J. et al. 1: 714 Fretwell, A. see Green, A.J. et al. Frey, J. 2: 335 Frey, N. see Semiatin, S.L. et al. Freyhardt, H.C. see Schultz, L. et al. Freyhart, W.C. see Brown, B.S. et al. Freyland, W. 3: 806 Friauf, J.B. 1: 409 Fridberg, J. et al. 1: 921 Fridman, Z.G. see Alisova, S.P. et at.; Matveeva, N.M. et nl. Friedel, G. 3: 403 Friedel, J. 1: 33, 547, 552,578,794,795; 2: 19, 259; see also Weitzer, F. et al. Friedman, I.L. see Kuhn, W.E. er al. Fnednian, S.L. see Welsch, G. et al. Friedrich, W see Laue, M. et al. Fneman, S.W. 2 584 Frier, P.S. ef al. 3: 264 Friesel, M. see Klippers, M. et al. Frijlink, G.P.A. see Sinke, W. et al. Fripan, M. see Sommer, F. et al. Fritsche, L. 1: 569 Fritscher, K. 2: 499 Fritzemeier, L.G. et al. 1: 974 Froes, F.A. see Imayev, R.M. et al.
896 Froes, F.H. 1: 861, 866, 867; see also Frefer, A. et al.; Khina, B.B. et al.; Li, W. et al.; Liu, C.T. et al.; MacKdy, R.A. et al.; Mukhopadhyay, D.K. et al.; Salishchev, G.S. et al.; Suryanarayana, C. et ul.; Ward, C.H. et al. Froes, F.H. and Suryanarayana, C, 3: 503 Froes, F.H. et al. 3: 645 Frohberg, G. 1: 757, 761, 763, 765; see also Bose, A. et al.; Hahn, H. et al.; Hunecke, J. et al.; Wever, H, et al. Froi~berg,G. see Hahn, E. et al. Froment, N. see Epelboin, I. et al. Froinmeyer, G. see Blum, M. et al., Knippscheer, S. et al.; Liminez, J.A. et al. Frommeyer, G.R. see Grolich, M. et al.; Wunderlich, W. et at. Frommeyer, G.R. et al. 2: 224, 240, 241 Frondel, C. 1: 343; see also Palache, C. et al. Frost, H.J. 1: 598,863,903,908,916,917, 918, 919; 2: 241; see also Scliulson, E.M. et al. Frota-Pessoa, S. see Lu, Z.W. et al. Froyen, S. see Yeh, C.-Y et al. Fruchart, D. 3: 803; see also Ayres de Campos, J. et al.; Fruchart, R. et al.; h a r d , 0. et al.; Obbade, S . et al.; Osterwalder, 3. et al.; Vert, R. et al.; Wolfers, P. et al. ~ o ~ s t r o rS.E. n , et al. Fu, C. see Lee, J. et al. Fu, C.L. 1: 65, 69, 70, 71, 72, 137, 144, 496,498,499,501,505,506,507,508, 527,540, 541,874, 875,876,882,884, 915, 920, 922; 2 58, 148, 149, 156, 167, 202, 204; see also George, E.P. et al.; Liu, C.T. et al.; Y 00, M.H. and Fu, C.L., Yoo, M.H. et al. Fu, C.L. and Painter, G.S. 3: 276, 280, 28 1 Fu, C.L. and Yoo, M.H. 3: 407,411, 422 Fu, M. and Shekhar, J.A 3: 737 Fu, Z. see Fecht, H.J. et al.; Hellstern, E. et al. Fuchino, S. et al. 2: 383 Fuchs, G. see Audouard, A. et al. Fuchs, G.E. 3: 650; see also Stoloff, N.S. Pt al. Fuchs, G.E. and Stoloff, N.S. 3: 325, 331, 332, 333 Fuchs, K.A. see Hoen~g,H.E. et al. Fueki, S. see Yamada, T. et al. Fuerst, C.D. see Herbst, J.F et al. Fuerst, C.D. et al. 3: 94 Fuggle, J.C. see Hoekstra, H.J.W.M. et al.; Sarma, D.D. et aE.; Speier. W et al.
Fujii9H. see Iwata, M. et al.; Koyarna, K. et al.; Kyogaku, M. et al., Saburi, T. et al.; Takabntake, T. et al. Fujii, T. see Mimura, T. et al. Fujii, Y et al. 1: 702
Author Index Fujimori, A. see Ogawa, S et al., Son, J.-Y. et al. Fujimori, H. et al. 3: 691 Fujimoto, H. see Takahashi, T. et al. Fujimura, S. see Sagawa, M. et al. Fujinaga, Y. et al. 3: 161 Fujita, F.E. see Muto, S. et al. Fujita, H, 1: 816; see also Luzzi, D.E. et al.; Mori, H. et al. Fujita, H. and Rawasaki, Y. 3: 411 Fujita, H. and Mori, T. 3: 422 Fujita, M. see Mori, H. et al. Fujita, S. 3: 41 1; see also Ikebuchi, M. et al.; Kawakami, Y. et al., Wu, Y.-H. et al. Fujita, S. et al. 2: 326 Fujita, T. see Takabatake, T. et al. Fujitani, S. see Nasako, K. et al. Fu.jitsuna, N. see Morita, A. et al. Fujiwara, E. see Kurosawa, K. et al. Fujiwara, F. see Kishida, T. et al. Fujiwara, K. see Iwasaki, H, et al. Fujiwara, T. 1: 71, 485; see also Hashimoto, M. et al., Nishitani, S.R. et ala;Yasuda, K. et al. Fujiwara, T. et al. 3: 621 Fujiwara, Y. see Kishida, T. et al. Fujiyarna, H. see Udoh, K.4. et al. Fujiyasu, H. see Shinohara, El. et al. Fukai, K. see Hishinuma, A. et al. Fukai, Y. 2 476 F u k a ~ i T. , see Tokizane, M. et al. Fukamichi, K. see Aoki, K. et al. Fukuda, H. see Koyama, K. et al. Fukuda, J. see Fuchino, S . et al. Fukunaga, 0. see Mishima, 0. et al. Fukunaga, T. see Mizu~ani,U. et al. Fukunda, T. 2: 523 Fukuota, K. see Yamasaki, T. et al. Fulap, G. et al. 2 51 1 Fulcher, G.S. 1: 690 Fulcher, M.R. see Allen, C,W. er al. Fulde, P. 1: 159 Fulton, J.M. see Patten, E.A. et al. Fultz, B. 1: 774, 776, 786, 854, 856; see also Anlage, S.M. et al.; Anthony, L. et al.; Kikuchi, R. et al., Ma, E. et al. Fumi, F.G. 1: 764 Funahashi, S. see Fujii, H. et al. Funakubo, H. 2: 558 Fundarnenskii, V.S. see Marusin, E.P. et al. Fung, K.K. see Zou, X.D. et al. Fung, K.K. et al. 1: 467; 2 176 Fu~ne~-Be~lisent, M.C. et al. 1: 667 Funtikov, A.I. see Altshuler, L.B. et al. Furdyna, J. see Ding, J. et al.; Jeon, H. et al. Furey, S.C. 3: 493 Furrer, D.U. see Eylon, D. et al. Furthniller, J. see Stadler, R. et al. Furubayashi, T. see Ishimoto, H. et al. Furuda, H. see Ohara, T. et al. Furukawa, H. 3: 581 Furushiro, N. 2 187 Futjisuna, N. et al. 631 Fuzhong, 2. see Adachi, G.Y et al. Gaal, I. et al. 1: 956 Gabdullin, N.K. see Salishchev, G.S. et al. Gaberson, P.C. see Singh, S.K. et nl.
Gachon, J.C. see Jorda, J.L. et al. Gachon, J.C. et al. 1: 64 Gaczi, P.J. 3: 669 Gaffet, E. 1: 701,792; see also Abdellaoui, M. et al.; Charlot, F. et al.; Gauthier, V. et al.; Gras, C. et al. Gafner, G. 2: 559, 569 Gage, P.R. 1: 998, 999, 1003; see also Bartlett, R.W. et al. GagnC, M. 2 133, 139 Gahn, U. 1: 783 Gahutu, D. see DiPasquale, J. et al. Gaillard, J.P. see Hewat, E.A. et al. Galachov, V.R. see Anisimov, V.I. et al. Galakov, V.R. see Elsukov, E.P. e t al. Galakov, V.R. et al. 3: 144, 145 Galasso, F.S. 2: 410, 411, 418 Galerie, A. see Kabbaj, M. et al. Galinski, G. see Maurer, R. et ul. Gall, K. et al. 3: 774 Gallego, L.J. et al. 1: 809 Galler, R. see Mehrer, H. et al. Gallerneau, F. 3: 298 Galli, G, 1: 195, 208 Galloway, N. see Schulze, M.P. et al, Galloway, N. et al. 2 400, 401 Galus, z. 3: 22 Galvm, G.J. see Palmstrorn, C.J. et al. Gambino, R.J. see Chaudhari, P. et al.; Holtzberg, F. et al. Gamble, R.P 2: 12 Gambone, M.L. 2: 292; see also Bain, K.R. et al., Larsen, J.M. et al. Gamo, T. see Mori~aki,Y. et al. Gandhi, C. 1: 918; see also Ashby, M.F et al. Gandy, T.H. see Cale, T.S. et al. Gangopad~yay,A.K. see Cornelius, A.L. et al. Gangopadhyay, A.K. and Schilling, J.S. 3: 103 Ganin, G. see Amsimov, V.I. et al. Gao, C. et al. 2 310 Gao, M. et al. 1: 993 Gao, W. see Lm, Z. et al. Gao, Y.Q. see Whang, S.H. et al. Gao, Y.Q. and Whang, S.H. 3: 413,419 Garber, J.A. 1: 874 Garbulsky, G.D. see Tepesch, P.D. el al. Garceau, W.J. 2: 624 Garceau, W.J. et al. 2: 624 Garcia, A. see Barbee, T.W. et al. Garcia, E. sec Corbett, J.D. et al. Garcia, J. see Ibarra, M.R. et al. Garcia, S. see Sanchez, J.L. et al. Garcia-Escorial, A. 1: 747, 748 Garcia-Landa, B. see Morellon, L. et al. Gardes, D. see Baudin, K. et al. Gardiniers, J.G.E. et al. 1: 183 Gardner, J.A. 1: 682 Garg, A. see Hebsur, M.G. et al., Kitabjiaii, P.H. et al.; Srinivasan, R. et al. Garg, A. et al. 3: 305, 494 Garg, S.P. see Raghavan, 'v et al. Garg, S.P et al. 3: 802 Garibay-Febles, V. see Garrido, J. and Orland, J. 3: 4 Garin, J. 1: 354
Authm Index Garin, J.L. 2: 18 Garland, C. see Kolawa, E. et al. Garlick, R.G. see Natbal, M.V. et al. Garmong, G. 2: 651 Garner, W.E. 1: 6 Garofalo, M. 1: 82 Garreau, M. see Epelboin, 1. et al. Garrido, F. et al. 1: 821 Garrison, K.C. see Palmstrom, C.J. et al. Garton, H.C. see Wieber, R.H. et al. Garwood, R.G. I: 835, 836 Gas, P. 2 606; see also Finstad, T.G. Gascom, F. and Sevov, S.C. 3: 129 Gasior, W. see Moser, Z. et al. Gaskell, P.H. 1: 742,743,750; 3: 687,690; see also Dubois, J.M. et al. Gasparrini, G. see Benci, S. et al. Gatos, H.C. 1: 181; see also Warekois, E.P. et al. Gaulin, B.D see Svensson, E.C. et a2. Gauthier, V. et al. 3: 645, 744 Gauthier, Y. 1: 611, 615, 616, 617; 3: 223 Gauthier, Y. et al. 3: 223 Gautier, F. 1: 24, 28, 29, 34, 36, 849, 850; see also Trbglia, G. et al. Gautier-Picard, P. see Sun, Y.Y. et al. C~V~ICS, 3 . 3: 864 Gavazza, S.D. see Barnett, D.M. et al. Gavigan, J.P. see Cadogan, J.M. et a(; Nu,B.P. et al. Gavriliuk, A.G. et al. 3 162 GavriIov, V.D. see Ryadchenko, V.M. et al. Gay. A.S. see Fraczkiewicz, A. et al. Gay, R.R. see Mitchell, R.W. et al. Gayanov, R.M. see Valiev, R.Z. et aE, Gaydosh, D.J. 1: 655, 864; 2: 202, 203, 205, 206; see also Crimp, M.A. et al.; Jha, S.C. et al.; Smialek, J.L. et al.; W~ittenb~rger, J.D. et al. Gay~osh,D.J. and Crimp, M.A. 3: 639, Gayle, F.W. 1: 455, 457; 2: 188, 190, 191; see also Goldman, A.T. et al.; Guryan, C.A. et al.; Heiney, P.A. et al. Gayler, M.L.V. 2 576, 577, 579, 580 Gayton, W.R. 1: 875 Ge, S.-L. see Yang, Y.-C. et al. Geballe, T.H. see Chin, G.Y. et al.; Mathas, B.T. et al., Menth, A. et al., Shen, Z.X. et al. Gebel, B. see Rubis, M. et al. Gebhardt, E. 2: 577 Gedevanishvili,S. and Munir, Z.A. 3: 735 Gedwill, M.A. see Brindley, W.J. et al.; S ~ ~ ~ lJ.L. e ket, al. Geertsma, W. see Saboungi, M.L. et al.; van der Marel, 6 . et al. Gegel, H.L. see Prasad, Y V.R.K. et al. Geibei, 6. see Thomas, F. eb al. Geibel, Ch. see Compf, F. et al. Geibel, G. see Berner, D. et al. Geichenko, V.V. see Larikov, L.N. et al. Geisler, A.H. 2: 308, 524 Geissberger, A.E. et al. 2: 621
Gelato, L. see Parthl, E. et al.; Zhao, J.-T. et al. Gelatt, C.D. Jr. see Williams, A.R. et al. Gelatt, C.D. Jr. et al. 1: 62, 63, 64, 604; 2: 479 Gel’d, P V. see Ayushina, G.D. et a/=; Balina, Ye.A. et ul.; Petrushevskii, M.S. et al. Gell, M. et al. 2: 5 Gellert, C.E. 1: 4 Genna, F see Bao, G. et al. Genta, V. see Kub~s~I~ewski, 0.et al. Gente, T. see Schneider, D. et al. George, E.P. 1: 591, 593, 595, 603, 866, 905, 906, 921, 928; 2: 27, 28, 59, 64, 139, 166, 204, 288; see also Baker, I, and George, E.P.; Heatherly, L. Jr. et al.; Liu, C.T. et al.; Schaefer,H.-E. et al.; Takasugi, T. et al.; Yoo, M.H. et al. George, E.P. and Baker, I. 3: 290, 291, 363, 368, 372, 373 George, E.P. et al. 1: 521, 530, 591, 593, 594,897,898,900,901,902,904,907, 909, 924; 2: 29, 39, 59, 64, 148, 156, 160, 161, 162, 164, 166, 167; 3: 444, Gerads, H. see Hilpert, K. et cld. Gerasimov, V.V. see Kablov, E.N. et al.
ee Binnig, G. et al. see Tonneau, A. et al. Gerlick, R. see Hellner, E. et al. Gerling, R. see Kestler, H. et al. Gerling, R. et al. 1: 743, 745; 3: 649 G e ~ a g n o ~E.i , see Benci, S. et al. German, R.M. 1: 646; 3 644; see also Bose, A. et al., Sims, D.M. et ul. German, R.M. et al. 2: 44 Germann, A. et al. 1: 221 Gernianioli, E. see Benci, S. et al. Gerold, U. and Herzig, Chr 3: 290 Gerold, V. 1: 40,565, 574; 2: 258; see also Berner, D. et al.; Wachtel, E. et al. Gerritsen, A N . I: 944 Gerstenberg, K.W. see Epperson, J.E. et al. Gharnaty, S. see Seaman, C.L. et at. Ghandehan, H.H. 2: 313 Ghernard, G. et al. 1: 353 Ghez, R. 2: 606, 608 Gkijsen, J. see K~mmacher,S. et al. Ghosh, A.K. see Gibala, R. et al. Ghosh, R.N. see Basoalto, H.C. et al. Ghuman, A.R.P. 2: 507, 511 Gialanella, S. et al. 1: 788; 2: 133: 3: 645 : 10; see also Anton, D.L. et al.; D~Pas~uale, 5 . et al.; Gell, M. et al., Mear, B.H. et al. Giamei, A.F. and Tschinkel, J.G. 3 542 Giamei, A.F. et al. 1: 529, 545; 2: 10, 22; 3: 74 Giannini, E. see Manfrinetti, P. et al. Gianozzi, P et al. 1: 157 Giauque, P.W. and Oberli, S. 3: 241 Giauque, W.F. 3 519, 536 Giauque, W.F. and MacDougal, D.P. 3 519, 522
897 Gibala, R. 1: 536, 925; Kim, J.T. et al.; Larsen, M. et al.; Misra, A.K. et al., Noebe, R. et al. Gibala, R. et al. 3: 486 Gibbons, P.C. 1: 463; see also Daulton, T.L. ef al.; Holzer, J.C. et al., Kelton, K.F. et aE.; Levine, L.E. et al.; Libbert, J.L. et al., Shield, et al. Gibbons, P.C. et al. 1: 463, 482 Gibbs, D. see Majkrzak, C.F, et al. Gibbs, H.M. see Jin, R. et al.; Ojima, M. et al.; Wemberger, D.A. et al. Gibbs, J.W. 1: 7, 167, 612 Gibbs, W.S. see Carter, D.H. et al. Gibbs, Z.P see Lilly, A.C. et ul. Gibson, E.D. see Leamy, H.J. et al.; McMasters, O.D. et al.; Verhoeven, J.D. et al. Gibson, J.M. see Hensel, J.C. et al.; Tung, R.T. et al. Gieseke, B. 3: 332 Gieseke, B. and Sikka, V.K. 3 329 Gieseke, B.G. 2: 41; see also Sikka, V.M. et al. Giessen, B.C. 1: 690, 703, 704, 705, 739, 740, 746; see also Bridges, F et al.; Calka, A. et al.; Chang, L.L. and Giessen, B.C.; Davis, S. et al.; Fischer, M. et al.; Predecki, P. et al.; Ray, R. et al.; Ritter, D.L. et al.; Ruhl, R. C. et al.; Sinha, A.K. et al.; Srivastava, P.K. et al. Giessen, B.C. et al. 1: 704, 735; 3: 56 Gifford, G.G. see Chao, P.C. et al. Ciggins, C.S. 1: 989; 2: 494, 495; see also Goebel, J.A. et al. Gigioli, G.C. see Jin, R. et al. Gigliotta, M.F.X. see Huang, S.C. et al.; Jacksan, M.R et al.; Menzies, R.G. et al.; Rowe, R.G. et al. Gigliotti, M.F.X et al. 3 544 Gignoux, D. 1: 441; see also Bauer, E. et al.; Blanco, J. A. et al., Vert, R. et al. Gignoux, D. and Schmitt, D. 3: 97, 183, 804 Giguere, A. et al. 3: 525, 527 Gil, A. see Mrowec, S . et al. Gil, J.M. see Ayres de Campos. J. et al. Gilbert, C.J. et al. 3 684 Gilbert, 3 . et al. 1: 577, 793 Gilbert, R.W et al. 1; 791, 813 Gilbon, D. see Njah, N. et al. Gilchrist, H.L. see Harbison, J.P. et al., Palmstrom, C.J. et al.; Sands, T. et al.; Tabatabaie, N. et al. Gilfrich, J.V. see Buehler, W.J. et al. Gilgen, M.H. see Salathb, R.P. et al. Giling, L.G. see Gardiniers, J.G.E. et al. Gill, B.J. see Taylor, T.A. et al. Gill, S.S. et al. 2: 621 Gillespie, C.C. 1: 3n Gillespie, D.J. see Tritt, T.M. ez al. Gilnian, J.J. 2: 238 Gilman, P.S. 2: 194; see also Das, S.K. et al. Gilmore, R.S. E: 206; see also Fleischer, R.L. et al. Gilp, B.F. and Desai, P.D. 3: 808
898 Gingell, A. see Strangwood, M. et al. Gingrich, N.S. see Wilkinson, M.K. et al. Giorgetti, C. see h a r d , 0. et al. Giorgi, A.L. 1: 214; see also Matthias, B.T. et al. Giorgi, A.L. and Stewart, G.R. 3: 45 Giorgi, A.L. et al. 3: 45 Girard, J.P see Defour, C. et al. Giraud, M. see Morin, P et al. Girgis, K. 1: 252, 377, 378, 385, 390; 3: 811, 812; see also Villars, P. et al. Girifalco, L.A. 1: 423, 565, 577, 849, 874; 2: 610 Girshick, A. see Mahapatra, R. et al., Vitek, V et al. Girshick, A. and Vitek, V. 3: 422, 424, 437, 445, 464 Giua, M. 1: 8, 9, 12, 13, 14 Giua-Lollini, C. 1: 8, 12, 13, 14 Giuzetti, G. see Amioti, M. et al. Givord, D. see Cadogan, J. M, et al.; Courtois, D. et al.; Deportes, J. et al.; itche ell, I.V. et al. Gjostein, N.A. 2: 608 Glade, S.C. and Thadhani, N.N. 3: 740 Gladun, A. see Quyen, N.H. et al. Gladun, V.P. 3: 818, 819, 822, 824, 825 Gladun, V.P. see Kiselyova, N.N. et al.; Savitskii, E.M. et al. Gladun, V.P. and Vashchenko, N,D, 3: 818,819,822,824,825,826,828,834 Gladyshev, V.P. see Kozlovskii, M.T. Gladyshevskii, E.I. 1: 374, 378, 406, 417; 2: 226,630; 3: 808; see also Grin, J.N. et al.; Grin’, YuN. et al.; Kalychak, Ya.M. et al.; Kuz’man, Yu.B. et al.; Levin, E.M. et al.; Skolozdra, R.V et al. Gladys~evskii,E.I. and Bodak, 0.1. 3: 806 Gladyshevskii, E.I. et al. 1: 394,408, 41 I; 3: 90 Gladyshevskii, R.E. 1: 364, 412; 2 630; see also ParthC, E. et al. Gladyshevskii, R.E. et al. 1: 413; 3: 93 Glaeser, W.A. 2: 596, 600; see alLw Heilmann, P. et al. Glakov, V.P. see Ananyn, V.M. et al. Glaser, W. see Schicktanz, S. et al. Glatz, A.C. 2: 462 Glatz, W see Appel, F et al., Clemens, H. et al.; Koeppe, C. et al. Glauber, R. 1: 775 Glaunsinger, W.S. see Guen, L. et al. Glazunov, S.G. see Kdybyshev, O.A. et al. Gleeson, B. see Munroe, T.C. and Gleeson, B. Gleiser, M. see Hultgren, R. et al. Gleiter, H. 1: 955; 2: 267; 3: 749, 750, 751, 752, 753; see also Bohn, R. et al.; Hermann, G. et al., Karch, J. et al. Gleseke, B.G. see Sikka, V.K. et al. Glossop, A.B. 2: 564 Glotzer, S.C. see Donati, C. et al. Glowacki, B.A. 1: 646 Glushko, P.I. et al. 1: 998 Gmelin, E. see Giguere, A. eb al. Gnanamoorthy, R. et al. 3: 334, 335
Godart, C. see Mazumdar, C. et al.; Nagarajan, R. et al.; Tominez, E, et al. Goddard, W.A. I11 1: 155 Godfrey, M.J. see Needs, R.J. et al. Godicke, T. 1: 719 Godinho, M. see Ayres de Campos, J. et al. Godlewska, E. see Mrowec, S. et al. Godlewski, K. see Jedlinski, J. et al.; Mrowec, S. et al. Goebel, J.A. 2: 54, 55 Goebel, J.A. et al. 2: 492, 495 Goedkoop, J.A. 2 309 Goedkoop, J.B. see Durr, H.A. et aE. Goela, J.S. et al. 2: 409 Goeltz, G. see Murani, A.P. et al. Goetz, R.L. see Seetharaman, V. et al.; Semiatin, S.L. et al. Goggin, D.E. see Hoffman, N.J. et al. Gogia, A.K. 2: 106, 107, 110; see also Banerjee, D. et al.; Koss, D.A. et al.; Muraleedharan, K. et al.; Nandy, T.K. et al. Gogia, A.K. et al. 2: 93, 101, 106, 107, 108, 110, 111, 112, 113, 114; 3: 494, 495 Gogtchadze, T.L. see Petrii, O.A. et al. Gogulya, M.F. see Batsanov, S.S. et al. Gohia, A.K. see Banerjee, L). et al. Gohring, E. see Artz, E. et al. Goken, M. 3: 227; see also Kempf, M. et al.; Weber, T. et al. Goken, M. et al. 3: 215, 228 Gokhale, A.B. see Schlesinger, et al. Golberg, D. et al. 3: 56 Goldberg, D. 1: 924 Goldman, A.I. 1: 482, 491; see rclso Bancel, P.A. et al.; Guryan, C.A. et al.; Libbert, J.L. et al., Majkrzak, C.F et al.; Shield, J.E. et al. Goldman, A.I. et al. 1: 161, 162, 463, 482 Goldman, E. 1: 989; 3: 542 Goldman, E.H. 2: 68, 69; 3 602; see also Darolia, R. et al. Goldman, J.E. 1: 444 Goldmann, A.I. see Shen, Z. et al. Goldschmidt, T. 2 647 ~ o ~ d s c h m i dV.M. t , 1: 10, 14, 16 Goldsmid, H.J. 2 453 Golec, C.G. see Sparks, C.J. et al. Golikov, V.M. see Borisov, V.T. et al. Golikova, M.S. et al. 3: 813, 814, 816, 834; Savitskii, E.M. et al. Golosov, N.S. see Popov, L.E. et al. Golubkova, G.B. see Mikhailenko, S.D. et al. Golubkova, T. see Ivanov, E. et al. Goman’kov, V.I. et al. 1: 726 Gomes, A.S.L. et al. 2: 421 Gbmez, A. see Perez-Campos, R. et al. Gbmez, A. see Yacaman, M.J. et al. Gomez, J.A. see Xia, S.K. et al. Gbmez, M. see Clarke, R.S. et al. Gomez-Polo, C. see Conde, F. et al. Gbmez-Sal, J.C. see Blanco, J.A. et al. Gompf, F. et al. 1: 154, 161 Gompper, G, 1: 614 Gomyo, A. see Kobayashi, K. et al. Gong, J.R. see Hussien, S.A. et al. Gong, H.Y. see Cheng, Z.H. et al.
Gong, W. see Hadjipanayis, G.G. et al. Gonis, A. 1: 25; see also Stocks, G.M. et al.; Turchi, P.E.A. et at‘. Gonis, A. et al. 1: 24, 29, 30, 34, 41, 849, 850 Gonsalves, K.E. 1: 650 Gonscr, U. 3: 169; see also Jink, J, et al. Goo, E. 1: 916; 3: 405: see also Park, K.T. and Goo, E. Goo, E. et al. 1: 916: 3: 404, 416 Goodell, P.D. 2: 484 Goodell, P.D. et al. 2: 478 Goodenough, J.B. 1: 441; 3: 806 Goodhue, W.D. et al. 3 787 Goodman, C.H.L. 1: 643 Goodman, D.A. et al. 1: 247 Goodrum, J.W. 2: 282 Goods. S.H. see Mills, M.J. et al. Goodwin, F.E. 2: 519, 520 Goodwin, G.M. 3: 572; see also Maziasz, P.J. et al. Goodwin, M.W see Reed, M.A. et al. Gopal, E.S.R. 1: 1019, 1020; see also Ghitralekha, J. et al. Gopalakrishnan, K.V. see Nagarajan, R. et al. Goralczyk, R. 2: 646 Gorbunov, V.L. see Vlasova, E.N. et al. Gordon, D.E. and Uiini, C.K. 3: 329,330 Gordon, D.E. et al. 3: 329 Gordon, R.G. 2 623 Gordy, W. 1: 101, 110,243,245,246,252 Gorelenko, YLLK.see Gladyshevskii, E.I. et al. Gorler, G.P. see Wilde, G. et al. Gornostyrev, Yu.N. 1: 534; see also Greenberg, B.A. el: al.; Indenbaum, V.N. et al., Mryasov, O.N. et al. Gor~ostyrev,Yu.N. et al. 1: 525 Goro, T. see Koyama, K. et al. Gorodetsky, S, see Vasiliev, M. et al. Goryunova, N.A. 1: 16, 350 Gorzel, A. see Palm, M. et al. Gosele, U. see Tan, T.Y. et al.; Tu, K.N. et al. Gosh, A.K. see Lornbard, C.M. et al. Goshchitskii, B.N. see Moseev, N.V. et al. Gossard, A,C. 3: 779; see also Bucher, E. et al.; Chemla, D.S. et al.; Dingle, R. et al.; Miller, D.A.B. et al. Goto, T. see Hashimoto, T. et al.; Koyama, K. et al.; Son, J.-Y. et al.; Thessicu, C. et al. Gottstein, G. see Ball, J. et ul.; Escher, C. et al. Gottstem, G. et al. 1: 864 Gotze, W. see ~ e n g t z e ~ ~U. u set , al. Gould, G. 1: 960 Goulette, M.J. see Hill, A.D. et al. Gouma, P.I. et al. 3: 476, 649 Gourdin, W.H. 3 738 Goutzoulis, A.P. 2: 432 Gouveia-Neto, AS. see Gomes, A.S.L. et al. Govern, C. see Larsen, D. and Govern, C. Govier, D. see Rawers, J.C. et al. Govinda Rajan, K. see Sahu, P.C. et al. Coward, G.W 2: 57,490,492 Goward, G.W. and Boone, D.H. 3: 566, 570
A u ~ Index h ~ ~ Grabke, H.J. 1: 987, 989, 994, 995, 999, 1005; see also Viefhaus, H. et al. Grabke, N.J. et al. 1: 927, 983, 987, 990, 991, 994, 1001, 1002, 1005 Graf, K. H. et al, 3: 31; see also Cox, G. et al. Graf, L. 2: 308 Graf, T. see Josda, J.L. et al. Graham, D.B. 2: 135 Graham, H.C. see Choudhury, N.A. et al. Graham, M.J. 1: 981, 987; see also Mitchell, D.F. et d. Graham, N.N. see Thadhani, N.N. et al. Graham, R.A. 3: 724, 740, 743; see also Anderson, M.U. et al.; Dunbar, E. et al.; Haminetter, W.F. et al.; Simonsen, I.K. et al.; Thadhani, N.N. et al. Graham, W. 1: 616 Grahle, P. see Artz, E. and Grahle P.; Artz, E. et al.; Schneibel, J.H. et al. Grahle, P. and Artz, E. 3: 656 Grala, E.M. 2: 60; 3: 502 Granato, A.V. 1: 874; 3: 694 Grandin, A.W. see Markiewicz, W.D. et al. Grandjean, F. see Hautot, D. et al. Graneli, B. see Brun, T.O. et al. Granovskii, A.B. et al. 1: 944 Granqvist, C.G. 2: 41 1 Granqvist, C.G. and B u h ~ a n R.A. , 3: 753 Granstaff, S.M. see Thoinpson, J.C. et al. Grant, M. see Thoinson, J.R. et al. Grant, N.J. see Giessen, B.C. et al., Predecki, P et al.; Ray, R. et al.; Ruhl, R.C. et al.; Srivastava, P.K. et al.; Ritter, D.L. et al. Gras, C. see Charlot, F et al. Gras, C . et al. 3: 744 Gratias, D. see Cahn, J.W. et aE.; Caillard, D. et al.; Devaud-Rzepski, J. et al.; Finel, A. et al.; Portier, R. et al.; Sanchez, J.M. et al.; Shechtman, D. et al.; Urban, K. et al.; Yu-Zhang, E;. et al. Gratids, D. et al. 1: 483 Grattidge, W. see Wawrousek, H. et al. Gratz, E. 1: 1028; see also Bauer, E. et al.; Hauser, R. et al. Graupner, €3. see Hammer, L. et al. Graupner, H. et al. 3: 220 Gravereau, P. see Reny, E. et al. Graves, J.A. 2: 296; see also Miracle, D.B. et al., Rhodes, C.G. et al.; Smith. P.R. et al.; Vassiliou, M.S. et al. Graves, J.A. et al. 2: 292, 293 Graves, R.S. see Williarns, R.K. et al. Gray, G.T. see Gray, G.T. I11 3: 368, 371, 372, 373; Albert, D.E. and Gray, G.T.; see ulso Noke, D.A.; Jin, Z. et al.; Maloy, n , et al. S.A. et al.; ~ u r a ~ e e d h a r aK. Gray, G.T. I11 and Embury, J.D. 3: 366, 368, 369 Gray, G.T. 111 et al. 3: 363, 368 Gray, J.S. see Du, H.L. et al. Gray, K.E. see Grimsditch, M. et al. Grebe, H.A. aad Thadhani, N.N. 3: 736 Greber, T see Osterwalder, J. et al. Stuck, A. et al.
Greegor, R.B. 1: 986 Green, A.J. et al. 3: 809 Green, G. et al, 3: 527, 529 Green, G.F see Zimrn, C.B. et al. Green, M.L. and Cohen, M. 3: 404,416 Greenberg, B.A. 1: 534; 2: 74; see aim Adrianovskii, B.P. et al.; Gornostyrev, Y1i.N. et al.; Indenbaurn, V.N. et al. Greenberg, B.A. and Gornostirev, Y .N. 3: 622 Greenberg, B.A. et al. 1: 533, 534, 545, 548; 2: 74, 78; 3: 413 Greene, J.E. see Adibi, F et al.; Bergstrom, D.B. et al.; Hultman, L. et ul.; Petrov, I. et al.; Ramatlath, G. et al.; Shin, S.M. et al. Greene, J.E. et al. 3: 664, 666 Greenfield, P. 2: 9 Greenough, R.D. 2: 400, 401; see also Abell, J.S. et al.; Galloway, N. et al., Jenner, A.G.1, et al.; Parvinmehr, A. et al.; Schulze, M.P. et al. Greenwood, G.W. 1: 918; see also Harris, J,E. et al., Mishra, R.S. et al. Greenwood, G.W. et al. 1: 918 Greenwood, N.N. and Earnshaw, A. 3: 21 Greer, A.L. 1: 960, 743, 745, 747, 748, 750, 751; 3: 701; see also Drehman, A.J. et al.; Highrnore, F.J. et al.; Rnowles, K.M. et al.; Kui, H.-W. et al.: Vittra, S. et al. Greer, A.L. et al. 1: 739 Gregan, G.P.J. ef al. 1: 735 Greggi, J, 2: 282; see also Baron, M. et al. Gregoire, J. see Sainfort, G. et al. Gregory, E. 2: 361; see also Stekly, Z.J.J. and Gregory, E., Summers, L.T. et al.; Walker, M.S. et al. Gregory, E. et al. 2: 357, 358, 359, 361, 362 Gregory, T.J. see Duddles, N.J. et al. Gregson. P.J. 2: 188, 190 Greidanus, F.J.A.M. 2 442, 450 Greig, D. see Blatt, F.J. et al. Gremer, E.S. see Levmstein, H.J. er al. Greiner, J.H. see Huang, H.C.W. et al. Grenga, H.E. et al. 2: 579, 580 Grenoble, D.E. 2: 586 Grensing, F.C. 1: 979,1000; see also Nieh, T.G. et al. Gresko, L.S. see Wakugawa, J.N. et al. Grewe, N. 1: 211, 215, 2x8 Grewen, J. 2: 175 Grey, F. see Dosch, H. et al., Krumniacher, S. et al. Gribulya, V. B. see Savitskii, E. M. et al. Grier, D.H. et al. 3: 251 Griessen, R. 2: 477, 478; see also Joss, W. et al.; Saloinons, E. et al. Grieveson, P see Alcock, C.B. et al. Griffin, D. et al. 3: 564 GriBth, A.A. 1: 86, 873 GriEth, W.M. et al. 2: 175, 178, 194 Griffiths, M. 1: 791, 813; see also Gilbert, R.W et al. Grigoneva, T. see Ivanov, E. et al. Grillo, D.C. see Jeon, H. et al. Grimddi, M.G. see Baeri, P, et al. Grimm, H. see Majkrzak, C.F et al. Gnmm, €3.0. and Sonunerfield, A. 3: 114
899 Grimmer, H. et al. 1: 598 Grimsditch, M. see Natesan, K. et al.; OkLzmoto, P.R. et al.; Rehn, L.E. et al.; Renusch, D. et al. Grimsditch, M. et al. 1: 815 Grimvall, G. 3: 808 Grin, J.N. 1: 412; 2: 630; six also Gladyshevskii, E.I. et al. Grin, J.N. et al. 1: 114, 408 Grin, J.N. and Glad~shevskii,R.E. 3: 808 Grin, Y see von Schnering, H.G. et al. Grin, Yu.N. see Sichevich, O.M. et al. Grin', Yu.N. 1: 364, 374 Grin', Yu.N. et al. 1: 374 Grinevich, G.N. see Kuzmenko, P.P et al. Grioni, M. see Pilo, Th. et al. Grioni, M. et al. 1: 136 Grivord, D. see Butler, Grivord, F sec> Butler, B. et al. Groenewald, T. 2: 562,570,572,649,652, 653 Grolich, M. et al. 2: 271 Groll, M. 2: 486; see also Heine, D. et al. Gronsky, R. see Krishnan, K.M. et al., Williams, D.B. et nl. Gronsky, R. et al, 1: 466 Gros, Y. 1: 63311, 688 Gross, K. see Chartouni, D. et al.; ZiitteL A. et al. Grossinger, R. see Katter, M. et al. Grossinger, R. et al. 2: 314 Grossmann, J. et al. 3: 543 Groth 1: 309, 310 Groult, D. see Provost, J. et al. Grove, A.S. 1: 997 Groves, W.O. 1: 875 Grudnen, N. see Becla, P. et al. Gruger, A. see Stefa~~ay, V. et al. Gruh12W. ,see Dahl, W et al. Grunblatt, G. see Gregory, E. et ~ r t . Grimdhoff, K.J. see Kunipfert, J. et al., Peters, M.et al. Grundy, P.J. 2: 449 Gruner, G. 1: 213, 214 Griinling, H.W. 2: 227, 228; see also Bauer, R. et at.; Singheiser, L. et al. Grupp, Ch. see Wurs~hum,R. et al. Grushko, 12. 1: 454; see also Wurschum, R.er al. Grushko, B. et al. 1: 112 Gruter, K. et al. 3: 670 Gryko, J. see R ~ ~ ~ c l i a n d rG.K. a n , et al. Grytsiv, V. I., see Tomashik, V.N. and Giytsiv, V.I. Gschneidner, K.A. 1: 247, 406; 3 806: see also Teaturn, E. et al.; Waber, J.T. et al.; Zimm, C.B. et al., Choe, W. et al.; Dan'kov, S.Yu. et al.; Korte, B.J. et al.; Levin, E.M. et al.; Niu, X.J. et al.; Pecharsky, V.K. and Gschneidner, K.A. Jr.; Pecharsky, V.M. et al.; Takeya, H. et al., Teatum, E. et al., voii Rdnke, P.J. et al.; Han, S.H. et al. Gschneidner, K.A. and Eyring, L. 3: 806 Gschneidner, K.A. Jr. and Pecharsky, V.K. 3: 519, 520, 523, 524, 525, 527, 530 Gschneidner, K.A. Jr. et al. 3: 807
Author Index
900 Gschwend, K. see McCoy, 3. et al.; Sato, H. et al. Gschwend, K, et al. 1: 775, 776, 786 Gu, Y. see Yama~e-Mitarai,Y. et al.; Yu, X.H. et al. Gu, Y. et al. 3: 69, 70 Gu, Y.M. 1: 569 Gualtieri, J.G. 1: 180 Guan, W.Y. see Ku, H.C. et al. R.W. Guard, R.W and Westbrook, J.H. 3: 66 Guay, D. see Blouin, M. et al. Gubleva, D.N. 1: 712, 713, 719; see also Pantelcimonov, L.A. et al. Gubser, D.U. 2: 367 Gudat, W see rumm mac her, S, et al. Gude, A. and Mehrer, H. 3: 290 Guder, S. et al. 3: 454 Guedo, J.Y. et al. 3: 416 Guen, L. et al. 1: 356 Guenais, B. see Guivar’c, A. et al. Guenin, G. et al. 1: 875; 3: 241 GuCrin, R. see Gu~var’c,A. et al. Guerm, R. see Guivarc’h, A. et al. Guertin, R.P see Meissner, G.W. et al. Guertler, W. 1: 16; 2: 577 Guguere, A. see Foldeaki, M. et al. Guha, S. see Schulson, E.M. et al. Gm, J.N. see Wang, R.H. et al. Guichard, D. see Baccino, R. et al. Guilfoyle, S.J. see Brown, J.D. ct al. Guillard, C. 2: 309 GL~illaunie,J.C. 1: 676 R. et al. GuillopC, M, see Rosato, V. et al. Guillot, J.P see Beaufort, M.F. Gmmaraes, A.P. see voii Ranke, Guinan, M.W. see Did2 de la R et al. G ~ i n a nM.W. , et al. 1: 808 Guinea, F. see Rose, J.H. et al. Guiraldenq, P. 1: 757; see also Labarge, J.-J. et al.; Poyet, P et al.
Gulbransen, E.A. see Berztiss, D.A. et al. Gulchman, A.L. see Kanonenko, V.L. et al. Guloy, A.M. see Corbett, J.D. et al.; Kwon, Y.U. et al. Culoy, A.M. and Corbett, J.D. 3: 89 Gulyev, B.B. and Piivleiiko, L.F. 3: 812, 822, 824 Gumbsch, P. see Ludwig, M. and Gumbsch, P.; Schroll, R. et al. Gumbsch, P and Schroll, R. 3: 772 Gumen, N.M. 1: 768 Guminski, C. 3: 22, 26, 27, 28, 30, 32, 33; see also Borgstedt, H.U. et al. Guminski, C. and Calus, Z. 3 22, 28. 30, 800
Gummel, H.K. see Sze, S.M. and Gummel, N.K. ~unnarsson,0 . 1: 130; see also Bagno, P. et al. Gunnarsson, 0. et al. 1: 132 Gunshor, R.L. see Jeon, H. et al.
Gfinter, S. sec Morris, D.G. et al. Gunther, K.Z. 3: 779 Gunther, S. see Morris, D.G. and Gunther, S. G~ntheradt,H. see Busch, G. et al. Guntherodt, G. 1: 875 Guntherodt, G. et al. 1: 217 Gunton, D.J. and Sauders, G.A. 3 258 Guo, C. see Barnes, A.C. et al. Guo, H. see Thomson, J. R. et al. Guo, H.Q. et al. 1: 743 Guo, J.T. see Liu, Z.G. et al. Guo, K.-J. see Wiley, J.D. et al. Guo, X.-Q. see Hong, T. et al.; Sluiter, M.H.F et al. Guo, X.-Q. et al. 2: 67, 68, 70, 876, 920 Guo, Y. see Tang, W. et al. Guowei, Z. see Sheng, L. et al. Gupta, B.K. 2: 490 Gupta, D. 1: 765; see also Fishman, S.G. et al. Gupta, D. and Ho, P.S. 3: 674 Gupta, D. et al. 1: 575, 762; Tt; 604 Gupta, K.P. 3: 803 Gupta, L.C. 3 103, 104; see also Mazumdar, C. et al.; Nagarajan, R. et al. Gupta, M. 2: 479,480,481; see also Belin, E. et al. Gupta, R.K. see Agrawal, R.C. and Gupta, R.K. Gurland, J. 1: 926 Gurry, R.W. 1: 245, 246, 247, 712; see also Darken, L.S. et al. Curyan, C.A. see Goldman, A.I. ef al. Guryan, C.A. e t al. 1: 475 Gusak, A.M. see Bushin, 1.N. et al. Gushchik, G.F see Yanson, T.I. et al. Gustafsson, V see Nygren, S. et al. Gutfieisch, 0 . see Kubis, M. et al.; Yartys, V.A. et al. Giither, V. see Knippscheer, S. et al. Guthrie, H.C. see Callegari, A. et al. Gutov, L.A. see Razuvayera, B.D. et al. Guttman, L. 1: 500, 504, 506, 1020 Guyot, P 1: 454,461,465,471,472,475, 482, 483; 2: 180, 186; see also Papon, A.M. et al. Guyot, P. et al. 1: 473 Guzey, L.S. et al. 1: 728 Guzik, A. and Pierre, J. 3: 104 Gyorffy, B.L. 1: 24,25,27; see also Ceder, G. et al.; Stocks, G.M. et al. Gyorffy, B.L. et al. 1: 27, 41, 47, 50 Gyorgy, E.M. see Cava, R.J. et al., Kwo, J. et al. Gypen, L.A. and Deruyttere, A. 3 359 Gysler, A. 2: 282; see also Ahrens, T. et al. Gyulal, 5. see ~aimstrom,C.J. et al. Gyurko, A.M. and Sanchez, J.M. 3: 68 Haake, F.K. see DeLuca, D.P. et al. Haas, C. see Otto, M.J. et al. Haasch, R.T. et al. 1: 982, 986, 989 Haase, M.A. et al. 2: 326, 339, 428, 429, 43 1 Haasen, P 1: 521, 915, 917: 2: 268; see also Beddoe, R.E. et al.; Grolich, M. et al.; Rembges, M. et al. Habel, U. see Yolton, C.F. et al. Habel, U. et al. 3: 649, 650
H a b e ~ e ~ eW.-U. r, et al. 3: 806
, D.V. et al., Gao, C. et al.; , G.C. et al. 3: 98 Haehl, W.D. 1: 721 Hafiier, J. 1: 56, 67, 70, 154; see also Stadler, R. et al. Hafner, J. eit al. 3: 798 Haftel, M. 1: 620 Hagel, W.C. 1: 575, 757, 762, 764, 765, 766, 768; 3: 805; see also Sims, C.T. et al. Hagel, W.C. and Westbrook, J.H. 3: 353 Hagen, M. and Finnis, M.W. 3: 769, 770, 771 Hagenmuller, P. see Cros, C. ef al.;
Hagiwara, M. see Inoue, A. et al. Hagiwara, R. see S a ~ b o n g T. ~ , et al. Hagston, W.E. see Duddles, N.J. et al. Hahn, H. et al. 1: 575, 849 Hahn, J.P see Kozubski, R. et al. Hahn, K.N. 2: 924; 2 58, 59, 60, 61, 64; see also Vedula, K. et al. Hahn, K.H. and Vedula Hahn, T. 1: 294, 363, 38 Hahn, Y.D. see Whang, Hahn, Y.D. and Whang, S.H. 3: 368 Hairre, A. see Defour, C. et al. Hajaligol, M.R. see Deem, S.C. et al., Mistier, R.E. et al., Straws, J.T. et al. Hajaligol, M.R. et al. 3: 504, 655 Hajashi, C. see ~ a r k i e ~ i cW.D. z , et al. Hajko, V Jr. see Macko, D. et al. Hakansson, P. see Baudin, K. er al. Halas, E. see Stekly, Z.J.3. et al. Halbwachs, M. 3: 802; see also Beretz, D. P t al. Haldaar, P. 2: 366; see also Motowidlo, L.R. et al.; Walker, M.S. et al. Hale, J.R. see Thome, R.J. et al. Halene, C. 2 485 Hall, C.W. see Coupland, D.R. et Hall, D.E. see Bertocci, U. et al.; R.E. et al. Hall, D.G. see Erdogan, T. et al. Hall, E L . 1: 540, 638, 640, 916; 2: 77, 78, 80, 81, 82, 85, 87, 95, 101; see also Banerjee, D. et al.; Briant, C.L. et al.; Huang, S.C. et al.; Huang, S.C. and Hall, E.L.; Livingston, J.D. and Hall, E.L.; Livingston, J.D. et al.; Rowe, R.G. et al.
et al. Ball, R.C. 2: 307 Hallais, J.P. see Martin, G.M. et al. Hallman, E.D. 1: 160; see also Svensson, E.C. et al. Halthuis, J.T. see Dariel, M.P et al.
Author Index Eam, R.K. 2 7, 258, 259, 261, 262, 265, 267, 268; see also Kear, B.H. et al. Hamada, S. see Chiba, A. et al. Hamada, T. see Oshima, R. et al. Hamada, Y see Inoue, H.R.P. et al. Hamada, Y. et al. 1: 836, 839, 840, 841, 842 Hamamoto, €3. see Sagawa, M. et al. Hamamoto, K. see Komatsu, K. et al. Hamby, C.J. 2 637, 643 Hamdi, A.H. see Kattelus, H.P. et al., Zhu, M.F. et al. Hamed, N. see Mish~ma~ Y. et al.; Mishima, Y et al. Hameed, M.Z. et al. 1: 800, 801, 802 Hamer, W.J. 2 652 Hamilton, C.H. see Rhodes, C.G. et al. Hammann, D.R. 1: 134; see also Harmon, B. N. et al.; Mattheiss, L.F and H a m a n n D.R. Hammelman, K.H. see Ishiyama, S, et al. Hammer, L. see Blum, V et al.; Graupner, Hi. et al.; Heinz, K. and Hammer, L. Hammer, 2. et al. 3: 220, 224 Hammer, W.N. see Baglin, J.E.E. et al. Hammetter, W.D. see Thadhani, N.N. et al. Hammetter, W.D. et al. 3: 744 Hammond, J.P. 2: 523 H a ~ m ~ nR.H. d , 1: 689; 2 610, 612 Hampel, C.A. 2: 518, 523, 524, 647 Harnpikian, J. see Chen Hampton, A.F. see StraEord, K.N. and H a m ~ t o nA,F Hampton, T.E. see Singh, S.K. et al. Hamrin, K. see Siegbahn, K. et al. Haii, H.N. see Park, S.J. et al. Han, J.-W. and Kamber, M. 3: 864 Han, S.B. see Pfeifer, H.U. et al. Han, X. see Hu, J. et al. Han, X.F see Wang, Y.Z. et al.: Yang, C.P et al.; Yang, F.M. et al. Han, X.F et al. 3: 94, 99 Han, Y.F. and Xing, Z.P 3: 609 Wan, Y.F. et al. 2: 41, 42, 46; 3: 493 Hanada, S. I: 537; s(3g also Chiba, A. et al., Hosoda, H. et al.: Lee, J.W. et al., Ogura, T. et al.; Takasugi, T. et al.; Watanabe, S. et aE.; Yoo, M.H. et al.; Yoshida, M. et a/., Yoshimi, K. et al. Hanada, S. et al. 2 218 Hanada, Y. see Inoue, H.R. et ul. Hanak, J. et al. 2 360 Hanamura, T. see Ikematsu, Y, et al.; Masahashi, N. et al.; Uernori, R. et al. Hanamura, T. et al. 1: 640; Hanawa, T. seo Tsutsumi, S Hancock, G.F. 2: 63; 3: 290 a ~ d M c D o ~ n eB.R. 1 , 3: 771 : 489, 494, 496; see also Wicholls, J.R. et al. Hancock, P. et al. 2: 496 Hatieman, D, 2: 51 1 Haner, A.N. 1: 616 Hanes, D.B. and Gibala, IQ. 3: 346 Hanitsch, R, see Mitchell, I.V. et al. Hank, J.J. 1: 875, 890 Hanke, G. see Rosenfeld, R. et al.
Hanks, R. and Faktor, M.M. 3: 786 Hanlon, J.E. see Wasilewski, R.J. et al. Hanneman, R.E. 1: 172 ~ a n n i g a n J.W. , see Schwarz, R.B. et al. Hanrahan, R.J. see Brady, M.P. et al. Hanrahan, R.J. Jr. 3 49 Hanrahan, R.S. Jr. et al. 3: 44, 45, 46 Hansen, D.A. 1: 325; see also Smith, J.F. and Hansen, D.A. Wansen, M. 1: 10, 101; 2 134, 309, 578, 589; 3: 785, 786 Hansen, P 2: 437, 441, 443, 450 Hanson, K. 2: 259, 260, 261, 262 Hao, S . X . et al. 1: 719 Hara, K. et al. 2: 330 Harada, H. see Gu, Y. et al.; Murakami, A. et al.; Yamabe, Y. et al., Yainabe-Mitarai, Y. et al.; Yu, X.H. et al. Harada, T. and Kuji, T. 3: 102 Harada, Y . et al. 841 Haran~uzo,I.Z. see Posgay, G. et al. Harbeke, G. et al. 2: 332 Harbison, J.P. see Sands, T. et al., Tabatabaie, N. et al. Harbison, J.P et al. 3: 783, 787 Harcourt, G.A. 2: 576 Harder, A. see Zintl, E. et al. Hardie, D. and Mclntyre, P. 3: 347 Hardman-Rhyne, K. et al. 2: 481 Hardouin Duparc, A. see Barbu, A, et al. Hardwick, D.A. 1: 651 Hardwick, D.A. et al. 3: 657 Hardy, C. 2 518, 524 Hardy, H.K. 1: 475 Hardy, V. see Provost, J. et al. Hare, J.P. see David, W.I.F. et al. Hare, R,1: 6, 15 Hargreaves, R. 1: 93, 94 Hargrove, W.H. 1: 671 Harker, D. 1: 167, 171; 2: 564 Harkness, H.H. see Zmm, C.B. et al. Har'kov, E.1. see Kuzmenko, P.P et al. Harland, C.L. and Davies, H.A. 3: 97 Hgrle, L. see Schubert, K. et al. Harman, G. 2: 653 Harrnelin, M. see Yct-Zhang, K. et al. Harmon, B.N. 1: 137, 143, 199 Harmon, B.N. et al. 1: 60 Harrnouche, M.R. 1: 885,886,920; 2: 55, 56 Harms, U. see Sin, 0. et al. Harper, J.M.E. see Charai, A. et al., Hdrnstriiin, S.E. et al., Svilan, V. et al. Harper, J.M.E. et al. 2 625 Harm, E.P. see Huang, H.C.W et al. Harris, G.L. 3: 808 Harm, I.R. 2: 479; 3: 804; see also Kubis, M. et al., Ragg, O.M. aiid Harris, I.R.; Willey, D.B. et al.; Yartys, V.A. et al.; Mitchell, I.V. et al.; Rozendaal, E. et al. Harris, I.R. and McGuiiiess, P.J. 3: 102 Harris, J.E. et al. 1: 918 Harris, J.J. 2: 432; see also Foxon, C.T. and Harris, J.J. Harris, J.M. et al. 2: 617 Harris, R. see Broomfield, R.W et al. Harm, K. et al. 2: 3 Harris, L.A. 2: 646
90 1 Harris, M.J. et al. 3: 258 Harris, P.G. see Todd, A.G. et al. Hank, S.W. see Baudry, A. et al. Hams, T.D. see Klopf, R.F. et ul, Harrison, A. see Birc~enall,C.E. et al. Harrison, J.G. see Heaton, R.A. et al. Hari"ioi1, W.A. 1: 78, 80 Hart, E.W. 1: 955 Hart, H.R. Jr. see Bean, C.P. er: al.; Fleischer, R.L. et aE. Harte, AS. et al. 3: 60 Harten, U. et al. J: 226 ~art~eld-Wunsch, S.E. 1: 925; see also Larsen, M. et al.; Misra, A.K. et al. Hartfield-Wunsch, S.E. et al. 3: 601 Hartig, C . see Lebensohn, R.A. et al.; Meckmg, H. et al. Hartig, H. see Bartels, A. et al. Hartman, P. 1: 167 Hartmann, H. 2 506 ~ a r t ~ a nH. n ,et al. 3: 34 Hartwig, K.T. et al. 2 651 Harutouni, S . see Bavarian, B. et al. Hasaka, M. sec Udoh, K.-1. et al. Hasan, F. see Cliff, G . et al. Hase, N. see Ohnaka, K. et al. Hasegawa, F see Kanaya, Hi. et al. Hascgawa, H. et al. 2: 335 Hasegawa, K. 2: 446, 447; see also Ohnaka, K. et al. Hasegawa, 0. see Yamaoka, T. et al. Hawegawa, S. 1: 735 Hasegdwa, s. et al. 2 216; 3: 315 Hashim, H. see Chakravorty, S. et al. Hashimoto, H. see Song, Z.Y. et al. Hashimoto, H. et al. 3: 653 Hashimoto, K. &seeKasaharu, K. et al.; Masahashi, N. et al.; Nobuki, M. et al.; Tsujimoto, T. and Hashimoto, K.; Uasumoto, Y. et al. Hashirnoto, K. et al. 1: 720, 928 Hashimoto, M. see Wakayarna, S. et al. Hashimoto, M. et al. 1: 604 Hashimoto, S. see Hiraga, K, et al. H ~ s h i m o tT. ~ , 1: 537; see also Kiniura, K. et al.; Shibuya, T. et al.; Takeuchi, S. et al.; Tokai, Y. et al.; Tomokiyo, A. et al. Hashimoto, T. et al. I: 534, 773, 777, 779; 3: 527 Hashmoto, Y. see Zwata, M. et al. Hashirnoto, Y et al. 2: 361 Hashin, 2;. 1: 202, 208 Hasoon, F.S. 1: 351. 356 H a s , G. 2 410: sec also Hunter, W.R. et al. Hasse, J. see ~ r e k k e l ~K.H. r , et al. Hasse, M.A. see Park, R.M. et al. Haszko, S.E. see Wernick, J.H. et al. Hata, S. see Kato, K. et al. Hatch, A.M. see Stekly, Z.J.J. et al. Hatcher, R.D. see Bakker, H. et al.; Welch, D.O. et al. Hatem, G. .see Jexinane, A. et al. Hatherly, M. see Stevens, G.T. et al. Hatt, B.A. 1: 410 Hatta, 1. 1: 777 Hauback, B.C. et al. 3: 48 Haubert, R.C. see Menzies, R.G. et al. Haubold, €3.-G. I: 40
Author Index
902 Haubold, T. see Bohn, R. et al. Hauck, 5. 1: 282, 298, 299, 305 Hauck, J. et al. 1: 277, 282, 284, 289, 292, 294, 296, 300, 301, 302, 305 Haucke, W. 1: 411; see Zintl, E. et al. Haufe, K. and Rahmel, A. 3: 707 Hausch, G. 2 : 3; see also Zogg, H. et al. Hauscr, J.J. see Boyce, J.B. et al.; Warren, W W. Jr. et al. Hauser, J.J. et al. 1: 484 Hauser, R. ,see Bauer, E. et d. Hauser, R. et al. 3: 104, 162 Haushalter, R. see Messerschmidt, U. et al. Haul, C. see Nicolas-Chaubet, D. et al. Nautojarvi, P. see Alatalo, M. et al.; Corbel, C. et al., Saannen, K. et al. Hautot, D. et al. 3: 100 Hauy, R.J. 1: 167 Haverkort, J.E.M. see Weegels, L.M. et al. Havinga, E.E. 1: 391, 392 Havinga, E.E. et al.1: 392, 393 Hawk, J.A. see Wilsdorf, H.G.E. et al. Hawkins, R.J. et al. 1: 32, 41 Hawthorne, F.C. 3: 3 Hawkridge, D.G. 2: 308, 309 Haworth, C.W. 1: 955; see also Hume-Rothery, W et al. Hay, P.J. see Chen, S.P. et al., Voter, A.F. et al.
Hayakawa, T. .see Ikeda, M. et al. Hayashi, I. et al. 2: 336 Hayashi, IS.see Nakano, T. et al.; Takai, M. et al. Hayashi, T. et al. 3: 300, 309, 310, 31 1 Haydock, R. 1: 31, 32, 33 Haydock, R. et al. 1: 81 Hayes, D.D. see Prater, J.T. et al. Hayes, D.T. 2 383 Hayes, R.R4. see Jeiig, Y.-L. et al. Hayes, R.W. 2: 115 Hayoun, M. see Rey-Losada, G. et al. Hayoz, J. see Pilo, Th. et a/. Hazelton, D.W. see Markiewlcz, W.D. et al., Schwall, R.E. et al.; Walker, M.S. et al. also Chou, C.T. et al.; Couret, A. et al.; Evans, D.J. et al.; Rao, S. et al., Sun, Y.Q. et al. Hazzledine, P.M. et al. 1: 521, 527, 540, 545 He, L. and Ma, 3: 723 He, L.X. 1: 455 He, L.X. et al. 1: 457, 467 He, M.Y. et al. 1: 927 He, P. see McGahan, W.A. et al. He, X.-M, et al. 3: 665 He, V. see Chen, H. et al.; Chu, F. et al.; Miller, M.K. et al.; Shen, T.D. et al.; Zheng, M. et al. He, Y. and Schwarz, R.B. 3: 683 He, Y. et al. I: 734; 3: 682, 683, 685, 686, 691, 701 Head, A.M. 1: 540, 542 Head, A.K. et al. I: 506, 508, 524, 536; 3: 383
Headley, T.J. see Baeslack, W.A. I11 et al.; Cieslak, M.J. et al. Hearst, M.A. 3: 864; sce also Chen, M. et al. Heathcote, J. see Rowe, R.G. et al. Heatherly, L. see Liu, C.T. et al. Heatherly, L. Jr. er al. 3 212, 223 Heaton, R.A. et al. 2: I30 Hebenstreit, E.L.D. et al. 3: 213 Hebenstreit, W see Gauthier, Y et al., Hebenstreit, E.L.D. et al. Hebsur, M.G. 1: 992, 994; see also Doychak, J. et al., Raj, S.V. e f al. Hebsur, M.G. et al. 1: 988, 992, 994; 2: 169; 3: 494 He, C . et al. 3: 53 Hecht, N.L. see Lee, J.1. et al. Hecht, R.J. 2: 1007; 2: 223, 290, 295; see also Corey, R.G. et al., Goebel, J.A. et al. Hecht, R.T. see Maloney, M.J. and Hecht, R.T. Hedenqvist, P et al. 3: 663 Hedin, L. 1: 130, 132, 138, 197 Hedman, J. see Seigbahn, K. et al. Heer, C.V. et al. 3: 520 Heess, F see Heine, D. et al. Heffner, R.H. 1: 215 Hegde, H. see Cadieu, F.J. et al.; Kamprath, H. et al.; Rani, R. et nl. Hegde, H. et al. 2: 315 Heger, G. see Qui~chini,M. et al. Hehemann, R.F. 2: 54, 55 Heheinaim, R.F et LJ. 2: 83.5 Hehenkamp, T. 1: 93,94 3: 280,28 1,282; see also Franz, M. et al.; Wolf€, J. et al. Heidsiek, H. 1: 782; see also Kohl, W et al. Heilmann, P. et al. 2: 599, 600 Heim, A. see Lottner, V et al. Heimann, P.A. see Eizenberg, M. et crl. Hem, R.A. 2: 367 Heine, C. et al. 1: 319 Heine, D. et al. 2: 637 Heme, V. 1: 319; see also Finnis, M.W and Heine, V., Haydock, R. et al.; Heine, C. et al. Heinemann, F see von Schnering, H.G. et al. Heiney, P.A. see Bancel, P.A. et al.; Lubensky, T.C. et al. Heiney, P.A. et al. I: 461, 482 Heinz, D.L. and Yealoz, R. 3 154 Heinz, K. see Blum, V. et al.; Hammer, L. et al. Heinz, K. and Hammer, L. 3: 224 Heinzig, M. see Shen, Z. et al. Heitrnan, P.W. see Chalterjee, A. et al. Helbiiig, W. et al. 1: 614 Heldt, D.T. see Lynch, R.J. et al. Heldt, L.A. 2 27 Helgeson, W.D. see Loubriel, G.M. et al. Helgesson, G. see Sjoberg, J. et al. Helle, A S . et al. 1: 645; 3: 647, 648 Heller, A. 2: 646 Hellner, E. 1: 317, 325, 341, 386, 387, 388; 3 6, 11; see also Donnay, J.D.H. et NE.; Fischer, W. et al.; Lima-de-Fana, J. er al. Hellner, E. and Pearson, W.B. 3: 798
Hellner, E. and Sowa, H. 3: 798 Hellner, E. et al. 1: 317, 332, 336, 341; 3: 798 Hellstern, E. see Fecht, H.J. et al. Hellstern, E. et al. 3: 61, 759 Hellwege, K.-H. and Hellwege, A.M. 3: 797 Helm, D. see Lutjering, G. et al. Helmersson, U. see Petrov, 1. et al. Helniholdt, R.B. 2: 314 Hemachalam, K. 2 360 Hemker, K. et al. 3: 297 Hemker, K.J. 1: 522, 527, 529, 545; 2: 42, 271; see also Balk, T.J. et al., Baluc, N. et al., Kumar, M. and Heinker, K.J.; Viguier, B. et al. Wemker, K.3. and Mills, M.J. 3: 443, 463 Hemker, K.J. et al. 1: 534, 546, 917; 2: 13, 3: 368, 371 Hemley, R.J. 2: 481 Hernley, R.J. and Mao, H.K. 3: 154 Heinpelmann, R. see Richter, D, et al.; Skripov, A.V. et al. Hempelmaim, R. et al. 3: 253 HeiiaK, G. see Mabru, C. et al.; Tonneau, A. et al. Henager, C.H. see Bruemmer, S.M. et al. Henager, C.H. et al. 3: 44 Henager, H. see Bruemmer, S.M. er al. Henderson, B. 1: 105 Henderson, P.J. see Quested, P.N. et al. Wendry, A. see Smith, S.D. et al. Henig, E.-T. 2: 54; ,see also Stadelmaier, H.H. et al. Henig, E.-T. et al. 1: 116, 117 Henisch, H.K. see Suri, S.K. et al. ~ e n ~ eD.l ,see Hauck, J. et al. Henley, C.L. 1: 463, 473, 477, 482, 483, 484; 2: 180; see also Leung, P W. et al., Widom, M. et al. Henlon, J.E. see Wasilewski, R.J. et al. Henning, R.W and Corbett, J.D. 3: 119, 120 Henning, T.A. see Ziinm, C.B. et al. Hennion, B. see Quilichini, M. et al. Hennion, M. see Mirebeau, 1. et al. Henry, J. see Dunlop, A. et al. Henry, M.F. see Gigliotti, M.F.X et al. Henry, M.F. et al. 1: 867 Hensel, F. see Rnecht, J. et al. Hensel, J.C. 3: 235 Hensel, J.C. et al. 3: 235, 787 Henshall, G.A. et al. 3: 552, 554 ensh hall, J.L. see Li, W.B. et al. Henson, H.M. see George, E.P. e f al. Henson, T.J. see Liu, C.T. et al. Henzel, F. see Schmutzler, R.W et al. Her, Y.C. see Wang, P.C. et al. Herb, C.K. 2 624; see also Garceau, W.J. ei al.
Herbst, H. see Fitzer, E. et al. Herbst, J.F. 2: 314; 3: 98, 165; see also Croat, J.J. et al. Herbst, J.F. see Fuerst, C.D. et d. Herbst, J.F. el al. 3: 529 Herd, S. 1: 696; see also Tu, K.N. et al. Heredia, F.E. 1: 540, 547, 915; 2: 25, 26 Heredia, F.E. et al. 2: 162, 296 Heredy, L.A. see Yao, N.P. et al. Herget, G. et al. 1: 164 Herlach, D. see Schaefer, H.-E et al.
A u t h ~ rIndex Herlach, D.M. 3: 683 Herlach, D.M. see Holland-Moritz, D.M. et al. Hennan, F 1: 23 I, 243; see also March, M.H. et al., Maurer, T. et al.; Oppeneer, P.M. et al. Herman, H. see Gilbert, J. et al.; Jardine, A.P et at., Koczak, M.J. et al., Tawari, R. et al. Hermann, A.M. see Mooney, G.D. et al. Hermann, C. 1: 309, 310, 313, 322 Hermann, G. et al. 1: 955 Hernando, A. see Conde, F. et al.; Yavari, A. et al. Herning, P.E. 3: 28 Herold, U. I: 746 Herpin, A. 1: 445 Herrasti, P. see Fatas, E. et al. Herrera, R. see Perez-Campos R. et al. Herres, N. see Richter, H.J. et al. Herring, C. 1: 918 Herring, R.A. 2: 160 Herron, S.J. see Saboungi, M.L. et al. Hershenov, B. see Wen, C.P. et al. Herte, M,S, see Shoemaker, J.R. et al. Hertel, J.A. see Rowe, J.R. et al. Hertz, 5. see Gachon, J.C. et al., Jorda, J.L. et al.; Notin, M. et al. Hertz, M. see Cunat, C. et al. Hertzberg, G. 1: 84 Hertzberg, R.W. see Hoover, W.R. and Hertzberg, R.W.; Quinn, R.T. et al. Hertzman, S. 1: 98, 99 Hertzriken, S.D. et al. 1: 765, 766, 767 Hervieu, M. see Provost, J. et al. Herzer, G. 1: 749; see also Cunat, Ch. et al. Herzig, C. I: 93, 575; see also Divinski. S.V. et al.; Koppers, M. et al., Kroll, S. et al.: ~ u r s c h ~ m R., et al. Herzig, C. et al. 3: 622, 773 Hcslop, J. 2: 10 Wess, D.W. see Fisk, Z. et al. Hessoii, J.C. et al. 1: 674 Heuer, A.H. see Maloy, S. et al. Heumann, T.H. 1: 103, 765 Heusler, F. 1: 15, 935; 2 303, 309 Hewat, E.A. et al. 1: 172 Hewett, C.A. see Xia, W. et aE. Hey, M.H. 1: 632 Heycock, C.T. 1: 5, 7 Heyd, A.R. see Park, D.G. et al. Heymer, G. 1: 70 Heyne, C.J. et al. 2: 373 Hibbard, W.R. Jr. 2: 259 Hickernell, F.S. 1: 875 Hickey, B.J. see Howson, M A . et al. Hickey, B.J. et al. 3: 691 Hicter, P. see Colinet, C. et al., Pasturel, A. et al. Hida, G.T and Lin, I.J. 3: 725, 734, 743 Hidouci, A. and Pelletier, J.M. 3: 574 Hiebl, K. see Weitzer, F. et al. Hiegl, F. and Sizmann, R. 3: 282 Hietaneii, J.R. 1: 171n; see also Wolf€, G.A. et al. Higashi, K. et al. 1: 749 Higashitanaka, N.see Nekano, T. et al. Higashiyama, N. see Inioto, T. et al. Higginson, A. 3: 62 Highmore, R.J. 1: 750
Highmore, R.J. et al. 1: 698, 699 Higuchi, H. see Tetsui, T. et al. Higuclii, T. see Murata, Y. et al. Hildemaii. G.J. see Griffith, W.M. et al. Hilgedieck, R. 1: 575 Hill, A.D. et al. 3: 66 Hill, D.D. see Rowe, J.R. et al. Hill, P.J. and Wolf€, I.M. 3: 73 Hill, P.J. et at. 3: 58, 59, 60, 62, 72, 73 Hillairet, J. 1: 782, 791; see also Balanzat, E. et al.; Beretz, D. et al., Doyania, M. et al. Hillenbrand, H.-G. et al. 1: 748 Hiller, M.W et al. 2 403 Hillert, M. sec Fridberg, J. et al. Hilliard, J. 1: 781 Hillman, H. 2: 361 Hilmer, 0. 1: 110 Hilpert, K. et al. 1: 69 Hilsch, R. 1: 734, 792 Hilscher, G. 2: 481; see also Hauser, R. et al. Hilsum. C. 3: 806 Hilton. A.R. 2 417 Hilziiiger, H.R. 1: 745; see also Cunat, Ch. et al. Himrnel, L. see Chang, Y.A. et al. Hindarn, H. 1: 912 Hindam, H.M. 1: 982, 990 Hines, A.L. see Okabe, T, et al.; Shires, P.J. et al.; Timmons, C.F et al. Hino, I. see Kobayashi, K. et al. Hinze, J.W see Choudhury, N.A. et al. Hippert, F see Bellissent, R. et al. Hippsley, C A . 1: 927; 2: 33; see also Strangwood, M. et al. Hippsley, CA. et al. 1: 927, 983, 986, 989 Hirabayashi, M. 1: 295, 303; 2: 564; sec also 1wasaki.H. et al., Sasaki, G. et al.; Takasugi, T. et al. Hirabayashi, M. et al. 1: 303 Hiraga, H. see Sagawa, M. et al. Hiraga, K. 1: 295, 303, 462, 483; see also Chen, L,C. et al.; Hirabayashi, M. et al.; Sagawa, M. et al.; Sasaki, G. et al. Hiraga, K. et al. 1: 453,467,477,478,483 Hirano, K. 1: 836, 838 Hirano, T. 2: 13, 39; see also Imai, M and Hirano T; Imai, M. et al.; Ito, K . et al.; Kimura, K. et al.; Nakamura. M. et al., Takeya, H. et al., Takeyarna, M. et aZ., Umakoshi, Y. et al. Hirano, T. and Kaise, M. 3: 235 Hirano, T. et al. 1: 643; 2: 39, 223 Hiraoka, T. see Takabatake, T. et al. Hirata, K. see Takeya, H. et al. Hiro, N. see Nasako, K. et al. Hirosawa, S. see Sagawa, RI. et al. Hirose, I(.see Mochizuki, Y et d. Hirose, R. see Zhao, Z.P et al.; Zhukovsky, A.Y et al. Hirose, Y see Surniyama, K. et al. Hirota, H. 2 307 Hirsch, P. 3: 361 Hirsch, P et al. 1: 463; 3: 384 Hirsch, P.B. 1: 498, 512, 513, 544, 547, 548, 550, 915; 2: 1 1 , 13, 27, 137, 265, 268; see also Chou, C.T. et al., Ezz, S.S. et al., Mitchell, T.E. et al.
903 Hirsch, P.B. and Sun, Y.Q. Hirschler, W. and Rocker, W. 3: 527 Hirth, J.P. 1: 524, 880, 881; see also Bruemmer, S.M. et al., Mitchell, T.E. and Hirth, J.P.; ~ i t c h eT.E. ~ ~ ,et al.; Sondhi, S, el al. Hirth, J.P. and Lothe, J. 3: 392, 407, 442, 450, 463 Hisatsune, K. see Ohta, M. et al.; Forter, W et al.; Porter, W.D. ef al., Udoh, K.-I. et al. Hishinuma, A. Hishinuma, A. et al. 3: 423 K. et al., Mimura, T. et al. Hlava, P.F. see Frear, D.R. et al.
Ho, C.T. et al. 2: 43; 3: 604, 609 Ho, C.Y. 1: 1019; see also Touloukian, Y.S. et al. MO,C.Y. and Li, H.H. 3: 813 Ho, F ,we Chetmg, N.PI crl. WO, J.C. see Collings, E.W. et al. Ho, J.C. ef al. 1: 1020 Ho, K. 1: 565; 2: 200 Ho, K. and Dodd, R.A. 3: 282 Ha. K.M see Kong, X.5. et al. Ho, P.S. 2: 606, 654; see also Gupta, D.R. et al.; Howard, J.K. et al.; Koster, U. et al. Ho, P.S. and Ruoff, A.L. 3: 241 Ho, T.L. 1: 471 Ho, T.L. et al. 1: 457 Hoagland, R.G. see Mitchell, T.E. et al.; Sondhi, S. et al. Hoaglund, R.G. see Baskes, M.I. et al. Hobbs, L.W 1: 983, 1007; see also Pint, B.A. et al. Hobgood, H. et al. 2: 327 Hobstetter. J.N. I: 758 Hochheimer, H.D. see Werner, A. et al. Hochman, H. see Poyet, P et al. Hochman, R.F. see Grenga, H.E. et al.; Kuczynski, G.C. et al., Okabe, T. et al. Hocking, L A . et al. 2: 63 Hodges, C.H. 2 610 Hodson, P.D. 1: 745 Hoehn, J. see Motowidlo, L.R. et al. Hoehn, J.G. see Wal~er,M.S. et al. Hoekstra, H.J.W.M. et al. 3: 149 Hoelke, W.C. see Swanson, M.L. et al. Hoenig, H.E. et al. 2: 388 Hoenig, M.O. 2: 363; see also Steeves, M.M. et al. Hofass, H. see Ronning, C. et al. Hoffman, N.J. et al. 2: 596 HoRman, W. see Gauthier, Y et al. Hoffmann, R.D. see Kuncser, V, et al.; Pottgen, R. Hoffrnann, W. see Wuttig, M. et ul. Hofler, G.E. see Baillargeon, J.N. et al. Hofman, C . 2: 572
904 Hofman, G.L. see Allen, C.W et al.; Birtcher, R.C. et al.; Rest, J. et al. W-ofinanii, A. see Guo, H.Q. et al. Hofmann, D. see Emst, F et al. HoErnann, G.L. see Birtcher, R.C. et al. Hofmann, M. see Campbell, S. J. et al. Hofmann, W 2 594 Hofmeister, W.H. see Anderson, C.D. et al.
Hogg, T.W. 1: 15 Hogmark, S. see Hedenqvist, P. et al. Hohenberg, P 1: 58, 78, 129, 196 Hohenberg, P. and Kohn, W. 3: 234 Hohls, J. see Hill, P.J. et al. Hohls, J. et al. 3: 62 Wohnke, D. and ParthC, E. 3: 89 Hokamoto, K. see Meyers, M.A. et al. Hoke, D.A. et al. 3: 736 Hokkeling, P. 2: 486; see also Havmga, E.E. et al.; Koch, A.J.J. et al. Holbrook, J.H. et al. 1: 974 Holdway, P. and Staton-Bevan, A.E. 3: 444 Holland-Moritz~D.M. et al. 3: 683 Holleck, H. 2: 410 Hollingsworth, E.H. ef al. 2: 177 Holly, K. see Danielewski, M. er al. Holly, K and Danielewski, M. 3: 563, 568 Holman, G.T. see Anderson, M.U. et al.; Thadhani, N.N. et al. Holt, B. see Dunmead, S.D. et al. Holt, D.L. see Bever, M.B. et al.; Nabarro, F.R.N. et al. Holtzberg, F see Guntherodt, G, et al.; Mook, H A . et al. Holtzberg, F. et al. 3: 531 Holz, A. 1: 960 Holzapfel, W.B. see Degtyareva, V.F. et al, Tissen, V.G. et a!. Holzer, J.C. 1: 454,461,465,479,480; see also Ali, N. et al.; Eckert, J. et al., Jeoiig, E.K. et al.; Levine, L.E. et al.; Sabes, P.N. et al. Holzer, J.C. et al. 1: 465, 477 Woman, C.G. 1: 760 Momma, M. see Kaneko, H. et ak. Honda, K. see Iwata, M. et al. Hondros, E. see Chou, C.T. et al. Iloness, C.W. see Hutchinson, W.B. et al. Honey, F.J. et al. 2: 492 Hon~ycombe,R.W.K. 1: 915; 2: 278,279; 3: 361 Hong, C. see Aning, A.O. et al. Hong, J. see Cben, M. et al. Hong, K.C. see Fetcenko, M.A. et al. Hong, M. see Hsieh, Y.F. et al.; Kwo, J. et al.; Majkrzak, C.F. et al.; Noh, D.Y. et al. Hong, M, et at, 3: 783, 784 Hong, M.-H. see ~amabe-Mitarai,Y. et al. Hong, M.-H. el al. 3: 341 Hong, Q.Z. 2: 621, 622; see also Hung, L.-S. et al.; Li, 5. et al.
Hong, Q.Z. et al. 2 622 Hong, S.Q. see Li, J. et al. Hong, S.I. see Gray 111, G.T. et al. Hong, T. 1: 70, 71, 72, 137, 507, 508; 2: 58, 157, 278; see also Asta, M. et al., Freeman, A.J. et al. Hong, T. et al. 1: 65, 66, 70 Hoiigu, H. see Sato, K. et al. Honjo, G. see Osakabe, N. et al. Honjo, N. see Yamabe, Y. et al. Honle, W. see von Schnering, H.G. and Hnle, W. Honma, T. see Murakami, A. et al. Honnell, R.E. see Carter, D.H. et al.; Gibbs, W.S. et al., Petsovic, J.J. et al. Hono, K. see Takasugi, T. et ul. Hooper, R.M. see Li, W.B. et al. Hoover, H.C. 2: 515 Hoover, L.H. 2 515 Hoover, W.R. and Hertzberg, R.W. 3: 415 Hopkins, A.A. 2: 515, 520, 523, 653 Hopkins, B.S. 2 518, 523, 649 Hopkins, M.A. 2 282 Hoppe, C. see Shield, J.E. et al. Hoppie, L.O. see Stekly, Z.J.J. et al. Hoppin, G.S. I11 2: 4 Worache, E.H. see Setton, N, et al. Horan, Y . see Jardine, A.P. et al. Hordubay, T.D. see Singh, S.K. et al. Hori, S. 2: 187 Horie, T. see Harnmetter, W.F. et al.; Nishtno, K.et al. Hone, Y see Simonsen, I.K. et al. Horihata, M, see Bolt, P,J. et al. Horikawa, H. see Golberg, D. et al. Horikoshi, Y. see Zhang, S. rt al. Horikoshi, Y et al. 2: 328 Horinaka, H, see Ya~amoto,T. et al. Horinaka, H. et al. 2: 330, 410 Horita, Z. see Oh-ishi, K. ct al. Horlock, A.J. see Blackford, J.R. et al. Horn, P.M. see Bancel, P.A. et al.; Birgeneau, R.J. and Horn, P.M.; Heiney, P.A. et al. Horn, P.M. et al. I: 482 Hornback, V.C. see Ramanath, G. et al. Hornbecker, N.F. 2 24 Hornbogen, E. 1: 839; 2: 267; see also Hillenbrand, H.-G. et al. Horner, T.J. see Haste, A.S. et al. Horner, I.J. et al. 3: 60 Hiirnstrom, S.E. see Charai, A. et al., Harper, J.M.E. et al. Homstrom, S.E. et al. 2: 625, 626, 629 Horstemeyer, M.F. see Gall, K. et al. Horstmann, R.E. see Ames, 1. et a/. Horton, J.A. 1: 498, 504, 509, 51 1, 590, 593, 593; 2: 20, 22, 23, 30; see also Baker, I. et al.; Cahn, R.W et al.; David, S.A. et al.; George, E.P. et al.; Liu, C.T. et al., McKarney, C.G. et al.; Santella, M.L. et al.; Schneibel, J.H. et al.; Schulson, E.M. et al.; VeyssiBre, P et al., Yoo, M.H. et al. Horton, J.A. et al. 1: 982, 983, 986 Horton, J.B. see Zoccola, J.C. et al. Horton, L.H. see Liu, C.T. et al. Horvath, J. et al. 1: 735 wloshmo, K. et al. 1: 564n, 574, 778
Hoshmo, T. see Asada, T. et al. Hosino, H. see Schmutzler, R.W. et al. Hosoda, H. see Matsuo, T. et al. Mishima, Y. et al. Hosoda, H. et al. 3: 59, 64 wlosomi, M. see Fujiwara, T. et at,; Nekano, T. ot al.; Shiniokawa, T. et al.
Hosoya, S. see Takahashi, T, et al. Hos~ain,Z. see ~ a g a r a j a nR. , et al. Hou, C.C. 1: 66, 70 Hou, D.H. see Shyue, J. et al. Hou, G.Y. see Huang~F. et al. HOU,M.-D. see Klaumunzer, S. et al. Hou, P.Y. see Renusch, D. et al. Hough, D. see Ledig, L. er at, Housseau, W. 1: 577 Houston, J.E. see Kiely, J.D. et al. Hove, M.V. see Baird, R. et al. Hovestreydt, E. 1: 243; see also Jorda, J.L. et al. Howard, J.K. et al. 2: 606, 608, 654 Howe, J.M. 2: 75; 3: 211, 222, 225; see Singh. S.R. and Howe, J,M, Howe, L.M. 1: 791, 806, 807, 809, 810, 811, 817; 2: 133; see also Motta, A.T. et al., Schulson, E.M. et al. Howe, L.M. et al. 1: 528,811,812,813; 2: 22, 133, 139 Howell, C.R. see Sikka, U.K. et al. Howell, M. see Tortorelli, P.F. et al. Howells, W.S. see Barnes, A.C. e r al.; Caciuffo, R. et al.; Price, D.L. et al.; Saboungi, N.L. et al.; Verkerk, P. et al.; Xu, R. et al. Howes, M.J. 2: 417 Howie, A. see Basinski, M.B. et al.; Hirsch, P. et al.; Wirsch, P.B. et al. Howson, M.A. see Hickey, B,J. et al. Howson, M.A. et al. 3: 691 Hoyt, J.J. see Amador, C. et al. Hsia, S.L. et al. 2: 620 Hsieh, H. 1: 817; see also Limoge, Y . et al. Hsieh, K.C. see Baillarge~n,J.N. et al.; Lin, Y.-C. et al. Hsieh, T.C. see Majkrzak, C.F. et al. Hsiung, L.M. and Nieh, T.G. 3: 409,413, 419, 421 Hsiung, L.M. and Stolog, N.S4 3: 326 Hsiung, L.M. et al. 3: 422 Hsu, F.S.L. see Kunzler, J.E. et al. Hsu, S,-E. see Lee, T.-S. et d. Hsu, S.E. et al. 2: 41 Hsu, S.S. see tlu, Z.W. et al. Hsu, Y.Y. et al. 3: 103 Hu, B.P. see Wang, Y.Z. et al.; Yang, C.P et al., Yang, F.M. et al, Hu, B.P et al. 3: 168, 169, 174, 176 Hu, C.D. 1: 130 Hu, C.K. 2: 624 Hu, C.Z. see Diag, D.H. et al.; Yaiig, W.G. et al., Yao, D.Z. et al. Hu, C.Z. et al. 3: 387 Hu, D. see Godfrey, A. et al. Hu, C . see Chen, S. et a/. Hu, G.X. see Chen, X.F. et al. Mu, C . X . et al. 2; 159, 162, 165
Hu, J. et al. 3: 94, 669 Hu, K. see Kaviani, K. ef al. Hu, W.K. 3: 108 Hu, Y.Z. et al. 3: 667 Hu, Z. see LOLLT. et al.; Luo, H. et al.; Yelon, W.B. and Hu, Z. Hu, Z. et ul. 3: 99 Hu, Z.M. see Brooks, J.W. et al. Hu, Z.O. see Liu, Z.G. et al. Hu, Z.W. et al. 1: 454 Hua, M.-J. 1: 590, 591, 901 Huang, B. see Qu, X. et al., Shingu, P.H. et al. Huang, C.Y. see Patten, E.A. et al. Huang, D.P. and Corbett, J.D. 3: 119, 120 Huang, D.P. e f al. 3: 120 Huang, F. et al. 3: 94 Huang, F.S. see Kao, C.H. et al. Huang, H,C.W et al. 2: 606 Huang, J.C. 2: 281, 282 Huang, P. see Qu, X. et al. Huang, (2. see Yang, Y et al. Huang, S. 1: 540 Huang, S.C. 1: 638, 640, 987, 993; 2: 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87; 3 503, 618; see also Hall, E.L. and Huang, S.C.; Shih, D.S. et al.; Taub, A.I. et al. Huang, S.C. and Chesnutt, J.C. 3 328, 351,442,477, 478, 594, 618, 648 Huang, S.C. and Hall, E.L. 3: 424, 598 Huang, S.C. et al. 1: 56411, 897, 8999;2: 35, 38, 39, 83, 156, 159 Huang, S.C. see Chang, K.M. et al. Huang, S.L. see Kao, C.H. et al. Huang, T. 1: 184 Huang, T.S. 2: 5 et al. et al. Hubbard, E L . se r, R.L. et al. Huber, F. I: 97 Huber, K.P. 1: 84 Huber, P. see Malik, M. e~ al. Hubler, G.K. er al. 1: 733 Huckstein, M.et al. 1: 671 Eiudak, 0. see Macko, D. et al. Hudd, R.C. 1: 477; 2: 177 Hudson, P.G. see Yeh, L.S.R. et al. Huebsch, J.J. see Akmc, M. et ut. Huerta-Ruelas, J. et al. 3: 667 Huffman, M. see Sargent, K. and HufTman M. Hiifner, S. 3 140, 576
Frangois, A. et al.; Lasalmon~e,A. et al.; Wardle, S . et al. Hug, G. et al. 1: 506, 527, 534, 545,551; 2: 77, 153; 3: 423,445,446,447 Hug, G. and VeyssiBre, P. 3: 422 Huggins, R.A. 1: 95; 2: 502, 510, 511; see also Wen, C.J. el ul. ~ u g ~ b a nT.~ 3: s , 122 Hughes, O.H. et al. 1: 356 Hugbes, T. et al. 2: 54 Hugo, F see Betz, U. et al. Hugo, F e8 al. 3: 542
Huguenin, D. see Doyama, M. et al. Huizer, E. 1: 743 Hull, A.W 1: 10 Hull, D. 1: 861; see also Warlimont-M~ier,B. et al. Hull, D.R. see Doychak, J. et al. Hull, G.W. see Buclier, E. et d. Hull, S. see Keen, D.A. and Hull, S. ~ u ~ ~ L.D. e t t ,see Somie§ki, B. et al. Hulliger, F 1: 268, 277, 280, 289, 382, 385, 390, 391, 396, 419, 434; see also Brandle, H. et al., Hafner, J. et al.; Siegrist, T. et al.; Villars, P. et al. Eiulsmann, S. see Peters, M. et al. Hultgren, R . 1: 920; 2: 564, 565 Hultgren, R. et al. 1: 68, 101, 102, 104, 107, 109, 110, 111, 112, 113, 114, 117, 119, 121, 662, 664, 666; 2: 607, 639, 643; 3 799 Hultman, L. see Adibi, F. et al.; Greene, J.E. et al.; Karlsson, L. et cd.; Petrov, I. et al. H ~ l t m a nL. , et al. 3: 665, 666 Hufts, W.L. see Fowler, C.M. et al. Humble, P. see Head, A.K. et al. Hume-Rothery, W 1: 13, 63, 102, 245, 319, 712; 2: 136; 3: 28, 114, 255 Hume-Rothery, W. and Raynor, G.V. 3: 811, 812 Hume-Rothery, W et al. 1: 63, 398, 566 Humphrey, D.L. 1: 991, 992, 993, 994; 2: 167, 168; see also Brady, M.P. et al.; Doychak, J. et al.; Siwalek, J.L. and Humphrey, D.L.; Sniialek, J.L. et al. Humphreys, C.J. see Fairbank, G.B. et al., Inkson, B.J. and Humphreys, C.J.; Wiezorek, J.M.K. and Humphreys, C.J., Wiezorek, J.M.K. et rrl. Humphreys, F.J. 2: 265, 268 Humphreys, F.J. and Hatherly, M. 3: 623 Humphreys, G. 2 521, 522 Humpston, G. 2: 521, 522 Hundley, M.F. see Canfield, P.C. et al.; Lacerda, A. e f al., Movshovich, R. el al. Hundley, M.F. et al. 1: 217, 218 Hunecke, J. see Wever, H. et al. Hiinecke, J. et al. 1: 565 Hung, H.H. see Rivers, 5.13. et al. Hung, L.S. 2: 610,614,616,619,621,630; sce also Mayer, J.W. et al., Nastasi, M. et al.; Ottdvrani, G. et al.; Pai, C.S. et al.; Saris, F.W. et al.; Wang, S.Q. et al.; Zheng, L.R. et al. Hung, L.S. e f al. 1: 706; 2: 610, 617, 618, 619, 625, 626, 627, 628, 629 Hung, P.M. see Harte, A.S. et al. Eiunsperger, R.G. see Mentzer, M.A. et al. Hunt, J.D. 1: 640 Hunt, N.E.J. et al. 2: 427 Hunt, P.J. see StaEord, K.N. et al. Hunter, W.R. et al. 2: 410 Huntington, H.B. 2: 134 Huntington, H.B. et al. 1: 575, 761, 765 Huntley, D. 1: 612 Huntley, M. see Pope, D.P. et al. Huntz, A.M. 1: 987, 990 Huntz, A.M. see Nicolas-Chaubet, D. et al. Hurley, D.P.F. see Coey, J.M.D. et al. Hurley, G.F. 2: 295
Hurly, J. and Wedepohl, P.T. 3: 75 Hussey, C.L. see Tierney, B.J. et al. Hussien, S.A. et al. 2: 423 Huston, E.L. see Goodell, P.D. et al. Hutchings, M.T. 3: 171 Hutchmgs, R. 1: 983, 987, 988, 989, 990, 095, 996; 2: 57 Hutchinson, J.W. see Bao, G. et al. ~utchinson,W.B. et al. 1: 787, 863 Hutten, A. see Berkowitz, A.E. et al. Huzuhara, T. see Tornokiyo, A. et al. Hwang, A.M.H. see Bi, Y.J. et al. Hwang, C.M. 1: 716 Hwang, R.Q. see Kiely, J.D. et al. Hwang, S. see Varma, A. et al. Hwang, S.J. see Chung, H.H. et al. Hwang, S.M. see Kang, B. et al.; Kim, J.K. et al.; Lee, T.K. et al. Hwang, W. see Kang, B. et al. Hwu, Y see Noh, D.Y et al. Hybertsen, M.S. 1: 130, 132 Hyde, B.G. 1: 352 Hyde, B.G. et al. 1: 359 Hyde, J.R. see Clarke, R.S. et al. H y m n , M.E. e f al. 3: 597, 618, 619 Hyzak, J.M. see Bain, K.R. et al. Iandelli, A. and Palenzona, A. 3: 26, 87 Iatsenko, T.K. see Hertzriken, S.D. et al.; Larikov, L.N. et al. Ibach, H. see Giessen, B.C. et al.; Wuttig, M. et al. Ibarra, M.R. see Morelion, L. et aZ. Ibarra, M.R. et al. 3 94 lbberson, R.M. see Davrd, W.X.F. et al. Ibe, G. see Rahl, W et al. Ibel, K. see Murani, A.P. et al. Ice, G.E. see Reinhard, L. et al.; Sparks, C.J. et al. Echihara, M. see ~ d a ~ a wK. a , et al.; Karnisada, Y. et al.; Nernbach, E. et al., Suzuki, K. et al. khikdwa, K. see Thompson, J.C. et al. Ichikawa, M. 2: 507 Ichikawa, T. 1: 479 Ichmo, K. see Kawakami, Y. et al. Idrestedt, I. see Sjoberg, J. et al. Iga, F see Kasaya, M. et al.; Ogawa, S. et al. &a, F. et al. 1: 217, 218 Igarashi, Masakki see Senba Hiroyuki et al. Iglesias, J.E. and Ste~nfink,H. 3: 531 Iglesias, J.R. see Martinez, G. et al. Ignat, M. see Guenin, G. et al. Ignatiev, A. et al. 1: 620 Iguchi, Y. see Yanagihara, M. et al. Iijuina, V. see Tachikawa, K. et al. Iikubo, T. see Shimizu, T. et al. lizumi,M. 1: 773, 777 Ikarashi, N. see Tanaka, M. et al. Ikawa, M. see Hori, Y. et al. Ikebuchi, M. et al. 3: 411 Ikeda, B. see Matsuno, S. et al. Ikeda, K. 1: 953, 954; see also Umakoshi, Y et al. Ikeda, M. see Oku, S. et al. Ikeda, M. et al. 2: 327, 336 Ikeda, T. see Nonaka, K. et al. Ikegami, T. see Ito, T. et al.; Matsuoka, T. et al.
906 Ikezawa, M. see Suzuki, R.O. et al. lkuma, Y. see Jin, M. et ul. Ikushima, K. et al. 3: 664 lkuta, 1. see Minemura, T. et al. Ilderem, V. 1: 652 llegems, M. see Tsang, T.S. and Ilegems, M. lles, P. see Cheung, N. et al. I k , R. and Durrani, S. 3: 263 h a , M.A. 3: 161 Il’nitskaya, O N . et al. 1: 352 llschncr, B. 1: 916, 917 lmaeda, C . see Mizutani, U, et al. Imai, M. and Hirano, T. 3: 161 lmai. M. et al. 3: 161 lmamnazarov, D,Kh. see Granovskii, A.B. et al. lmamura, U.see Takeuchi, H. et al. Imanaka, N.see Adachi, G.Y et al. lmayev, R. see Salishchev, G. et al. Imayev, R. et al. 3: 630 lmayev, R.M. 1: 654, 657; see also Kaybyshev, O.A. et al.; Salishchev, G.S. et al. Iinayev, R.M. et al. 1: 925; 3: 623, 624, 625, 626, 651 Imayev, V see lrnayev, R. et al.; Salishchev, G, et al. lmayev, V.M. 1: 654, 657; see also lmayev, R.M. et al., Salishchev, G.S. et al. Imer, J.-M. see Grioni, M. et al. lmmarigeon, J.P. 1: 987, 996 Imoto, T. see Nasako, K. ef al. Irnoto, T. et al. 3: 108 Emura, T. see Salca, H. et al. Inaba, M. et al. 3: 671 Inaba, T. see Tokizane, M. et al. lnal, O.T. 1: 924 Inden, G. 1: 38, 849, 850 Inden, G. et al. 1: 40 Indenbaum, V.N. stv Greenberg, B.A. et al. Indenbaurn, V.N. et al. 1: 534 Ingersent, K. 1 : 457 lngrarn, D.C. see Oh, J.E. et al. Ingrani, J.C. see Strutt, P.R. et al. Inkson, B.J. 3: 445, 464 Inkson, B.J. and Clemens, H. 3: 651 Inkson, 13.5. and Humphreys, C.J. 3: 410, 445, 464 lnnoue, A. see Terauchi, M. et al. Xno, H. 1: 35; see also Rimura, K. et al. lnokuchi, €3. see Mori, M. et al., Takahashi, T. et al. Inornata, K. see Tomokiyo, A. et al. Inou, A. see Koch, C.C. et al. lnoue, A. 1: 735; 3: 682, 691; see also Chen, H.S. et al., Higashi, K. et al., Hisaga, I<. et al.; Kawaniura, Y. et al., Kim, S.G. et al.; Mortan, A.R. et al.; Makino, A. et al., Masumoto, Y et d,; Matsubara, E. et al.; Srtzuki, I<. et al.; Tsai, A.P. et al., Zhang, T. et al. Inoue, A. and Gook, J.S. 3: 682 lnoue, A. et al. 1: 454, 467, 692, 735, 736, 743; 3: 397, 682, 691, 701 Inoue, H. see Markiewicz, W.D. et al.
lnoue, H.R. et al. 3 447 lnoue, H.R.P et al. 2: 159, 164 Inoue, K. see Chu, J.P. et al., Kiyoshi, T. et al.; Ohnaka, K. et al.; Tachikawa, K. et al. lnoue, K. et al. 2 371, 374 Inoue, N. see Conn, R.W. et al. Inoue, Y see Kaiiiisada, Y. et al. Inouye, A. et al. 2 39 Inouye, H. 1: 913, 914, 924, 2: 199 lntermann, A. et al. 3: 669 liiui, H. see Ikebuchi, M. et al.: Ishikawa, K. et al., Ito, K. et al.; Kishida, K. et al.; Moriwaki, M. et al.; Oh, M.H. et al.; Pdidar, V et al.; Shimokawa, T. et al.; Siegl, R. et al.; Vitek, V et al.; Yaiiiaguchi, M. and Inul, H., Yamaguchi, M. et al. Inui, H. et al. 1: 505, 528; 2 75,79.81, 87. 159, 162, 165; 3: 275, 407, 408, 414, 422, 444, 446, 455, 459, 773 Toffe, A.F. 2: 453, 461, 469 Iono, L E . see Wolff, I.M. et al. Ipser, H. 1: 113 lreiie, E. see Hu, Y.Z. et al. Irkaev, S.M. see Gavriliuk, A.G. et al. Irving, G.N. et al. 1: 995 Irving, P.E. 2: 217 Irwin, R.B. see Chittipeddi, S . et al. Isawa, T. see Mitao, S. et al. Iseler, G.W. 2: 327 l s ~ b a s h iT. , 2: 335: see also Sano, €3. et al. lshida, A. see Shinohara, K. et al.; Taker, A. and Ishida, A. Ishida, K. 1: 118; see aZso Kainuma, R. et al.; =do, G. et at.; Tanaka, M. et al. lshida, K. et al. 1: 846, 860 lshida, M. see Karnigaki, K. et al. Ishida, Y. see Hashimoto, M. et aE.; Wakayama, S . et al. Ishidaira, T. see Nishi, Y et al. Ishigura, T. see Matsunaga, S. et al. Ishihara, K.N. see Dela Torre, S.D. et al.; Mukhopadhyay, N.K. et al.; Skingu, P.H. et al.; Taguchi, K. et al. lshihara, T. see Ding, J. et al. Ishii, T. see Watako, Y. et al. lshii, Y. 1: 445 Ishikawa, H. see Sakai, T. et al. Ishikawa, H. et al. Iwakura, C.et al. lshikawa, K. see lnui, H. et al. Ishikawa, K. et al. 3 459 Ishikawa, M. see Jorda, J.L. et al. Ishikawa, M. et al. 2: 336 Ishimasa, T. see Mori, M. et al. lshimoto, H. et al. 3: 522 Ishino, Y. see Nagasawa, A. et al. Ishioka, S. see Koiwa, M. et al. Ishitam, A. see Mochizuki, Y. et al. Ishiyama, S. et al. 3: 61 Ishizaki, K. 2: 225; see also Celis, P.B. et al. Ishizawa, Y. see Otani, S. et al. Ishizawa, Y. et al. 1: 954, 955 Isikaw, Y. see Sato, K. et al. ls~arnga~iev, R.K. see Valiev, R.Z. et al. Isnard, 0. see Obbade, S. et al. Isnnrd, 0. et al. 3: 98 Isobe, K. see Miyazaki, T. et al.
Isobe, S. see Maki, K. et al., Nishiyama, Y. et al.; Shimizu, T. et al. Isobe, S . and Noda, T. 3: 593 Isoda, Y see Kaibe, H.T. et al. Isomoto, T. and Stoloff, N. S. 3: 330 Isper, H. see Leubolt, R. et al. Itaini, T. see Kitajima, M. et al. Itaya, Y . see Matsuoka, T. et al. Ito, K. see Enui, H. et al.; Moriwaki, M. et al., Tanaka, Y et al.; Yamaguchi, M. et al. lto, K. et al. 2: 217; 3: 367, 454, 455, 457, 458, 775 lto, M. see Takizawa, H. et al. lto, T. see Ashida, A. et al.; Inoue, A. et al. Ito, T. et al. 2: 339 Ito, W. 3: 665 Itoh, C. see M o t o j i ~S. ~ ,et al. Itoh, G. see Sears, J.W. et al. Itoh, K. see Inoue, K. et a/.; Kiyoshi, T. et al.; Ohara, T. et al.; Tachikawa, R. et al. Itoh, Y . see Imoto, T. et al. Itoi, K. see Suzuki, H. et al. Ivanko, A.A. 2: 252 Ivanov, E. see Mikhailenko, S.D. et al.; xu, Y. et al. lvanov, E. et al. 1: 706 Ivaiiov, G.A. 1: 956 Ivanov, M.A. see Greenberg, B.A. et al. lvanov, 0.3. see Vedernikov, M.V. et a/. Ivanov, V.I. see Kaybyshev, O.A. et al. lvanova, G.V. et al. 3: 94 hey, D.G. 2: 476 Ivtchenko, V.V see Dan’kov, S.Yu, et al. Iwaki, T. see ~ o r i w a ~Y.i , et al. Iwakura, C. 2: 509; see also Ikoma, M. et al.; Sakai, T. et al. Iwakura, C. et al. 2: 509 lwarnura, H. see Kagawa, K. et al., Kagawa, T. et al. Iwanaga, H. see Motojima, S. et al. Iwasa, Y. see Williains, J.E.C. et al.; Zhao, Z.P et al.; Zhukovsky, A.Y et al. Iwasaki, H. see Fujinaga, Y. et al. Iwasaki, H. et al. 1: 118; 3: 161 Iwasaki, K. see Zhu, S.M. et al. Twase, N. see Inaba, hl. et al. Iwata, M. et al. 3: 183 Iwata, S. see Di, G.Q. et al. Iwata, T. see Doyarna, M. et al. Xzerman, S.A. et al. 3: 822, 835 lzurni, 0. 1: 40, 85, 539, 564, 591, 598, 599,603,604,896,897,898,899,907, 908, 914,924, 928, 970,973; 2: 18,27, 28, 29, 30, 38, 39; see also Aoki, K. and Izumi 0.;Hanada, S. et al.; Hasegawa, S . et at., Kawabata, T. et al.; Li~i,Y. et al.; Masahashi, N. et al.; Ogura, T. et al.; Stoloff, N.S. et al., Takasugi, T. et al.; Watanabe, S. et al., Yoo, M.H. et al. Jablonski, D.A. 2: 25 Jabra, A. see Chao, P.C. et al. Jaccard, D. see Takabatake, T. et al. Jaccarino, V. et al. 1: 217 Jacinto, M.A. see Fritzeineier, L.G. et al. Jack, D.H. 2: 278, 279
A u t ~ Index o~ Jackson, A.G. et al. 3: 812 Jackson, K.A. see Perdew, J.P et al. Jackson, M.R. 3: 65, 488, 490, 492; see also Bewlay, B.P. et al.; Bewlay, P.A. et al.; Gigliotti, M.F.X et al., Taub, A.I. et al. Jackson, M.R and Bewlay, B.P. 3: 550, 555, 557, 558 Jackson, M.R. et al. 3: 490,491,544,548, 658 Jackson, R.O. et al. 1: 794 Jackson, T.J. see Wangts, C.H. et al. Jackson, T.N. see Basile, D.P et al. Jacob, G. see Martin, G.M. et al. Jacob, I, et al. 1: 613 Jacob, M. see Datta, P.K. et al. Jacobi, H. 1: 566, 568; see also Eibner, J.E. et al. Jacobi, H. and Stahl, R. 3: 232, 237 Jacobs, K.T. 2: 502 Jacobs, M.H. see Dowson, A.L. et al. Jacobs, M.H. et al. 2: 178 Jacobs, R.B. 1: 184 Jacobs, T.H. see Grosssnger, R, et al. Jacobsen, K.W. et al. 1: 79 Jacobson, D.C. see Tung, R.T. et al. Jacobson, D.M. 2: 521, 522 Jacobson, L.A. I: 531, 916; 2: 147; see also Hanrahaii R.J. Jr.et al.; He, Y. et al.; Shechtman, D. aiid Jacobson, L.A. Jacobson, N.S. 1: 998 Jacobson, R.E. see Henager, C.H. et al. Jacobson, S. see Hedenqvist, P. et al. Jacques, R. see Vijh, A.K. et al. JaRee, R.I. 2: 317 Jagodzinski 1: 279 Jam, A. see Pint, B.A. et al. Jain, H. see Alamgir, F.M. et al.; Jin, 0. et al. Jain, M. 1: 647 Jain, M. and Christma, T. 3: 739, 740 Jake, C.E. et al. 3: 332 Jakob, B. see Blau, B. et al. Jaksic, M. 2: 482 Jakubovics, J.P. see Sun, H. et al. Jakubowicz, J. and Jurczyk, M. 3: 98 Jalota, S.K. 2: 521 Jambor, J.L. 1: 627, 629, 630, 631 James, A.W. 2: 83 Janies, A.W. and Bowen, P. 3: 334 JiLmes, M.R. see Cox, B.N. et al. James, W.J. 1: 440, 447, 448; see also Hardman-Rhyne, K. et al.; Luo, H. et al. Jamieson, J.C. 1: 184 Jamison, K. et al. 1: 618 Janak, J.E. see Moruzzi, V.L. et al. Janak, J.F 1: 138 Janak, J.F see Moruzzi, V.L. et al. Janaway, G.A. see Diaz de la Rubra, T. et al. JHnecke, E. 1: 10 Jang, H. 1: 614; see also Shih, D.S. et al. Jang, H. et al. 2: 68 Jang, J. 1: 788 Jang, J.M. and Jeng, S.M. 3: 582 Jang, J.S.C. and Koch, C.C. 3: 759 Jang, T.S. 2: 315, 316 Janghorban, K. 1: 809 Janner, A. 1: 453
Janng, G. and Bach, H. 3: 29 Janossy, A. see Stephens, P W. et al. Janot, C. see Dong, C. et al. Janowski, G.M. 1: 645 Janschek, P. see Appel, F et al. Jansen, H.J.F see Min, B.I. et al. Jansen, H.J.F. and Freeman, A.J. 3: 234 Jansen, J. see Zandbergen, H.W et al. Jansen, W.B. see Hafner, J. et al. Janssen, T. 1: 453; 3 387 Janssen, Y. see Koyama, K. et al. Jaiisson, B. see Andersson, J.-0. et al. Jaouen, C. I: 811, 817 Jaouen, C. et al. 1: 810, 816 Jarczyk, G. see Blum, M. et al. Jardine, A.P et al. 1: 642 JariC, M.V. 1: 481, 483; 3: 389 JariC, M.V. et al. 1: 491 Jarrnoljuk, J.P. 1: 41 I; see also Grin, J.N. et 01. Jarosevich, E. see Clarke, R.S. et al. Jarvinen, A. see Fugleberg, S. et al. Jaskierowicz, G. see Barbu, A. et al., Dunlop, A. et al. Jastrab, A. see Zirnni, C.B. et al. Jaswal, S.S. 1: 154 Jaswari, M.A. see Foreman, A.J.E. et al. Jaswon, M.A. and Dove, D.B. 3: 417 Jaszczak, J.A. see Ho, T.L. et al. Jaszczak, J.A. et al. 1: 191, 457 Jaulmes, S. see GhCmard, G. et al. Jaumot, F.E. 1: 498 Jayakumar, O.D. see Kulshreshtha, S.K. et al. Jayarman, A. 3: 154 Jayashankar, J.S. et al. 3: 657 Jayne, D.T. see Aikin, B.J.M. et aE. Je, J.H. see Noh, D.Y. et al. Jean, A. see Boily, S. et al. Jeanjaquet, S. see Menezes, S. et al. Jedlinski, J. 1: 983, 990; 2: 57; see also Mrowec, S, et al. Jedlinski, J. et ul. I: 987, 990 Jeitschko, W. 2 314; see also Evers, C.B.H. et al., Kaiser, J.W. and Jeitschko, W.; Prots, Y.M. et al. Jellinghaus, W 2: 308, 309 Jeniran, W.A. see David, S.A. et al. Jena, A. et al. 1: 787 Jena, P see Press, M.R. et al. Jeiia, P. et al. 1: 681 Jeng, S.M. 2: 290; see ulso Jang, J.M. and Jeng, S.M. Jeng, Y.-L. et al. 3: 320 Jenkins, J.F see Sergev, S.S. et al. Jenkins, M.L. 1: 806, 807; see also Black, T.J. et al. Jenkins, M.L. et al. 1: 806 Jenkinson, D. see Datta, P.K. et al. Jenkinson, H.A. see Mentzer, M.A. et al. Jenks, C.J. see Shen, Z. et al. Jennane, A. et al. 1: 577 Jenner, A.G.I. see Galloway, N. et al., Parvinmehr, A. et al. Jenner, A.G.I. et al. 2: 399 Jennings, G. see Campuzano, J.C. et al. Jensen, J. 1: 441; see also Sullivan, C.P. et al. Jensen, W.B. 1: 367, 427, 428, 433; 3: 3 Jeon, H. see Ding, J. et al. Jeon, H. et al. 2: 326, 428, 430
90’7 Jeong, E.K. et al. 1: 479 Jepsen, D. et al. 1: 611 Jepsen, 0. 1: 58; see also Bagno, P et al. Jesse, A. et al. 2: 648 Jewett, P.R. 1: 973 Jewett, T. see Schneider, D. et al. Jewctt, T.J. see Percpezko, J.H. et al. Jex, H. see Herget, G. et al. Jha, I.S. et al. 1: 671 Jha, S.C. et al. 1: 655, 926; 2: 61, 128 Jia, C.L. et al. 3: 103 Jian, S.K. see Martin, P.L. et al. Jiang, P.Z. see Olson, C.G. et al. Jiang, X.L. see Hu, Z.W et al. Jiang, Z.L. see Yang, C.P. et al. Jiles, D.C. 2: 318, 390, 400 Jirnersez, J.A. et al. 3: 656 Jin, M. et al. 3: 225 Jin, 0. et al. 3 688, 689, 690, 696 Jin, P see Yang, H.S. et al. Jin, R. et al. 2: 412 Jin, S.M. see Zhou, B. et al. Jin, Z . and Bieler, T.R. 3: 409, 419 Jin, Z. and Gray, G.T. 111 3: 364, 365 Jin, Z. et al. 3: 368, 413 Jing, K.L. see Wan, X.J. et al. Jink, J. et al. 3: 687 Johan, Z. et al. 1: 628 Johansson, B. see Eriksson, 0. et al. Johansson, C.H. 2: 562, 564 Johansson, G. see Seigbahn, K. et al. Johansson, M.P see Karlsson, L. et al. John, R. see Larsen, J.M. et al. Johnson, A.D. 2: 432 Johnson, D.D. 1: 197; see also Althoff, J.D. et al., Asta, M.D. and Johnson, D.D.; Gyorffy, B.L. et al.; Sluiter. M. et al.; Stauntoii, J.B. et al.; Stocks, G.M. et al., Turchi, P.E.A. et al. Johnson, D.D. et al. 1: 41, 47, 50 Johnson, D.R. see Kishtda, IS.et al. Johnson, D.R. et al. 3: 545 Johnson, G.D. see Goodhue, W.D. et al., Vojak, B.A. et al. Johnson, G.K. see Saboungi, M.L. et al. Johnson, G.K. and Saboungr, M.L. 3 254 Johnson, H.H. ,see Lilieiifeld, D.A. et al.; Nastasi, N.et al. Johnson, J.R. see Sandrock, G.D. et al. Johnson, K.H. 1: 137; see also Eberhart, M.E. et al. Johnson, L.A. 2: 241 Johnson, L.A. et al. 2: 17, 18 Johnson, L.B. Jr. 2: 577, 585, 588 Johnson Matthey and Company Limited 2: 569 Johnson, Q. see Smith, G.S. et al. Johnson, R. 1: 5, 7; see also Dosch, H. et al., Krummacher, S. et al. Johnson, R A . 1: 82; 3: 241 Johnson, R.A. see Baskes, M.I. and Johnson. R.A. Johnson, R.E. 2: 328 Johnson, R.L. 2 597 Johnson, T.J. et al. 2: 124 Johnson, R.N. see Natesan, K. and Johnson R.N. Johnson, R.W and Garland, C.M. 3: 61 Johnson, T.P. see Dowson, A.L. et al. Johnson, W.C. and Cahn, J.W 3: 418 Johnson, W 1: 773, 780
908 Johnson, W. et al. 1: 94 Johnson, W.C. 1: S85, 586, 867 Johnson, W.L. 1: 690, 692, 693, 694, 695, 696,698,700,701,703,704,105,733, 734,736,748,792,817; 2 611; 3: 682, 703; see also Askenazy, P. et al.; Atzmon, M. et als,Cheng, Y.T. et a/,, Clemens, B.M. et al.; Cotts, E.J. et al., Eckert, J. et al.; Fecht, H.J. et al.; Lee, M.C. et al., Liu, B.X. et al., Meng, W.J. et al.: Yeh, X.L. et al. Johnson, W.L. see Bruck, H.A. et al., Conner, R.D. et al.; Gitbsrt, C.J. et al.; Hellstern, E. et al,; Kim, Y.J. et al.; Seki, Y. and Johiison, W.L. Johnson, W.L. et al. 1: 705, 735 Johnston, D.C. see Moodenbaugh, A.R. et al. Johnston, T.L. see Beardmore, P. et al.; Thornton, P.M. et al. D. et al. Jones, A. see Dtivey, S. et al.; Ignatiev, A. et al. Janes, A.C. and O’Brian, P. 3: 669 Jones, C. see Farkas, D. and Janes, C. Jones, C. and Farkas, D. 3: 774 Jones, C.N. see Fairbank, G.B. e f al. Jones, E.C. see Sharp, J.W. et al. Jones, F.W 1: 947 Jones, H. 1: 13, 21, 309, 319, 703, 705, 944; SFP also Blackford, J.R. et al.; Greenwood, G.W. et (41.; Li, Y. et al.; Midson, S.P. et al.; Mishra, R.S. et al.; Prakash, U. et al.; Sadananda, K . et al.; Surytmarayana, C. and
Sadananda, K. et al. Jones, I. see Green, A.J. et al. Jones, I.P. 1: 540, 550; see also Kazantzis, A V et al.; Konitzer, D.G. et al., Millett, J.C.F. et al.; Mgan, H.W. et al.; Perez, J.F. et al.; Rong, T.S. et al.; Yu, H.F. et al.; Zhu, W. et al. Jones, J,W. see Worth, B.D. et al. Jones, P.E. see ~lemeiis,H, et al. Jones, P.E. and Eylon, D. 3: 326 Jones, P.E. et al. 3: 593 Jones, P.J. 1: 534 Jones, R.B. see Narris, J.E. et al. Jones, R.D. see Denner, S.G. et al. Jones, R,M. 2: 200 Jones, R.O. E: 59
Jongebreur, R. see Buschow, K.H,J, e f al.; van Engen, P.G. et al. Jonson, M. see Gunnar~soii~ 0. et al. Jonsson, H. 1: 479 Joos, R. see Kestler, ET. et al.; Knippscheer, S, et al. Jorda, J.L. see Grioni, M. et al. Jorda, J,L, et al. 1: 399; 3; 56 Jordan, A.S. 1: 675, 676 Jordan, AS. et al. 1: 352 Jordan, J.F. see Albright, S.P. et al. Jordnn, J.L. see Weiney, P.A. et al. Jordan, K. see ~ampisi,I.E. et al.
A ~ t h Index o~ Jordan, L.T. see El-Masxy, N.A. et al. Sordan-Sweet, J. see Svilan, V. et al. Jory, D.B. 1: 996 Jose-Yacai~an,M. see Perez-Campos, R. et al. Josephy, Y. 2: 486 Joshi, V.A. see Baaerjee, D. et al. Joslin, D.L, et al. 3: 645 Josse, C. see Gauthier, V. et al. Joubert-Bettan, C.A. et al. 1: 359 Joud, J. 1: 61 1; see also Molinari, C. et al. Joussst, J.C. see Audouard, A. et al. Joy, T3.C. see George, E.P. et al. Joyce, B.A. see Foxon, C.T. and Joyce, B.A. Joyix, V. see AbeIl, J.S. et al. Juan, Y.M. and Kaxiras, E. 3: 439 Juang, F -Y. see Ojima, M. et al. Juhasz, A. see Lacy, D.E. et al. Julien, J.P. see Singh, D.J. et al. Jumonji, K. et al. 3 368 Jung, I. 1: 91S, 921, 922 Sung, I. et al. 1: 916>917, 318, 919, 922, 923; 3: 297, 311, 313 Junqua, N. et al. 1: 574 Jurczyk, M, see ~akubowicz,J. and Jurczyk, M. Jurgenesen, H. see Gruter, K. et al. Jurs, P.C. and Isenhour, T.L. 3 822 KabbG, M. et al. 2: 168 Kabl, W. see Clemens, H. et ul. Kablov, E.N. et al. 3: 542 Kaburagi, M, 1: 35, 295 Kache~myer,C. see Rogachev, A S . et al. Kachi, S . 1: 845; see also Murakami, Y. et al. Kachvr, E.V. see Kornilov, 1.1. et al. Kacprzak, L.see Massalski, T.B. et al. Kad, B. 2: 87 Kad, B.K. see Dao, M. et al.; Hazzledins, P.M. et al. Kad, B.K. et al. 3: 458 Kadowaki, K. see Takeya, H. et al. Kaesche, F. see Eggeler, G. et al. Kaess, U . see Kimmerle, F. et al. Kagamida, M. et al. 1: 282 Kagan, LK. I: 726 Kagawa, E. see Celis, P.B. et al. Kagawa, K. et al, 2: 342 Kagswa, T. et al. 2: 421, 425 Kagayama, T. et al. 1: 221 Kahan, P.J. see Carter, G.C. et al. Kahilov, 1.Kh. see Granovskii, A.B. et al. Kahn, J.S. see Smith, D.K. et al. Kahng, D. see Atalla, M. and Kahng, D. Kahora, P.M. see Chittipeddi, S . et al. Kahveci, A.I. 1: 991, 992, 993; 2: 124; see also Welsch, G. et al. Kai, T. see Nishi, Y. et al. Kai, W. see Chu, J.P. et al. Kaibe, H.T. 2: 329; see ~ r b oOhsugi, 1.J. et al. Kaibe, H.T. et al. 2: 329 Kaibyshev, 0,A. see Imayev, R.M. et al. Kaida, S. see Ikoma, M. et al. Kaieda, Y. 1: 645 Kainuma, R, see Ishida, K. et al. Kainuma, R. et al. 1: 846
Kaise, M. see Wirana, T. and Kaise, M, Kaiser, H. 1: 13 Kaiser, J.H. et al. E: 137
Rakehi, K. 3: 302 Kaldis, E. see Haberineier, H.-U. et al. Kale, G.B. see ~ l i a i i u n i u ~ hKy ,. et al. Kalinina, O.T. see M~khai~enko, S.D. et al. Kalisher, M.H. see Patten, E.A. et al. KaMngal, C.G. 3: 326 Kallingal, C.G. see Smith, T.R. et al. Kallingal, C.G. et al. 3: 326, 327 Kailoniatm, A.C. see Choy, T.C. et al. Kalogirou, 0. see Hu, a. et al. Kalogirou, 0. et al. 3: 94 Kalongji, G. 1: 857 Kalos, M.M. 1: 128; see also Binder, K. et al. Kalupn, P.A. et ul. 1: 457 Kalychak, Ya.M. et al. 3 101 Kalychak, Y.M. see Vert, R. et al. Kamada, H. et al. 2: 327 Kamata, K. ct al. 2: 650 Katnata, K.-Y. see Degawa, T. et al. amene et sky, E.A. see Askenazy, P. er al. Kamigaki, K. et al. 3: 783, 784 Kamijima, A. see Shinoura, 0. et Kamisada, Y. see Fuckino, S . et a Tachikawa, K. et al. ~ a m i t a k ~ h a rW.A. a , see Neumann, D.A. et al. Kamiya, K. see Man, M. et al. Kamm, J.L. and Mifligan, W.W. 3: 66 Mampe, S. see Larsen, D.E. Jr. et ad. Kamprath, H. et al. 2: 316 Kamprath, N. see Cadiera, F.J. et al. K a ~ p ~ ~ rA.T. t h ,see Brown, B.S. eb al., Grimsditch, M. et al.; Kirk, M.A. et al. Kan, B. et al. 3: 724 Kan, YaS. see Aieksandrov, B.N. et al. Kanai, T. see Kawabata, T. et al. Kanamori, J. 1: 35, 295 Kanamori, S . see Mori, M. et al. Kanata, K. see Kamisada, Y. et al. Kanaya, H. et al. 2: 622 Kancheev, 0.D. 2 12, 14 Kanda, H. see Kagamida, M. et al. cane, R.D. 1: 993 Kaneda, T. see Shirai, T. et al. Kaneko, H. et al. 2; 307 Kaneko, K. see Ikeda, M. et al. Kaneko, M. see Hiraga. K. et al. Kaneko, T. 3: 803 , CiL et al. 3: 154, 160 J,-S. see Seam~n,C.L. et al. Kang, 5.S. see Dong, C. et al.; Dubois, J.M. et al. ~ ~ n ~H.C. ~ hseei Gregory, , E. et al. Kannan, V.C. see Ch~ttipe~di, S . et al. Kanno, M. 2: 188; see also Suzuki, ET. et PI. Kano, W. 2: 336 Kanonenko, V.L. et al. 1: 662
909 Kanzawa, Y. 1: 774; see ulso Uzuka, T. et al. Kanzawa, Y. et al. 2: 565 Kao, C.H. et al. 2: 621 Kao, C.R. and Chang, Y.A. 3: 771 Kao, W.H. see Brennan, P.C. et al. Kapikka, A. 2: 958 Kapur, V.K. 2: 330,423 Kapusta, Cz. see Lord, J.S. et al.; Zhou, R.J. et al. Karasawa, T. see Ohkawa, K. et al. Karasev, K.A. see Shemyakin, V.S. et al. Karashima, S. see Watanabe, T. et al. Karch, J, et al. 3: 750 Karchenko, V.K. see Arbuzov, M.P. et al. Karim, Kh.R. 1: 956 Karkina, L.Ye. see Greenberg, B.A. et al.; Indenhaurn" V.N. et al. Karla, I. et al. 3: 104 Karlsen, C.E. see Summers, L.T. et al. Karlsson, L. et al. 3: 663, 665 Karlsson, S.E. see Seigbahn, K. et al. Karmonik, C. see Skripov, AV. et al. Karnthaier, H.P. 1: 498, 505, 528, 540; see also Baluc, N. et al.; Hnzzledine, P.M. et al.; Korner, A. et al.; Mills, M.J. et al. Karnthaler, H.P. et al. 3: 443 Karolik, A.S. 1: 956, 959, 960 Karpe, N. see Greer, A.L. et al. Karpov, Yu.G. see Kourov, N.I. et al. Karsanov, G.V. see Zakbarova, A.M. et al. Klarsten, K. 1: 5 Kasahara, K. et al. 2 82 X-asaya, K. see Takeuchi, H. et al. Kasaya, M. see Iga, F. et al., Ogawa, S. et at, Kasaya, M. et al. 1: 217 Kaschner, G. see Mukhopadhyay, J. et al. Kasm, M.B. 1: 955, 956 Kashina, T A . see Rezukhina, T.N. et al. Kasiraj, P. see Schwarz, R.B. et al. Kaskel, S . and Corbett, J.D. 3: 120 Kasper, J.S. 1: 277, 278, 409, 445, 473, 474,939 Kasper, J.S. et al. 3: 130 Kassem, M A . 1: 638 a s ~ ~ aA, n ,see Heden~vi~t, P. et al. Kassner, M.E. and Peterson, D.E. 3: 802 Kasuya, T. 1: 945; see also Fraas, K. et al.; lga, F et al., Kasaya, M. et al.; Ogawa, S. et al. Kas'yanova, A.V. see Rezukhina, T.N. et al. Katano, S, 1: 773, 777 Kalaoka, N. see Suzuki, K. et al. Kataoka, T. 2: 344 Kataoko, M. see Asada, T. et al. Katayama, K. ,see Mabuchi, H. et al. Katayama, T. 2: 446, 447 Kalayama-Yoshida, H. see Kido, G. et al.; Takahashi, T. et al. Katerbau, K.H. see Jenkins, M.L. et al. Katheder, H. see Bruzzone, P. et al. Kato, A, see Inoue, A. et al.; Sakai, T, et al. Kato, H. see Karnigaki, K. et al.; Sugiyama, K. et al. Kato, K. see Imoto, T. et al. Kato, K. et al. 2: 342
Kato, M. see Jumonji, K. et al. Kato, N. see Yamaguchi, A. et al. Kato, Y. see Kurosawa, K. et al. Katsnel'son, A.A. 2: 218 Katsut, A. 2: 327, 330, 448; see also Kamada, H. et al. Katsura, S. 1: 35, 295 Kattelus, H.P. 2: 605 Kattelus, H.P et al. 2 623, 624, 626 Katter, M. 2: 315; see also S ~ h n i t z ~K, e, et al.; Schultz, L. et al.; Wecker. J. et al. Katter, M. et al. 1: 746; 2 314, 316 Kattner, U. see Ellner, M. et al. Kattner, U.R. 2: 95 Kattner, U,R. et al. 1: 638, 639 Katz, A. 1: 457 Katz, J.L. 2: 586 KauRman, G.B. 1: 7 Kaufinan, A&. 2: 361 Kaufman. H.R. and Robinson, R.S. 3: 665 Kaufman, L. I: 100, 830, 831 Kaufman, M.J. 1: 454; see also Bendersky, L. et al.: Cotton, J.D. et al.; Frasier, F.R. et al.; Jayashankas, J.S. et al.; Jones, S.A. et al.; Kest~~r-Weykamp, H.T. et al. Kaufnian, M.J. et al. 1: 638 Kaufmann, M.J. see Krishnan, P. and Kaufmann, M.J. Kaufmann, R. 1: 579 Kauppinen, H. see Alatalo, M. et al. Kauzlarich, S,M. see Chan, J.Y. et al. ~ a u z ~ aW. ~ n1: ,731,736 Kaveci, A.I. see Welsh. G. and Kaveci, A.I. Kaveh, M. see Weger, M. et al, Kaviani, K. et al. 2 425 Kawabata, T. 1: 40, 539 Kawabata, T. et al. 1: 532, 915; 2: 77; 3: 423 Kawaguchi, H. see Takeuchi, H. et al. Kawakami, M. see Yamamoto, T. et al. Kawakami, T. see Oku, S. et al. Kawakami, Y. see Wu, Y.-H, et al. Kawakami, Y. et al. 2: 428, 429 Kawamoto, J. see Kido, Y. et al. Kawamura, Y see Asahi, H. et al.; ISagawa, K. et al.; Kagawa, T. et al.; Taiiaka, H. et al.; Wakita, K. et al. Kawamura, Y. et al. 3: 685 Kawanaka, A, see Nekano, T. et al. Kawanabe, T. 2: 449 Kawase, M. 1: 542; see also Saka, H. et al. Kawasaki, Y. see Fujita, H. and ISawasaki, Y. Kawasima, A. see Masumoto, Y. et al. Kawasurni, I, et al. 2: 329 Kawata, S. see Kino, T. et al.; Kobayashi, K. et al. Kawaura, H. see Nishino, K. et al.; Nishitani, S.R. et al. Kawazoe, H. see Masahashi, N. et al. Kawazoe, Y. see Sluiter, M.H.F. et al. Kaxiras, E. see Boyer, L-L. et al.; Juan, Y.M. and Kaxiras, E.; Wagh~are, U.V Kaxiras, E. and Duesbery, M.S. 3: 439 Kaybyshev, Q.A. et al. 1: 653
Kayser, F.X. 1: 874, 876, 887, 888; 2: 203; see also Leamy, H.J. et al.; Stassis, G . et al. Kaysser, W.A. see Lnag, R, et al., Leyens, C. et al., Maurer, R.et al. Kazakov, V.A. see C'llzey, L.S. et al. Kazakovd, E.F. see Tsurikov, V.F. et al. Kazantzis, A.V et at, 3: 415, 419, 461 Kazior, J. see Gialan~~la, S. et aE. K-azmerski, L.L. see Coutts, T.J. et al. Kedr, B. see DiPasquale, J. et al, Kear, B.H. 1: 498,525, 528,546,882,896; 2: 13, 17, 24, 25, 137; see also Breinan~E.M, et al.. Das, S,K. et al.; Giamei, A,F. et al,; M~hrabian,R, et al., Pearsoti, D.D. et al. Kear, B.H. and Piearmy, B.J. 3: 302, 942 Kear, B.H. and Strutt, P,R. 3: 7.54 Kear, B.H. and Wilsdorf, H.G.F. 3: 372 Kear, B.H. et d. I: 528,989; 2: 17, 18, 19, 22; 3: 404, 501 Kear, W.R. 1: 907 Kearley, C. see Price, D.L. et al. Kearns, M.W see Johnson, T.J. et al. Kecskes, L.J. see Niiler, A, et al. F.J. et al. Keegan, M. ,we Trambly de Laissnrdiere, G. et al. Keeler, J.H. 2 134 Keeling, L. et al. 3: 32 Keen, D.A. and Hull, S. 3: 251 Keene, D.E. see Aimstrong, R.D. et al. Kehler, B.A. see He, X.-M, et al. Keil, M. et al. 1: 164 Keim, T.A. see Gamble, B.B, et al, Keir, William I: 3n Kek, S. see Ellner, M. et al. Kek, S. et al. 1: 121, 122 Kekalo, I.B. 2: 650 Keller, K.R. 1: 875, 890 Keller, R. see Stefanou, N. et al. Kelley, K,K. see Hult~ren,R. et al. Kelley, M.J. see Todd, A.C. et al. Kelley, P. see Campisi, I.E. et al. Kelly, A. 2: 258; see al.ro Fairbank, C.B. et al.; Groves, C.W. and Kelly, A, Kelly, M.J. see Chittipeddi, S , et al.; Haydock, R. et al.; Wickenden, D.K. Kelly, T.F. see Kim, Y -W. et al. Kelly, T.J. 2: 83, 87, 88; see also Austin, C. M. et al. Austin, C.M. and Kelly, T.J.; Austin, C.M. et al. Kelton, K.F. 1: 453, 454, 457, 461, 463, 465,467,470,473,477,479,480,482, 891, 743; 3: 379, 693, 755; see also Ali, N. et al.; Daulton, T.L. et ad.; Gibbons, P.C. et al.; Holzer, J.C. et al.; Jeong, E.K. et al.; Levine, L.E. et al.; Libbert, J.L. et al.; Sabes, P.N. et al., Shield, J.E. et al. Kelton, K,F, see Foster, I<. et al. Kelton, K.F. et al. 1: 454, 463 Kematic, R.J. 1: 109 Kembaiyan, K.T. 1: 651 Kemeny, T. ,see Balogh, J. er al. Kempf, M. see GGken, M. et al. Kempf, M. et al. 3: 227
910 Kenan, R.P see Chiang, K.H. et al. Kendig, M. see Menezes, S. et al. Kenshole, G. see Rockett. A. et LIE. Kentzinger, E. and Schober, N.R. 3: 281 Keppens, V. see Sales, B.C. et al. Keramidas, V.G. see Harbison, J.P et al., Sands, T. et al., Tabatabaie, N. et crl, Kerans, R.J. 3: 326 Kerdja, T. see Boily, S. et al. Kerl, R. et al. Kerley, G. see Price, D.L. et al. Kern, J. 1: 40 Kern, R. 1: 172, 180, 181 Ken, W.R. see Mendiratta~M.G. et al., Semiatin, S.L. et al. Kerry, S. see Chave, R.A. et ul. Kes, P.H. see Palstra, T.M.M. et al. Keshwoto, K. see Soboyejo, W 0. et al. Keskar, N. R. see Chelikowsky, J. R. et al. Kessler, H.D. 2: 77, 81, 82 Kestel, B.J. see Kirk, M.A. et al., Meng, W.J. et al. Kestler, €3. see Bartels, A. et al.; Chatterjee, A. et al.; Clemens, H. et al.; Knippscheer, S. et WE., LeHolm, R. et al. Kestler, H. et al. 3 635, 637, 638, 651 Kestner-Weykam~,H.T. et al. 2 95, 98 Khachaturyan, A.G. 1: 26, 44, 851, 853, 855; see also Clien, L. et al.; Chen, L.-Q. et al. Khachin, V.N. see Matveeva, N.M. et al., Pushin, V.G. et al.; Savvmov, A.S. et al.; Tokarev, V.N. et al. Khachin, V.N. et al. 1: 715, 717 Khadkikar, P.S. see Rigney, J.D. et al. Khadkikar, P.S. et al. 1: 896; 2: 30 Khali, A. see Hartwig, K.T. et al. Khalim, A.A.R. et al. 1: 956, 957 Khan, A. see Cook, J. et al.; Lee, E.W. et al., Pregger, B.A. ef al. Khan, A S . see Barrett, C.A. et al., Corey, R.G. et aI. Khan, T. 3: 544; see also Blavette, D. et al.; Caron, P. et al., Naka, S. et al. Khan, T. et al. 3: 841, 843 Khan, Y. 2 3 10 Khangoakar, P.R. 1: 108 Khanna, S.N. see Jena, P. et al. Khantha, M. see Cserti, J. et al., Vitek, V. et al. Khantha, M. el al. 1: 523, 547, 548, 549, 915, 916; 2: 25, 26, 153, 154; 3: 439, 447,464 Khatee, A. et al. 3: 62 Khera, S.K. see Bhanumurthy, K. et al. Khimich, Y.P see Miroshnikov, M.M. et al. IUlina, B,B. et al. 3: 730. Khobaib, M. 2: 124; see also Balsone, S.J. et al. Khomenko, I.A. see Rogachev, AS. et al. Khryapov, V.T. et al. 2: 418, 432 Khvostantsev, L.G. see Brazhkin, V V. et al. Kiang, L.L. 1: 265 Kichigin, V.1. 2 502 Kidd, J.M. see Fowler, P.H. Pt al. Kidin, I. 1: 782 Kido, G. et al. 1: 136
Author Index Kido, Y. et al. 1: 650 Kidson, G.V. 1: 768 Kieback, R. see Scholl, R. et al. KieRer, J.C. see Boily, S. et al. Kieffer, R. see Nowotny, H. et al. Kiely, C.J. see Mullan, CA. et al. Kiely, J.D. et al. 3: 228 Kieschke, R.R. et al. 1: 861 Kiesielowski-Kemmerich, C, see Cox, G. et al. Kiewit, D.A. see Lautenschlager, E.P. et al. Kihara, J. see Yen, B.K. et al. Kikegawa, T. see Fujinaga, Y. et al. fmai, M. et al. Kikuchi, J. see Ikushima, K. et al. Kikuchi, M. 2: 75; see also Cliff, G. et al., Takeyama, M. et al.; Yamabe, Y. et al. Kikuchi, R. 1: 30, 38, 496, 499, 502, 511, 610, 613, 775, 782, 786; see also Gschwend, K. et al.; McCoy, J. et al.; Sato, H. et al. Kikuchi, R. et al. 1: 856; 2: 565 Kim, A. see Wang, Y. et al. Kim, A.S. see Rarnamurty, U. et al. Kirn, A.S. and Camp, F.E. 3: 98 Kim, C.K. see Jordan, A S . et al. Kim, D. see Rifkin, J.A. et al. Kim, D.K. see LaSalvia, J.C. et al. Kim, G.M. see Schaeffer, J. et al. Rim, H.M. see Kim, Y.-W et al.; Mendiratta, M. et al. Kim, I. see Hunecke, J. et nl. Kim, I.K. see Baldi, R.W. et al. Kim, I.S. 1: 926 Kim, J. see Noebe, R. et al. Kim, J.G. 1: 967, 970, 973; 2: 208 Kim, J.H. see Kirn, J.K. et al., Lee, T.K. et al. Kim, J.K. et al. 3: 653 Kim, J.T. 1: 536; 2: 58 Kiin, J.T. and Gibala, R. 3: 366 Kim, J.T. et al. 1: 927 Kim, J.Y see Foster, K. et al. Kim, K.-B. see Choi, C.-H. et al. Kim, M.J. and Fianagan, W.F 3 74 Kim, M.S. 1: 701; see also Chiba, A. et al. Kirn, N.J. 2: 188, 190; see also Bye, R.L. et al.; Chung, H.H. et al.; Kim, J.K. et al. Kim, N.J. et al. 2: 188, 190, 194 Kim, O.K. see Forrest, S.R. et al. Kim, P.A.S. see Ramamur~hy,U. et al. Kim, S. and Smith, G.D.W. 3: 64 Kim, S. et al. 3: 64 Kim, S.G. see Inoue, A. et al. Rim, S.G. et al. 1: 735, 736 Kirn, S.M. 1: 562, 564, 565, 566, 572 Kim, S.W. see Kang, B. et al. Kim, T.H. see Kozubski, R. et al. Kirn, Y. see Lopatin, S. et al. Kirn, Y.-G. see Lee, J.-Y. et al. Kim, Y.G. and Lee, J.-Y. 3: 78 Kim, Y.J. et al. 3: 685 Kirn, Y.K. see Sadwick, L.P. et al. Kim, Y.K. et al. 1: 651; 2: 621 Kirn, Y.-M. see Semiatin, S.L. et al. Kim, Y.-W 1: 864, 916, 927, 928; 2: 33, 76, 83, 86, 159, 163, 170; 3: 446, 618, 630, 634, 648
Kim, Y.-W et al. 1: 733; see also Chan, K.S. and Kim, Y W.; Chen G.L. et al., Diiiiiduk, D.M. et al., Diniiduk, D.M. et al.; Gouma, P.I. et al.; Jin, Z. et al., ~ e n d i r a t t aM.C. , et al., Venkateswara Rao, K.T. et al.; Yoshihara, M. and Kiiii, Y.-W Kim, Y W. and Dimiduk, D.N. 3: 368, 472,477,478,480,481,484,618,648 Kim, Y.W et al. 3: 477 Kinball, C.W. 1: 411; 2: 317 Kirnrnerle, F. et al. 3: 48 Kimoto, M. see Imoto, T. et at.; Shibuya, A. et al. Kimura, A. et al. 3: 363 Kimura, H. and Masumoto, T. 3: 695 Kimura, H.M. see Inoue, A. et al. Kimura, K. 1: 470,485, 875, 878; 2 148, 149; see also Hirano, T. et al. Kimura, M. see Masahashi, N.el al. Kimura, S.4. see Watanabe, T. et al. Kinchin, G.H. 1: 784, 803 King, A.N. 1: 902; 2: 28; see also Yoo, M.H. and King, A.H. King, H.E. Jr. see Penney, T. et al. King, H.W. 1: 105, 106, 115 King, J.C. see Fowler, C.M. et al. King, 0. see Erdogan, T. et al. King, R.C. 1: 107, 108, 109 King, W.E. 1: 804 Kingery, W.D. 1: 100 Kingman, D.D. see Dunniead, S.D. et al. Kino, T. et al. 1: 960 Kinoshita, C. see Eguchi, T. et al.; Liu, H.C. et at. Kinoshita, C. et al. 1: 577 Kinoshita, K. 2: 328 Kinsman, K.R. see Hehemann, R.F. et al.; Lonmer, G.W. Kirby, J.E. 1: 534 Kirby, R.K. see Touloukian, Y.S. et al. Kircher, C.J. see Huang, N.C.W. et al. Kircher, T.A. 1: 997, 998, 999; see also Pregger, B.A. et al. Kirchmayr, N. see Belin-Ferre, E. et al. Kirchmayr, H.R. 3: 30 Kirchtnayr, H.R. and Poldy, C.A. 3: 535 Kirchner, H.O.K. 1: 540 Kirk, M.A. 1: 804, 805; see also Black, T.J. et al. Kirk, M.A. et (d. 1: 792, 804, 805, 808 Kirkpatrrck, S. see Velick?, B. et al. Kiselyova, N.N. 3: 815 Kiselyova, N.N. 1: 270; 3: 812, 814, 815, 818, 824, 825, 827, 831, 832; see also Degtyaryov. Yu, I. et al.; Golikova, M.S. et al.; Kravchenko, N.V. et al.; Savitskii, E.M. et al., Zemskov, V.S. et al. Kiselyova, N.N. and Burkhanov, G.S. 3: 812, 821, 824, 825, 829 Kiselyova, N.N. and Gladun, V.P. 3: 812, 824 Kiselyova, N.N. and Kravchenko, N.V. 3: 812, 813, 814, 818. 824 Kiselyova, N.N. and Savitskii, E.M. 3 812. 824, 826, 829 Kiselyova, N.N. et al. 3: 812, 813, 814, 815, 819, 824
Author Index Kishida, K. see Inui, H. et al.; Paidar, V et al.; Siegl, R. et al. Kishida, K. et al. 3: 414 Kishida, T. et al. 2: 382 Kishizawa, T. see Hao, S.M. et al. Kisly, P.S. et cd. 1: 996. 998 Kiss, S. SLY Posgay, G. e t al. Kiss, S.J. see Durovic, D. et al. Kissinger, H.E. see Brimhall, J.L. et al. Kistmp, C.J. and Schuster, H.U. 3: 238 Kita, K. see Higashi, K. et al. Kitabjian, P.H. et al. 3 304 Kitaev, A.Y. see Kaiugin, P.A. et al. Kitaigorodskii, AS. 3 7 Kitajima, M. et al. 1: 668 Kitajima, S. see Kinoshita, C. et al. Kitano, Y. see Komura, Y. and Kitano, Y
Kitano, Y et al. 1: 410 Kitano, K. and Pollock, T.M. 3: 368 Kitano, K et al. 3: 371 Kitaoka, Y. see Kido, G. et al., Kobajashi, T. et al., Kyogaku, M. et al., Nakamura, W.et al. Kitazawa, H. see Tang. J. et al. Kittel, C. 1: 57, 198, 440; 2: 389 Kittl, J.F. 1: 843 Kiyasawa, T. see Koguchi, N. et al. Kiyoshi, T, see Tnoue, M.et at. Riyoshi, T. et al. 2: 369 Kjeins, J.K. see Andersen, N.H. et al. KkdgeS, C.-P 2 411 Klahn, D.H. 2: 250 Klakov, M.P. et al. 2: 326 Klaniut, C. see Bussiere, J.F et al. Klanski, J.L. see Nic, J.P et al. Klansky, J.L. et al. 3 580 Klar, E. 3: 644 Klassen, M. see Clemens, H. et al. Klassen, T. see Oehring, M. et al., Yan, Z . et al. Klaumunzer, S. 1: 821; see also Carrido, F. et al. Klaumunzer, S. et al. 1: 821, 3: 270 Klaus, S. see Liminez, J.A. et al. Klavins, P. see Chan, J.Y. et al. Klee, H. 1: 107 Klee, P. 3: 232 Kleijn, C.R. 3: 669 Klein, B.M. 1: 208; see also Chubb, S.R. et al.; Galakov, V.R. et al.; Mehl, M.J. et al.; Osburn, J.E. et al. Klein, C.F see Ayer, R. et al. Klein, D. see Charlot, E. et al. Klem, J. see ~isenmann,B. and Klem, J., Kosonocky, W.F. et al. Klein, J.D. see Cogan, S.F et al. Klein, 0. and Baker, I. 3: 368, 373 Klein, R. 1: 453 Klein, T. et al. 1: 484 Kleinennan, V.I. see Goman'kov, V.I. et al. Kleinman, L. see Zhu, M.J. et al. Kleinstuck, K. see Nghiep, D.P. et al.; Quyen, N.H. et d. Kleitz, M. 1: 95; see also Fabry, P. et al. Klemn, M, 3: 379; see also Yu, D.P. et al. Klemens, P.G. 1: 944, 1025, 1026; see also Touloukian, Y.S. et al. Kleinent, W Jr. 1: 104; see also Duwez, P et al.
Klement, W. Jr. et al. 1: 792 Klemm, W. 3: 114 Klepner, S.P see Huang, H.C.W. et al. Kleppa, O.J. 1: 64, 107, 108, 109 Klirna, S.J. see Brindley, P.K. et al. Klimker, W. 1: 875; see also Rosen, M. et al. Klimker, N.et al. 1: 889, 890 Klinger, L. see Rabkin, E. et al. Klissurski, D. see Radev, D.D. and Klissurski, D. Kloc, L. see Fiala, J. et al. Klopp, W.D. 2: 228 Klosek, P. 3: 805 Klower, J. 1: 917, 926, 989 Klug, D.D. et al. 1: 702 Klug, F.J. et al. 3: 542 Klug, K. see Natesan, K. et al. Kluge, M. see Schubert, K. et al. Klyuyeva, T.B. see Arkharov, V.I. et aE. Knabl, W. see Chatterjee, A. et al.; Clemens, H. et al. Knacke, 0. 1: 109, 110, 111, 112, 121; see also Barin, I. et al. Knapp, J.A. 1: 454; see also Mirkarimi, P.B. et al. Knapton, A.C. 2: 590, 649 Knaster, M. see Panicker, M.P.R. et al. Knaul, D.A. et al. 3: 480 Knecht, J. et nl. 1: 319 Knibloe, J.R. see Sikka, V.K. et al.; Wright, R.N. et al. Knibloe, J.R. et al. 3: 504 Knippmg, P. see Laue, M. et al. Knippscheer, S. et al. 3: 635 Knippscheer, S, see Blum, M. et al. Knoch, K.G. see Coey, J.M.D. et ul. Knoeseii, D. see Pretonus, R. et al.; Rozgonyi, G.A. et al. Knol, K.S. see Caster, D. et al. Knoll, M. see Espe, W. et al. Knorr, D.B. see Fox, T. et al. Knotek, 0. 2: 410 Knott, H.W 1: 393, 394 Knott, J.F. see Soboyejo, W.O. et al. Knowles, K.M. et al. 1: 453 Knudsen, J.M. see Petersen, J.F. et al. Knudsen, M. 3: 667 KO, T. 1: 835 KO,W.H. 1: 956 Kob, W see Donati, C. et al. Kobayashi, K. et al. 2: 336 Kobayashi, K.F. see Nishitani, S.R. et al. Kobayashi, M. see Abe, M. et al., Inui, H. et al.; Jeon, H. et al., Oh, M.H. et al. Kobayashs, N. see Zhang, S. et NI. Kobayashi, T. see Nakamura, N. et al. Kobayashi, T. et al. 2: 431, 3: 162 Kobayashi, Y. see Fuckmo, S. et al. Kobyllun, A.N. see Alisova, S.P. et al. Koch, A.J.J. et al. 2: 309 Koch, C. 1: 788 Koch, C.C. 1: 638, 640, 701, 706, 925; 2 29, 30, 38; 3: 645, 753; see also Cho, Y .S. and Kocls, C.C.; Davis, R.M. et al.: Jang, J.S.C. and Koch, C.C.; Liu, C.T. et al.; ~ c D e r m o t tB.T. , and Koch C.C., Stolof?, N.S. et al.; Suryanarayana, C. and Koch, C.C. Koch, C.C. et al. 1: 700, 734; 2: 17, 18; 3: 68 1
91 1 Koch, E. 1: 317, 325; see also Hellner, E. et al. Kocli, E.F. see Livingstoii, J.D. et al. Koch, F see Intermann, A. et al. Koch, 5.141. 1: 568; see also Koenig, C. et al. Koch, J.M. et al. 1: 569 Koch, T. see Shani, Y et al. Kocherzhinskii, Yu.A. see Svechnikov, V.N. et al. Kock, E. see Hellner, E. et al. Kock, W and Kneringer, G. 3: 631, 632 Kocks, U.F. 1: 921: 2: 269: 3: 371; see also Bacon, D.J. et al. Kocks, U.F see Mecking, H. et al. Kocks, U.F. et al. 3: 361, 371 Koczak, M.J. et al. 1: 559 Kodama, H. see Yasuda, K. et al. Kodash, V.U. see Kisly, P.S. et al. Koebrugge, G.W 1: 743; see also de Vries, J. et al. Koehler, J. 1: 784 Koehler, J.S. 1: 15, 496, 693, 803; 2: 263 Koehler, W.C. see Cable, J.W et al. KoeHing, D.D. 1: 130, 132, 199 Koenig, C. 1: 568; see also Koch, J.M. et al. Koenig, C. et al. 1: 569 Koeppe, C. see Clemens, H. et al. Koeppe, C. et al. 3: 633, 634, 637, 762 Koernerb, N.see Intermann, A. et al. Koestner, R.J. see Reed, M.A. et al. Koflat, D.D. et al. 1: 479 Kofstad, P. 1: 988, 997; 3: 717 Kofstad, P. et al. 1: 980, 993 Koga, K. et al. 2: 328 Kogachi, M. 1: 726; 2: 566; see also Kirn, S.M. et at. Kogachi, M. and Haraguchi, T. 3: 277, 282 Kogachi, M. et al. 1: 561, 565, 566 Kogelnik, H. 2: 339 Koguchi, N. 2: 332; see also Takahashi, S. et al. Koguchi, N. et al. 2: 329 Koh, IS.see Ohtani, T. et al. Kolianoff, J. et al. 3: 258 Kohga, Y see Kornatsu, K. et al. Kohl, W. et al. 1: 782 Kohler, B. see WolfT, 5. et al. Kohlhaas, R. see Rocker, W, aiid Kohlhaas, R. Kohn, W. 1: 24, 26, 58, 59, 78, 129, 157, 196; see also Ho~enberg,P. and Kohn, W. Kohn, W. and Sham, L.J. 3: 234 Kohno, 0. see Kamisada, Y. et al. Koide, T. see Yanagihara, M. et al. Koike, J. see Xu, G.-B. et al. Koike, J. et al. 1: 695, 809, 81 1, 815 Koiwa, I. see Osaka, T. et al. Koiwa, M. see Nonaka, K. et al.; Yasuda, H. et al. Koiwa, M. et al. 1: 768 Koizunii, H. see Yainabe-Mitdrai, Y. et al. Koizumi, Y. see Murakami, A. et al.; Yamabe, Y et al. Kojim, K. see Showaki, K. et al. Kojima, T. see Hara, K. et al.; Ohsugi, I.J. et ~11, Kojima, Y, see Ochiai, S. et al.
912
Author Index
Kojnok, J. see Szaza, A. and Kojnok J. Kokorin, V.V. et al. 3: 691 Kolatschek, K. see Eliner, M. et al. Kolawa, E. see Dorner, W. et al.; Kattelus, H.P. et al.; Nieh, C.W. et al.; Pokela, P.J. et al.; So, F.C.T. et al. Kolawa, E. et al. 2: 616,625,626,627,628 Kolinsky, P.V 2: 417, 430 Kolsky, H. 3: 695 Kohter, B.H. see Veer, F.A. et al. Komarek, K.L. 1: 94, 95; 3: 845, 849; see also Ettenberg, M. et al.; Johnson, W. et al.; Leubolt, R, et al. Komarova, LA. see Vlasova, E.N. et al. Komatsubara, T. see Kagayama, T. et al. Komem, Y. see Balluffi, R.W. et al.; Lahav, A. et al. Komissarova, L.N. see Kiselyova, N.N. et al. Korniya, S. see Yamakoshi, S. et al. Komoda, T. see Ogawa, S. et al. Komori, K. see Ikorna, M. et al. Konioriia, K. see Kido, G. et al. Komura, Y. see Kitano, Y . et al., Yoshimoto, N. et al. Komura, U.and Kitano, Y. 3: 415 Kondo, J. 1: 212 Kondo, K. see Abe, M. et al. Kondo, S. see Iwasaki, H. et al. Kondo, Y. see T a k ~ u ~ hR. i , et al. Kondrat’ev, V.K. see Bulycheva, Z.N. et ul.
Kondrat’ev, V.V. see Pushin, V.G, et al. Kong, L.-S. see Yang, Y.-C. et a/, Kong, X. see Qu, X. et al. Kong, X.J, et al. 1: 130, 132 Konigstern, C. and Spallart, M.N. 3: 671 Konishi, H. 1: 777, 780 Konishi, T. 2: 649 Konitzer, D. see DiPasquale, 5. er al. Konitzer, D.G. see Angers, L.M. et al.; Darolia, R. et al.; Kaufman, M.J. et al.; Rowe, R.G. et al.; Soboyejo, W.O. et al.
Konoshi, T. see Son, J.-Y- et al. Konstanchuk, I.G. see Bokhonov, B.B. et al. Kontorova, T. see Frenkel, J. and Kontorova, T. Konyaev, Yu.S. see Smirnova, N.L. et al. Koo, C.H. see Hon, W.P. et al. Koon, N.C. 2: 314 Kopasz, C. see Ttxrn6czi, T. et al.
Koppenaal, T.J. 2: 261 Koppenaal, T.J. aitd Kuhlnian-Wilsdorf, D. 3: 359 Kopsky, V. and Litvin, D.B. 3: 444 Kopylov, V.J. see Segal, V.M. et al, Korchugonov, B.N. see Ryadchenko, V.M. et al.
Koren, U. see Sham, Y . et al. Korenivski, V. see Madurga, V, ei ul. Konnko, P.S. 1: 1008 Korinko, P.S. et al. 1: 1007 Korner, A. 1: 496,498,503,505,528,540, 542, 543, 546, 548, 552; 3: 443; see also Hazzledine, P.M. et al. Korner, A. and Schoeck, G. 3: 443, 461 Korner, A. et al. 3: 443 Kornilov, 1.1. 1: 712, 714, 947 Kornilov, 1.1. et al. 1: 712, 715 Kornilova, 2.1. see Barinov, S,M. et al. Korshunov, V.N. 3: 26 Kortan, A.R. see Hong, M, et al., Hsieh, Y.F. et al.; Rabe, K.M. et al. Kortan, A.R. et al. 1: 457, 468, 483 Korte, B.J. see von Raiike, P.J. et al. Korte, B.J. et al. 3: 519, 527, 529 Korth, G.E. see Rabin, B.H. et al.: Suryanarayana, C. et al. Korth, @.E. and Williamson, R.L. 3: 739 Kortovich, C.S. see Culbertson, G. and Kortovich, C.S. Kosaka, T. see Matsumara, T. et al. Koshelev, I. see Renusch, D. et al. Koskenmaki, D.C. et al. 1: 454, 475 Kosonocky, W.F. 2: 232 Kosonocky, W.F. et al. 2: 232 Koss, D.A. 2: 107, 282 Koss, D.A. et al. 2: 93, 106 Koster, G.F 1: 33 Koster, U. see Wollgarten, N. et al. Koster, U. 1: 465, 746, 747, 748; see also Hillenbrand, EX.-G. et ul, Kijster, U. er al. 1: 749; 2: 608 Koster, U. 3: 755 Kaster, W 1: 719, 721; 2: 308, 309, 310 Kostic, E. see Durovic, D. et al. Kostorz, G. see Beddoe, R.E. et al.; Biihrer, W et al.; Epperson, J.E. et al.; Kozubski, R. et al., Kral, F et al.; Reinhard, L. er al. Kostyanaya, O.V. 1: 722, 723, 728; see also Matveeva, N.M. et al. Kosuge, H. 2: 175, 176, 177; see also Miki, I. et al. Kosug~,M. 1: 41; see also Tso, N.C. et nl. Kosugi, M. et ul. 1: 41 Kotaka, 1. see Wakita, K. et al. Kottcke, M. see Bium, V et al. Kottke, T. see Wiiler, A. et al, Kotyukov, V.1, see Kutolin, S A . et al. Kotzyba, G, see Pattgen, R. Kou, X.C. see Grossinger, R. et al., Klatter, M. et al. Kougli, J.R. see Ferreira, A. et al. KOUI,K. see Son, J.-Y. et al. Kourov, W.I. see Shcherbakov, A.S. et al. Kourov, N.I. et al. 1: 949 Kouvel, J.S. 1: 441, 445, 939, 948, 949; 2 304; 3: 804; see also Said, M.R. et al. Kovac, 1. 2: 175 Kovachev, V. see Bussiere, J.F et al. Kovalev, A.I. et aE. 1: 1021 Kovneristy, Yu.K. see Alisova, S.P. er al.; Khachin, V.N. et al.; Matveeva, N.M. et al. Kowaka, M. see Fujii, Y. et al. Koya, A. see Kiinura, A. el al. Koyama, K. et al. 3: 99, 162 Kozakar, T. see Miyazaki, T. et ul.
Kozen, A. see Kato, K. et al. Kozhanov, V.N. see Skripov, A.V. et al. Kozin, L.F 3: 29 Kozin, L.F et al. 3: 26, 33 Kozlov, E.V 1: 711; see also Popov, L.E. et al. Kozlovskii, M.T. et al. 3: 25, 30 Kozubski, R. 1: 777, 778; see ulso Cadeville, M. et al., Leroux, C. et al. Kozubski, R. et al. 1: 777, 778 Kraft, R.W. see Quinn, R.T. el al. Kraft, W 2: 650 Kraftmakher, Y. 3: 278 Krajewski, J.J. see Cava, R.J. et al.; Ikushima, K. et al., Siegrist, T. et al.; ZLindbergen, H.W. et al. Krakauer, H. 1: 138; see also Chen, J. et al.; Feldman, J.L. et al., Mehl, M.J. et al.; Pickett, W.E. et al.; Singh, D. et al.; Wang, C.S. et al.; Wei, S.-H. et al.; Wimmer, E. et al. Krakauer, H. et al. 1: 59 Kral, F. et al. 3: 443, 454 Kramer, E.J. 1: 957 Kranier, M.J. see Akinc, M, er al.; Meyer, M.K. et al., Shen, 2;. et al.; Shield, J.E. et al. Kramer, P. 1: 457 Kritmer, P. er al. 3: 216 Kramer, U. 2: 224; see also Nghiep, D.P. et al.; Quyen, N.H. et al. Krasij, M. see Goebel, J.A. et al. Krasulin, Y.L. see Barinov, S.M. et al. Kratockvil, J. and Libovieky, S. 3: 326 Krause, J.T. see Chen, H.S. et al. Krautle, H. see Su, L.M. et al. Kravchenko, N.V see Kiselyova, N.N, et al., Savitskii, E.M. et al. Kravchenko, N.V et al. 3: 813, 814, 815 Krebs, J J asee Pnnz, G.A. and Krebs, J.J. Kreige, O.H. 2: 8 Kremser, T. see Wunder~ich,W. et al. Kren, E, 2: 318 Krepski, R.P. 2 519 Kress, J.D. 1: 79 Kress, W. 1: 150, 1019 Kresse, G. see Stiidler, R. af al. Kressel, H. 1: 960 Kreutz, T.J. see Aebi, P et al. Krier, C.A. see Regan, R.E. et al. Krill, C.E. see Eckert, J. et al. Krill, G. see Isnard, 0. et al. Kripyakevich, P.1, 1: 374, 403, 405, 411, 413,415,417; see ~ladyshevskii,E.I. et al.; Kuzma, J.B. et al.; Vorosbilov, Yu.V. et al. Kripyakevich, P.I. et al. Krisement, 0. 1: 398 Krishna~oorthy,A. et al. 3: 670 ~ r i s ~ n a m u r t N. h ~see , Gasg, S.P. et al., Raghavan, V et al. Krishna~urthy,S. see Das, G, and Krishna~urthy,S. Krishnan, K.M. see Anlage, S.M. et al., Gronsky, R.et al. Krishnan, K.M. et al. 1: 454 Krishnan, R. see Garg, S.P et al., Raghavan, V. et al. Knvitskii, V.P. see Muller, Ch. et al., Ziesche, P. et al. Krtvoglaz, M.A. 1: 28, 761, 767
913 Kriz, R. see Ternes, J.K. et al. Kroeiner, H. 2: 335 Kroger, F.A. ,we Panicker, M.P.R. et al. Kroll, D. 1: 614 Kroll, S. et al. 3: 290 Kronberg, M.L. 3: 405, 415 Kronmiilles, H. 2: 484; see also Guo, H.Q. et al. Kropotova, N.V. see Semenova, A.D. et al. Kroto, H.W see David, W.1.F et al. Krotov, 1.1. see Vanyukov, A.V et al. Krshizhanovski~,R.E. J: 806 . see Wessels, J.F. et al. Krueger, D.D. et al. 1: 655 Kriiger, J. see Cox, G. et al. Kruger, P. et al. 1: 132 Kruisman, J.J. see Vitek, V et al. Krursman, J.J. et al. 1: 598, 600, 601, 602; 2 28 Uruml, T. et al. 3: 442, 443 Krumrnacher, S . see Dosch, H. et al. Krummacher, S . et al. 3: 218 Kruinme, J.P 2: 437, 441. 443 Krus~~i-E~bauni, L. see Harper, J.M.E. et al. Krusin-Elbaum, L. et al. 2: 623 Krutenat, R.C. 2: 291 Kryliouk, 0. et al. 3: 669 Kszanowski, J.E. 1: 586, 587, 591, 594, 865, 866 Ku, H.C. see Hsu, Y.Y. et al. Ku, W.C.et al. 3: 103 Ku, R. 1: 612 Kubaschewsk~,0 . 1 : 69,70, 101, 107, 110, 115, 117, 681; 2: 642; 3: 800; see also Barin, I. et al.; Elford, L. et al. Kubaschewski, 0. and Catterall, J.A. 3: 799 Kubaschewski, 0. et al. 1: 97; 3 799 Kubm, L.P. et al. 1: 552 Uubis, M. et al. 3: 98 Kiibler, , 63, 69, 71, 137, 920; see also r, T. et aE,, Terakura, K. et a ams, A.R. et al. Kiibler, J. et al. 1: 71 Kubo, €3. 2: 836, 838 Kucherenko, L.V. 2: 983, 987 Kuczynski, G.C, et al. 2: 564 Kudman, 1. see Dismukes, J.P. et al. Kudo, T. 1: 35, 295 Kudryavtsev, Y.V and Semenova, E.L. 3: 56 Kuech, T.F. see Lau, S . S . et al. Kuentzler, R.K. see Waterstrat, R.M. et al.
Kuhlmann-Wi~sdorf,D. 2: 261; 3: 361, 362, 363 Kuh~mann-~ilsdorf, D. see Mays, C.W. et ul. Kuhn, W.E. et al. 3 727 Kuhrt, C. see Katter, M. et al., Schrclpf, H. et al.; Schultz, L, et al. Kuhrt, Cb. et al. 3: 527 Kui, H.W et al. 3: 685 Kui, H.-W. et al. 1: 733
Kuiper, A.E.T. see Duchateau, J.P.W.B. et al.; van der Kolk, C.J. et al. Kuji, T. see Aizawa, T. et al.; Harada, T. and Kuji, T. Kukimoto, H. see Mara, K. et al., Yasuda, T. et al. Kulalovskii, V.D. s’ee Klakov, M.P. et 01. Kulcke, H. see Jin, R. et al. Kulp, D.T. et al. 1: 820 Ku~shreshtha,S.U. et al. 3: 78 Kumagai, M. 2 424, 425 Kumagai, T. see Abe, E. et al.; T a k e y a ~ aM. , et al. Kumar, D. et al. 1: 610 Kumar, U.S. 1: 530, 647, 864, 912, 924, 925; 2: 156, 1.57, 158, 159, 160, 161, 163, 164, 165, 166, 167; 3: 328, 361, 485, 486,495, 657; see also Brown, S.A. et al.; DiPietro, M.S. et al.; Douin, J. et al., Eberhart, M.E. et al., Pang, L. and Kumar, K.S., Whittenberger, J.D. et al. Kumar, K.S. et al. 2: 62, 64, 65, 165, 294 Kumar, M. 1: 775; see also Balk, T.J. et al. Kumar, M. and Hernker, K.J. 3: 443,444, 447 Kumar, M. and Vasudevan, V.K. 3: 418 Knmar, P. 2: 213 Kumar, R. see Saboungi, M.L. et al.
Kummerle, E. et al. 3: 279, 281, 282, 298 Kumnick, A. 2: 291 Kumok, L.M. see Larikov, L.N. et al. Kumpfwt, J. et al. 2: 1 10, 112, 11 3 Kunc, K. 1: 156 Kunc, K. et al. 1: 152 Kuncser, V. et al. 3: 100 Kung, H. 1: 588, 590 Kung, H. et al. 1: 586, 587, 588 K~nitomi,N. see Yarnada, T. et al. Kunzi, U. see Busch, G et al. Kunzler, J.E. 1: 16 Kunzler., J.E. et al. 1: 16; 2: 351, 360 Kuo. K. see He, L.X. et al. Kuo, K.H. 1: 467,470,484,491; see also , L.X. et al.; Zou, X.D. et al. 1.
Kuokka~a,V.-T. and Schwarz, R.B. 3: 697 Kuper, A.B. et al. 1: 564, 575, 767 Kupreev, V.P see Kablov, E.N. et al. Kuramoto, E. 1: 498, 528, 529, 534, 547; 2: 17, 22. 24, 25; see also Suzule, K. E.; Takeuchr, S. et al. Kuranov, A. et al. I: 788 Uurath, P. see Lofvander, J.P.A. et al.; Vasudevan, V.K. et al. Kurdjutnov, G.V. 1: 828 Kurihasa, S . 2: 507 Kurihara, T. see Hashimoto, T. et al. Kurimoto, T. see Shibuya, A. et al. Kuriyama, W. see Sakai, T. et al. Kurmaev, E.Z. see Anisimov, V.I. et al., Galakov, V.R. et al. Kurnakov, N.S. 1: 7, 8, 9, 14
Kurnakov, N.S. et al. 2: 559, 562 Kuroishi, K. see Kiyoshi, T. et al. Kurosawa, K, et al. 2: 410 Kursumovic, A. see Toloui, B. et al. Kurti, N. et al. 3: 522 Kurtzig, A.J. see Sherwood, R,C. et al. Kuruvilla. A.K. 1: 970; 2: 28; see also Stoloff, N.S. et al. Kuruvilla, A.K. et al, 1: 970 Kusaba, K. see Fujinaga, Y. et al. Kusaka, K. 1: 925 Kussman, A. 2: 308 Kussmann, D. et al. 3: 93, 95 Kusunoki, H. see Oota, A. et al. Kutolin, S.A. and Kotyukov, V.I. 3: 812, 824 Kutolin, S.A. et al. 3: 812, 824 ~ u y a m a J. , see Shingu, P.H. et al. Kuylenstierna, U. see Anderson, S. et al, Kuzma, J.B. Pt al. 1: 412 Kuz’ma, Yu.B. 2: 312; see also Gladyshev~kii,E.I. et al., I ~ n i t s k a ~ a , 0.N. et al.; Markrv, V.Ya. et al., Voroshilov, Yu.V. et al. Kuz’ma. Yu.B. et al. 1: 394 Kuzrnenko, P.P. et al. I: 765 Kuz’min, M.D. see Mikitin, $.A, et al. Ruz’min, M.D. and Tishm, A.M. Kuznetsov, A. see hayev, R. et al. Kuznetsov, A.V see Imayev, R.M. et al.; Salishchev, G.S. et al. Kuznetsov, F.A. see Zemskov, V.S. et al. Kuznra, J.N. see Wowchak, A.M. et al. Kwo, J. see Hong, M. et al.; Hsieli, Y.F. et al.; Majkr~ak,C.F et al. Kwo, J. et d.3: 667 Kwok, C.-K. see Pokela, P.J. et at. Kwok, S.P. 2: 621 Kwok, T. 2: 654 Kwon, S . see Cogan, S.F. et al. Kwon, Y.U. see Corbett, J.D. et al. Kwon, U.U. et al. 3: 89 Ky, H.G. see Csegory, E. et al. Kyogaku, M. et al. 1; 217 Kyvik, T. see Ulvens~en,J.M. et al. Laag, R. see Maurer, R. et al. Laag, R. et al. 1: 640 Laan, G.V. see Dun, H.A. et a/. Labarge, J.-J. et al. 2: 568, 572 LabbC, J. 1: 578, 794, 795 Laborde, 0. see Thomas, 0. et al.; Traverse, A. et 61. Labrador, M. see Mondielg, D. et al. : 260, 261, 263, 268, 273,
Lacerda, A. see Canfield, P.C. ei al.; Movshovich, R. et al. Lacerda, A. et al. 1: 217 Lachenal, R.see Joubert-Bettan, C.A. et al. Lacroix, C. see Martmez, G. et al. Lacy, D.E. et al. 2: 643, 644 Laegsgaard, E. see Sprunger, P.T. et al. Lagasde, P 1: 955; see also Sadoc, A. et al. Lagerborn, J. et al. 3 744 LaGoues, F.K. see Dimos, 2). et al. Lahav, A. 2: 621 Lahee, A.M. see Harten, U. et al. Lahin, S.K. and Gupta, D,
914 Lahrman, D. see Field, R. et al. Lahnan, D.F. 2 56,57; see also Darolia, R. et al. Field, R.D. et al.; Kumar, K.S. et al.; Schneibel, J.H. et al. Lahrman, D.F. et al. 2: 59, 61, 66 Lai, C.C. see Ku, H.C. et al. Lai, J.K.L. see Sun, J. et al. Lai, M.O. see Lu, L. and Lai, M.O.; Lu, L. et al. Lai, S.L. see Allen, C.W et al., Ramaaath, G. et al. Lame, T. see Saarinen, K. et al. Laird, C. see Bonda, N.R. et al. Lakin, D.M. 2 573 Lakso, G.E. 1: 507; 2 218 Lall, C . et al. 1: 547; 2 24 Lam, N.Q. 1: 88, 792, 817; see abo Devanathan, R. et al., Lutton, R.T. et al.; Shoemaker, J.R. et al., Zhu, H. et al. Lam, N.Q. et al. 1: 792. 817,819, 820 Lam, P. 2: 481 Lamarche, A.M. et al. 1: 356 Lamarche. G. see Lamarche, A.M. et al. Lambert, M. see Launois, P. et al. Lambert-Andron, B. see Courtois, D. et al. Lambeth, D.N. see Lee, L.-L. et al. Lambin, Ph. see March, N.H. et al. Lambrecht, A. see Brekkeler, K.H. er al. Lambrecht, W.R.L. see Amador, C. et al. Lamparter, P. et al. 1: 740 Lampman, S.R. 2 165, 367 Lamprecht, I. see Lechner, R.E. et al. Lanchester, P.C. 1: 1021 Lancon, F. 1: 482 Lancon, F et al. 1: 482 Landau, D. see Helbmg, W. et al., Schweika, W. et al. Landau, L.D. and Lifshitz, E.M. 3: 222 Lander, G.H. see Gschneidner, K.A. Jr. er al.; Stirling, W.G. et al. Landesman, J.P. 1: 295, 304, 579 Landesman, J.P et al. 1: 579 Landgraf, F.J.G. et al. 2: 316 Lando, Z.P. see Hu, J. et al. Landolt-Bornstein, 3: 241 Landsberg, P.T. 3: 805 Lang, C.I. 3: 61 Lang, C.I. see Towle, N.R. et al. Lang, C.I. and Dowyle, R.A. 3: 74 Lang, (2.1. et al. 3: 74 Lang, J. see David, J. et al. Lang, N.D. 1: 78 Langdon, T.G. 1: 918,925 Lange, H. et al. 2 622 Langreth, D.C. 1: 130, 663; see also Ashcroft, N.W. and Langreth, D.C. Lapasset, G . see Naka, S. et al.; Petez, L. et al. Lapertot, G . see Thessicu, C. et al. LaPlaca, S . see Heiney, P.A. et al. LaPlaca, S.J. see Penney, T. et al. Lapimova, R.V. see Sichevich, O.M. et al. Larbalestier, D. 2 353 Larchev, V.I. and Popova, S.V. 3: 161 Larese, J.Z. see Shapiro, S.M. et al. Larikov, L.N. see Bushin, I.N. et al.; Fal’chenko, V.M. et al.
Author Index Larikov, L.N. et al. 1: 757,761,763, 765, 766, 767, 768; 3: 290, 297, 300, 311, 312, 320, 805
Lark, K.A. see Prasad, Y V.R.K. et al.; Semiatin, S.L. et al. Larpin, J. see Gautiuer, V . et al. Larpm, J.P. see Gauthier, V et al. Larsen, D. see McQuay, P.A. and Larsen, D.; Seo, D. et al. Larsen, D. and Govern, C. 3: 592, 594, 595, 597
Larsen, D.E. 3: 595, 596, 601 Larsen, D.E. see Rishel, L.L. et al.; Seo, D.Y. et al. Larsen, D.E. et al. 2 85: 3: 597. 598 Larsen, J.M. 2 122, 123; see Worth, B.D. et al. Larsen, J.M. et al. 2: 93, 113. 289. 290, 292, 293; 3: 328, 338, 479, 480,483 Larsen. M. see Noebe, R. et al.: Rowe. R.G. et al. Larsen, M. et al. 1: 925 Larsen, P.K. see Verbeek, B.H. et al. Larsen, S.E. et al. 3: 328 Larson, A.C. see Cromer, D.T. et al.; Waber, J.T. et al. Larson, D.J. see Kim, S . et al. Larson, D.J. and Miller, M.K. 3: 419,478 Larson, J.M. et al. 3: 759 Lartigue, C. see Percheron-Guigan, A. et al. Laruelle. P see Rivet, J. et al. LaSalle, J.C. see Das, S.K. et al. LaSalle, J.C. et al. 2 194 Lasalmonie, A. 2 82; see also Hug, G. et al., Loiseau, A. and Lasalmonie, A. Lasalmonie, A. et al. 1: 496,497. 502,509: 2 58 LaSalvia, J.C. et al. 3: 736 Lasjaunias, J.C. see Berger, C. et al.; ZougmorB, F. et al. Laskaris, T . 2: 378, 379 Laskaris, T.E. see Gamble, B.B. et al. Lassen, N.O. 3 267 Lasser, R. 2: 477 Lasserre, A. see Reynauld, F. et al. Latanision, R.M. see Eberhart, M.E. et al. Lau, S.S. 2 606, 614, 616, 630; see also Harris, J.M. et al.; Hung, L.-S. et al.; LIU,B.-X. etal.; Marshal1,E.D. etal.; Mayer, J.W. et al.; Ottaviam, G. et al.; Par, C.S. et al.; Snni, I . et al.; Tsaur, B.Y. et al.; Tseng, W.F. et al.; Xia, W. ef al. Lau, S.S. et al. 2 621 Laudise, R.A. 2 395 Laue, M. et al. 1: 10 Lauf, R.J. 2: 637, 643 Laughlin, D.E. 1: 852.853,854,855, 867; 2 150; see also Cheng, S.F. et al.; Chnstian, J.W. and Laughlin, D.E.; Nesbit, L.A. and Laughlin, D.E.; Lee. Li-Lien er al.; Okamoto, H. et al., Subramanian, P.R. et al. Laugier, A. 1: 676 Laugier, J. see Chamberod, A. et al., Nkl, L. et al.; Paulevt, J. et al. Laumond, Y. see Tixador, P. et al. Launois, P. et al. t: 467 Laurent, Y. see David, J. et al.
Lautenschlager, E.P. see Hughes, T. et al. Lautenschlager, E.P. et al. 2 61 Lavemia, E.J. see Jeng, Y.-L. et al. Laves, F. 3: 3, 4, 404, 415,416, 812 Laves, F.J. 1: 14, 107, 112, 248, 249, 254. 281, 294, 309, 392,40511,
409,961
Lavine, M.C. 1: 181; see also Warekois, E.P. et al. Lavoie, C . see Svilan, V . et al. Lavrentev, V.L. 3: 800 Law, C.C. 1: 504,2: 59, 60, 64, 294; see also Miracle, D.B. et al.; Russell, S.M. et al. Law, C. and Blackburn, M. 3: 298 Lawandowski, J. see Maloy, S. et al. Lawler, J.F. see Nu, B.P. et al. Lawless, K.R. see Poon, S.J. et al. Lawley, A. et al. 2: 206 Lawnizak-Jablonska, K. see Czyzyk, M.T. et al. Lawrence, J.M. see Hundley, M.F. et al., Movshovich. R. et al. Lawrence, S . and Giles, C.L. 3: 857 Lawrence, J.M. et al. 1:214 Lawson, K.J. 2 499; see also Nicholls, J.R. et al. Lay, K.W. see Fleischer, R.L. et al. Lazareth, 0.W Jr. see Welch, D.O. et al. Lazarev, E.M. see Bannov, S.M. er al. Lazarevic, D.P. see Vidoz, A.E. et al. Lazarus, D. 1: 765; see also Gupta, D. et al., Kuper, A.B. et al. Le, D.H. et al. 1: 41 Le, H.Q. see Goodhue, W.D. et al. Leake, J.A. 1: 743; see also Wangts, C.H. et al.; Yang, R. et al. Leamy, H.J. 1: 509, 510, 511, 733; 2 203 Leamy, H.J. and Dirks, A.G. 3: 666 Leamy, H.J. et al. 1: 496, 497, 504, 507, 875, 886, 887
Lebedeff, Y.E. 2 517 Lebenbaum, D. see Atnnony, V . et al. Lebensohn, R.A. and Tom. C.N. 3 418 Le Beyec, Y. see Baudin, K. et al.; Dammak, H. et al. Leboeuf, M. see Morris, D.G. and LeboeuC, M.; Moms, D.G. et al., Morris, M.A. and Leboeuf, M. Lebowitz, J.L. see Binder, K. et al. Lebrat, J.-P. et al. 3: 656 Le Breton, J.M. see Steyaert, S . et al. Le Caer. G. see Cunat, C. et al.; Dubois, J.M. et al. Leccabue, F. see Sanchez, J.L. et al. LeChatelier, H. 1: 7 Lechner, R.E. et al. 3: 257 Lechtman, H. 1: 3 Leciejewicz, J. 1: 394 LeClair, S.R. see Jackson, A.G. et al.; Pao. Y .H. et al. Leclerq, G. see Bronoel, G. et al. Leclerq, L. see Bronoel, G. et al. Ledbetter, H.M. et al. 2: 284 Lederich, R. see Lu, G.-Y et al. Lederich, R.J. see Soboyejo, W.O. et al.; Ye, F et al. Ledig. L. et al. 3: 104 Lee. B.R. see Chao. P.C. et al. Lee, B.W. see Lopez de la Torre, A. et al.; Seaman, C.L. et al. Lee, C.-C. et al. 2 654
Author Index Lee, C.M. see Neurnann, J.P. et al. Lee, C.P. et al. 2: 343 Lee, C.-S. see Cheng, J. et al. Lee, C.-Y. and Lin, K.-L. 3: 671 Lee, C.S. see Sun, J. et al. Lee, D.N. see Choi, C.-H. er al.; Park, S.J. et al. Lee, D.N. and Kim, H.S. 3: 647 Lee, D.W. see Berkow~tz-Mattuck,J. et al. Lee, E.H. see Liu, C.T. et al. Lee, E.W. 1: 991, 992, 993; 2: 390; see also
Cook, J. et al. Lee, E.W et al. 1: 1007 Lee, H.G. 1: 994 Lee, H.-J. see Choi, C.-H. et al. Lee, J. 1: 652 Lee, J. et al. 1: 611, 616, 617 Lee, J.-H. see Rozgonyi, G.A. et al. Lee, J.H. aiid Thadhaiii, N.N. 3: 725 Lee, J.I. et al. 3: 486 Lee, 3.K. 1: 920; see also Yoo, M.H. and Lee, J.K., Yoo, M.H. et al. Lee, J.K. and Yoo, M.H. 3: 418,419 Lee. J.O. see Cava, R.J. er al. Lee, J.S. 1: 1000 Lee, J.W. et al. 3: 418 Lee, J.-Y. et al. 2: 133 Lee, K.J. 1: 727 Lee, K.N. 1: 988, 996 Lee. K.N. and Worrel, W.L. 3: 59 Lee, L.-L. et al. 3: 566 Lee, M.C. et al. 1: 690 Lee, P.W 2: 178 Lee, P.Y. 1: 701 Lee, R.W. 2: 314; 3: 102; see also Croat, J.J. et al. Lee, S. see Rainanath, G. et al. Lee, T. see Campisi, I.E. et al. Lee, T.K. see Kirn, J.K. et al. Lee, T.K. et al. 3: 653 Lee, T.L. see Chen, L.J. et al. Lee, T.S. see HSU,S.E. et al. Lee, T.S. et al. 3: 368, 494 Lee, T.Y.R. see Touloukian, Y.S. et al. Lee, Y.H. see Ojirna, M. et al. Lee, Y.T. see Kunipfert, J. et al.; Peters, M. et al. Lees, M.R. 1: 221; see also Roy, S.B. ef al. Lefebvre, F. 1: 791; see also Motta, A.T. et al. Lefebvre, S. see Quilichini, M. et al. LeFevre, B.G. 2: 282 Le Fevre, B.G. see Mitchell, R. et al. Legg, G.J. 1: 1021 Leggieri, G. see D'Anna, E. et al. Legnini, D.G. see Das Gupta, A. et al. LeGoues, F.K. see Gas, P et al. Legrand, B. see Meunier, I. et al., Rosato, V. er al. Legrand, P. see Diniitrov, C. et al.; Dunlop, A. et al. Legros, M. see Paidar, V et al. Legros, M. et al. 3 450, 463 Legvold, S. 1: 446, 447; 2: 390, 391 khmann, C. and Ziesche, P. 3: 808 Lehmann, H.W. 2 332 Lehnert, G. 2: 491 LeHolm, R. et d.3: 651 Lehtinen, B. see Inoue, A. et al. Lehtonen, M. see Lagerbom, J. et al.
Lei, J.L. see Qin, Y.L. et al., Yang, W.G. et al. Lei, M. see Migliori, A. et al. Leidheiser, H. 2: 506, 507, 511 Letdheiser, H. et al. 2: 507 Leighly, H.P. Jr. see Jackson, R.O. et al. Leinfelder, K.F 2: 565 Leinfelder, K.F. et al. 2: 565 Leisure, R.G. see Foster, K. et al.; Migliori, A. et al. Leitgeb, R. see Ceding, R. et al. Leithe-Jasper, A. see Coey, J.M.D. et al.: Weitzer, F et al. Leithe-Jasper, A. et al. 3: 100 Lele, S. 1: 471, 750; see also Cliattopadhyay, K. et al., Muraleedharan, K. et al., Ramachandra Rao, P et al. Lelovic, M. see Zbang, B. et al. Lcmaigaan, C. 1: 791, 814, 817; see also Motta, A.T. et al. Lemaire, L.P. see Kear, 33.11. et al. Lemaire, R. see Deportes, J. et al. Lernkey, F.D. 2: 651, 654; see also Pearson, D.D. et al. Lemkey, F.D. and Machlin, I. 3: 307 Lemoine, V. et al. 1: 986, 987 Lernoisson, E. see Ansara, I. et al. Lernski, J. see DiPasquale, J. et al. Lenz, K. 2: 402, 405 Leon, J.T. see Wright, C.R.A. et d. Leonard, B.F. 1: 629; see also Desborough, G.A. et al. Leonelli, R. see Bigot, J.Y. et al. Leon-Escarnilla, E.A. see Corbett, J.D. et al. Leon-Escarnilla, E.A. and Corbett, J.D. 3: 90 Leonowicz, M. see Manaf, A. et al. Leontis, T.E. 2: 651 Lepeshkov, I N . 1: 7 Lequette, R. sec Guyot, P et al. Lequeux, M.J. see Lasalmonie, A. et al. Lerch, B.A. see Noebe, R.D. et al. Lerf, R. 1: 531, 532; 2: 150, 151, 153, 159, 162, 164, 165; see also Morris, D.G. and Lerf, R., Morris, D.G. et al. Lerf, R. and Morris D.G. 3: 444 Le ROUX,R. 1: 886,889: see also Cabarat, R. et al. Le Roy, J. see Moreau, J.M. et al. Leroux, C. see Cadeville~M. et al. Leroux, C. et al. 1: 511, 947 LeSar, R. see Najafabadi, R. et al. Lesbats, P. see Paris, D. et al., Weber, D. et al. Leslie, W.G. 2: 11 Lesoult, G. see Grossiord, C. et al. Lessrnann, E. see Maier, K. et al. Lester, L.F. see Chao, P.C. et al. Lesueur, D. 1: 567, 693, 733, 750; see also Audouard, A. et al., Barbu, A. et al., Dimitrov, C. et al., Damrnak, H. et al., Dunlop, A. et al. Lesueur, D. and Dunlop, A. 3: 270, 271 Letant, A. see Allernand, J. et al. Lethuillier, P. see Yavari, A. et al. Letzig, D. see Palm, M. et al. Leubolt, R. et al. 1: 566 Leung, P. see Boily, S . et al.
91 5 Leung, P.K. 1: 741 Leung, P.W et al. 1: 482 Leuser, J. 2: 559, 560, 564 Leutheusser, E. 1: 679 Levalois, M. see Defour, C. et al. Leverant, G.R. see Kear, B.H. et al.; Paslay, P.R. et al. Leverant, G.R. and Duhl, D.N. 3: 302 Levey, F.C. et al. 3: 56, 77 Levi, A.F.J. see Hensel, J.C. et al. Levi, C.G. see De Graef, M. et al., Heredia, F.E. ef al.; Hyman, M.E. et al.; ~ c C u l l o u ~C. h , et al., Rosler, J. et al., Valencia, J.J. et al. Levin, E.M. et al. 1: 949,950; 3: 533, 534 Lcvin, E.S. see Ayushiiia, G.D. et al.; Petrushevskii, M.S. et al. Levin, K. 1: 214 Levin, L. see Wein, W. et al. Levine, D. 1: 480 Levine, D. and Steinhardt, P.J. 3: 389 Levine, D. et al. 1: 461 Levine, L.E. see Gibbons, P.C. et al.; Holzer, J.C. et al. Levine, L.E. et al. 1: 465, 477, 480 Levinsky, Yu. V. 3: 155, 159, 161 Levtnson, L.M. 2: 654 Levtnstein, H.J. et al. 2: 224 Levit, V.I. aiid Kaufnian, M.J. 3: 371 Levttov, L.S. 1: 463; see also Kalugin, P.A. et al. Levshin, G.A. 1: 69 Levy, J. see Paris, D. et al. Lewandowski, J. see Maloy, S. et al. Lewandowski, J.J. 3: 491; see also Bewlay. B.P. et al.; Hardwick, D.A. et al.; Mendiratta, M.G. et al., Rigney, J.D. and Lewaiidowski J.J., Rigaey, J.D. et al. Lewandowski, J.J. and Lowhaphandu, P. 3: 368 Lewandowski, J.J. et al. 2: 64 Lewis, B.G. 1: 735 Lewis, D. see Du, H.L. et al. Ley, L. see Cardon, NI. and Ley L. Leyenaar, F.A. see Geissberger, A.E. et al. Leyeiis, C. et al. 3: 570, 571, 577 Leys, M.R. see Weegels, L.M. et al. L'HBritier, P. see Fruchart, R. et al. LI, C.F see Wemberger, D.A. et al. Li, C . G . see Froes, F.H. et al. Li, C.H. 2: 217 Li, C.-H. see Chen, N.-Y. et al. Li, D. see Park, D.G. et al. Li, F. see Yu, S. et al. Li, F.H. sce Yaiig, Y. et al. Li, G.B. see Dong, C. et al., Sui, H.X. et al. Li, €3. 1: 655; see also Yang, Y.-C. et al. Li, H. and Chaki, T.K. 3: 492 Li, €3. et al. 2 136 Li, H.-D. see Huang, L.J. et al. Li, H.H. 3: 805 Li, H.-S. 2: 310. 314, 315; .see also Cadogan, J.M. et al.; Courtois, D. et aE., Han, X.F et al., Hu, B.P. et al.; Margarian, A. et al.; Wang, Y.Z. et al., Yang, F.M. et al. Li, H.S. and CadOgdn, J.M. 3: 179 Li, H.S. et al. 3: 94 Li, J. see Olowolafe, J.O. et al.
916 Li, J. et al. 2: 417, 606, 609, 622 Li, K. see Larsen, J.M. et al. Li, M. 2: 155; see also HLI,Y.Z. et al. Li, N.C. see Zhang, Z. et al. Li, S.H. see Han, Y.F. et at. Li, S.J. see Chen G.L. et al. Li, T. 3: 731; see also Chen, S. et al. Li, T.F. .Tee Wu, W.T. et al. Li, W. see Sury~narayana,C. et al. Li, W. et al. 3: 758 Li, W.B. et al. 1: 913 Li, X.F see Tichelaar, F.D. et al. Li, X.F. and Fan, T.Y. 3: 397 Li, X.G. see Chiba, A. et al. Li. X.-P et al. 1: 128 Ln, X.Z. I: 467; see also He, L.X. et al. Li, Y. see Zhang, T. et al. Li, Y et al. 1: 736 Li, Y.H. see WO, T.L. et al. Li, Y Y. 1: 504 Li, Y.-Z. see L1, T. Li, Z.C. see Whang, S.H. et al. Li, Z.W. and Morrish, A.H. 3: 98 Li, Z.X. see Whang, S.H. et al. Liang, B. see Shen, B.G. et al. Liang, C.C. 2: 510 Liang, J. see Tang, W. et al. Liang, J.K. see Huang, F et al.; Rao, G.H. et al. Liang, J.M. see Chen, L.J. et al. Liang, K.M. et al. 1: 913 Liang, K.S. see Rivers, S.B. et al. Liang, R.C. see Ho, J.C. et al. Liang, W. see Nash, P.G. et al. Liang, W . ~ 1.: 718, 727 Liao, J.J. see George, E.P. et al. Liao, L.X. et al. 2 315 Liao, P.R. see Vanyukov, A.V. et al. Libbert, J.L. 1: 478 Libbert, J.L. et al. 1: 473,477 Libby, W.F. see Darnell, A.J. et al. Libei-nlm, D.S. see Das et al. Libsch, S.F. et al. 2: 307 L i c h ~ e n b e F. ~ ~see , Ziittel, A. et al. Lichter, B.D. 1: 118 Lide, D.R. 1: 920; 3: 809 Ltdiard, A.B. 1: 761, 762, 765 Liebau, F see Lima-de-Faria, J. et al. Liebeman, D.S. see Fishman, S.G. et al.; Gupta, D. et al.; Schmerling, M.A. et al.; Wechsler, M.S. et al. Liebei-mann, H.H. 1: 733, 735 Liebertz, H. see Wollgarten, M. et al. Liebetrau, J, see Sikka, V.K. et al. Liebowitz, H. I: 882 Liebscher, H. 3 670 Lied, A. see Dosch, H. et al. Lieke, W. see Steglich, F. et al. Lienard, G, see Crucq, A. et al. Lieser, K H . I: 108 Lifshitz, E.M. see Landau, L.D. and Lifshitz, E.M. Lifshitz, I.M. 1: 866; 2 10, 263 Lightsey, G.L. see Shires, P.J. et al.; Timmons, C.F et al. Lile, D.L. et al. 2 335 Li~ienfe~d, D.A. I:454; see also Nastasn, M. et al. Lilienfeld, D.A. et al. 1: 454 Lifl, J.V. et al. 3: 841 Lillibridge, S . see Campbell, J.P. et al.
Author Index Lilly, A.C. see Deevi, S.C. et al. Lilly, A.C. et al. 3: 503 Liloyd, D. see Palmer, I.G. et al. Lim, S.P. et al. 1: 215; 2: 448 Lima-de-Faria, J. 1: 282; 3: 4, 5, 6, 7, 8, 10, 11, 12, 16 Lima-de-Faria, J. et al. 1: 300, 313, 317 Liming, W. see Dong, Z. et al. Lirnoge, Y. see Bocquet, J.L. et al. Limoge, Y. et al. 1: 817 Limoncelli, E.V. see LaSalle, J.C. et al. Lin, C.C. see Heaton, R.A. et al. Lin, C.J. 1: 733 Lin, C.S. see Sekhar, J.A. et al. Lin, D. 1: 88, 928, 989; 3: 505; see also Liu, Z. et al. Lin, D.L. et al. 2: 30 Lin, G. see Yu, S. et al. Lin, H. see Lin, D.L. et al. Lin, I.J. see Hida, G.T. and Lin E.J. Lin, J. 1: 453; see also Chen, M. et al. Lin, J.C. see Kattner, U.R. et al., Perepezko, J.H. et al. Lin, T. see Hiirnstrom, S.E. et al. Lin, T.L. and Zhang, Y. 3: 352 Lm, W. see Freeman, A.J. et al. Lin, W. et al, 1: 68, 69, 72, 137, 144 Lin, X. et al. 3: 56 Liii, Y see Yu, S. et al. Lin, Y.-C. el al. 2: 465 Lince, J.R. 2: 621 Lince, J.R. and Williams, R.S. 3: 783 Lincoln, F.J. see Hiraga, K. et al. Lrncofn, G.A. see Vojak, B.A. et al. Lind, D. see Jamison, K. et al. Lindau, I. see Shen, Z.X. et al.; Yeh, J.J. and Lindau 1. Lmdberg, B. see Seigbahn, K. et al. Lindberg, J.F. see MoEett, M.B. et al. Lmdberg, P A P . see Shen, Z.X. et al. Lindblad, N.R. see MeCarron, R.L. et al. Linde, J.O. 2: 562, 564 Lindemann, F.A. 1: 1023 Lmdgren, I. 1: 130; see also Seigbahn, K. et al. Lindley, W.T. see Alley, G.D. et al.; VOjak, B.A. et al. Lindsey, T.F see Klhachaturyan, A.G. et al. Lindstrom, R.M. see Neumann, D.A. et al. Ling, F. 2: 584 Ling, I. see Tomkiewicz, M. et al. Ling, R.G. see Belin, C. and Ling, R.G. Liiih, N.T. see Massies, J. and Linh, N.T. Linker, M.H. 2: 560, 571, 572 Linse, V see Waehtel, E. et al. Lintula, P see Lagerborn, J. et al. Liou, K.Y. 1: 784, 785, 800 Liou, S.H. see Koon, N.C. et al. Lipkin, D.M. and Clarke, D.R. 3: 573 Lipowsky, R. 1: 614, 615 Lipsitt, H.A. 1: 535, 654, 928; 2: 73, 83, 91, 93, 98, 102, 113, 115, 200, 203, 205; 3: 361; .see also Beday, B.P. et al.; Bewlay, P.A. et al.; Chang, K.M. et al., Dimiduk, D.M. et al.; Fbscher, R.L. et al.; Jackson, M.R. et al.; Martin, P.L. et al.; Mendiratta, M.G. et al.; Sastry, S.M.L. and Lipsitt, H.A. 3: 328 Shecht~an,D.
et al., Sheetman, R. et al.; Szaruga, A. et al. Lipsitt, H.A. et al. 1: 534, 535,654,919; 3: 417, 648, 707 Lipson, A, 1: 566, 568
Lissberger, P.H. 2: 438, 448 List, R.S. see C a m p u ~ n oJ.C. , et al.; Olson, C.G. et al. List, R.S. et al. 11: 136 Listovnichiy, V Ye see Semenova, E.L. et al. Litvin, D.B. see Kopsky, V. and Litvin, D.B. Litvin, E.G. see Larikov, L.N. et al. Litvinov, V.S. see Kuranov, A. et al. Lm, B.X. see Huang, L.J. et al. L~u,B.X. et al. 1: 733; 2 610 Liu, C. et al. 3: 771 Liu, C.H. see Fulap, 6. et al. Liu, C.T. 1: 498, 509, 511, 591, 593, 595, 603, 866,905,906,913,914, 924,925, 927, 928, 970, 986, 989, 993, 995; 2: 18, 20, 22, 23, 27, 28, 29, 30, 31, 32, 33,34,38,39,43,44,59,64, 137, 142, 143, 204, 288; 3: 656; see also Cahn, R.W. et al., Chang, Y.A. et al.; David, §.A. et al.; Deevi, S.C. et al.; George, E.P. et al.; Heatherly, L. Jr. et al.; Horton, J.A. et al.; Joslin, D.L. et al.; KO&, C.C. et al.; M c K a ~ ~ ~ , C.G. et al.; Maziasz, P.J. et al.; Nieh, T.G. et al.; Pike, L.M. et al.; Schneibel, J.H. et al.; Sikka, V.K. et al.; StoIoK N.S. et al., Taub, A.I. et al.; Veyssi&re,P. et al.; Wang, J.N. et al.; Whang, S.H. et al.; White, C.L. et al.; Yoo, M.H. et al. Lm, C.T. and Horton, J.A. Jr. 3: 316 Liu, C.T. and Pope, D.P. 3: 373,492,493, 656 Liu, C.T. and Sikka, V.K. 3: 604 Liu, C.T. et al. 1: 433, 591, 593, 594, 595, 897, 899,900,901,905,911,914,921, 924, 926, 928, 970, 971, 983, 989; 2: 17, 18, 27, 28, 29, 30, 38, 39. 43, 44, 204, 224; 3: 492, 604, 613, 614, 652, 654, 657, 843 L ~ uF. , see Baker, I. et al.; Nagpal, P. et al. Lm, G.C. see Wang, Y.Z. et al.; Yang, F.M. et al. Liu, H. 1: 784, 785 Liu, H.C. et al. I: 706 Liu, W.K. see Li, H.S. et al. Liu, H.L. see Han, X.F. et al. Liu, I.M. see Chu, J.P. ex al. Liu, J.Z. see Campuzano, J.C. et al. Liu, J.P. et al, 3: 100 Liu, L.Z. see Seaman, C.L. et ul. Liu, M.T. 1: 669 Liu, N.C. see Kamprath, €3. et al.; Wang,Y. et al.
Liu, Q.L. see Huang, IF. et al. Lm, R. see Olson, C.G. et al. Liu, S. see Chen, C.H. er al. Lm, S. see Compaan, A. et al. Lm, T.L. 1: 265
917 Liu, T.S. 2 217; see also Hsu, S.E. et al. Liu, W 1: 465 Liu, Y. 1: 611 Liu, Y. see Alonso, T. et al. L ~ uY , . et al. 1: 527, 528, 550; 3: 454 Lm, Y.F. see Chen, D. et al. Liu, Z. et al. 3: 572, 770, 772 Liu, Z.C. see Chen G.L. et al. Liu, Z.G. et al. 3: 758 Liu, Z.X. see Lin, C. et al. Livet, F. 1: 40 Livingston, J.D. 1: 916, 957; 2: 353; 3: 454, 461; see also Liu, Y. et al. Livin~ston,J.D. and Hall, E.L. 3: 415 Livingston, J.D. et al. 1: 537; 2: 238, 239; 3: 356 Ljubarsky, S.V. see Miroshnikov, M.M. et al. Ljungcrantz, H. see Karlsson, L. et al. Llanos, J. see von Schnering, H.G. et al. Lloyd, D.J. see Cbaturvedi, M.C. et al., Mukhopadhyay, N.K. et al. Lloyd, J.R. 2: 654 Lloyd, P. 1: 26 Lo,J. see Bionne, S. and Lo, J. Loboda, T.P. see Tsurikov, V.F. et al. Lobzov, M A see Alisova, S.P. et al.; Matveeva, N.M. et al. Locci, I.E. ,we Bowman, R.R. et al., Brady, M.P. et al.; Doychak, J. et ul.; Lewando~ski,J.J. et al., Raj, S,V et al. Locci, I.E. et al. 2: 61, 63. 64, 282 Lock, D.G. et al. 1: '137 Loeb, V.A. see Chambers, S A . and Loeb, V.A. Loeff, P.I. see Beke, B.L. et al. Loeff, P.I. et al. 2: 613 L o ~ e rF. y 2 410 Lofgren, E.J. see Frier, P.S. et al. Lofvander, J.P.A. see Court, S.A. et al.; De Graef, M. et al.; Valencia, J.J. et al. Lofvander, J.P.A. et al. 1: 53.5; 2: 103 Logan, J.C. 2: 402 Logan, R.A. see Hunt, N.E.J. et al.; Tsang, W.T. et al. Lograsso, T.A. see Shen, Z. et al. Loh, B.T.M. see Yoo, M.H. and Loh, B.T.M. Lohberg, K. 2: 95 Lohmann, M. see Knippscheer, S . et al. L o h ~ a n nW. , see Graf, K.H. et al. Lohneysen, H.V. see Gerrnann, A. et al. Loiseau, A. 2: 82; see iiLo Hug, G. et al., Leroux, C. et al.; Petez, L. et at.; Ricolleau, C. et al. Lomar, W.M. 1: 57111 Lombard, C.M. et al. 637 Loinmasson, T.C. see Long, G.J. see Hautot, D. et al. Lornpado, A. see Goela, J.S. et al. London, B. 2: 83 Long, J.V.P. I: 631 Long, P. see Fang, R.-Y et al. Long, R.G. 2 617. 618 Lonsdale, K. 1: 309 Loomis, RA. 1: 574 Loong, 6.-IS.see Stassis, C. et al. Lopatin, S . et al. 3: 671
Lopez de la Torre, A. et al. 1: 221 Lopez-Lopez, M. see Huerta-Rraelas, J. et at. Lopez-Rivera, S.A. see Hughes, 0.H. et al. Lo Piccolo, B. 3 443 Lord, J.S. et al. 3: I62 Lorenz, U. see Appel, F. et al., Wagner, R, et al. Lorenzelli, N. see Barbu, A. et al.; Dammak, H. et al., Dunlop, A. et al. Lorenzen, B. see Schillinger, W. et (d. Loretto, M . H ~1: 512, 527, 533, 536, 551, 983, 987, 988, 989, 990, 995, 996; 2: 57, 58, 98; see also Carnpany, R.G. et al.; Davey, S. et al.; Godfrey, A. et al.; Gouxna, P.X. et al.; Smith, L.S. et al.; Court, S.A. et al.; Fraser, H.L. et al.; Harneed, M.Z. et al.; Head, A.K. et al.; Johnson T.J. et al., Lofvander, J.P.A. et al., Kim, Y.-W. et al.; Sears, J.W. et al. Loretto, M.H. and Wasilewsk~,R.J. 3: 356 Loria, E. 3: 503 Loria, EA. see Sikka, V.K. et al. Lorich, A. see Clernens, H. et al. Lorimer, G.W 2: 282; sec also Cliff, G. et al.
Lorirner, G.W et al. 1: 835, 841 Lorusso, S. see Battaglin, G. et al. Losbichler, P. see Mitterer, C. et al. Losch, W. 1: 907 Lash, W.1: 604 Loshchinin, Yu.V. see Kovalev, A.I. et al. Lothe, J. 1: 524, 880, 881 Lothe, J. see Wirth, J.P. and Lothe, J. Lotspeich, J.E. 2: 330 Lottner, V. et al. 3: 253 Lou, W. see Taniguchi, S. et al. Lou, L. et al. 2: 218 LOU,T. et al. 3: 758 Loubradou, M. et al. 3: 413 Loubriel, G.M. et al. 2: 421, 432 Louchet, F. see Brechet, Y. et al., Viguier, B. et al. Louie, S.G. 1: 59, 130, 132, 133 Lours, P. 1: 551; see also Caillard, D. et al. Lovas, A. see Tarnbczi, T. et al. Love, A.E. 1: 874 Lovett, D.R. 2 329 Low, G.G. 1: 449,450 Low, T.R. 2: 510 Lowden, R.A. see Stinton, D.P. et al. Lowell, C.E. 1: 979, 980, 986, 987, 989, 990, 1008, 1009, 1010; sec also Barrett, C A . et al.; Hebsur P.M.G. et al.; Khan, A.S. et al.; Santoro, G.J. et al. Lowell, C.E. et U!, 1: 1007, 1008 Lowenthal, G.C. 1: 944 I Lowery, J.L. see Hanrahan R.J. Jr. et al. Lowhaphandu, P, and Lewandowsk~,J.J. 3: 684 Loxton, C.M. see Haasch, R.T. et al. Lozovskaia, A.V. see Larikov, L.N. et al. Lozovskaya, O.V. 2: 595, 596, 601 Lu, B.-H. 1: 745 Lu, D. see Eow, K . et rrl. Lu, D.C. and Pollock, T.M. 3: 56,58,370 Lu, G.H. see Wen, J.Q. et al.
Lu, G.Y. et al. 3: 345 Lu, H. et al. Qu, X. et al. Lu, I(.3: 755; see also Koch, C.C. et al.; Lu, L. et al. Lu, L. see Zhang, D.-L. et al.. Lu, L. and Lai, M.0. 3: 727,731 Lu, L. et al. 3: 730 Lu, M. see Hemker, K. et al. Lu, W.X. See Chen, N.-Y. et al. Lu, Z.W. see Galakov, V.R. et al.; Yeh, C.-Y et al. Lu, Z.W. et al. 1: 41, 61, 65, 67, 68, 69 Luan, K. see Hu, J. et al. Lubbehuseny M. and Mehrer, H. 3: 290 Lubell, M.S. see Dresner, L. et al. Lubensky, T.C. see Levine, D. et al. Lubensky, T.C. et al. 1: 461, 482; 3: 389 Luborsky, F.E. 1: 439, 449, 683, 734; 389 Luby, S. see D'Anna, E. et al. Lucas, G. see Rowe, R.G. et al. Luches, A. see D'Anna, E. et al. Luck, J.M. see AshraE, J.A. et al. Luck, R. 1: 97 Lucke, K. 1: 782 Liicke, R. see Kohl, W et al. Ludecke, C. see F r o ~ ~ e y eG.R. r , et al. Luderndn, E.M. see Zirnni, C.B. P I al. Ludl-uun, J. see Zhukovsky, A.Y. et al. Ludwig, M. and Gumbscb, P. 3: 769, 770 Lue, J.W. see Dresner, L, et al. Luedecke, C. see Fro~meyer,G. et al. Luevano, A.J. see Chellman, D.J. et al. Lugscheider, E. et al. 2: 221, 223 Luhman, T.S. 2: 358; ,Tee also Larson, J.M. et al. Luhvich, A A . 1: 956, 959, 960 LUI, G. see Sahoo, M. et al. Lui, S.-C. et al. 1: 137, 138; Lukas, H.L. 1: 69; 2: 54; see also Henig, E.-T. et al. Lukasak, D.A. 2: 107; see nlso Koss, D.A. et al. Lukashenko, G.M. et al. 1: 120 Lumsden, J.B. see Mysko, D.D. et al. Lunden, A. 3: 255 Lundqvist, B.I. 1: 197; see also ~ u n n d r ~ s o 0. n , el al. Lundqvist, S. 1: 129, 130, 132, 138 Lunm, V. see Nefedov, V. et al. Lunn, B. see Duddles, N.J. et al. Luo, H. see Ding, J. ef al.; Jeon, H. et al. Luo, H. et al. 3: 98 Luo, H.L. see Chao, C.C. et al. Luornajgrvo, M. see Suni, 1. et al. Lupinc, V. et al. 3: 297, 298, 300, 595 Lur, W. see Chen, L.J. et al. Luscher, E. see Huck~tein,K. et al. Luth, H. 3: 220, 221 Luthra, K.L. 1: 987, 992, 993, 999; see also McKee, D.W and Luthra, K.L. Luthy, W.see Salath~,R.P et al. Lutjering, G. 2 108; see also Ahrens, T. et al. Lutjenng, G. et al. 2: 112 Luton, M.J. see Garg, A. et al.; Whittenberger, J.D. et al. Lutskaya, N.V. 1: 724; see also Alisova,
S.P.et 611. Lutsyk, V.I. see Kiselyova, N.N. et al., Manzanov. Ye., E. et al.
918 Lutterotti, L. sec Gialanella, S. et al. Luttinger, J.M. 1: 128 Lutton, J.N. see Dresner, L. et al. Lutton, R.T. see Shoemaker, J.R. et al. Lutton, R.T. et al. 1: 88, 799 Lutwak, M. 2: 573 Luzzi, D.E. 1: 694, 733, 751, 809, 817, 820; see also Inui, H. et al., Kulp, D.T. et al., Sinkler, W et al. Luzzi, D.E. and Meshii, M. 3: 271 Luzzi, D.E. et al. 1: 812, 816, 817; 3: 415, 419 Lyapin, A.G. see Brazhkin, V V. et al. Lyer, S.S. see Thompson, R.D. et al. Lynch, D.W. see Olson, C.G. et al. Lynch, R. 2: 519, 520 Lynch, R.J. et al. 1: 928 Lynden-Bell, R.M. and Michel, K.H. 3: 254, 258 Lynn, K.G. see van der Kolk, G.J. et al. Lytle, F W 1: 986 Lyubitov, Yu.N. 1: 681 Lyuboserdova, J.E. see Matveeva, N.M. et al. Lyzo, A.P. see Muslimov, I.S. et al.
Ma, B.M. see Huang, M.Q. et al. Ma, C,L. see Yang, C.P. et al. Ma, E w e He, L. and Ma, E. Ma, E. see Miura, H. et al. Ma, E. et al. 3: 676, 749 Ma, Q. and Clarke, D.R. 3: 573 Ma. S. see Han, Y.F et al. Ma, T.-J. see Fang, R.-Y. et al. Ma, X.-L. see Zhang, D.-L. et al. Ma, Y. see Stern, E.A. et al. Ma, Z. see Ramanath, G. et al. Mabru, C. et al. 3: 337 Mabuchi, H. 2: 157, 158, 164; see also Hirukawa, K. et al. Mabuchi, H. et al. 2: 156, 157, 158, 160, 164, 165, 168; 3: 570, 577, 580 McAlister, A.J. 1: 47, 48; see also Bendersky, L.A. et al., Jones, S.A. et al. McAlister, A.J. et al. 1: 465 McAlister, S.P 1: 671 McAllister, K.G. see Austin, C.M. et al. MacAlpine, S. see Southwell, G. et al. McAndrew, J.B. 2: 77, 81, 82, 91 McCallum, R.W see Branagan, D.J. and McCallum, R.W.; Shield, J.E. et al. McCamont, J.W. see Bartlett, R.W. et al. McCance, A. 1: 8 McCandless, B.E. see Shafaman, W.N. et a/. McCann, J.F. 2: 5 11 McCarron, R.L. 3: 580 McCarron, R.L. et al. 1: 986, 987, 989, 990; 2: 57 McCartney, D.G. see Blackford, J.R. et al. McCarty, K.F see Mirkarimi, P.B. et al. McCaulay, J.P Jr. see Neumann, D.A. et al. McClellan, K. ,see Chu, F. et a/. McClelland, R.W see Vojak, B.A. et al. McClusky, R. et al. 2 377 McCombie, C.W 1: 762, 763; 2: 134; see also Elcock, E.W. and McCombic, C.W. McConnell, M. see Donnelly, N. et al.
McCooeye, D.P. see Howe, L.M. et al. McCord, D. 3: 31 McCormack, R. see Asta, M.D. et al. McCorrnick, A.S. see McCormick, S. et al. McCormick, M. see McCormick, S. et al. McCormick, P.G. 3: 727 McCormick, P.G. see Alonso, T. et al.; Schaffer, G.B. and McCormick P.G. McCorrnick, S. et al. 1: 924 McCoy, J. et al. 1: 782 McCrombie, C.W. 1: 575 McCullough, C. see Choi, B.W. et al.; De Graef, M. et al.: Hyman, M.E. et al.; Valencia, J.J. et al. McCullough, C. et al. 1: 638, 640; 2: 75, 83; 3: 594, 618, 619, 648 McCurne, R.A. 2: 308, 309 McDaniel, T. see McGahan, W.A. et al. McDaniel, T.W and Victoria, R.H. 3: 805 McDerinott, B. see Davis, R.M. et al. McDermott, B.T and Koch, C.C. 3: 731 McDonald, A.S. 2: 559, 560, 561, 562, 563, 568, 570, 571 McDonald, W see Smathers, L). et al. McDonald, W.K. 2 363 McDonnell, B.R. 2: 63; see also Hancock, G.F and McDonnel, B.R. MacDougal, D.P. ,Tee Giauque, W.F. and MacDougal, D.P. MacDougal, M.H. see Rohatgi, A. et al. McDougall, P.G. 1: 835 McDowell, A.F see Kiinmerle, I?. et al. McElroy, D.L. see Williains, R.K. et al. McEvily, A.J. see Bhowal, P.R. et al. McEwan, J.J. and Biggs, T. 3: 55 McEwan, W.S. 2: 507, 511 MacEwen, S.R. see Schulson, E.M. et al. McGahan, W.A. 2: 438, 439 McGahan, W.A. et al. 2: 435, 437, 443, 446 McGhie, A.R. see Zhu, Q. et al. McGill, I.R. 2: 489; see also Coupland, D.R. et al. McGill, I.R. and Selman, G.L. 3: 63 McGuiness, P.J. 2: 479; see also Hams, I.R. and McGuiness, P.J., Rozendaal, E. et al. McGuire, G.E. see Hsia, S.L. et al. McGuire, T.R. see Holtzberg, F et al. McHenry, M.E. see Dunlap, R.A. et al. Machatschki, F 3: 3 Machida. S. see Ito, T. et al. Machlin, E.S. 1: 604, 850; see also Lee, C.-C. et at. Machlin, E.S. et a/. 1: 242 McHenry, M. see Huang, M.Q. et al. Machon, L. 1: 917, 926, 927; see also Wunderlich, W McInerney, E.J. 3 664 Maciolek, R.B. see Bowers, J.E. et al. Mack, D.J. 1: 843 McKamey, C.G. 1: 927, 928; 2 199, 288; 3: 654; see also Koch, C.C. et al., Liu, C.T. et al.; Sikka, V.K. et al.; Stoloff, M.S. and McKarney, C.G.; Tortorelli, P.F et al. McKamey, C.G. et al. 1: 597, 907, 928, 1003, 1004; 2: 203, 205; 3: 502, 503, 572, 613, 614, 654, 663, 713 Mackay, A.L. 1: 473,480
Mackay, C.A. see Clarke, M. et al. MacKay, R.A. see Brindley, P.K. et al., Ndthal. M.V. et al. MacKay, R.A. et al. 1: 926; 2: 289 McKee, D.W. 1: 987, 988, 993, 996, 999: 2: 81, 82, 247, 251; see also Huang, S.C. et al. McKee, D.W and Fleischer, R.L. 3: 59 McKee, D.W and Luthra, K.L. 3: 577 McRelvey, A.L. et al. 3: 335, 336 McKelvey, A.L. see Badrinarayanan, E=. et al.; Campbell, J.P et al. Macken, P.J. 2: 594 Mackenzie, J.K. 1: 828, 829 Mackenzie, R.A.D. 1: 586, 903 McKeman, E. see Strauss, J.T. et al. McKernan, J.E. see Mistler, R.E. et al.; Strauss, J.T. et al. McKernon, J. see Deevi, S.C. et al. Mackey, B. see Sikka, V.M. et al. Mackin, T. and Yang, J. 3: 585 Mackintosh, A.R. 1: 441, 960 Macko, D. et al. 1: 457 McLam, M.E. see Okabe, T. et ai. MacLaren, J.M. see Eberhart, M.E. et al.; Woodward, C. et al. MacLaren, J.M. et al. 1: 285 McLarnan, T.J. 1: 242, 354, 356 McLLaughlin, D.L. see Loubriel, G.M. et al. McLaughlin? E.A. see Moffett, M.B. et al. McLean, D. 1: 591 McLean, M. 1: 868; 3: 542; see also Basoalto, H.C. et al., Cheng, G.H. et al.; Chou, C.T. et al.; Morsi, K. et al., Quested, P.N. et al. McLean, M. and Strang, A. 3: 306, 321 Macleod, H.A. x e Jin, R. et al. McMahon, C.J. Jr. 1: 897 McManamy, T.J. see Dresner, L. et al. McManus, G.M. 1: 762 McMasters, O.D. see Clark, A.E. et al.; Verhoeven, J.D. et al. McMasters, O.D. et al. 2: 394, 395 McMeeki~g,R.M. see Anton, D.L. et al., Bao, G. et al.; Sigl, L.S. et ai. McMichael, R.D. see Beimett, L.H. et al.; Herbst, J.F. et al. McMichael, R.D. et al. 3: 691 McMillan, P.F. see Ra~achandran,G.R. et a/. MacNairn, J.S. see Uhlig, H.H. et al. McNallan, M.J. 2: 218 McNally, C.M. see Nieh, T.G. et QE. McNaney, J.M. see Hong, M.-H. et al. McMeill, D.J. 2: 329 McNiff, E.J. Jr. see Flukiger, R. et a/. McPherson, D.J. 2: 134 McQuay, P.A. 1: 644; 2: 86; see also Dimiduk, D.M. et al., Martin, PL. et al., Semiatin, S.L. et al. McQuay, P.A. and Larsen, D. 3: 592 McQuay, P.A. and Sikka, V.K. 3: 503, 619 McQuay, P.A. et al. 2: 75; 3: 591, 594, 598, 619 Macqueron, J. see Elgueta, J. et al. McRae, E. 1: 614, 618 McRae, E. et ul. 1: 618; 3: 223 McRae, E.G. and Malic, R.A. 3: 222 McShane, H.B. see Morsi, K. et al.
A u t h o ~_Index McSkimin, H.J. and Andreatch, P. 3: 241 McTiernan, B.J. see Moll, J.W. et al. McWhan, D.B. see Kwo, J . et al.; Majkrzak, C.F. et al. Madar, R. see Fruchart, R. et al.; Thomas, 0. et al. Madar, R. and Bernard, C. 3: 669 Madelung, 0. 1: 319, 1026; 3: 806 Mader, S. see Seeger, A. et al. Mader, W. see Willman, N.et al. Madhava, M. see Calka, A. et al.; Giessen, B.C. et al. Madhukar, A. see Kaviani, K. et al. Madru, D, see Watako, Y. et al. Madsen, P.V. see Bertocci, U. et al. Madurga, V. et al. 3: 691 Maeda, H. see Xnoue, K. et al.; Kiyoshi, T. et al. Maeda, H. et al. 3: 527 Maeda, T. PI al. 2: 87 Maeda, Y. see Minemura, T. et al. Maehara, Y. 1: 925 Maeland, A.J. see Hauback, B.C. et al.; Kimmerle, F. er al. Maeland, A.J. and Libowitz, G.G. 3: 48 MaenpHa, M. see Suni, I. et al.; von Seefeld, H. et al. Maenpaa, M. et al. 2: 623 Maex, K. see Tung, R.T. et al. Maex, K. and Van Russum, M. 3: 808 Magerl, A. see Potzel, U. et al.; Salomons, E. et al.; Zabel, H. et al. Magerlem, J.H. see Callegari, A. et al. Maggiore, C.J. see Hanrahan, R.J. Jr. et al., Schwarz, R.B. et al. Magrni, M. et al. 3: 730. Maglia, F. see Bertolino, N. et al. MagnCli, A. 1: 405n, 409 MagiiCli, A. see Anderson, S. et al. Mah, T. see Lee, J.I. et al. Mahajan, S. 1: 844; 3: 805; see also Bask, D.P. et al.; Chin, G.Y. et al., Christian, J.W and Mahajan, S.; Shi, X. et al. Mahajan, S. and Chin, G.Y. 3: 422 Mahan, J.E. 2: 617, 618 Mahapatra, R. see Cook, J. er al.; Lee, E.W. et al. Mahapatra, R. et al. 3: 414, 424 Mahler, D.B. 2: 583, 585, 586 Mahler. D.B. et al. 2 577 Mahoii, G.J. 2: 75 Mahoney, M.W 1: 656 Maier, C.U. 2: 484 Maier, K, see Schaefer, H.-E. et al. Males, K, et al. 3: 290 Maier, U. see Burger, J.P. et al. Mai~ander,L. see Dosch, H. et al. Mains, E.F. see Markiewicz, W.D. et al. Nairy, C. see Balanzat, E. et al. Maisano, J.R. see Bryant, J.D. et al. Maita, J.P. see Bucher, E. et al. Maitrepierre, P.L. 1: 735 Majer, G. see Kimmerle, F. et al. Majewski, J.A. 1: 56; see also Hafner, J. et al. Majkrzak, C.F et al. 1: 450 Majni, G. see D'Anna, E. et al. Majumdar, B.S. see Boehler, C.J. et al. Majumdar, S. see Sampathkuinaraii, E.V. et al.
Makarova, G.M. see Ivanova, G.V. et al. Mak, A.A. et al. 2: 432 Makarnel, D. see Bihain, 0. et al. hlakarov, E.S. 1: 413; 2: 31 I Makh~ouf,S.A. see Xu, Y. et al. Maki, K. et crl. 1: 987, 993 Maki, N. see Fuchino, S. et al. Maki, T. 1: 835 Makin, M.J. 2: 259, 262, 281 Makm, P.L. 2: 188, 190 Makinen, J. see Alatalo, M. et al.; Saarmen, K. et al. Makino, A. see Suzuki, K. et al. Makino, A. et al. 1: 735; 3: 702 Makita, T. and Nagasawa, A. 3: 56 Makovicky, E. 3: 7; see Liina-de-Farm, J. et al. Makroczy, P. see Macko, D. et al. Maksimov, 1.1. see Batsanov, S.S. et al. Makyta, M. et al. 2: 507 Malagelada, J. see Baro, M. et al. Malakondaiah, G. 2: 121 Malaman, B. see Francois, M. et al.; Venturini, G. et d. Malas, J.C. see Prasad, Y.V.R.K. et al. Malcom, J.A. 1: 867 Malhotra, M.L. et al. 2: 584 Malic, R. 1: 614, 618; see also McRae, E. et al. Malic, R A . see McRae, E.G. and Malic, RA.; McRae, E.G. et al. Malik, M. see Restall, J.E. et al. Malik, M. et ul. 2: 495 Malik, R.J. 3: 805 Malik, S.K. 1: 217 Malins, A.E.R. see Brown, J.D. et al. Maliszewsky, N.C. see Neumaiin, D.A. el fd. Mallery, J.H. 2: 134 Maloney, M.J. 1: 1007; 2: 290, 295; see also Doychak, J. et al. Maloney, M.J. and Hecht, R.T. 3: 583 Maloy, S.A. and Gray 111, G.T. 3: 372, 373 Maloy, S.A. et al. 2: 221; 3: 334; 3: 356, 368,455 Malozenioff, A.P 1: 814 Malrieu, J.P et al. 1: 319 Malterre, D. see Grioni, M. et al. Maluff, E.I. and Cahn, R.W. 3: 54 Malyus~itskaja.Z.V. 3: 161 Malzfeldt, W. see Horn, P.M. et al. Manaf, A. see Otam, Y. et al. Manaf, A. et al. 1: 748, 749 Manaila, R. et al. 3: 665 Manchester, F.D. 3: 802 Mandal, R.K. 1: 471 ~ a n d a r i n oJ.A. , 1: 626, 627, 628, 630, 631 Mandrus, D. st~eSales, B.C. et al. Mandrus, D.G. see He, Y. et at. Manfrinetti, P. see Canepa, F. et al. Manfrinetti, P. et al. 3: 101 Manghi, F et al. 1: 132 Manfi, et al. 3: 58 M ~ n n a e r J.P. ~ ~ ,see Wong, M. et al.; Nsieh, Y.F et al.; Noh, D.Y. et al. Mannan, Kh.M. 1: 956 Mannan, S.K. 1: 647; see also Kumar, K.S. et al., Whittenberger, J.D. et al, Mannhart, J. see Dimos, D. et al.
919 Manninen, M. 1: 79; see also Puska, M.J. et al. Manning, J.R. 1: 574; see also Kuper, A.B. et al., Kupper, A.B. et al. Manos, J. see West, J. et al. Mansfield, M. see Needs, R.J. et al. Mansur. L.K. 1: 818 Mansuri, Q.A. 2: 604, 628 Mantovani, S. s p e Nava, F. et al. Mantravadi. N, 2: 200, 206 Manz, M. 1: 93 Manzanov, Y. E. et al. 3: 824 Mao, H.K. 2: 481; see also Hemley, R.J. and Mao H.K. Mao, W. see Yang, J. et al. Maple, M.B. 1: 216; see also Lacerda, A. et al., Lopez de la Torre, A. et al., Meissner, G.W et d . Seaman, C.L. et al. Marais, T.K. see Li, J. et al.; Pretorius, R. et al. Marancik, W. see Adam, E. et al., Sanger, P. et al. Marasinghe, G.K. see Luo, H. et al. March, N.H. 1: 58, 129, 674, 677, 682, 683, 684; see also Singh, R.N. and March, N.H. March, N.H. et al. 1: 682, 682 Marchionni, C. see Brechet, Y et al. Marchionni, M. see Lupinc, V et al., Lupnic, V. et al. Marchut, L. see Buck, T.M. et al. Marcinkow§ki, M.J. 1: 496,497,498, 500, 504,505, 507, 508, 509,521, 535, 588, 804, 913; 2: 218; 3: 422 Marcinkowski, M.J. et al. I: 15, 505, 529; 2: 19, 29, 74 Marcus, H.L. see Fine, N E , et al. Marcus, M.A. 1: 477 Marcus, P.M. I: 71; sec also Jepsen, D. et al.; Sondericker, D. et al. Marcus, R.B. see Calbrick, C.J. and Marcus, R.B. Marder, A.R. see Banovic, S.W et al. Mardix, S. 1: 279. 293, 345 Marezio, M. et al. 3: 56 Marfaing, Y. 2: 326 Margalit, S. see Lee, C.P. et al. Margarian, A. see Li, H.S. et al. Margarian, A. et al. 3: 94 Margolin, H. 2: 75; see also Sahin, 0.et al. Margulis, W see Gomes, A.S.L. et al. Mariano, A.N. 1: 172, 174, 17811, 184; see also Warekois, E.P. et al. Marillo, J. see Dunlop, A. et al. Markete, W.T. et al. 3: 426 Markiewicz, W.D. s e Miller, ~ J.R. et al. Markiewicz, W.D. et al. 2: 366, 367, 371 Markiv, V.Ya. 1: 718, 720 Markiv, V Ya. et al. 1: 727; Markosyan, A.S. see Hauser, R. et al. Markov, V.T. 1: 366 Marquardt, B. see Semiatin, S.L. ei al. Marquardt, B.J. 2: 599; see also Gray, G.T. III et al.; Krueger, D.D. et al.; Rowe, R.G. et al.; Wessels, J.F et al. Marquardt, B.J. et al. 1: 535; 2: 91. 93, 102, 122 Marr, J. see Saboun~i,M.L. et al.
920 arraza, R. see Ferro, R. et al. Marrijnissen, G. see Duret, C. et al. Marschall, J. see Choi, B.W. et al. Marsc~ner,J, see Paufler, P, et al. Marsh, R.E. see Shoemaker, D.P. et al. Marsh, R.E. and Shoemaker, D,P. 3: 254 Marshall, D.B. see Cox, B.N. et a/,
Marshall, G.R. see Xliughes, 1.R. et al. Marshall, G.W. 1: 574 Marshal arshal artelli artens Martin, artin, artin, G. 1: 701, 792, 817, 821; see also Bellon, B. et al.; Massobrio, C . et al., Yu-Zhang, K. et al.
Mason, D.P. see Cibala, R. et al. Mason, D.P. and Van Aken, D.C. 3: 486 Mason, H. Jr. see Levinstein, H.J. et al. Mason, P.W. 2: 396 Mason, T.E. et 1.11. 1: 218 Mass, S.C. see Shapiro, S.M. et al. Massalski, T.B. 1: 21,43,44,63, 105, 106, 115, 137, 638, 689, 843, 886; 307, 316, 607, 608, 613; 3: 67 547, 800; see also Murray, J.L. et al.; Qkamoto, H. et al.; Pei, S. et al. 147; 3: 22, 33, 85, 670, 701 Massaro, T.A. 1: 594 Mask, G. 1: 352 Massies, J. and Linh, N.T. 3: 783 Massing, W. 2: 506 Masso~rio,C. 1: 817, 820 Massobrio, C. et a/* 1: 817 Mastuda, K. see Nakagawa, Y.G. et ale
~ a t s u ~ oS ,t see ~ ,Kato, Nakamura, M. et al. Matsumoto, T, see Aoki, K. et al. Matsumoto, Y. see ~ i s h i ~ a , ~ a t s u m L ~ M. ~ o see , h i , H. Matsuna~a,S. et al. 1: 680
~ a s a h ~ sN. h ~et, al. Matsuo, S, see Mori, M. et al.
Matsuoka, 1". et al. 2: 339 Matsushita~A. see Tang, J. et al. ~ a t s u u r aT. , see Mori, M. et al. M a t s ~ L ~Y. a , see Sagawn, M. et al. Matsuyama, W. see Tak~hashi,T. et al. Mattas, R,F. 2 648 and Fame, G. 3: 732 Matteazz~,P. et al. 3: 724, 743 M a t t ~ ~W.C.M. ~s, see de a t t h ~ i ~L.F. s , 1: 67, 13 attheiss, L.F. and ~ a ~ m aD.R. n ~ 3:, 139, 140 (I
Martin, J,E. 1: 184; see also Owen, N.B. er al. Martin, J.F. 1: 117 Mar~in,J,J, 1: 1027, 1028 Martin, J.L. It 547; see also Baluc, N. et al.; Bonneville, J. et al.; Kruml, T.
Antoii, D.L. et al.; Dimiduk, D.M. et al.; Hard~ick,D.A. et al.; Patterson, R.A. et al.
Martin, P.M. see Sharp, J.W. et al. Martin, R. see ~ a l m s t r o iC.J. ~ , et al. Martin, R.M. 1: 156, 218; see also Li, x.-P. et al. Martin, T.P. 1: 681, 682
Martynov, A.J. I: 232,233, 243, 260,261, 424
R,S. et al. 1: 591, 593, 594; see ako
ee Craig,
Takasugi, T. et al. M a s a ~ ~N. s ~et ial. ~ 1: 638,657, 725, 898, 899, 900, 901, 925, 927; 2: 29, 30, 39, 86, 87 Masaki, K. see Nishitani, S.R. ef al.
Masuda, Y . 1: 1024 Masuda-Jindo, K. 1: 70, 569, 604, 955 Masumoto, K. 2: 332; see also K a w a s u ~ i , I. et al., Takahashi, S . et al. Masumoto, T. 1: 735; see also Aoki, K. et al.; et al.; Higashi, K. et al.; et a/.; Inouye, A. ct al.; , Y. et al.; Kim, 5.C. et al.; Kortan, A.R. et al.; Makino, A. et al.; Masum et al., Matsubara, E. et al.; et al.; Sumki, K. et al.; Terauchi, M, et ul.; Tsai, A.P. et al.; Zhang, T. et al. Masumoto, T. and Maddin, R. 3 694 ~ a s u m o t oY. , see Mishina, T. et al. Masumoto, Y. et ad. 1: 743 Mataga, P,A. see Sigl, L.S. et al. Matare, H.F, I: 958 Mateer, R.S. 2: 586 Matejczyk, D.E. 1: 973 Mathew, J.G.H. see Bigot, J.Y. et al.; Smith, 23.19. et al. Mathias, B.T. 3 811, 812 Mathias, B.T. et al. 1: 16 Mathis, K. see Hafner, J. et al.; Villars, P. et al, Matialu, X . see Boone, D.H. et al. Matiasovsky, K. see Makyta, M. er al. Matini, L. see Guilde, J. et al. Matlokhova, L.A. see Matveeva, N.M. et al. Matson, L. see Rigney, J.D. et al. Matson, R. see Albin, D.S. et al. Matsubara, E. et al. 1: 479 Matsuda, T. see Inoue, A. er al. Matsugi, K. Ip: 331, 338 Matsuka~a,T. see Mori, M. et al. Matsumara, T. et al. 3: 161 Matsumoto, K. see ~ashimoto,T. et al.; ~ a k a T. ~ et o al. ~ Matsumoto, T. see Matsumara~T. et al. Tang, J. et aL Matsumoto, N. see Inoue, A. et al. Matsumoto, 0.see Miyasa~i,S. et al.
also Giorgi, A.L. art al. Matthias, B.T. et al. 3: 45
et al. Matusita, K. sep Komatsu, T.et al. Matuszyk, W. et al. 1: 927; 2: 42; 3: 341, 347 Matveeva, N.M. 1: 711,722, 723, 728; see also Khachin, V.N. et al.; Savvinov, A.S. et al. Matveeva, N.M. et al.
sky, R.M. 2: 595, 596, 601
mat^^^^ H. see CJka, A. et al. Mau~uin1: 309
Maurer, R. et al. 1: 640, 773 M~LKW T., et d.2: 440 Maurette, M. see Fleissher, R.L. et al. M a u ~ c eD, , see ~ o u r t n e yT.H. ~ and Maurice D. Mauriee, D.A. and Court Maurice, D.A. 1: 647 Maurice, D,A, and Court ~ a v i t y J.T. , see Sikka, V.K. et al. Max, N. 1: 278 Maximenko, E A . see Larikov, L.N. et al. Maxwell, D.C. see Balsone, 5.J. et al.; Larsen, J.M. et al. Mayadas, A.F. 1: 956 M a y ~ ~ nS.M. a , see Srimath, Mayer, N.see Betz, U. et al.; et al.
Author Index Mayer, I, et al. Ip: 30 Mayer, J. see Csanady, A. et al.; Fahnle. M. et al.; Urban, K. et al. Mayer, J. and Eahnle, M. 3: 277,281,282 B. et al.; Colgan, E. et al.; Hong, Q.Z. et al.; Hung, L.S. et al.; Li, J. et al.; Lilienfeld, D.A. et al., Nastasi, M. et al.; Olowolafe, J.0, et al.; Ottaviani, G. et al.: Palmstro~,C.J. et al.; Zheng, L.R. Mayer, J.W. et al. 2: 610, 616, 617
Maynau, D. see Malrieu, J.P. et al. Mayou, D. see Klein,T. et al.; Trambly de Laissardiere, G. et al. Nays, C.W. et al. 3: 226 Mazdiyasni, S. 1: 1008; 2 295; see also Mendiratta, M.G. et al. Mazdiyasni, S . et al. 1: 66, 720; 157, 161 Maaiasz, P.J. see Alexander, D.J. et al.; Castagna~A. et al.; LIU, C.T. et al. Maziasz, P.J. and Liu, C.T. 3: 652 Maziasz, P.J. et al. 3: 503, 572, 613, 614, 652 Mazumdar, C. see Nagarajan, R. et al. Mazumdar, C. et al. 3: 104 Mazurkiewicz, B. 2 503 Mazzoldi, P. see Battaglin, G. et al. Mbaye, A.A. see Ferreira, L.G. et al.; Wei, S.-H. et al.; Zunger, A. et al. Mead, C.W. 1: 629 Meaden, G.T. 1: 945 Mearns, D. 1: 130 Mecking, H. 1: 917; 2: 81; 3: 361; see also Koeppe, C. et al.; Lebensohn, R.A. et al. Meckmg, fi. and Estrin, Y. 3: 361 Meckison, C.D. see Sun, H. et al. Medlin, D.L. 3: 410; see also Mirkarimi, P.B. et al. Medlin, D.L. et al. 3: 410 Meeking, H. see Bartels, A. et al. Meeks, M.B. 2: 402 Meetsma, A, et al. 1: 391 Megura, S. see Kamisada, Y. et al. Meiil, M,J. 1: 130, 195, 196, 207, 208; see also Boyer, L.L. et al.; Chen, J. et al.; Feldman, 3.L. et al.; Osburn, J.E. et al.; Siglas, M. et al. Mehl, M.J. et al. 1: 137, 195, 196, 198, Mehl, R.F. 1: 3, 773 Mehrabian, R. 2: 178; see also Cao, H,C. et al.; Choi, B.W, et al., Hyman, M.E. et al.; McCullough, C. et al.; Rosler, J. et al.; Valencia, J.J. et al. Mehrabian, R. et al. 2; 178 Mehrer, H. 3: 275, 422; see also Dorner, W. et al., Eggersrnann, M. et al.; Kroll, S . et al.; Kummerle, E. et al.; Maier, K. et al.; R u ~ m e lG. , et al. Mehrer, H. et al. 3: 282 Mei, L. see Hu, J. et al. Mei, X.B. see Yang, Y. er al. Mei, Y. see Dong, Z. et al.
Meier, G.W. 1: 980, 982, 987, 988, 989, 991, 992, 993, 994, 997, 999, 1000, 1008; 2: 17, 57, 227; see also Ashary, A. et al.; Betziss, D. et al.; Perkins, R.A. et al., Schaeffer, J. et al. Meier, G.H. et al. 1: 991, 992; 2: 81, 82 Meier, G. and Pettit, F 3: 576 Meier, G.N. see Berztiss, D.A. et al.; Perkins, R.A. and Meier, G.H. Meier, S.A. see Busch, G. et al. Meiiiers, L.G. 2: 335; see also Lile, D.L. et al. Meinhardt, D. 1: 398 Meinhardt. H. 2: 491; see also Lugscheider, E. et al. Meisel, K. 1: 399 Meisner, G.P. see Chen, B. et al.; Morelli, D.T. and Meisner, G.P. Meissner, G.W. et al. 1: 217 Meisterle, P 1: 782 Mejia-Lira, F. see M o r ~ n - L ~ p eJ. z ,et al. Melander, A. 2: 275, 276 Melendres, C.A. 2: 510 Meleshko, L.L. see Guzey, L.S. et al. Meli, F. see Chartouni, D. et al.; Schlapbach, L. et al.; Zuttel, A. et al. Meli, F. et al. 2: 486 Melloch, M.R. 2: 421, 422 Melmed, A.J. 1: 453 Melnyk, E.W. 1: 410 Melone, S. see Caciufl'o, R. et al. Melton, K1. see Duerig, T. et al., Goo. E. et al. Meltsner, K.J. 3: 857 Menand, A. 3: 424, 478; see also NercPartaix, A. and Mennand, A. Mendeleev, D.I. 1: 12 Mendes, P.J. see Ayres de Campos, J. et al. Mendicino, M.A. et al. 3: 668 Mendiratta, M.A. et al. 3: 488, 492 Mendiratta, M.G. 1: 504; 2: 115,200,203, 204, 205, 295; see also Dimiduk, D.M. et al.; Hazzledine, P.M. et al.; Henshall, G.A. et al.; Martin, P.L. et al.; Mazdiyasni, S. et al.; Rigney, J.D. et al.; Miracle, D.B. and Mendiratta, M.G.; ~arthsarthy,T.A. et al.; Rao, S.I. et al.; Subrarnanian, P.R. et al. Mendiratta, M.G. and Dimiduk, D.M. 3: 541, 545, 547, 550 M e n d i ~ ~ tM.G. t ~ , et al. 1: 536, 907; 2: 200, 203, 204, 205, 206, 295; 3: 368 Mendoza, W.A. and Shaheen, S.A. 3: 94 Mendoza-Zelis, L. see Traverse, A. et al. Menezes, S. et al. 1: 653 Meng, R.L. see Sun, Y.Y. et al. Meng, W.J. see Cotts, E.J. et al. Meng, W.J. et al. 1: 698; 2: 133 Meng, X.M. see He, L.X. et al. Mengucci, P, see D'Anna, E. et al. Menon, J. 1: 467 Menovsky. A.A. see Palstra, T.M.M. et al.; Sinnema, S. et al. Menth, A. 2; 313 Menth, A. et al. 1: 217 Mentzer, M A . et al. 2: 413, 432 Menzies, R.G. et al. 3: 544 Meray, G. see Kosonocky, W.F. et al.
92 1 Mercer, C . see Soboyejo, W.O. et al. Meriel, P. 1: 445 Merk, R. see Ronning, C. et al. Merlo, F. 3: 87, 95; see also Bruzzoiie, G. et al.; Fornasini, M.L. and Merlo, F. Merlo, F and Fornasini, M.L. 3: 87 Merlo, R. see Canepa, F. et al. Mermm, N.D. 1: 200,207,480,485, 1018 Merrel, R. et al. 2: 490 Merrick, H. see Seo, D.Y. et al. Merrick, H.F. see Larson, 3.M. et al. Merzhanov, A.G. 3: 725, 736; see also Podlesov, V V et al. Meschede, D. see Steglich, F. et al. Meschter, P.J. 1: 64, 65, 66, 67, 70, 499, 530, 1003, 1004, 1007; 2: 156, 157, 219, 224, 295; 3: 574; see also Carlsson, A.E. and Meschter, P.J. Meschter, P.J. et al. 3: 28 Meshii, M. 1: 694,733,751,809,814,815, 817, 819; see also Devanathan, R. et aE., Koike, J . et al.; Luzzi, D.E. et al., Xu, G-B. et al. Messer, R. see Dais, S. et al. Messerschmidt, U. see Feuerbacher, N. et al.; Guder, S. et al.; Rosenfeld, R. et al.; Urban, K. et al. Messerschmidt, U. et al. 3: 356, 346 Messick, L. see Lile, D.L. et al. Messmer, R.P 1: 156, 604, 907 Metahi, H. see Kanzawa, V. et al. Methfesse~,M. 1: 133; see also Alouani, M. et al.; Asta, M. et al. Metbfessel, M. et a/. 1: 133 Methfessel, M.S. see Verbeek, B.H. et al. Metropolis, N. et al. 2: 86 Metzger, P.H. 1: 172 Metmacher, C. see Feuerbacher, M. et al. Meunier, I. et al. 3: 224 Meuris, M. see de Potter, M. et at. Memtin, M. see Weber, D. ~t al. Meyer, B. see FPhnle, M. et al. Meyer, G. see Reppich, B. et al. Meyer, J. 2: 646; see also Ra~ganathan.S. et al. Meyer, L.W. see Hoke, D.A. et al.; LaSalvia, J.C. et al. Meyer, M.K. see Akinc, M. et al. Meyer, M.K. and Akinc, M, 3: 486, 490 Meyer, M.K. et al. 3: 486, 490 Meyer, 0 . 1: 579 Meyers, M.A. 1: 649; see also Ferreira, A. et al.; LaSaIvia, J.C. et al. Meyers, M.A. et al. 3: 724, 738 Meyers, P see Fulap, G. et al. Mezei, F. see Murani, A.P. et al. Miani, F. see Matteazzi, P. et al. Miceli, P.F. see Youngquist, S.E. et aE. Michael, J.R. see Baker, I. et al.; Frear, D.R. et al. Michaelsen, C. see Yan, Z. et al. Michal, G.M. see Lewandowski, J.J. et al. Michalak, J.T. see Arajs, S . et al. Michaut, B. see Elolmaiin, W. et al. Michel, D-3. see Feng, C.R. et al. Michel, E.G. see Castro, G.R. et al. Michel, K.H. see Lyndeii-Bell, R.M. mid Michel, K.H. Michor, H. see Hauser, R. et al.
922 Midson, S.P. et al. 2: 188 Miederna, A.R. 1: 116, 117,234,247,257, 258,264,269,366,381, 568, 569, 604, 612; 2: 609, 610; see also Buschow, K.H.J. et al.; de Boer, F.R. et al., Loeff, P.I. et al., van der Kolk, G.J. et al. Miedema, A.R. et al. 1: 116, 117, 385, 390, 391 Miers 1: 309 Miglion, A. et al. 3: 697 Mihalisin, J.R. 2 283 Mihama, K. see Saito, Y. et al. Mika, IS. 1: 282; see also Hawk, J. et al. Mikalopas, J. et al. 1: 41, 65, 68, 69 Mikami, 0. 2: 343 Mikawa, T. see Shirai, T. et al. Mikes, T.L. see Hunter, W.R. et al. Mikhailenko, S.D. et al. 3: 758 Miki, 1. et al. 2: 175, 176 Mikkelsen, J.C. Jr. see Boyce, J.B. et al. Parfitt, L.J. et al.; Zhang, S. et al., Zhang, S.J. et al. Mikkola, D.E. and Cohen, J.B. 3: 411 Mikkola, D.E. et al. 2: 148, 163, 166, 168 Mikucki, B. see Housh, S. et al. Mikulla, R. et al. 3: 439 Miles, M.H. 2: 507, 511 Milillo, F.F. see Mayer, S.G.B. et a/. Milke, J.G. see Knaul, D.A. et al. Miller, A.C. see Alamgr, F.M. et al. Miller, A.E. see Allen, C W et al.; Lebrat, J.-P. et al. Miller, A.J. et al. 1: 875, 890, 891 Miller, B.I. see Shani, Y et al. Miller, D. see Lang, C.1. et al., Towle, M.R. et al. Miller, D.J. 2: 51 1 Miller, D.S. 1: 505 Miller, E. see Ettenberg, M, et al.: Johnson, w. et al. Miller, G.J. 3: 113, 841; SCE also Choe, W. et al. Miller, M. see Hilpert, IS.et al. Miller, M.K. 1: 504, 590, 591; 2: 30; see also Larson, D.J. and Miller, M.K. Miller, M.K. et al. 3: 687 Miller, N.C. see Wunt~ngton,H.B. et al. Miller, R. see Perkins, R.A. et al. Miller, S.A. 1: 733 Mi~ler,W.H. 1: 5 Miller, W.S. see Hughes, I.R. et al.; Palmes, I.G. et al. Millett, J.C.F. et al. 3: 635 Milligan, W.W. see Lynch, R.J. et al.; Mikko~a,D.E. et al.; Zhang, S. et al.; Zhang, S.J. et al. Million, A. see Hewat, E.A. et al. Millot, F see Nicolas-Chaubet, D. et al. Mills, M.J. 1: 521, 522, 529, 536, 545, 548, 587; 2 271; 3: 421; see also Baluc, N. et al.; Gouma, P.I. et ul.; Hemker, K.J. and Mills, M.J.; Hemker, K.J. et al.; Medlin, D.L. et al.; Srinivasan,
Author Index R. et al.; Wiezorek, J.M.K. et al.; Yoo, M.H. et al. Mills, M.J. and Miracle D.B. 3: 772 Mills, M.J. et al. 1: 88, 548, 586, 587, 903; 3: 772 Milnes, A.G. see Basile, D.P. et al. Mimoto, H. see Fujita, S. et ul. Mimura, T. 2: 334; see also Abe, M. et al. Mimura, T. et al. 2 334 Min, B.I. see Xu, J-H. et al. Min, B.X. et al. 1: 69, 71, 138 Min, S.-H. see Choi, C.-H. et al. Minakawa, K . see Mitao, S. et al.; Tsuyama, S. et al. Minakuchi, D. see Kishida, T. et al. Miiiani~g~wa, S. see Kogachi, M. et at. Minas, C. 2 380 Minemura, T. see van der Kolk, G.J. et al. Mineniura, T. et al. 1: 843, 846; 2: 610 Miner, R.V. see Locci, I.E. et al. Minervini, J.V. see Steeves, M.M. et al.; Thorne, R.J. et al. Ming-Jian, H. see Sieloff, D.D. et al. Minonishi, Y. 2: 102, 103 Minonishi, Y. et al. 1: 523, 535, 544 Minikov, A.V. see Kokorin, V.V. et al. Minomura, S. see Asaunii, IS,et al. Minoniura, S. e f al. 3: 161 Minonishi, Y 3: 450 Minonishi, Y. and Yoo, M.H. 3: 424 Minor, R.V see Locci, I.E. et al. Mintmire, J.W. 1: 88 Miodownik, A.P. 1: 696, 698; 2 192, 639; see also Saunders, N, et al. Miracle, D.B. 1: 419, 536, 1008; 2: 54, 58, 59, 63, 295, 556; 3: 361; see also Anton, D.L. et al.; Boehler, C.J. et al., Dimiduk, D.M. er al., Farkas, D. et al., Mazdiyasm, S. et al.; Mills, M.J. and Miracle D.B.; Mills, M.J. et al.; Subrarnanian, P.R. et al. Miracle, D.B. and Darolia, R. 3: 45 1,494, 656 Miracle, D.B. and Mendiratta, M.G. 3: 485,486, 490, 494 Miracle, D.B. et al. 2: 59, 289, 290 Miraglia, S. see Baudry, A. et al.; Isnard, 0, et al.; Obbade, S. et al. Mirebeau, I. et al. 1: 37 Mirkarirni, P.B. et rrl. 3: 672 Mirkin, 1.L. 2: 11, 14 Mirkin, I.L. and Kancheev, O.D. 3: 313, 314 Miroshnikov, M.M. 2: 417, 432 Miroshnikov, M.M. et al. 2 409 M i r s h a ~ s A.R. , 1: 655 Misaki, M. see In& H. et al.; Oh, M.H. et al. Misawa, T. see Kimura, A. et al. Misch, L. 1: 410 Misemer, D.M. 2 440, 441, 444 Misenheimer, M.E. see Robertson, J.L. e f al. Mishima, 0. 1: 183; see also Klug, D.D. et al. Mishima, 0. et al. I: 181, 702 Mishima, Y. see Hayashi, Tohru. et al.; Hirano, T. et al., Matsuo, T. et al., Miura, S. et al., Ochiai, S. et al.;
Otsuka, K. et al.; Suzuki, T, et al.; Tounsi, B. et al. Mishima, Y, et al. 1: 499, 915; 2 23, 24, 36, 37, 38; 3: 49, 352, 353, 355 Mishin, Y. see Herzig, C. et al. Mishm, Y. and Farkas, D. 3: 769, 770, 771 Mishina, T. et al. 2: 412 Mishra, A.K. el al. 1: 671 Mishra, R.S. 2 103, 115, 116, 117, 118, 119, 120, 121; see also Banerjee, D. et af.; McFadden, S.X. et al., Nandy, T.K. et al. Mishra, R.S. et al. 1: 918 Mishurda, J.C. 2: 75; see also Perepezko, J,H. et ul. Mishurda, J.C. and Perepezko, J.W. 3: 648 Misra, A. see Larsen, M. et al.; Mitche~~, T.E. and Misra, A. Misra, A.K. see Stafford, K.N. et al. Misra, A.K. et al. 1: 925 Misra, R.D.K. 1: 743 Misra, S. et al. 1: 97 Misroch, M. see Oesterreicher, H. et al. Missell, F.P. see Landgraf, F.J.G. et al., Moreau, J.M. et al. Mistler, R.E. et al. 3: 504, 655 Mitani, S. see Fujimori, W. et al. Mitao, S. see Tsuyama, S. et al. Mitao, S. et al. 2 78, 84, 85 M i t c ~ e l D.F ~ , et al. 1: 990 Mitchell, I.V. et al. 1: 434 Mitchell, J.A. see Tierney, B.J. et al. Mitchell, J.B. see Guinan, M.W. et al. Mitchell, J.R. see Berkowitz, A.E. et al. M i t c ~ e ~K.W l , et al. 2: 330, 423 Mitchell, M.R. see Vassiliou, M.S. et al. Mitchell, N, see Bruzzone, P. et al. Mitchell, R. et al. 3: 54 Mitchell, T. 1: 784, 785 Mitchell, T.E. see Doychak, 5. et al.: Liu, H.C. et al.; Maloy, S.A. et al.; Unal, 0. et al. Mitchell, T.E. and Hirth, J.P. 3: 418 Mitchell, T.E. and Misra, A. 3: 456 Mitchell, T.E. et al. 3: 363, 437, 455, 457, 458 Mitomi, 0. see Wakita, K. et al. Mitra G h e ~ ~ w aA. t , et al. 2: 310 Mitra, N.R. see Jha, IS. et al. Mitra, R. see Sadananda, K. et al. Mitsuishi, I, see Yasuda, T. et al. Mitsuyu, T. see Ohkawa, K. et al. Mitteau, R. see Ball, J. et al. Mitterer, C. et al. 3: 664 Miura, H. et al. 3: 676 Miura, K. see Takizawa, H. et al. Miura, S. see Matsuo, T, et al., Mishima, Y. et al.; Suzuki, T. et al. Miura, S. et al. 2: 36, 37 Miura, Y. see Ishimoto, W. et al. Miura, Y. and Watana~e,N. 3: 444, 447 Miyahara, T. see Yanagihara, M. et al. Miyajiina, H. see Yuasa, S. et al. Miyajima, T. see Akiinoto, K. et al. Miyake, H, see Sugiyama, K. et al. Miyamoto, Y. see Futjisuna, M.et al. Miyarnura, H. see Sakai, T. et al. Miyasaki, S. et al. 1: 544 Miyashita, H. 2: 507
Author Index Miyashita, T. see Nishiyama, Y et al. Miyata, S . see Takabatake, T. et al. Miyauchi, M. 2: 621 Miyauchi, T. see Horinaka, H. et al.; Yaniamoto, T. et al. Miyazaki, S. see Hosoda, H. et al. Miyazaki, T. et al. 1: 717, 718 Miyazaki, Y. see Fujii, H. et al. Miyoshi, T. see ~ashimoto,T. et al. Miyazu, T. see Kobajashi, T. et al. Mizogmchi, T. see Uchitomi, N. et al. Mizokawa, T. see Son, J.-Y et al. Mizuhara, Y. see Gnanamoorthy, R. et al.; Masahashi,N. et al. Mizuhashi, K. see Cava, R.J. et al. Mizukarm, 1. 2 175, 176, 177 Mizuno, 0. 2: 327 Mizutani, U. 1: 63, 105; see also Nakamura, Y. and Mizutani U. Mlavsky, A.Z. 1: 191 Moberly, W.J. et al. 2 251; 3: 416 Mobilio, S. see Czyzyk, M.T. et al. Mochizuki, Y. et al. 3: 669 Mock, R. 1: 875 Mocuta, C. see Durr, W.A. et al. Modder, I.W. see Bakker, H. et al. Moe~wyn-Hughes,E.A. 2: 580 Moffatt, W.G. 3: 786 Moffett, J.C. see Ryge, G. et al. Moffett, M.B. et al, 2: 403 Mohamed, F.A. and Langdon, T.G. 3: 308 Mohammad, S.N. see Park, D.G. et al. Mohamed, F.A. et al. 1: 918 Mohammed Yonsuf, see Sahu, P. Ch. et al. Mohan Rao, P.V. et al. 1: 1024 Mohling, VV. see Lange, H. et al. Mohri, T. 1: 784; see also Kikuchi, R. et al.; Takizawa, S. et al.; Terakura, K. et al. Mohri, T. et al. 1: 40, 41, 850 Mohs, R. I: 97 Moine, P 1: 81 1; see also Junqua, N. et al. Moisy-Maurice, V. 1: 579 Mokhosoev, M.V. see Manzanov Ye, E. et al. Mokrovskii, N.P. see Blum, A.N. et al. Molarius, J.M. see Kolawa, E. et ul. Molchanova, E.K. 2 95 Molchanova, L.V. et al. 1: 7 I8 Mole, C.J. 2: 382 MolBinan, A.C. see Van Ommen, A.H. et al. Molhat, G. 1: 540, 547,549,915; see also Caillard, D. et al., Clement, N. et al.; Paidar, V. et al. Molenat, G. see Rasner, H. et al. Molinari, C. et al. 1: 612 Molinari, A. see Gialanella, S . e f al. Moll, J.H. 2: 178; see also Habel, V. et al. Moll, J.W.et al. 3: 649, 650 Moll~nd-Moritz,E. see Sievering, A. et al. Moller, H.J. 1: 521 Moller, W 1: 692 Molnar, R.J. see Moustakas, T.D. et al. Molotilov, B.V. 2: 470 Molotkov, A. see Bondarev, B. et al. Molotskiy, M.I. 1: 960 Moncevicz, A. et al. 1: 82, 88
Mondal, A. see Shafarman, W.N. et al. Mondieig, D. et al. 2: 637 Moiidolfo, L.F. 2: 175 Monier, J.C. 1: 172, 180, 181 Monkhorst, H.J. 1: 132, 198 Monod, P. see Bellissent, R. et al. Monroe, P.J. 2 237 Montgomery, D.B. 2: 375; see also Steeves, M.M. et al.; Thome, R.J. et al. Montilla, K.L. 3: 742 Moodenbaugh, A.R. et al. 1: 300 Moodie, A.F. 1: 345 Moody, N.R. see Angelo, J.E. et al.; Baskes, M.I. et al. Mooij, J.H. 1: 484; 2: 157 Mook, H.A. et al. 1: 159 Mookerjee, A. see Kumar, D. et al. Moon, J.R. 1: 836 Moon, R.M. 1: 441,447 Mooney, G.D. see Albin, D.S. et al. Mooney, G.D. et al. 2: 423 Mooney, M. see Feder, R. et al. Moore, J.J. 1: 646 Moore, J.J. and Feng, H.J. 3: 645, 725 Moore, P.B. 3: 10 Mooren, M.M.W. see Gardiniers, J.G.E. et al. Moorhouse, S. see Postans, P.J. et al. Moorjani, K. see Rubmstein, M. et al. Moorman, J.O. see Takeya, H. et al. Moose, C.A. 2: 56 Mooser, E. 1: 183,238,241,242,357,419, 420, 431,434 Moran, J.B. 1: 537; 2: 238; 3 415 Morhn-Lbpez, J. 1: 610, 611, 613, 614, 619; see also Kumar, V. et al., Mukherjee, S. et al. Morrin-Lbpez, J. et al. 1: 610 Morbiolo, R. see Malik, M. et al. Mordovetz, N.M. and Rachek, A.P. 3: 161 Moreau, J.M. see Allemand, J. et al.; Landgraf, F.J.G. et al. Moreau, 5.141. et al. 1: 399; 2 316 Morefield, G. see McCormick, S. et al. Morelli, D.T. see Chen, B. et al, Morelli, D.T. and Meisner, G.P 3: 106 Morellon, L. see Ibarra, M.R. et al. Morellon, L. el al. 3: 533, 534 Moret, F. see Baccino, R. and Moret, F.; Baccmo, R. et al. Morgam, G. see Fleischer, R.L. et al. Morgan, A.E. see Pan, A.I. et al. Morgan, D. see Althoff, J.D. et al. Morgan, D.V. 2: 417 Morgan, E. 2: 519 Morgan, E.R. 2: 203 Morgan, J.T. see Prasad. Y.V.R.K. et al, Morgenstern, K. et al. 3: 218 Morgner, H. 1: 956 Morhange, J.F. see Ojima, M. et al. Mori, H. 1: 816; see also Kiyoshl, T. et al.; Luzzi, D.E. et al. Mori, H. et al. 1: 816 Mori, K. see Sato, K. et al. Mori, M. et al. 2: 624; 3: 139 Mori, N. see Kagayama, T. et al., Matsumara, T. et al.; Tatebayashi, T, et al., Watako, Y et al. Mori, T. see Fujita, H. and Mori, T.
923 Mori, Y see Akimoto, K. et al., Ikeda, M. et al. Morian-Lopez, J.L. see Sanchez, J.M. and Morian -Lopez, J .L . Moriarty, J.A. see Xu, W. and Moriarty, J.A. Morii, Y. see FujL H. et al. Morikawa, H. see Ikematsu, U.et al. Morinaga, M. see Murata, Y. et al. Morinaga, M. et al. 2: 80 Morillo, J. see Dunlop, A. et al. Morimura, T. see Kimura, A. et al. Morin, P. et al. 3: 183 Morinaga, M. see Harada, Y et al. Morishita, K. see Ysunasaki, T. et al. Morita, A. et al. 3: 631 Morita, M. see Kishida, T, et al. Morita, T. 1: 35 Moriwaki, M. see Inui, H. et al.; Ito, K. et al. Moriwaki, M, et al. 3: 459 Monwaki, Y. et al. 2: 486 Morkawa. H. see Uemori, R. et al. Morkoc;, H. 2 425 Morkoc, H. see Park, D.G. et al. Morokhovets, M.A. et al. 2 468 Moroni, E.G. 1: 203 Morosin, B.Y. see Hammetter, W.F. et al. Morris, D. 1: 521, 528 Morris, D.G. 1: 503, 505, 51 1, 531, 532, 540, 747; 2: 148, 150, 151, 153, 158, 159, 162, 164, 165. 206; sep also Koch, C.C. et al.; Lerf, R. and Morris D.G., Morris, M.A. and Morris, D.G.; Perez, J.F et al. Morris, D.G. and Gunther, S . 3: 444, 447 Morris, D.G. and Leboeuf, M. 3: 461 Morris, D.G. and Lerf, R. 3: 446 Morris, D.G. et al. 3: 422, 424, 442 Morris, J.W. see Khachaturyan, A.G. et al. Morris, J.W. Jr. 2: 259, 260, 261, 262; see also Mohamed, F.A. et al. Morris, M.A. 1: 647; 2: 206; 3: 413 Morris, M.A. spe Luster, J. and Morris, M.A. Morris, M.A. and Leboeuf, M. 3: 419 Morris, M.A. and Morris, D.G. 3: 417 Morris, R.A. see Nevitt, M.V. et al. Morrish, A.H. 1: 748; 3: 521 Morrish, A.H. see Li, Z.W. and Morrish, A.H. Morrison. R.S. see Hart~e~d-WL~nsch, S.E. et al. Morrogh, H. 2: 524 Morrow, G.R. 2 377 Morsi, K. et al. 3: 657 Morton, A.J. see Head, A.K. et al. Mnrrup, S. 1: 733 Moruzzi, V.L. 1: 71, 138; see also Finstad, T.G. et al.; Gelatt, C.D. Jr. et al., Sanchez, J.M. et al.; Sondericker, D. et al., Williams, A.R. et al. Moruzzi, V.L. et al. 1: 137, 203; 3: 808 Moseev, N.V et al. 1: 569, 579 Mosel, B.D. see Pottgen, R. Moser, H. 1: 1019, 1020 Moser, J. see Bergstrom, D.B. et ~11. Moser, N. see Guo, H.Q. et al.
924 Moser, P. see Corbel, C. et al.; Dimitrov, C. et al.; Doyama, M. et al.; Sitaud, B. et al. Moser, Z. 1: 66, 70 Moser, Z, et al. 1: 673 Moses, R.T. see Fowler, P.H. et aE. Moseychuk, A. see Vasiliev, M. et al. Moshchalkov, V. 1: 21 1 Moss, R.W. see Prater, J.T. et al. Moss, S.C. 1: 28, 50, 463, 482; see also Chen, L.C. et al.; Price, D.L. et al.; Reinhard, L. ef al.; Robertson, J.L. et al. Moss, T.S, 3 805 Motojima, S. et al. 1: 652 Motoki, M. see Ohtani, T. et al. Motowidlo, L.R. 2 366; see also Gregory, E. et al.; Walker, M.S. et al. Notowidlo, L.R. et al. 2: 366 Motrya, S.F. et al. 1: 355 Mott, F.R. 1: 944 Mott, N.F. 1: 21, 484, 670, 671 Mott, N.F. and Nabarro, F.R.N. 3: 352 Motta, A.T. 1: 791, 813, 814, 817 Motta, A.T. et al. 1: 791, 812, 814, 815 Moudden, H. see Coldman, A.I. et al. Mountain, R.D. 1: 479, 679, 680 Mount~eld,K.R. 1: 1020, 1021 Moustakas, T.D. 3: 782, 783 Moustakas, T.D. see Dismukes, J.P. and Moustakas, T.D. Moustakas, T.D. et al. 3 786 Mouturat, P. see Sainfort, G. et al. Movshovick, R. et al. 3: 162 Moya, G. see Jennane, A. et al. Moyer, T.D. 1: 923 Moze, 0. 3 173 Moze, 0. and Buschow, K.H.J. 3 99 Mozer, B. see Cahn, J.W. et al.; Fowler, H.A. et al.; Gratias, D. et al. Mozer, B. et al. 2: 95 Mrowec, S. 1: 990; 2 57; see also Jedlinski, J. et al. Mrowec, S. et al. 1: 983, 987, 990; 2: 57 Mryasov, O.N. and Freeman, A.J. 3: 439 Mryasov, O.N, et al. 3: 439 Muddle, B.C. see Wu, M.H. et al. Mueller, A. et al. 1: 998, 999, 1003; 2: 296 Mueller, F.M. see Fowler, C.M. et al,; Verbeek, B.W. et al.; Weger, M. et al. Mueller, W.J. 2: 649 Mueller, M.H. 1: 393, 394; see also Birtcher, R.C. et al. Mueller, R.R. see Ayer, R. et al. Muench, G. see Sellmyer, D.J. et al. Mughrabi, H. 3: 361 Mugnoli, A. see Fornasini, M.L. et al. MuhZ, S. and Mendez, J.M. 3: 664 Muhlbacher, E.T. see Karnthaler, H.P. et al, Mublstein, C.C. see Venkateswara Rao, K.T. et al. Mujumbar, M.S. see Ho~brook,J.H. et al. Mukai, T. see Higash, K. et al.; Kinoshita, C. et al. Mukaibo, T. see Hori, Y. et al. Mukasyan, A.S. see Vama, A. et al. Mukasyan, A.S. et al. 3: 723 Mukerjee, D. see Sjngh, A.K. et al. Mukherjee, A.K. 1: 656, 917,925; see also Bendersky, L,; Chou~hury,A. et al.
M c ~ a d d ~S.X. n , et al.; Mukhopadhyay, J. et al., Stoner, S.L. et al.; Valiev, R.Z. et al.; Vandervoort, R.R. et al.; Yang, H.S. et al. Mukherjee, A.K. et al. 1: 916 Mukherjee, K. 2: 55; see also Prasad, A. et al. Mukherjee, S. 1; 611 Mukherjee, S. et al. 1: 611 Mukhopadhyay, D.K. et al. 3: 757, 7558 Mukhopadhyay, J. er al. 1: 656 Mukhopadhyay, N.K. et al. 1: 463,465, 471, 477 Mukimov, K.M. et al. 2 446 Muldawer, L. 3: 240 Muldawer, L. and Feder, R. 3 55 Mulder, A.L. et al. 1: 747 ~ u ~ ~ n gW. e r2, 310 Mulford, R.A. 1: 896, 904 MulhofT, A. and Muller, G. 3 670 Mullan, G.A. et al. 3: 672 Mullauer, J. see Appel, F. et al.; Wagner, R. et al. Muller, A.E. see Pretorius, R. et al. Muller, Ch. see Ziesche, P. et al. Muller, Ch. et al. 1: 137 Muller, F, see Elford, L. et al. Muller, F.A. 2 576 Muller, G. see Dasbach, D. et al. Miiller, H. see Belin-Ferre, E et al. M ~ ~ ~3,esee r , Grioni, M. et al.; Gueramian, M, et al.; Jorda, J.L. et ul.; Obbade, S. et al. Miiller, K. 1: 747 Miiller, K.A. 2: 352 Miiller, K. see Graupner, H. et al. Miiller, K.H. see Kubis, M, et al. Miiller, R.H. et al. 3: 98, 102 Muller, M.A. 3: 274, 280, 281, 286, 287 Miiller, M.A. see Valeeva, A.A. et al. Muller, R, see Agyeman, IS,et al. Muller, S. 1: 471 Miiller, W. and Volk, K. 3: 255 Mullins, D.R. see Overbury, S. et al. Mullins, D.R. and Overbury, S.H. 3: 219 Mullins, P. 1: 612, 616 Miillrnann, R. see Pattgen, R. Miillner, M, see Herget, G, et al., Keil, M. et al.; Tietze, f-i. et al. Miillner, P. and Pirouz, P. 3 408 Mtillner, P. et al. 3: 408 Mundy, J. see Liu, C. et al. Mundy, J.N. see Liu, C.T. et al. Muair, Z.A. 1: 646; 3: 723, 725, 732, 734; see Bertolino, N. et al.; Feng, A. and Munir ZA,; Gedevanishvili S.; Xue, H. et al. Munir, Z.A. and Anselmi-Tamburini~U. 3: 725 Mumratham, N.R. see Biggs, B.D. et al. Munjal, V. 2: 271, 274, 281; see aho Ardell, A.J. et al. Munroe, P.R. 1: 535, 536, 539, 913, 920, 924; 2: 157; 3: 60 see also Baker, I. et al.; Nagpal, P. et al.; Schulson, E.M. et al. Munroe, P.R. and Kong, C.H. 3: 60 Munroe, T.C. and Cleeson, B, 3: 570, 574-5 Munson, D. 2: 175, 174, 177, 180
Mum, T. 3: 390 Murakami, A. and Tsunoda, Y. 3: 74 Murakami, A. et al. 3: 66 Murakami, H. see Yamabe, Y, et al.; Yama~e-Mitarai,Y. et al. Murakami, M, 2: 606, 621; see also Huang, W.C.W. et al. Murakami, M. et al. 2: 627 Murakami, Y. 1: 843; see also Golberg, D. et al. Murakami, U.et aE. 1: 504 Muraleedharan, K. see Banerjee, D. et al.; Gogla, A.K. et al.; Rowe, R.G. et al. Muraleed~aran,K. et al. 1: 844 2: 95, 98, 101, 107; 3: 594 Murramatsu, S. see Okamoto, A. et al. Murani, A.P. et al. 1: 440 see also Eizenberg, M. et al.; Krishnamoorthy, A. et nl. Murasaki, S. see Mizutam, U. et al. Murata, Y . et a6. 1: 907; 2: 224 Murphy, A.J. 2: 576, 577, 578, 583, 589 Murphy, D.W. see Christides, C. et al. Murphy, E.V. 2: 133; see also Cameron, D.J. et al. Murphy, R,A. see Alley, G.D. et al.; Vojak, B.A. et al. Murphy, R.J. 1: 733 Murphy, W.H. see Walston, W.S. et al. Murray, B.W see Goela, J.S. et al. Murray, F.S. see Markiewicz, W.D. et al. Murray, J.L. 2 75, 91, 569; 3: 802; see also Angers, L.M. et al.; Massalski, T.B. et id.; Nash, I?. et al. Murray, J.L. et al. 2: 572 Murthy, A.S. see Dimitrov, D.V. et al.; Hadjipanayis, G.C. et al. Murthy, K.S. see Mohan Rao, P.V. et al. Murthy, M.R.L.N. see Satya Murthy, N.S. et al. Murty, B.S. and Rangannthan, S. 3: 727, 731 Murty, K.L. see M o ~ a ~F.A. ~ d et, al. Murugesh, L. et al. 3: 342, 645 Musa, S.O. see Hickey, B.J. et al.; Howson, M A . et al. Muschik, T. see Ernst, F. et al. Museux, E, see Bronoel, G. et af. Musher, J. 2: 623 Muslimov, I S . ~t al. 2: 523 Musso, E. 1: 735 Musson, C.W. see Shah, D.M. et al. Mutasa, B. see Farkas, D. et al. Mutasa, B. and Farkas, D. 3: 226, 771 Muto, S. see Oshima, R. et al. Muto, S. et al. 3: 56 Muto, T. 1: 567 Mutoh, Y. see Gnanamoorthy, R. et al. Mutz, A.H. see Vreeland, T. Jr. et al. Myali~gul~ev, G. see Nikitin, S,A. et al. Mydosh, J.A. 1: 440; 3: 692; see also Palstra, T.M.M. et al. Myers, M.A. see Woke, D.A. et al. Myles, C.W. see Carlsson, A.E,et al. Mys'kiv, M.G. see ~ a r e c l i ~ y u8k. ,S . et ai. Mysko, D.D. et al. 2: 161
Nabarro, F.R.N. 1: 87,918; 2 19; 3::363, 440
Author Index Nabarro, F.R.N. and de Villiers, H.L. 3: 298, 308; 3: 361, 372 Nabarro, F.R.N. et al. 3: 361, 362 Naburgh, E.P see Reader, A.H. et al. Nachman, L.F 2: 307 Nachtrab, W.T. see Eylon, D. et al. Nadgorny, E M . 1: 523 Nadgornyi, E. 1: 875; 2: 148 Nadis, S. 3: 864 Nagahama, K. see Miki, I. et al. Nagahora, J. see Higashi, K. et al. Nagai, H. see Asahi, H. et al.; Deportes, J. et a[.; Matsuoka, T. et al.; Takahashi, T. et al. Nagano, M. see Aoki, R. et al. Naganuina, M. s m Wakita, K, et al. Nagaoka, M. see Uchitomi, N. et al. Nagarajan, R. see Mazumdar, C. et al. Nagarajan, R. et al. 3: 103 Nagasawn, A. et al. 2: 55 Nagashima, M. see Takasugi, T. et al. Nagashima, Y see Degawa, T. et al. Nagata, K. see Takeuchi, H. et al. Nagata, M. see Karnisada, Y. et al.; Ogawa, T. et ai, Nagayama, K. see Kimura, K. et al. Nagel, D.J. 1: 135, 137, 141, 143 Nagel, G. see Sajovec, F et al. Nagender Naidu, S.V. and Rania Rao, P. 3: 802 Nagorson, G. 1: 322 Nagle, J. see Razeghi, M. et al. Nagpal, P 2 144; see also Baker, I. et ul., Schmidt, B. et al. Nagpal, P. and Baker, I. 3 358 Nagpal, P et al. 2: 64 Naguchi, 0. see Wee, D.M. et al. Nahashima, Y. see Kamata, K. et al. Naidu, S.V.N. see Mohan Rao, P.V. et al., Muraleedharan, K. et al. Naik, R. see Pulatunda, S.K. et al. Naimon, E R . see Ledbetter, H.M. et al. Naimushin, S.G. see Molchanova, L.V. et al. Nail Salem, N, 1: 745 Najafabadi, R, see Wang, H. et al. Najafabadi, R. et al. 1: 88 Naka, S. see Douin, J. et al., Khan, T. et al.; Petez, L. et al. Naka, S. and Khan, T. 3: 843 Naka, S. et al. 1: 860; 3: 450,478, 597, 841 Nakabayashi, Y see Fuchiiio, S. et al. Nakada, Y. 2 11 Nakada, T. and Kunioka, A. 3: 672 Nakagawa, Y. 1: 682, 683 Nakagawa, Y.G. see Caron, P. et al. Nakagawa, Y.G. et al. 2: 85 Nakagome, H. 2: 343 Nakahara, S. see Chin, G.Y. et al. Makahigashi, K. 1: 726; 2 566; see also Kogachi, M. et al. Nakai, Y see Yamada, T. et al. Nak~jima,H. see Nonaka, K. et al. Nakajima, K. see Shirai, T. et al. Nakamichi, T. 1: 953, 954 Nakamoto, T. see Inui, H. et al.; Ito, K, et al. Nakamura, A. see Fujiwara, T. et al.; Inui, W.et al., Nishitani, S.R. et al. Nakamura, H. see Kyogaku, M. et al. Nakamura, H. et al. 2 230
Nakamura, K. see Kaneko, H. et al.: Yata, M. et al. Nakamura, M. 1: 645,875,877,878, 879, 880, 881, 882; 2: 145, 149, 221; see also Abe, E. et al.; Hashimoto, T. et al.; Hirano, T. et aE.; Kirnura, K. et al.; Takeuch, S. et al.; Takeyama, M. et al.; Tsutsumi, S. et al. Nakamura, M. et al. 1: 875 Nakamura, N. see Hiratio, T. et al.; Kobayashi, T. et al. Nakamura, N. et al. 2: 432 Nakamura, T. see Inoue, A. et al.; Yokoyama, N. et al. Nakarnura, Y. 1: 441; see also He, X.-M. et al.; Sunxyama, K. et al. Nakamura, Y and Frame, J.J.M. 3: 804 Nakamura, Y and Mizutani, U. 3: 144 Nakamura, 2. see Watanabe, S. et al. Nakanishi, N. 1: 842; see also Murakami, Y. et al. Nakanishi, T. see Ishikawa, M. et al. Nakano, H. see Aizawa, T. et al. Nakano, T. see Behgoziti, A. et al.; Hagihara, K. et al., Umakoshi, Y. et al.; Yasuda, H.Y. et al. Nakano, T. et al. 3: 460 Nakao, M. et al. 2: 343 Nakashima, T. see Umakoshi, Y. and Nakashima, T.; Umakoshi, Y. et al. Nakata, N. see Hishinuma, A. et al. Nakata, T. see Koga, K. et al. Nakata, Y see Tadaki, T. et al. Nakata, Y. et al. 1: 840, 841 Nakayama, R. and Takeshita, T. 3: 102 Nakayarnd, Y . 2: 157, 158, 164; see also Hirukawa, K, et al.; Mabuchi, H. et al. Nakazawa, T. see Yarnabe-Mitarai, Y. et al. Nam, S.W. see Kim, J.K. et al. Namjoshi, S.N. and Thadhani, N.N. 3: 731, 745 Nana, S. see Cortie, M.B. et al. Nanao, S. see Koflat, D.D. et al. Nanbu, K. see Mimura, T, et al. Nandy, T.K. 3 370; see also Banerjee, D. et al.; Gogia, A.K. et al.; Muraleedharan, K. et al.; Singh, A.K. et al. Nandy, T.K. and Banerjee, D. 3: 494 Nandy, T.K. et al. 2: 95, 115 Naoe, M. 2: 449 Napaletano, M. see Canepa, F er al. Narasirnhan, M.C. 1: 733 Narayan, J. sec Tiwari, P et al. Nardone, V.C. et al. 2 65, 66, 290, 294 Narita, A. 1: 35 Narita, N. and Takamura, J, 3: 422 Nartova, T.T. see Barinov, S.M. er al. Narumiya, Y see Shinoura, 0. et al. Narwankar, P see Hu, J. et al. Nasako, K. et al. 2 486 Nash, P. 3: 802; see aZso Davies, C.K.L. et al., Desch, P.B. et al.; Ma, E. et al. Nash, P. and Schwarz, R.B. 3: 700 Nash, P. et al. 2: 54, 55; 3: 739 Nash, P.G. 1: 718, 727; see nlso Smith, J.F et al. Nash, P.G. et al. 2: 156
925 Nason, D. 1: 610 Nastasi, M. 1: 817; see also Colgan, E. et al.; He, X.-M. et al.; Hung, L.S. et al.; Lilienfeld, D.A. et al.; Saris, F W. et al. Nastasi, M. et al. 1: 694, 733; 2 610 Nasu, S. see Shiiigu. P.H. et al. Nasunjilegal, B. see Yang, F.M. et al. Natesan, K. 1: 982, 983, 986, 989; 3: 573, 613, 707, 708, 717; see also Renusch, D. et al.; Tiearney, T.C. Jr. and Natesan K,; Tortorelli, P.F and Natesan, K. Natesan, K. and Cho, W.D. 3: 573 Natesan, K. and Johnson, R.N. 3: 572 Natesan, K. and Tortorelli, P.F. 3: 503, 572, 613 Natesan, K. et al. 3: 573 Nathal, M.V. 2: 63, 64; see also Darolia, R. et al.; Garg, A. et cd., Locci, LE. et al.; Noebe, R.D. et al.; Whittenberger, J.D. et al. Nathal, M.V. et al. 2: 10 Nathan, V. 2: 419 Nathans, R. 1: 441, 444; see also Shirane, G. et al. Naundorf, F. see Dimitrov, C. et al. Nava, F see Amioti, M. et al. Nava, F. et al. 2: 616, 617 Navarathna, A. see Cadieu, F.J. et al.; Hegde, H. et al.; Rani, R. et al. Navis, S. see Wein, W. et al. Navrotsky, A. 1: 95 Nawata, K. see Ito, T. et al. Nayak, P.U. see Misra, S. et al. Nayeb-Hasherni, A.A. and Clark, J.B. 3: 802 Nazmy, M. 3: 600 Nazmy, M. see Lupinc, V. et al. Neale, F.E. 1: 677 Neckel, A. 1: 93, 137 Needhani, J. 1: 633 Needleman, A. see Baskes, M.I. et al., Christman, T. et al. Needs, R.J. et al. 3 226 NCel, L. 1: 15, 439, 935; see also PaulevC, J. et al. NCel, L. et al. 1: 560 Nefedov, V et al. 1: 613 Nefedova, M.V. see Tissen, V.G. et al. Negishi, A. see Nakano, T. et al. Neighbors, J.R. I: 204, 206 Neikirk, D.P see Block, T.R. et al. Neipce, J.C. see Charlot, F. et al., Gras, c. et al. Neisius, H. see Lugscheider, E, et al. Neite, G. 2 269, 271, 273, 274, 281, 284; see also Pottebohm, H. et al.; Wallow, F. et al. Nekkanti, R.M. and Dirniduk, D.M. 3: 552, 658 Nelen, J. see Clarke, R.S. et al. Nelin, G. see Skold, K. and Nelin, G. Nelissen, A.J.M. see Dirks, A.G. et al. Nelson, D.R. 1: 473, 474, 479, 480; see also Steinhardt, P.J. et al. Nelson, D.R. and Spaepen, F. 3: 700 Nelson, J.S. see Baskes, M.I. et al. Nembach, E. 2 263, 269, 271, 273, 274, 275, 279, 281, 283, 284; see also
926 Pottebolim~H. et al., Rgsner, H. et al.; Wallow, F. et al. Nernbach, E. et al. 2: 268 Nemec, C. see Kallingal, C.G. et al. Nemoshkalenko, V V. see Ziesche, P. et al. Nemoto, M. 1: 860; see also Oh-ishi, K. cJt al., Tian, W.H. et al. Nenno, S. 1: 618, 831, 837; 2: 22, 55; see also Enami, K. et al., Nagasawa, A. et al.; Pak, H.R. et al., Saburi, T. et al. Neobe, R.D. and Lerch, B.A. 3: 347 Nerc-Partaix, A. and Mennaiid, A. 3 424 Neri, R. 1: 457 Nerses, V see Huntington, H.B. et al. Nes, E. 2: 188 Nesbrt, L.A. 1: 867 Nesbit, LA. and Laughlin, D.E. 3: 418 Nesbit, S. et al. 3: 78 Nesbitt, E.A. see Sherwood, R.C. et d . Nesbitt, E.A. et al. 2 313 Nesbitt, J.A. 1: 979, 980, 1009, 1010; see also Doychak, J. et al. Nesbitt, J.A. and Barrelt, CA. 3: 502 Nesbitt, J.A. et al. 1: 987, 989, 991, 1009; 3: 502 Nesbitt, R.A. see Sherwood, R.C. et al. Nesper, R. 3 231, 252 Nesper, R. see von Schnering, H.G. et al.; Wengert, S. et al. Netherwood, P. Jr. see Niiler, A. et al. Neugebauer, J.M. see Gaal, 1. et al. Neuman, M.R. 1: 956 Neuniann, D.A. see Christides, C. et al. Neumann, F. 3: 439 Neumann, J,P 1: 565,572, 573; 2: 200; see ulsa Chang, Y .A. and Neumann, J.P.; Chang, Y.A. et al. Neumann, J.P. et al. 1: 565, 571; 2: 54, 200 Neumann, M. see Galakov, V.R. et al. Neumann, P, see Von Keitz, A. et al. Neumann, P and Sauthoff, G. 3: 451 Neumark, G.F. et al. 2: 428 Neumeier, J. see Movshovich, R. et al. Neumuller, H.W. see Dumas, J. et al. Neuringer, L.J. see Willianis, J.E.C. et al. Neves, S. see Escher, C . et al. Neveu, C . 3: 443 Neveu, N. 1: 505 Neville, F.H. 1: 5, 7 Neville, R.C. 2: 421 Nevitt, M.V 1: 11, 107, 385, 386, 387, 392 Nevitt, M.V. et al. 1: 394 Newcoinb, S. see Gialanella, S. et al. Newey, C.W.A. 2: 58, 61 Newkirk, H.W. see Smith, D.K. et al. Newkirk, L.R. 1: 703, 704 Newinan, D.W. 2: 426,427 Newman, D.M. see Barns, G.S. et al. Ney, E.P. see Frier, P.S. et al. Ngan, H.W. 1: 496, 501, 502, 514, 550 Ngan, H.W. et al. 1: 541, 550 Nghiep, D.M. see Quycn, N.H. el al. Nghiep, D.P. et al. 2: 224 Nguyen, M. see Compaan, A. et al. Nguyen-Manh, D. see Trambly de Laissardiere, G. et al.; Znam, S. et al. Nguyen, N.M. see Carrabba, M.M. et al. Ngwenya, K.P. and Wolff, I.M. 3: 65
Author Index Niarchos, D. see Kalogirou, 0. et al. : 168, 169; see also Klansky, J.L. et al., Mikkola, D.E. et al.; Parfitt, L.J. et al.; Zhang, S. et al. Nic, J.P et al. 2 157, 159, 161, 163, 165, 166, 168 Nicholas, T. see Balsone, S.J. et al.; Smith, P.R. et al. Nicholls, J.E. see Duddles, N.J. et al. Nicholls, J.R. 1: 922; 2: 489, 496, 499; see also Hancock, P. et al. Nicholls, J.R. et al. 2: 492, 495, 498 Nichols, K.B. see VOJak, B.A. et al. Nicholson, D.M. see Ceder, G. et ul., Gonis, A. et al.; Gyorffy, B.L. et al.; Johnson, D.D. ef al.; Sluiter, M. et al.; Stocks, G.M. et al., Turchi, P.E.A. et al. Nicholson, D.M. et al. 1: 65, 70, 499; 2: 155, 157 Nicholson, D.M.C. et al. 3: 685, 689 Nicholson, J.M. see Ray, U. et al. Nicholson, L.K. 1: 1021 Nicholson, M.M. 2 510 Nicholson, R.B. 2: 258; see also Hirsch, P et al. Nickel, H. see Hilpert, K. et al. Nickl, J.J. see Sprenger, H. et al. Nicklow, R.M. 1: 441, 447; see also Nook, H A . et al.; Rowe, J.M. et al. Nicol, A.D.I. 2: 177 Nicolai, H.P. see Wagner, R. et al. Nicolas-Chaubet, D. et al. 1: 983, 990 Nicolet, M.-A. 2 605, 606, 614, 616, 630; see also Cheng, Y.T. et al.; Cheung, N.et al., Dorner, W et al., Finetti, M. et al.; Harris, J.M. et al.; Johnson, W.L. et al.; Kattelus, H.P. et al.; Kolawa, E. et al.; Liu, B.X. et al.; Maenpaa, M. et al., Nieh, C.W et al.; Olowolafe, J.O. et al., Poketa, P.J. et at.; So, F.C.T. et al.; Suni, I. et al.; Tsaur, B.Y. et al.; von Seefeld, €3. et al.; Zhu, M.F et al. Nicolet, M . 4 . and Lau, S.S. 3 675 Nicoll, A.R. 2: 492, 493 Nieh, C.W see Kolawa, E. et al., Meng, W.J. et al. Nieh, C.W. et al. 2 625 Nieh, T.G. 1: 656, 927, 987, 1000, 1003, 1004, 1005, 1007; 2 213; see ulso Chou, T.C. et al.; Hsiuiig, L.M. and Nieh, T.G.; Hsiung, L.M. et al., Lm, C.T. et al., Pharr, G.M. et al., Wang, J.N. et al. Nieh, T.G. et al. 1: 656; 3: 43, 47 Niehus, H. 3: 214,215,218; see also Blum, R.P. et al.; Morgenstern, K. et al. Niehus, H. and Achete, C. 3: 218 Niehus, H. and Comsa, G, 3: 214 Niehus, H. et al. 1: 612, 615; 3: 219 Nielsen, B. see van der Kolk, G.J. et al. Nieminen, R.M. see Alatalo, M. et at., Puska, M.J. et al. Niepce, J.C. see Charlot, E. et al. Niessen, A.K. 1: 366; see also de Boer, F.R. et a!.; van der Kolk, G.J. et al. Niessen, K. see de Boer, F.R. et al. Nieuwenhuys, G.J. 1: 440; see also Palstra, T.T.M. et al.
Nigarn, A.K. see Germann, A. et al.; Mazuindar, C. et al. NiggIi, A. see Donnay, J.D.H. et al. Niggli, P. 1: 309, 310; 3: 3, 4 Nigmetova, RSh. see Kozin, L.F NIH Technology Assessment Conference Statement 62: 590 Niihara, K. see Suzuki, IS. and Niihara IS. Niiler, A. et al. 3 736 Niina, T. see Koga, K. et al. Niki, Y. see Flanagan, T.B. et al. Nikitin, S.A. see Annaorazov, M.P et al. Nikitin, S.A. and Tishin, A.M. 3: 527 Nikitin, S.A. et al. 3: 174, 177, 527, 529 Nimmagadda, R. et al. 3: 667 Ninomiya, 3. see Tachikawa, K.Y. et al. Ninorniya, T. 1: 491 Nisenoff, M. 2 383 Nishi, H. see Yokoyama, N. et al. Nishi, M. sec Ando, T. et al. Nishi, Y. et al. 1: 473 Nishida, I. 2: 329; see also Kaibe, H.T. et al., Kawasurni, I. et al.; Ohsugi, I.J. et al.; Tokushima, T. et al. Nishida, IS.et al. 2: 420 Nishida, N. see Ishimoto, H. et al. Nishida, T. see Nakao, M. et al. Nishida, T. and Urabe, K. 3: 736 Nishigori, S. s ~ Takabatake, e T. et al. Nishijima, Y. see Shinohara, E(. et al. Nishikawa, K. 2 345 Nishimura, C. 1: 928; 2: 29, 44 Nishimura, C. and Lm, C.T. 3: 656 Nishimura, K. see Hashimoto, T. et al. Nishinaga, T. see Tanakd, M. et al. Nishino, K. et al. 3: 600 Nishio, K. see Imoto, 3'. et al. Nishmko, T. 3: 411 Nishisako, T. see Ikebuchi, M. et al. Nis~itani,S.R. see Fujiwara, T. et al. Nishitani, S.R. et al. 1: 454, 457, 655; 2: 157, 165 Nishnyama, N. see Inoue, A. et al. Nishiyama, Y. et al. 2: 84, 88 Nishiyama, 2;. 1: 828 Nishizawa, J. 2: 335 Nishizawa, J. et al. 2: 326, 335 Nishizawa, T. 1: 118; see also Ishida, K. et al.; Kainuma, R. et al. Nissen, H.U. see Beeli, C, et al. Nissen, H.U. et al. 1: 457 Nitsche, R. 1: 356 Niu, X.J. et al. 3: 527 Niwano, M. see Yanagihara, M. et al. Nix, W. see Wright, W.J. et al. Nix, W.D. 1: 527, 916, 925; 2: 42, 63; 3: 227; see also Gtiken, M. et al., Hemker, K.J. et al.; ~ i t a b j i a nP.H. , et al., Memker, K.J. et al. Nix, W.D. and Clemens, B.M. 3: 226 Nixdorf, J. 2 653 Nixon, T. and Pond, R.C. 3: 410 Njah, N. 1: 786 Njali, N. et al. 1: 784, 785 Nobili, C. 1: 999 Noble, F.W. 1: 534 Nobuki, M. 2 86; see also Hashimoto, IS. et al. Nobuki, M. et al. 3: 623 Noda, T. 3: 593; see also Isobe, S. and Noda, T.; Nishryama, Y. et al.
Author Index Noda, Y 1: 777, 780; see Shapiro, S.M. et al. Noebe, R. see VeyssiZre, P and Noebe, R. Noebe, R.D. 1: 536, 540,926; 2: 58, 63, 278, 290; see also Bowman, R.R. et al.; Cotton, J.D. et al.; Doychak, J. et al.; Garg, A. et al.; Johnson, D.R. et al.; Kim, J.T. et al.; Kitabjian. P.H. et al.; Kitano, K. et al.; Larsen, M. et al.; Loccs, I.E. et al.; Raj, S.V. et al., Srinivasan, R. et al.; Wliittenberger, J.R. et al.; Wright, R.N. et al. Noebe, R.D. and Walston, W.S. 3: 300, 305,308,316,494,602,604,656,657 Noebe, R.D. et al. 1: 620,982,990; 2 294; 3: 330, 331, 347, 361, 368, 656, 808 Noguchi, 0. see Wee, D.-M. et al. Noguchs, 0. et al. 2: 35, 36 Noguchi, S. see Oota, A. et al. Noguchi, T. see Fujita, S. et al. Noguchi, Y. see Matsuoka, T. et al. Nohara, A. 1: 522, 544; see Saka, H. et al. Nojinia, S. see Tanaka, H. et al. Nol, H. see Weitzer, F et al. Nolas, G.S. see Tritt, T.M. et al. Nonaka, K. et al. 3: 290 Noonan, J. 1: 616 Noonan, J.R. see Davis, H.L. and Noonan, J.R. Noonan, J.R. and Davis, H.L. 3: 219, 225 Nordberg, R. see Siegbahn, K. et al. Nordheim, R. and Grant, N.J. 3: 74 Nordman, J.E. see Wiley, J.D. et al. Nordstrorn, L. see Eriksson, 0. et al. Nori, N. see Koyarna, K. et al. Norman, J.H. et al. 3: 584 Norman, M.R. 1: 130, 132; see also Wang, C.S. et al. Norskov, J.K. 1: 78; 2 479; see also Jacobsen, K.W et al. Norstrorn, H. see Nygren, S. et al.; Ostling, M. et al. Norstrom, H. et al. 2: 623, 624 North, D.M. see Enderby, J.E. et al. Northwood, D.O. 2: 476; see also Ding, Y. et al. Norton, M.G. 3: 760; see also Suryanarayana, C. and Norton, M.G. Norton, P.R. see Patten, E.A. et al. Nosova, G.I. 2 271, 282 Notin, M. see Cunat, C. et al. Notin, M. et al. 1: 95 Notkin, A. see Bondarev, B. et al. Notkin, A B . see Greenberg, BA. et al. Notomi, S. see Abe, M. et al. Notten, P.H.L. 2: 486 Nouet, G. see Braisaz, T. et al. Noufi, R. see Albin, D.S. et al., Mooney, G.D. er! al., Tuttle, J.R. et al. Noufi, R.N. see Bhattacharya, R.N. et at. Nourbakhsh, S. see Sahin, 0. et al. Novak, P. see Winter, M. et al. Novikov, 1.1.see Khalim, A.A.R. et al. Novoselova, A.V. see Molchanova, L.V. et al. Nowack, L. 2: 564 Nowak, W. see Fitzer, E. et al. Nowick, A. 1: 778, 782 Nowick, A.S. see Feder, R. et al.
Nowicki, R.S. 2: 624; see also Harm, J.M. et al. Nowik, 1. see Atzmony, V et al. Nowotny, H. et al. 2: 226 Noya, A. 1: 651; see also Takeyama, M. et al. Nozaki, H. see Isl~zawa,Y. et al. Nozieres, J.P. see Allemand, J. et al. NoziGres, J.P see Landgraf, F.J.G. et al.; Moreau, J.M. et al. NoziZres, P. 1: 128, 215 Niicker, N. 1: 159, 160 Nugent, L.J. 3: 30 Nuhfer, N.T. see Martin, P.L. et aE. Nukm, A. see Kagamida, M. et al. Numakura, H. see Nonaka, K. et a/. Nurnasawa, T. see Nashimoto, T. e~ al. Numazawa, T. see Hashimoto, T. et al. Nunes, C.A. see Perepezko, J.H. et al. Nunes, C.A. et al. 3: 657 Nurmikko, A.V. see Ding, J. et al., Jeon, H. et al. Nusair, N. see Vosko, S.H. et al. Nusirnomcs, M.A. see Kunc, K. et al. Nygren, S. see Ostling, M. et al. Nygren, S. et al. 2: 624 Nzula, M. see Lang, C.I. et al. Obbade, S. see Wolfers, P. et al. Obbade, S. et al. 3: 98, 99 Oberli, S. see Giauque, P.H. and Oberli, S. Obermyer, R.T. see Huang, M.Q. et al. Oblak, A.F 1: 528 Oblak, J.M. see Giamei, A.F. et al.; Kear, B.H. et at, O’Brien, W.J. 2: 590; see also Lcinfelder, K.F et al. Obrowski, W. 3 55 Ochaia, S. et al. 2: 9, 214 Ochiai, A, see Fraas, K. et al. Ochiai, S. see Mishima, Y. et al., Suzuki, T. et al. Ochiai, S . et al. 1: 92, 725, 726, 727, 925; 2: 34, 35, 650; 3: 66 Ocko, B.M. see DiMasi, E. et al. Oda, K. see Otsuka, K. et al. Oda, 0. see Triboulet, et al. Odani, K. see Abe, M. et al.; Yokoyama, N. et al. Odette, G.R. see Cao, H.C. er al., Rowe, R.G. et al.; Venkateswara Rao, K.T. et al. O’Donnell, K. see Coey, J.M.D. and O’Donnell, K. Oe, K. see Takeuchi, H. et al. Oehme, G. 1: 96, 97, 666, 671 Oehring, M. 3: 687; see also Appel, F. et al.; Imayev, R.M. et al.; Yan, Z. et al. Oehring, M. and Haasen, P. 3: 687 Oehring, M. et al. 3: 624, 652, 762 Oehrli, M.L. see Shoemaker, J.R. et al. Oelsner, O.W. 1: 635 Oertel, C.G. see Ledig, L. et al. Oesterreicher, H. et al. 2: 312 Offer, S. see Atzmony, V et al. Ogawa, K. see Yata, M. et al. Ogawa, R. see Kamisada, Y. et al.
927 Ogawa, S. 1: 277; 2: 564; see also Iwasaki,H. et al.; Yamaguchi, S. et al. Ogawa, S. et al. 2: 564, 567; 3: 138, 149 Ogawa, T. see Fuchins, S. et al.; Kamisada, Y. et al.; Ohara, T. et al. Ogawa, T. et al. 2: 383 Ogino, Y. see Yamasaki, T. et al. Oglesby, P.L. see Dickson, G. et al. Oglesby, P.L. et al. 2: 585 Ogletree, D. see Baird, R. et al. Oguchs, M. see Inoue, A. et al. Oguchi, T. see Hong, T. et al.; Min, B.I. et al.; Mohri, T. et al.; Takizawa, S. et al.; Terakura, K. et al.; Xu, J.-H. et al. Oguey, C. 1: 483 Oguma, S. see Yoshizawa, Y. et al. Ogura, M. see Fuchino, S. et al. Ogura, T. et al, 1: 591, 593; 2: 27 Oguro, K. see Sakai, T. et al. O~urtani,T. 1: 768 Oh, A. see Inui, H. et al. Oh, J.E. et al. 2: 621, 626, 627 Oh, J.M. 2: 218 Oh, K.H. see Park, S.J. et al. Oh, M.H. see Inus, H. et al.; Nishitani, S.R. et al. Oh, M.H. et al. 3: 363 O’Handley, R.C. see Dunlap, R.A. et al.; Mitra ~ h e ~ a w aA. t , et al. Ohara, A. see Kishida, T. et al. O’Hara, K.S. see Walston, W-S. et al. Ohara, T. et al. 2: 383 Ohashi, K. see Fidler, J. et al. Ohashi, K. et al. 2: 335 Ohashi, W 1: 454,457; see also Syaepen, E;. et al. Ohba, K. see Inoue, A. et al. Ohi, M. see Ohara, T. et al. Ohis, M. see Semiatin, S.L. et al. Oh-ishi, K. et al. 3: 316 Ohishi, K. see Tsuji, H. et al. Ohkawa, K. et al. 2: 326 Ohki, A. see Shibata, N. et al. Ohkl, A. et al. 2 326 Ohl, M. et al. 2: 371 Ohls, M. see Semiatin, S.L. et al. Ohmori, Y 1: 835 Ohnaka, K. et al. 2 343 Ohnenstetter, M. see Johan, E. et OZ. Ohnishi, T. see Yokoyama, N. et al. Ohno, H. see Hasegawa, H. et al. Ohno, K. and Kino T. 3 282 Ohnurna, S. see Fujimon, H. et d. Ohoba, Y see Ishikawa, M. et al. Ohring, M. 3: 669 Ohsaka, K. see Lee, M.C. et al. Ohsawa, K. see Yamada, T. et al. Ohshima, K. I: 49; see aZso Ohtani, T. et al. Ohsugi, I.J. et al. 2: 329 Ohta, M. 2: 282, 565; see also Hisatune, K. er d. Ohta, M. et al. 2: 282, 565 Ohta, Y 2: 392; see also Caron, P et al. Ohtani, T. e f al. 3 756 Ohtera, K. see Higashi, X;. et al.; Inoue, A. et al. Ohtsuka, H. see Hashmoto, T. et al. Ohtsuka, M. 3 417
Author Index Ohtsuka, S. see Tatebayashi, T. et al. Ohuchi, K, see Bolt, P.J. et al. Oikawa, H. see Takahashi, T. et al. Oishi, T, see Suzuki, R.O. et al. Ojima, M. et al. 2: 412 Ok, H.N. 1: 748 Okabe, T. 2: 584; see also Grenga, H.E. et al.; Shires, P.J. et al.; Timmons, C.F. et al.; Tsutsurni, S. et al. Okabe, T, et al. 2: 584 Okabe, T.H. see Suzuki, R.O. et al. Okabe, Y. see Taka~iashi,T. et al. Okada, M. see Maeda, T. et al. Okada, T. see Hashimoto, T. et al. Okajima, M. see Iwasaki, H. et al. Okamoto, A. et al. 3 667 Okarnoto, H. 3 802; see Massalski, T.B. et al.; Murray, J.L. et al.; Villars, P. et al. Okamoto, H. and Massalski, T.B. 3: 802 Okarnoto, H. and Tanner, L. 3: 38, 49 Okamoto, €3. et al. 2: 562, 563 Okamoto, J.K. see Anthony, L. et al. Okamoto, M. see Oku, S. et al. Okamoto, P.R. 1: 809,814, 815,817, 819; see also Devanathan, R. et al.; Koike, J. et al.; Lam, N.Q. et al., Meng, W.J. et al.; Motta, A.T. et al.; Nastasi, M. et al.; Rehn, L.E. et al.; Xu, G.-B. et al. Okamoto, P.R. et al. 1: 694, 695; 2: 133 Okamoto, Y. see Mochizuki, Y. et al. Okamura, H, 2: 417 Okayama, V. see Thomasson, Y. et al. Okazaki, H. see Shomojo, F. and Okazaki, H.; Tadaki, T. et al. Okazaki, I(.2: 178; see also Kini, N.J. et al. O’KeeRe, M. 1: 352, 354; see also Hyde, B.G. et al. Oku, S. sec Kato, K. et al. Oku, S. et al. 2: 343 Okuda, M. see Kurosawa, K. et al. Okuda, S. 1: 785 Okuda, T. see Honnaka, H. et al. Qkuno, Y. see Nishiza~a,J. et al. Olander. A. 1: 834 Olander, D.R. 1: 813 O’Larey, P. see Smathers, D. et al. Olbright, G.R. see Jin, R. et al. Oleksin, O.J. see Bodak, 0.1. et al. Oligschleger, C. and Schober, H.R. 3: 694 Oliveira, M. see Ansara, 1. et al. Oliver, B. see Whittenberger, J.D. et al, Olives, B.F 2: 18, 28, 39, 87; see also ASaJS, S. et al.; George, E.P. et aL; Johnson, D.R. et al.; Kad, B.K. and Olives, B.F. Olives, J. 1: 528, 529, 545, 550, 551, 552 Oliver, J.O. 1: 498, 505 Oliver, W.C. 1: 656, 928; 2: 27,29, 33, 39, 40, 213; 3: 503; see also George, E.P. et al.; Porter, W.D. et al., Sparks, C.J. et al.; Stoner, S.L. et al. Ollitrault-Fichet, R. et al. 1: 353 Olowolafe, J.O. see Blanpain, B. et al. Olowolafe, J.O. et al. 2: 616, 617, 618, 622, 624, 625 Olson, C.G. et al. 1: 136
Olson, G.B. 1: 831, 832; 3: 473, 841; see also Christian, J.W. et al.; Cohen, M. et at,
Olsson, M. see Hedenqvut, P. et al. Olsson, N.A. see Temkm, H. et al.; Tsang, W.T. et al. Onia, K. see Nakano, T. et al. O’Mahony, J.D. see Donnelly, N. et al. O’Malley, M.W see Loubriel, G.M. et al. Ornelianenk~,V.V. see Larikov, L.N. et al. Omlor, R. see Szaruga, A. et al. Omnes, F. see Razeghi, M. er al. Ondracek, G. see Jesse, A. et al. O’Neal, J.E. see Shih, D.S. et al. Onnes, H.K. 1: 16; 2: 351 Ono, K. 1: 875, 887; see also Ishimoto, H. et al.; Suzuki, R.O. et al. Ono, S. see Takasugi, T. et al. Ono, T. see Chiba, A. et al. Onoda, G.Y. et al. 1: 481 Onodera, A. see Fujii, Y. et al. Onodera, H, see Yokoyarna, N. et al. Onofrio, G. see Lupinc, V. et al.; Lupnic, V. et al. Onstott, E.1. 2: 647 Onuki, Y see Kagayama, T. et al.; Kobajashi, T. et al. Oomi, G. see Kagayama, T. et al.; Watako, Y. et al. Oonk, H.A.J. see Mondieig, D. et al. Oota, A. et al. 3: 56 Opila, E.J. 3: 44 Oppeneer, P.M. see Maurer, T. et al. Oppeneer, P.M. et al. 2: 440, 441 Oppenheimer, F. see Frier, P.S. et al. Orange, G. see Liang, R.M. et al. Orchard, J. 1: 784, 785, 786 Orent, T. et al. 1: 619 Omeci, A. see Asta, M.D. et al. Ormerod, J. see Rozendaal, E. et al. Oron, M. see Shani, Y. et al. Orowan, E. 2 258, 260; 3: 418 Ortega, R.J. see Madurga, V, et al. Ortel, C.G. see ~ o s c h r n ~I.n ~et, al. Ortiz, C. see Epperson, J.E. et al. Osaka, T. see Takai, M. et al. Osaka, T. et al. 3: 671 Osakabe, N. et al. 3: 222 Osantowski, J.F 2: 409; see also Hunter, W.R. et al. Osawa, T. see Kagamida, M. et al. Osborne, M. see Zirnm, C.B. et al. Osburn, J.E. see Burton, B.P et al., Nehl, M.J. et al. Osburn, J.E. et al. t: 196 Osgerby, S. 1: 927 Osgood, R.M. Jr. see Willner, A.E. et al. Oshima, R. see Muto, S. et al. Oshima, R. et al. 3: 56 Osipenko, L A . see Kokorin, V.V. et al. Osipov. V.V. see Khryapov, V.T. et al. Ossi, P.M. 1: 809 Ostenson, J.E. see Sevov, S.C. ef al.; Verhoeven, J.D. et d. Osterwalder, J. see Aebi, P. et al., Schlapbach, L. et al.; Stuck, A. et al. Osterwalder, J. et al. 2: 479; 3: 148 Ostling, M. see Norstrom, H. et al., Nygren, S. et al. Ostling, M. et al. 2: 623, 624
Ostlund, S. 1: 491; see also Levine, D. et al. Osugi, R. see Ohashi, K. et al. Oszlanyi, G. see Stephens, P.W. et al. Otani, S. see Ishizawa, S. et al. Otani, S. et al. 1: 954 Otaiii, Y . see Coey, J.M.D. et al., Sun, H. et al.; Yuasa, S. et al. Otani, Y. et al. 1: 749 Oti, J.A. see Yu, K.O. et al. Oti, J.A. and Yu., K.O. 3 602, 603, 604 otsuka, K. see Colberg, D, et al.; Miyasaki, S. et al. Otsuka, K. and Ren, X . 3: 56, 404 Otsuka, K . and Wayman, C.M. 3: 55 Otsuka, K. ef al. 3: 55, 56, 72 Qtsuka, M, see ino on is hi, Y et al. Otsuki, M. 2: 158, 164, 165 Ott, E.A. 3: 366 Ott, H.R. 1: 211, 213, 215, 216, 217, 219 Ott, H.R. et al. 3 45 Ott, J. see Horvath, J. et al. Ottaviani, G. 1: 999 Ottaviani, G. et al. 2 616, 617, 619, 622 Otto, A. see Knippscheer, S. et al. Otto, M.J. et al. 1: 950 Ou, B.L. 2: 188 Overbury, S. 1: 612, 616 Overbury, S. et al. 1: 616 Overbury, S.H. see Mullins, D.R. and Overbury, S.H. Overhauser, A.W. 1: 440 Overhoff, H. see Knecht, J. et al. Overs, M.P. see Taylor, T.A. et al. Ovsyuk, V.N. 2: 418,419, 432 Owczarski, W.A. see Erickson, J.S. et al. Owerm, D. see Nieh, T.G. et al. Owen, D.M. see Conner, R.D. et al. Owen, J.R. 2: 510 Owen, N.B. et al. 1: 184 Owens, K.E. see Meschter, P.J. et al. Ownby, G.W. see Hammer, L. et al. Oya, Y. 2: 214; see also Mishima, Y. et al.; Noguch~,0. et al., Oehiai, S. et al.; Suzuki, T. et al.; Wee, D.-M. et al., Yadagawa, NI. et al.; Yodogawa, Y. et al. Ozerova, M.I. see A ~ O S OVVYa. , et al. Ozeryansky, G.M. see Gregory, E. et al., Schwall, R.E. et al.; Summers, L.T. et al.
Qzlbilen, S. 2: 188 Pablenko, S.G. see Elagin, V.1 et al. Paccard, D. see Moreau, J.M. et al. Paccard, L. see Landgraf, F.J.G, et al.; Moreati, J.M. et al. Pack, J.D. 1: 132, 198 Packer, C.M, 1: 997 Padalia, B.D. see Mazumdar, C. et al.; Nagarajan, R. et al. Padella, F. see Magini, M. et al. Paden, B. see Choi, B.W. et al. Padgett, R.A. see George, E.P. et al.; White, C.L. et al. Pa~enba~ger, G.C. 2: 590 PaEet, M.T. 1: 987, 993 Page, R.A. see Davidson, D.A. et al.; Davidson, D.L. er al.
929 Pai, C.S. see Hung, L.-S. et al.; Lau, S . S . et al.; Marshall, E.D. et al., Suni, I. et al. Pai, C.S. et al. 2: 628; 3: 666 Paidar, P. 1: 496 Paidar, P. et al. 1: 498 Paidar, V 1: 528, 925; 3: 442, 454, 463, 4450; see also Ezz, S.S. et al.; Yamaguchi, M. et al. Paidar, V. et al. 1: 529, 547, 549,914,915; 2: 17. 23, 24, 25, 27, 137, 164, 165; 3: 446,450, 463, 772 Paine, B.M. 1: 705 Paine, R.M. 3: 45, 47; see also Booker, J. er al. Paiiie, R.M. et al. 1: 999 Painter, G.S. 1: 907; 2: 204; see also Liu, C.T. et al. Painter, T.A. see Steeves, M.M. et al. Pak, H.R. et al. 1: 843; 2: 22 Palache, C. et al. 1: 626, 629, 630 Palacios, E. see Obbade, S . et al. Palenzona, A. see Canepa, F. et al.; Fornasini, M.L. et aZ.; Iandelli, A. and Palenzonn, A.; Manfriiietti. P. et al. Paliard, M. see Cuedo, J.Y. et al. Paljevic, M. 1: 997 Palm, M. et al. 3: 368 Palmer, I.G. et al. 2: 175, 187, 188 Palmer, R. et al. 1: 782 Palmer, S.B. see Butler, B. et al. Palmier, E. see Defour, C. et al.; Wang, Z.G. et al. Palmstrom, C.J. see Olowolafe, J.P. et al.; Sands, T. et al. Palmstrom, C.J. et al. 2: 617, 621; 3: 783, 784, 786 Palstra, T.M.M. see Mason, T.E. et al. Palstra, T.M.M. et al. 1: 220; 3: 174, 177 Palumbo, G. see Cheung, C. et al. Pamplin, B.R. 1: 347,351,356; 2: 323; see also Hughes, O.H. er al., Spencer, R.M. et al. Pan, A.I. et al. 2: 623 Pan, E.T.4. see Finetti, M. et al.: Zhu, M.F. et al. Pan, H. see flu, J. et al. Pan, M.G. see Wan, X.F. et al. Pan, W.Y. see Yang, F.M. et al. Pan, Q. see Uang, U.C. et al. Pan, T. see Ray, U. et al. Pan, Y.C. 1: 927 Pan, Y.C. et al. 1: 983, 986, 989 Panasyuk, V.V. see Yartys, V.A. et al. Panchenko, N.M. see Eremenko, V.N. et al. Pandey, A.K. see Jha, S.C. et al. Pang, L. see K m a r , K.S. and Pang, L. Pang, L, and Kumar, K.S. 3 613, 614 Paiii, M. see Canepa, F. et al.; Merlo, F. et al. Panicker, M.P.R. et al. 2: 511 Panish, M.B. 3: 781; see also Hayashi, I. et al. Panizzieri, R. see Sanchez, J.L. et al. Pankey, T. see Davey, J.E. and ank key, T. Panko, T.I. et al. 1: 955 Pankratov, N.A. 2: 418 Pannier, J. et al. 1: 260
Panova, J. see Farkas, D. et al. Paiiteleimonov, L.A. 1: 712, 713, 719 Panteleinaonov, L.A. et al. 1: 721 Pao, Y.H. et al. 3: 812, 824 Paoli, T.L. see Epler, J.E. et al. Paoluzi, A. see Ibarra, M.R. et al. Pa~aco~stantopou~os, D. 3: 774 ~apaconstantopoulos,D.A. 1: 135, 195; 3: 143; see also Chubb, S.R. et al.; Mehl, M.J. et al., Sigalas, M. et al.; Siglas, M. et al.; Singh, R.J. et al.; Trambly de Laissardiere, G, et al. Papadopolos, Z. see Kramer, P et al. Papaleo, R. 2: 278 Papanicolaou, N.A. et al. 2: 626 Papirov, 1.1. er al. 1: 960 Papon, A M . et al. 3: 308, 309 Papov, A.M. 1: 956 Paradiso, E. see Magini, M. et al. Parathasanathy, TA.see Dimiduk, D.M. et al. Pardy, J.R. I: 716 Pareti, L. see Ibarra, M.R. et al. Parette, G. see Mirebeau, I. et al. Parfitt, L.J. et al. 1: 987, 994; 2: 168, 169 Parida, B.K. and Nicholas, T. 3: 337 Parikh, M. see Rozgonyi, G.A. et al. Parinello, M. see Carr, R. and Parrinello, M. Paris, D. see Weber, D. et al. Paris, H.G. see Mitchell, R. et aE. Pans, H.G. et al. 2: 186 Parish, P.C. and Erdogan, F.J. 3: 325 Park, D.G. et al. 3: 668 Park, H.K. 2: 311, 314 Park, J,-G, 1: 221 Park, J.K. 2: 282 Park, K.T. 1: 916 Park, K.T. and Coo, E. 3: 416 Park, R.M. see Meurnark, G.F. et al. Park, R.M. et al. 2: 326 Park, S.J. et al. 3: 647 Parker, B.A. 2: 175, 184 Parker, E.H.C. 2: 421, 425; 3: 779 Parker, F.T. see Berkowitz, A.E. et al.; Oesterreicher, H. et al. Parker, M.R. 2: 435, 438 Parker, S.G. 2: 328 Parker, S.P 3 165 Parkes, A. 1: 5; 2: 516 Parkey, B.A. see Tibballs, J.E. et al. Parkinson, M.et al. 2: 507 Parks, D.M. and Ahzi, S. 3: 371 Parks, R. see Lawrence, J.M. et al. Parks, T.C. see Alonso, T. et al. Parog, M.D. 1: 768 Parrinello~M. 1: 133, 196, 208; see also Ercolessi, F. et al.; Kohanoff, J. et al.; Wengert, S. et al. Parns, P.A. 2: 178 Parsey, J.M. Jr. see Goldman, A.I. et al. Parsonage, N. and Staveley, L.A.R. 3 246 Parsons, J.R. see Swanson, M.L. et al. Parsons, W.S. see Tomhewicz, M. et al. Pa~thasarathi,A. 1: 99Q; 2 58 Parthasarathy, T A . see Dimiduk, D.M, et al., Mendiratta, M.G, et al.; Rao, S.I. et al.: Subrama~ian,P.R. et al.
Parthasarathy, T.A. et al. 1: 88, 523, 529, 536; 3: 305, 317452 Partht, E. 1: 302, 303, 304, 343, 345, 347, 348, 350,351,353, 354,367,417,568; 3: 7; see also Fornasini, M.L. et al.; Gladyshevskii, R.E. et al.; JoubertBettan, C.A. et al.; Linza-de-Faria, J. et al.; N~wotny,H. et al.; Sauvage, M. et al.; Zhao, J.-T. et al. ParthB, E. see Gladyshevskii, R.E. et al.; Hohnke, R. and Parthi, E.; Schob, 0. and Parthk, E. ParthB, E. and Chabot, B. 3: 90 Partht, E. et al. 1: 356, 360, 40311 Partridge, P.G. 2: 298 Parvmmehr, A. ,Tee Jenner, A.G.I. et al. Parvinmehr, A. et al. 2: 403 Pascoe, R.T. 2: 58, 61 Pashley, D.W. 2: 564; see also Hirsch, P. et al. Pashley, D.W. et al. 1: 534; 3: 413 Pasianot, R. 1: 88; see also Farkas, D. et al. Pasianot, R. et al. 1: 523, 527, 541; 2: 21; 3: 768 Paslay. P.R. et al. 3: 302 Pasotti, G.see Blau, B. et al. Pasturel, A. see Burton, B. et al.; Burton, B.P. et al.; Colinet, C, et al.; Le, D.H. et al.; Manh, et al. Pasturel, A. et al. 1: 114, 115, 117 Pasztor, G. see Blau, B. et al. Patankar, S.N. see Hardwick, D.A. et al. Patel, S. see Ciien, D.Y. et al. Pathare, V.M. see Vedula, E(. et al.; Whittenberger, J.D. et al. Patnaik, B. see Rozgonyi, G.A. et al. Patnaik, P.C. 1: 987, 996; 2: 227, 228 Paton, N.E. see Rhodes, C.G. ei al. Paton, N.E. and Backofen, W.A. 3: 424 Patnck, D.K. and Van Aken, D.C. 3: 657 Patrick, F.E. see Slack, S.S et aZ. Patten, E.A. et al. 2: 419 Fatten, J.W. 2: 45; see also Prates, J.T. et al. Patterson, A.L. 1: 277, 278, 302, 303, 304 Patterson, R A . et al. 1: 655 Patterson, W. see Jeon, H. et al. Pattnaik, A. see Skowronek. C.J. et al. Patton, W. see Green, G. et al. Patton, W.G. see Ziinm, C B , et al. Paufler, P 1: 252, 351, 911, 913, 915, 921; 2: 238, 252; see also Nghiep, D.P. et al.; Quyen, N.H. et al. Paufler, P. er al. 2: 238 Paul, J. see Chen, W et aL Paul, J.D.H. see Appel, F et al. Paul, J.D.H. et al. 3: 618 Paulevk J. 1: 63311, 688; see also Nhel, L. et al. PaulevB, J. et al. 1: 63311, 688 Paulikas, A.P. see Campuzano, J.C. et al.; List, R.S. et al.; N a ~ ~ s aI nS. , et al., Olson, C.G. et al. Pauling, L. 1: 11, 278, 280, 281, 282, 294, 301, 302, 305, 319, 453, 662; 2: 9; 3: 179; see also Bergman, G. et al.; Shoemaker, D.P. et al. Paulsen, C. see Berger, C. et al. Paulson, W. 1: 781 Pauly, H. et al. 3: 238
930 Paumier, E. see Dufour, C. et al., Toulemonde, M. et al. Paunovic, M. et al. 3: 671 Pauthenet, R. see Bertaut, E.F. et al.; NCel, L. et al. Pavone, R. see Cianozzi, P. et al. Pawlak, Z. see Jackson, A.C. et al. Paxton, A.T. see Manh, et al., Pasturel, A. et al. Paxton, A.T. and Entwisle, A.R. 3: 418 Paxton, A.T. and Pettifor, D.G. 3: 58 Paxt on, D,M . see Kestner-Weykamp, H.T. et al. Payne, A.P. 2: 610, 613 Payne, J.E. and Desai, P.D. 3: 808 Peacock, D.E. see Tien, J.K. et al. Pearah, P.J. see Baillargeon, J.N. et al. Pearsall, T.P. see Temkin, H. et al. Pearson, D.D. see Anton, D.L. et al.; Giamei, A.F. et al. Pearson, D.D. et al. 3: 314, 317 Pearson, J. see Okamoto, P.R. et al.; Rehn, L.E. et al. Pearson, W.B. 1: 10, 109, 115, 183, 204, 205,238,241,242,249,252,277,279, 281,304, 317,336, 341, 357,363, 364, 365,366,367,370,378,386,387,388, 391,397, 398,399,403,407,419,420, 422,431,434,473,477,712,713, 876, 885, 887; 2 311; 3: 4, 841, 845’; see also Hellner, E. et al. Pease, D.M. see Russell, S.M. et al. Pease, R.S. 1: 784, 803 Peat, A.J. SPP Westbrook, J.H. et al. Pecharskij, V.K. see Gladyshevskii, E.I. et al. Pecharshy, V.K. see Bodak, 0.1.et al. Pecharsky, A.O. see Choe, W. et al.; Pecharsky, V.K. et al. Pecharsky, V. see Zimm, C.B. et al. Pecharsky, V.K. see Choe, W et al.; Dan’kov, S.Yu. et al.; Gladyshevskii, E.I. et al.: Gschneidner, K.A. Jr. and Pecharsky, V.K.; Korte, B.J. et al.; Levin, E.M. et al.; Niu, X J . et al.; Takeya, H. et al.; von Ranke, P.J. et al. Pecharsky, V.K. and Cschneidner, K.A. Jr, 3: 519, 520, 524, 525. 527, 530, 531, 533, 534 Pecharsky, V.K. et al. 3 527, 529 Peck, W.F Jr. see Cam, R.J. et al.; Siegrist, T. et al.: Zandbergen, H.W. et al. Peck, W.W. Jr. see Ikushima, K. et al. Pederson, M.R. see Erwin, S.C. et al.; Perdew, J.P. et al. Pedraza, D.F 1: 570, 797, 798, 809, 817, 818; see also Chen, F.C. et al. Pedraza, D.F. et al. 1: 570 Peercy, P.S. see Doyle, B.L. et al., Poate, J.M. et al. Peets, C.S. see Wieber, R.H. et al. Pei, S. et al. 1: 65, 68, 72, 505 Peierls, R. I: 504, 873 Peiniger, M, see Dasbach, D. et al. Peisl, J. see Dosch, H. et al., Voges, D. et al. Peker, A. 1: 692 Peker, A. and Johnson, W.L. 3: 682 Pekker, S. see Stephens, P.W. et al.
Author Index Pelissier, J. 1: 577 Pellegrmi, P.W. see Tung, R.T. e? al. Pellerin, F see Baccino, R. et al. Pelletier, J.M. see Hidouci, A. and Pelletier, J.M. Peltier, J. see Das, S.K. et al. Peltner, H.E. 1: 93 Pelton, A.D. and Schmalzried, H. 3: 154, 155 Pelton, A.D. and Thompson, W.T. 3: 154, 155 Pelton, A.R. see Tanner, L.E. et al. Pendry, J.B. see MacLaren, J.M. et al. Peng, Z. see Suryanarayana, C. et al. Penisson, J.M. see Chamberod, A. et al.; Loubradou, M. et al. Peiiney, T. e i al. 1: 875 Penny, T. see Mook, H.A. et al. Pennycock, S.J. see Baker, 1. et al. Penrose, R. 1: 480 Penton, R.J.T. et al. 3: 337 Penzo, C . see ParthC, E. et al. PCpm, H. see Boily, S. et al. Pepm, P. see Samfort, G. et al. Peralta, P. see Chu, F. et al. Percheron-GuCgan, A. et al. 2 510 Perdew, J.P 1: 130, 132, 138 Perdew, J.P. et al. 1: 130, 197, 203, 204 Perdok, W.G. 1: 167 Pesepezko, J.H. 1: 69, 638, 733, 861, 862; 2: 75, 219, 221; see also Allen, W.P. et al., Boettinger, W.J. et al.; Doyle, B.L. et nl.; Giiken, M. et al.; Lee, M.C. et al.; Nunes, C.A. et at., Sakidja, R. et al.; Thomas, R.E. et al.; Wiley, J.D. et al. Perepezko, J.H. et al. 2: 95; 3: 487, 488 Peretti, E.A. see Fretague, W.J. et al. Perevezentsev, V.N. e f al. 1: 925 Pereyra, R. see Hanrahan R.J. Jr. et al. Perez, J.F. et al. 3: 462 Perez-Campos, R. et al. 1: 467, 483 Perez-Ramirez, J.C. see Perez-Canipos, R. et al. Peric, Z. see Borgstedt, H.U. et al. Perin, R. 2 377 Perkins, F.C. 3: 44, 45 Perkins, J. see Wu, M.H. et al. Perkins, R.A. 1: 988, 991, 996, 997; see also Meier, G.H. et al. Perkins, R.A. and Meier, G.W. 3: 575, 576, 579 Perkins, R.A. et al. 1: 987, 988, 992, 993, 994, 995, 1002; 2 81, 82 Perrier de la Bathe, R. see Allemand, J. et al. Perrin, I.J. 3: 483 Perrin, R. see ten Kate, H.H.J. et al. Pershan, P.S. 2 440; see also DiMasi, E. et al. Persson, P.A. 2: 275, 276 Pertrovic, J. 3: 317 Pessa, M. see Saarinen, K. et al. Pestman, B.J. et al. 1: 88, 598, 600, 603 Peter, F see Jia, C.L. et al. Peters, B. see Frier, P. S. et al. Peters, E.M. see Somer, M. et al. Peters, J.A. 1: 655; see also Bassi, C. et a/* Peters, R.see Grin, J. et al.; Somer, M. et al.; von Schnering, H.G. et al.
Peters, M. see Kumpfert, J. et al.; Leyens, C. et al.; Welpman, K. et al., Winkles, P.J. et al. Peters, M. et al. 2 124 Petersen, A. 3: 74 Petersen, A. see Czybulka, A. et al. Petersen, G.F. see Schneibel, J.H. et al. Petersen, J.F. et al. 2: 306 Petersenn, A.V. 3: 238 Petersenn, A.V. and Sehuster, H.U. 3: 238 Peterson, D.E. see List, R.S. et al. Peterson, E.E. 1: 5’94 Petersson, C.S. see d’Heurle, F.M. et al.; Norstrom, H. et al.; Nygren, S. et al.; Ostling, M. et al. Petersson, S. see Baglin, J.E.E. et al. Petez, L. e? al. 2: 159, 161, 165 Pethick, C.J. see Fisk, Z , et al. Petit, J. see Mabru, C. et al.; Sainfort, G. et ul.; Tonneau, A. et al. Petit, J.I. see Paris, H.G. et al. Petrich, R.R. 1: 700, 701; see also Schwarz, R.B. et al. Petrii, O.A. et al. 2: 504 Petroff, Y. see Grioni, M. et al. Petrov, 1. see Adibi, F et al.; Bergstrom, D.B. et al.; Greene, J.E. et al. Petrov, I. et al. 3: 666 Petromc, J.J. 1: 391, 997, 998; 2: 223, 295; 3: 454, 455; see also Carter, D.H. et al.; French, J.D. et al., Gibbs, W.S. et al.; Maloy, S. et al.; Ramamurthy, U. et al., Sadanaiida, M. et al., Schwarz, R.B. et al.; Unal, 0. et aE.; Vasudevan, A.K. and Petrovic, J.J. Petrovic, J.J. and Vksudevan, A.K. 3: 454,455, 485, 657 Petrovic, J.J. et al. 1: 998, 999, 1008; 2: 290 Petrovich, A. 2: 361 Petrushevskii, M.S. et al. 2 54 Pettersen, K. see Ulvensoen, J.H. et nl. Pettifor, D.G. 1: 14, 31, 33, 36, 56, 57, 81, 238,241,243,266,267,269,381,385, 389,390,393,396,400,419,420,422, 425,426,429,431,432,433,434,435, 436,437, 683, 713, 923, 931; 2: 157, 610; 3: 90, 235, 236, 237, 811, 841, 842; see also Beer, N.and Pettifor, D.G.; Lm, C.T. et al.; Manh, et al.; Nicholson, D.M. et aI.; Znam, S. et al. Pettifor, D.C. and Aoki, M. 3: 215 Pettifor, D.G. and Podloucky, R. 3: 236 Pettifor, D.G. et ul. 1: 80 Pettit, F. see Meier, G. and Pettlt, F. Pettit, F.S. 1: 980,982,983,987,989,992, 993, 994, 996, 1008; 2: 17, 494, 495; see also Ashary, A. et al., Berztiss, D.A. et al.; Kear, B.H. et al.; Schaeffer, J. et al. Petty, M. see Brown, J.D. et al. Petukhov, V.V. see Kiselyova, N.N. et al.; Yrrdiaa, N.V. et al.; Zeniskov, V.S. et al. Petyukh, V.M. see Semenova, E.L. et al. Petzow, G. 1: 717, 718, 719, 726; 2 577; see also Henig, E.-T. et al.; Laag, R. et al.; Stadelmaier, H.H. et al. Petzow, G. and Effenberg, G. 3: 801
A ~ t h Index o~ Peyghambarian, N. see Jin, R. et al., Ojinia. M. et al. Pfaffenberger, J. 2: 306 Pfahler, K. see Horvith, J. et al. Pfann, W.G. 2: 395; 3: 545 Pfeifer, H.U. et al. 1: 721 Pfeiler, W. 1: 777, 778, 779, 782; see crlso Schrank, J. et al. Pfeiler, W. et al. 1: 782 Pfister, J.C. see Rosencher, E. et al. Pfbllmann, Tk. see Oehring, M. et al. Phan, 1. 1: 532; see also Wardle, S. et al. Phan~Co~rson. I. 3: 413, 419 Phanl, M.K. see Binder, K. et al. Pharr, G.M. et al. 3 43 Philibert, J. 1: 829 Philips, D. see Baeslack, W.A. I11 et al. Phillips, D. see Howe, L.M. et al. Phillips, J.C. 1: 265, 357, 358,419; 2: 610; see alsu Andreoni, W et al.; Machlin, E.S. et al.; Rabe, K.M. et al.; Villars, P et al.; Zhang, S.B. et al. Phillips, J.E. see Shafarrna~,W.N. et al. Phillips, N.E. 1: 1019. 1020 Phillips, R. 1: 484 Phillips, R. et al. 1: 71 Phillips, R.B. see Gibbons, P.C. et al. Phillips, R.B. et al. 1: 485 Pliillips, T.E. see Rubinstern, M. et al. Phillips, V.A. 1: 948; 2: 271, 283 Philofsky, E. 2 654 PhragmCn, G. 1: 10, 13; 2: 177 PhragmCn, G. see Westgren, A.F and PhragmBn, G. Piao, M. see Otsuka, K. et al. Piatetsky-Shapiro, 6. see Fayyad, U.M. et al. Piatti, G. and Bardy, M. 3: 61, 78 Piatti, G. and Pellegrini, G. 3: 76 Picard, C. see Nicolas-Chaubet, D. et al. Piccari, L. 2: 412 Pichoir, R. 2: 490; see also Duret, C. et al., Merrel, R. et al. Pickart, 5.3. 1: 441, 444 Pickens, J.R. 2: 159, 163, 164, 165 Pickering, M.A. see Goela, J.S. et al. Pickett, J.J. 2: 361 Pickett, W.E. 1: 134, 195, 196, 197; see also Erwin, S.C. et al., Singh, D. et al., Wang, C.S. et al. Pickett, W.E. et al. 1: 130 Pickus, M.R. 2: 360; see also Dariel, M.P et al. Piearcey, B.J. see Kear, B.H. and Piearcey B.J Piearcey, B.J. and Terkelsen, B.E. 542 Piearcey, B.J. and VerSnyder, F.L. 542 Pieczonka, T. see Gia~anel~a, S. et al. Piel, H. see Dasbach, D. et al. Pier, D. see Michell, K. et al. Pierce, C.B. see List, R.S. et al. Piercy, G.R. 1: 791, 792, 803, 804 Pierre, J. see Guzik, A. and Pierre, J.; Karla, I. et al. Pierron-Bohnes, V see Cadeville, M. et al.; Kozubski, R. et al., Vennkgues, P. et al. Pierson, H.0. 3: 669, 670 Piesbergen, U. 1: 1022 Pignatel, G. see Nava, F. et al.
Pike, L.M. 1: 925; see also Chang, V.A. et al., Liu, C.T. et al. Pike, L.M. et al. 3: 288. 359 Pilling, J. 3: 803 Pilo, Th. et al. 3: 139 Pinch, H.L. see Wen, C.P. et al. Pineart, A. 2: 277, 278; see also Taillard, R. et al. Pines, B.J. 1: 761, 762, 763, 764, 766, 768 Pines, D. 1: 128, 132; see also Fisk, Z. et al. Pinkerton, F.E. see Croat, J.J. et al., Fuerst, C.D. et al. Pinnel, M.R. 2: 653 Pinski, F.J. see GyorRy, B.L. et al.: Johnson, D.D. et of.; Sluitcr, M. et al.; Sluiter, M.H.F et al.; Staunton, J.B. et al.; Stocks, G.M. et al.; Turchi, P.E.A. et al. Pint, B.A. 1: 983, 986, 987, 990, 1007 Pint, B.A. et al. 1: 986, 987 Pintschovius, L. et al. 1: 158 Piotrowski, A. 2: 503 Piotrowski, J. 2: 419; see also Becla, P. ei al. Piratmska, 1.1. see Senchenko, A.A. et al. Pirouz, P 3: 415 Pirouz, P. see Miillner, P and Pirouz, P Pischik, B.N. see Savitskii, E.M. et al. Pitinan, K.C. see Abell, J.S. et al. Pitner, W.R. see Tierney, B.J. et al. Pitney, K.E. 2: 653 Pitsch, W 1: 38, 783, 849, 850 Pitt, C.W. 2: 413 Planes, A. see Elgueta, J. et al. Plaskett, T.S. see Penney, T. et al. Plattner, H. see Howe, L.M. et al. Platzgummer, E. see Gauthier, Y. et al. Pleier, S. see Bliim, M. et al. Plimpton, S.J. see Donati, C. et al. Ploog, K. 3: 779 Ploog, K. see Chang, L.L. and Ploog, E(. Plotnikova, N.P see Hertzriken, S.D. et al. Plurnmer, E.W see Lui, S . 4 . et al. Poate, J.M. see Hensel, J.C. et al., Tung, R.T. et al. Poate, J.M. et al. 1: 732; 2: 630 Pocholle, J.P. see Razeghi, M. et al. Podbel’skii, V.V see Degtyaryov. Yu, I. et al. Podlesnik, D.V. see Willner, A.E. et al. Podlesov, V.V. et al. 3: 736 Podloucky, R. 1: 14, 56, 57, 420; see also Guo, X.-Q. et al., Pettifor, D.G. and Podloucky, R., Sluiter, M. et al., Sluiter, M.H.F. et al., Stadler, R. et al.; Steinernann, S.G. et al.; Vogtenhuber, D. and Podloucky, R. Podtykan, V.P see also Glushko, P.I. et al. Poerschke, R. see Dimitrov, C. et al. Pogosov, V.K. see Bulycheva, Z.N. et al. Pohl, A. see Eriksson, 0. et al. Poiblaud, G. see Martin, G.M. et al. Poizat, J.C. see Baudin, K. et al. Poize, S. 1: 991, 994; .see also Streiff, R. and Poize, S. Pokela, P.J. see Dorner, W. et al.; Kolawa, E. et al.
93 1 Poker, D.B. see Pal, C.S. et al.; Xia, W et al. Pokrovskii, B.I. see Kiselyova, N.N. et al. Polik, J. 1: 958 Poldy, C.A. see Kirchmayr, H.R. and Poldy, C.A. Poliak, R.M. 1: 80 Polishtuk, D.F. see Larikov, L.N. et al. Politis, C. 2: 610 Polk, D.E. 1: 735, 741; 3: 687; see also Calka, A. et al.; Davis, S. et al.; Fischer, M. et al.; Ciessen, B.C. et al.; Sinha. A.K. et al. PoIIniann, J. see Kruger, P. ez d. Pollock, T. and Argon, A. 3: 297,302,303 Pollock, T.M. see De Graef, M. et al.; Eow, K. et al.; Kitano, K et al.; M u r a l e e d ~ ~ r K. ~ n et , al., Risliel, L.L. et al.: Shi, X. et al.; Walston, W.S. et al. Polmear, 1.J. 2: 175, 176 Polotskaya, R.1. see Lukashenko, G.M. et al. Polvani, R.S. 1: 921, 927; see also Strutt, P.R. e f al. Polvani, R.S. et al. 1: 537, 925; 2: 61 Polyakova, N.A. 2: 282 Pomeroy, A.R. see Campbell, LA. et al. Pomey, 6. 1: 98 Pornpe, G. 1: 956 Pond, R.C. 1: 598; 3 410; see also Hirth, J.P and Pond, R.C.; Nixon, T. and Pond, R.C. Ponge, D. and Gottstein, 6 . 3: 639 Ponornarenko, V.P. see Khryapov, V.T. et ul. Ponomarev, M.V see Greenberg, B.A. et al. Pontikis, V. 1: 820; see also Kubin, L.P et al.; Massobrio, C. et al., Rey-Losada, C. et al. Pontonnier, L. see Bawdry, A. et al. Ponyatovsky, D.G. see Degtyareva, V.F. and Ponyatovsky D.G. Ponyatovsky, E.G. 1: 702, 703; see also Aptecar, I.Z. et al. Degtyareva, V.F. et al. Tissen, V.G. et al. Ponyatovsky, E.G. and Belash, I.T. 3: 155, 161 Poole, P.H. see Donati, C. et al. Poon, S.J. I: 703, 704, 705; see also Biggs, B.D. et al., Cassada, W.C. et al.; Chen, H. et al., He, U.et al.; Johnson, W.L. et al.; Koflat, D.D. et al.; von Lohneysen,H. et al. Poon, S.J. et al. 1: 454 Pope, D.P 1: 498,521, 528,530,540,547, 548, 896, 914, 915, 928; 2: 18. 25, 26, 27, 32, 162, 224, 231; 3: 440; see also Bonda, N.R. et al.; Chu, F and Pope, D.P., Cserti, J. et al., Ezz, S.S. et al.; George, E.P. et al.; Heredia, F.E. et al., Inui, H. et al., Johnson, L.A. et al., Khantha, M. et al.; Lall, C. et al., Liu, C.T. and Pope D.P., Luzzl, D.E. et al.; Mahapatra, R. et al.; Paidar, P et al.; Paidar, V. et al.; Takasugi, T. et al., Tichy, 6. et al.; Urnakoshi, U et al.; Vitek, V
932 et al., Wee, D.M. et al., Whang, S.H. et al., Wu, Z.L.; Yamaguchi, M. et al. Pope. D.P. and Ezz, S.S. 3: 305, 361, 371,
440, 708 Pope, D.P. et al. 3: 545, 546 Pope, W.J. 1: 10 Popescu, R. see Manaila, R. et al. Popov, E.V 3: 818 Popov, L.E. et al. 1: 496, 502 Popov, S.A. see Khryapov, V.T. et al. Popova, S.V. 1: 707; see also Larchev, V.I. and Popova S.V. Popova, T.V. see S a ~ u z i n a R.G. , Poppe, U. see Cox, G. et al. Popplewell, J. see Keeling, L. et al. Windle, P.L. et al. Poquette, G.E. 1: 511 Porsch, F see Degtyareva, V.F. et al. Porter, A.J. see Ricks, R.A. et al. Porter, D.A. 1: 391 Porter, D.A. and Easterling, K.E. 3: 225 Porter, W.D. 1: 433, 907; 2; 157; see also George, E.P. et al.; Inui, H. et al.; Schneibel, J.H. et al.; Sparks, C.J. et al., Viswanathan, S. et al. Porter, W.D. et al. 1: 66; 2: 157, 165 Porter, W.J. see Larsen. J.M. et al. Porter, W.J. 111. see Jones, P.E. et al. Portier, R. see Finel, A. et al.; Traverse, A. et al.; Yu-Zhang, K. et al. Portier, R. et al. 1: 466 Posgay, G. et al. 1: 743 Pospelov, G.S. 3: 818 Postans, P.J. et al. 2: 88 Posternak, M. see Krakauer, H. et al. Postikov, A.V. see Galakov, V.R. et al. Postogvard, G.I. see Glushko, P.Z. et ai. Potgieter, €3. see WO@ I.M. et nl. Potori, M.V. see Motrya, S.F. et al. Pottebohm, H. et al. 2 272 Potter, D.I. 1: 504, 649; see also Mayer, S.G.B. et al. Potter, H. 1: 618 Potter, H.C. and Blakely, J.M. 3: 218 Potter, H.H. 2: 308 Pottgen, R. 3: 95 Pottgen, R. see Kuncser, V. et al.; Prots, Y.M. et al. Pottgen, R. et al. 3: 95 Potts, J.E. see Cheng, H. et al. Potzel, U . et al. 3: 253 Pouch, J.J. see Oh, J.E. et al. Pouchard, M. see Cros, C. et al.; Kasper, J.S. et al.; Reny, E. et al. Poulsen, S. see Sanger, P et al. Pouranan, I?. see Zhang, L.Y. et ol. Pourrahimi, S. see Williams, J.E.C. et al. Povolo, F. 1: 791 Powell, R.W. see Touloukian, Y.S. et al. Powers, J.M. see MoBett, M.B. et al.; Powers, W.0. 2: 157, 160, 162; see also Mysko, D.D. et al.; Turner, C.D. et al. Poyet, P. et al. 1: 98 Pozyukov, Yu.P. see Muslimov, I.S. et al. Prakash, U. see Baligidad, R.G. et al. Prakash, U. et al. 1: 502, 924 Prasad, A. et al. 2: 565, 566 Prasad, L.C. 1: 676, 678 Prasad, R. see Chattopadhyay, K. et al.
Author Index Prasad, Y.V.R.K. see Sagar, P.K. et al., Sundar, R.S. et al. Prasad, Y.V.R.K. et al. 2: 125, 127 Prassidec, K. see Christides, C. et al.; David, W.I.F. et al. Prater, J.T. et al. 2 492, 495 Pratt, A.S. see Willey, D.B. et al. Pratt, J.N. 1: 64; see also Little, L. et al. Pratt, P.L. 2 217 Preble, J. see Cainpb, I.E. et aE. Predecki, P. et al. 1: 105 Predel, B. 1: 96, 97, 102, 103, 104, 105, 107, 108, 109, 113, 114, 666, 671; see also Ellner, M. et al.; Moser, Z. et al.; Rosa, C.J. et al.; Sommer, F. et al.; Witlman, N. ef al. Predel, B. and Madelung, 0. 3: 799, 802 Pregger, B.A. et al. 1: 1007 Prescott, R. 1: 981, 987; see also Mitchell, D.F. et al. Presland, A.E. 2: 5G4 Press, M.R. et al. 2: 481 Press, W*H.et al. 1: 198, 208 Presthus, R. see Zribi, A. er al. Preston, G.D. 2: 576 Pretorius, R. 2: 606, 609; see also Li, J. et al. Pretorius, R. et al. 2 606, 608, 609, 630 Preuhs, J. see Gromann, J. et al. Preuss, E. see Wuttig, M. et al. Prevender, T.S. see Schiltz, R.J. Jr. et al. Prewo, K.M. see Bhowal, P.R. et al.; Nardone, V.C. et ul. Price, A. 3: 602 Price, D.L. see H~mpelmann,R. et al.; Lim, S.P. et al.; Reijers, H.T.J, et al.; Richardson, J.W. et al., Rowe, J.M. ef al., Saboungi, M.L. et al. Price, D.L. and Saboungi, M.L. 3: 254 Price, D.L. and Skold, K. 3: 247 Price, D.L. et al. 1: 681; 3: 246, 255 Price, G.D. see Burdett, J.K. et al. Price, P.B. 3: 419 Price, P,B. see Fleischer, R.L. et al. Price, P.E. and Kohler, S.P. 3: 646 Price, S.L. see Burdett, J.K. et al. Pneskorn, J.N. et al. 2: 505 Prigogine, I. 1: 675, 678 Primus, A. see Hoenig, H.E. et al. Prince, A. 1: 248; see also Villars, P. et al. Prince, A. et al. 2: 561, 566, 567, 568; 3: 802 Prince, E. see Hardman-Rhyne, K. et al. Prince, M.Y. see Waber, J.T. et al. Pnns, 3.14. see Coster, D. et al. Prinz, G.A. and Krebs, J.J. 3: 783 Priolo, F. see Baeri, P. et al. Probst, H.B. 1: 1008 Proch, D. 2 376 Prochazka, S. see Klug, F.J. et al. Proctor, C.S. see Vittra, S. et al. Proft, J.L. see Moberly, W.J. et al. Proske, G. see Lutje~ng,G. et al. Protasov, V.I. see Moseev, N.V. et al. Prots, Y.M. et al. 3: 95 Proust, J.L. 1: 8 Provost, J. et al. 3: 263, 267, 269 Pryce, G.J. see Gill, S.S. et al. Przeorski, T. see Herzig, C. ef al.; Herzig, Ch. et al. Pryzybylsks, H. 1: 132
Psycharis, V. see Hu, Z. et al.; Kalogirou, 0. et al. Pugachev, N.S. see Glushko, P.I. et al. Pugh, S.F. 2: 166; 3: 240 Puiido, E. see Yavari, A. et al. Pupin, J.P. 1: 186 Puppel, D. see Reppich, B. et al. Purdy, G.R. 1: 867; see also Cornwell, L.R. et al. Puschl, W. 1: 782 Pushin, V.G. et al. 1: 716 Puska, M.J. see Alatalo, M. et al.; Jacobsen, K.W. et al. Puska, M.J. et al. 1: 78, 79 Putiliii, A.N. see Balagurov, A.Y. et al. Puzei, I.M. 2: 306, 307 Puziewicz, J. 1: 627, 628 Pyka, N.see Goldman, A.1. et al. Qi, M. see Sui, 5I.X. et al. Qi, Q. see Coey, J.M.D. et al.; Leithe-Jas~er,A. et al. Qian, Y. see Qu, X. et al. Qiang, C.J. see Tadaki, T. et al. Qigong, C. see Dong, 2;. et al. Qin, P. see Chen, N.-Y et al. Qin, Y.L, et al. 3: 397 Qing, W.D. .we Ymg, F.N. et al. Qiu, J. see Haase, M.A. et crl. Qiu, S.Y. 1: 483 Qu, X. et al. 1: 552 Quadakkers, W. see Wagner, R. et al. Quadakkers, W.J. see Singheiser, L. et al. Quad~ieg,P 1: 174, 178, 180, 183 Queirolo, G. see Nava, F. et al. Queaeau, V. and Sevov, S.C. 3: 115, 116, 117, 122 Quested, P.N. et al. 3: 306 Quilichini, M. et al. 1: 161, 162 Quin, P, see Chen, N.-Y. et al.; Y an, L.M. et al. Qumn, R.T. et al, 2: 651 Quivy, A. see Qui~ichini,M. et at., Yu-Zhang, K. et al. Quyen, N.H. see Nghiep, D.P et al. Quyen, N.H. et al. 2: 224 Raab, R. see Potzel, 'U. et al. Raaijmakers, I.J. see Cale, T.S. et al. Rabe, K.M. et al. 1: 265, 419,435, 486; J: 812 Raberg, W. see Shen, 2;. et al. Rabin, B.H. see Bose, A. et al.; Wright, R.N. et al. Rabin, B.H. and Wrrght, R.N. 3: 654 Rabin, B.H. et al. 3: 736 Rabkin, E. et al. 3: 225 Rachek, A.P see Mordovetz, N.M. and Rachek A.P Rachinger, W.A. 1: 504, 512, 832
Radev, D.D. and Klissurskl, D. 3: 758 Radhakrishna, A. see Baligidad, R.G. et al. Radhakrishna, P. see Murani, A.P. et al. Radmilovic, V. e f al. 2: 191 Ragg>O.M. and Warris, I.R. 3: 97 Raghavan, V. 2: 225, 226; 3: 803 Raghavan, V. et al. 3: 803
Author Index Raghuram, A.C. see Nimmagadda, R. et al. Rahman, A. see Limoge, Y et al. Rahmel, A. 1: 992,999; see also Becker, S. et al.; Haufe, K. and Rahmel A.; Kofstad, P et al. Rai, A.S. see Allen, C.W. et al. Rainbacher, A. see Weitzer, F. et al. Rainford, B.D. see Murani, A.P. et al. Rainville, M,W. 1: 791,806,807,809,810, 81I, 817; 2: 133; see also Wowe, L.M. et al. Raisson, C.see Baccino, R. et al. Raj, S.V see Bowman, R.R. et al.; Doychak, J. et al., Garg, A. et al., Whittenberger, J.D. et al. Raj, S.V. et al. 1: 988, 994, 1001, 1006; 2 61 Rajan, K. see Alman, D.E. et al., Kallingal, C.G. et al., Smith, T.R. et al. Rajasekharan, T. 1: 385,390,454; see also El Boragy, M. et al. Rajeshwar, K. 2: 511; see also Bhattacliarya, R.N. et al. Rakennus, K. see Saarinen, K. et al. Raleigh, D. see Nenezes, S. et al. Ralph, B. 2 19,22,23, 190,270,271,278; see abo Brandon, D.G. et al. Ram, M.L. see Sen, S. et al. Ram, S.V and Barrett, J.R. 3: 591, 594, 595, 598 Rania Rao, P. 2: 121 Ramachandran, G.R. et al. 3: 131 Ramachandra Rao, P. 1: 475, 737, 750 Ra~achandraRao, P. et aE. 1: 668, 680 Ramamurthy, U. et al. 3: 486, 344, 345 Rainan, A. 2: 156 Ramanarayanan, T.A. 1: 95 Ramanath, G. see Alien, C.W. et al. Ramanath, G. et al. 3: 664, 666, 669, 674 Ramasesha, S. 3: 250 Ramaswaimi, B. 1: 504, 505, 528; 2: 20, 22 Ramaswami, M. see Cale, T S , e f al. Ramaswamy, S. see Levine, D, et al., Lubensky, T.C. et al. Ranibaldi, G. see Ferro, R. et al. Ramdohr, P. 1: 635 Ramesh, R. see Harbison, J.P. et al., Sands, T. et al. Ramirez, A.P see Mason, T.E, et al. Ran, A. 1: 615; see also Farkas, D, and Ran, A. Ran, A.P. see Mason, T.E. et al. Rand, D.A.J. 2: 646 Rand, M.H. 3: 800 Rand, W.H. see Gsamei, A.F et al. Randall, R. et al. 2 358 Randhawa, H. 3: 663 Randin, J.P 3: 231 Raney, M. 1: 15; 2: 647 Ranganath, S. et al. 1: 925 Ran~anathan,S. 1: 467, 598; see also Brandon, D.G. et al.; Chattopadhyay, K. et al.; ~ukhopadhyay,N.K. et al.; Murty, B.S. and ~dnganathanS.; Swamy, V.T. et al.; Van Tendeloo, G. et al. Ranganathan, S . et al. 1: 471 Rangarajan, V. 1: 598, 600; 2 27 Range, K.-3. 2: 572
Rani, R. see Cadieu, F.J. et al.; Hegde, H. et al. Rani, R. et al. 2 316 Rao, B.K. see Jena, P. et al.; Press, M.R. et al. Rao, B.M.L. et al. 2: 510, 511 Rao, B.S. see Noebe, R.D. et al. Rao, G. see Luzzi, D.E. et al.; Tang, W. et al. Rao, G.H. et al. 3: 100 Rao, K.R. 2: 453 Rao, K.V see Kim, K.S. et al.; Koskenmaki, D.C. et al.; Madurga, V. et al. Rao, M. and Soffa, W.A. 3: 413 Rao, P.K. see Baligidad, R.G. et al. Rao, S. see Dimiduk, D.M. et al.; Woodward, C. et al. Rao, S. et al. 3: 407, 424 e Parthasarathy, T.A. et al.; Simmons, J.P et al. Rao, S.1. et al. 1: 82,8%,506; 2: 58; 3: 455, 457, 769. 770, 772 Rao, V.R. see Baligidad, R.G. et al. R ~ QV.U. , see Wallace, W.E. et al. Rao, V.U.S. see Craig, R.S. et al. Rapp, G.R. see Roberts, W.L. et al. Rapp, M. see Urban, K. et al. Rapp, R.A. see Cockeram, B. and Rapp, R.A., Cockeram, B.V. et al., Kofstad, P. et al.; Mueller, A. et al.; Zheng, M. et al. Rapson, W.S. 2: 562, 570, 572, 649, 652, 653; 3: 54, 75 Rarada, H. see Yamabe-~itarai,U.et al. Rashupkin, V.1. see Degtyareva, V.F. et al. Rasktn, D. and Smith, C.H. 3: 698 Rasmussen, D.R. see Kung, H. et al. Rasorenov, S.V. see Kanel, G.I. et al. Rastogi, P.K. 2: 272 Rath, C. see Blum, V. et al. Rathbun, L. see Li, J. et al. Ra~henau,G.W. 2 306 Rathore, H. see Lee, C.-C. et al. Ratzmann, P.M. see Zinim, C.B. et al. Rau, J. see Bruzzone, P. et al. Raub, E. 1: 721; 2: 564, 565 Raud, S. see Pokela, P.J. et al. Rauh, R.D. 2: 511; see also Carrabba, M.M. et a/. Raunau, W see Niehus, H. e f al. Raupp, G.B. see Cale, T.S. et al. Rausch, J.B. 1: 874, 876, 887, 888 Ramchenbach, H.S. 2: 421 Ravichandran, G. see Tong, W. and Ravichandran G. Ravichandran, K.S. 2: 122, 123; 3: 338, 347 Ravichandran, K.S. and Larsen, J.M. 3: 338 Raviprasad, K. see Chitsalekha, J. et al. Ravot, D. see Chapon, L. et al.; Reinders, P.H.P. et al. Raw, P.M. 3: 61 Rawatt, R. see Sampa~hkumaran,E.V. et al. Rawers, J.C. et al. 3: 731 Rawlings, R.D. 1: 548, 565, 922, 957; 2: 24, 35 Ray, A.E. 2: 310
933 Ray, I.L.F. 1: 506, 507, 509, 521, 537; see also Cockayne, D.J.H. et al.; Crawford, R.C. and Ray, I.L.F. Ray, I.L.F. et al. 1: 496, 497; 2: 202 Ray, M. see Patten, E.A. et al. Ray, M.A. see Shin, S.M. et al. Ray, R. 1: 735,748; see also Jha, S.C. et al. Ray, R. et al. 1: 735 Ray, U. et al. 3: 663 aaybould, D. see Brown, A.M. et al.; Bye, R.L. et al.; Das, S.K. et al.; Kiin, N.J. et al.; LaSalle, J.C. et al.; Skinner, D.J. et al.; Zedalis, M .S. et al. Raynor, G.V. see Prince, A.A. et al. Rayl, M. see Rehwald, W. et al. Raymond, T. 2 572 Rayne, J.A. 1: 1020, 1021, 1024 Raynor, D. 2: 272, 276, 283, 284 Raynor, G.V. 1: 1311, 105, 107, 252, 386, 387, 712; 2: 175, 177, 308; see also Little, L. et al.; Prince, A. et al. Razeghi, M. et al. 3: 669 Razuvayera, B.D. et al. 2: 567, 572 Read, T.A. 1: 832, 834; Wechsler, M.S. et al. Read, W.T. 1: 954; 2 238; see also Eshelby, J.D. et al. Reader, A.H. et al. 2: 618, 619 Rebbah, A. et al. 1: 351 Rebstock, H. see Seeger, A. et al. Rechenberg, H.R. see Landgraf, F.J.G. et al. Rcckman, A.P.F.M. see Van Ommen, A.H. et al. Redfield, A.C. 1: 88 Redjai, E. 1: 352 Reed, M. see Kryliouk, 0. et al. Reed, M.A. et 1-11. 2: 327 Reed, R.S. 2: 403 Reed, S. 3: 592, 619 Reed, T.B. see Terauchi, W. et al. Reed, W.A. see Testardi, L.R. et al. Reeder, W.J. see Bewlay, B.P. et al. Refsnes, J. see Andresen, A.F. et al. Regan, R.E. et al. 3: 574 Regazzoni, G. 1: 737, 739 Regel, A.R. see Blum, A.N. et al. Rehn, L.E. see Allen, C.W. et al., Birtcher, R.C. et al., ~ r ~ m s d i t cM. h, et al.; Koike, J. et al.; Meng, W.J. et al., Okamoto, P.R. et al.; Scheuer, U. et al. Rehn, L.E. et al. 2: 133 Rehwald, W. et al. 1: 875 Reichardt, W 3: 159, 160; see atm Gompf, F et al.; Pintschovius, L. et al.; Sievering, A. et al. Reichenberger, W. see Hoenig, H.E, et al. Reichle, K.J. sec Valeeva, A.A. et al. Reichman, B. see Fetcenko, M.A. et al. Reid, C.N. 3: 424 Reid, J.S. see Kolawa, E. et al.; Pokela, P.J. et al. Reid, L.F. 1: 669 Reidinger, F. see Cava, R,J. et: al. Reif, R. 1: 652 Reihsner, R. 1: 782; see also Pfeiler, W et al. Reijers, H.T.J. see Saboungi, M.L. et al. Reijess, W.T.J. et al. 1: 661, 686: 3: 254
934 Reijers, R. see Price, D.L. et al. Reilly, J.J. see Sandrock, G.D. et al. Reim, W, 2 448; see also Brandle, H. et al. Reinders, P.H.P see Fraas, XI. et al. Reinders, P.H.P. et al. 1: 212 Reinhard, L. see Biihrer, W. et al., Turch, P.E.A. et al. Reinhard, L. er al. 1: 37, 44, 46 Reinhardt, A. see Hellner, E. et al. Reinke, P see Ronning, C. et al. Reinsch, B. see Stadelmaier, H.H. and Reinsch, B. Reinsch, H. 2: 647 Reinshagen, J.H. see Shaw. K.G. and Remshagen, J.H., Sikka, V.K. et al., Vikka, V.K. et al. Reip, C . 2 . 1: 916, 922 Reiso, 0. ,see Westengen, W. et al. Reitz, C.D. 2: 586 Rembges, M. et al. 2: 282 Reniillieux, J. see Baudin, K. et al. Remmele, W. 2 295 Rempel, A.A. see Valeeva, A.A. et al. Remy, L. see Lupmc, V. et al. Ren, J. et al. 2: 426 Ren, X. see Otsuka, K. and Ren, X. Renker, B. see Tietze, Hi. ef al. Rentenberger, C. see Karnthaler, H.P et at. Rentzeperis, P.J. 3: 10 Rentzsch, S. see Klaumiinzer, S . et al. Renusch, D. see Natesan, K. ei al. Renusch, D. et at. 3: 573 Reny, E. et al. 3 131 Reppich, B. 2 273 279, 280. 281 Rest, J. ~t al. 1: 791, 815, 816 Restall, J.E. 2: 228, 490, 492; see also Nicholls, J.R. et al. Restall, J.E. et d. 1: 996; 2: 490 Restorff, J.B. see Clark, A.E. et al. Reuhl, K. see Wendt, H. et al. Reukers, W.M. 2: 605 Reuss, A. 1: 202 Reuss, S. 2: 65, 85, 224 Revelos, W.C. 1: 993; 2: 289, 292; see also Larsen, J.M. et al., Smith, P.R. et al. Revie, R.W. 2: 217 Reviere, R.D. see Johnson, D.R. et al. Rey-Losada, C. et al. 1: 581 Reynauld, F. et al. 1: 498 Reynolds, C.L. see Malhotra, M.L. et al. Reynolds, G.H. see Norman, J.H. et al. Reynolds, J, see Hume-Rothery, W. et al. Reynolds, J.E. 1: 832 Reynolds, H.L. and Morris, J.W. 3: 671 Reznikow, V.J. see Segal, V.M. et al. Rezukhina, T.N. et al. 1: 64 Rhee. W.H. see Sahin, 0. et ctl. Rhim, W.K. see Rim, Y.J. et al. Rhines, F.N. 3: 155 Rhodes, C.G. 1: 653; 2 117, 292, 651; 3: 584; see also Graves, J.A. et al.; Marquardt, B.J. et al.; Martin, P.L. et al.; Smith, P.R. et al.; Vassiliou, M.S. et al. Rhodes, C.G. et al. 2: 91, 127, 128; 3: 584 Rhodin, T. see Ignatiev, A. et al. Rhyne, J.J. 2 390, 391; see also Hardman~R~iyne, K. et al.
Rhys-Jones, T.N. 2 494 Riabov, V.R. see Larikov, L.N. et al. Ribbe, P.H. 1: 626, 633 Ribbens, A. see Graf', K.H. et al. Ricci, M. see Blau, B. et al., Bruzzone, P. et al. Ricco, M. see Cristof'olini, L. et ul. Rice, J. see Motowidlo, L.R. et al. Rice, J.A. see Walker, M.S. et al. Rice, J.R. 2: 166 Rice, J.R. and Thompson, R. 3: 241 Rice, R.W. 1: 927 Richards, M.J. 1: 849, 850, 851, 852, 853, 854 Richards, T.W. 1: 115 Richardson, G.E. see Chatterjee, A. et al. Richardson, J.W see Birtcher, R.C. et al., Reijers, H.T.J. et al. Richardson, J.W Jr. see Reijers, N.T.J. c't al. Richardson, J.W et al. 3: 257 Richel, H. see Delavignette, P. et al. Richie, R.O. see M~~rugesh, L. et al. Richier, C. see Natesan, K. et al. Richman, R.H. I: 845; 2 150; see also Bolling, G.F. and Richman, R.H. Richter, D. see Hempel~ann,R. et al., Springer. T. and Richter, C. Richter, D. et al. 2: 484 Richter, H. see Sprenger, H. et al. Richter, H.J. et al. 3: 786 Richter, M. 3: 97 Rickard, T.W 1: 3 Ricker, R.E. I: 967,970; see also Bertocci, U. et al. Ricker, R.E. et al. 1: 927, 967, 969, 970, 971, 972 Ricks, R.A. et al. 1: 867 Ricolleau, C. et al. 1: 780 Ridder, S.D. 1: 740; 2: 185 Rider, R. see Ryadchenko, V.M. et al. Ridley, M. and Pops, H. 3: 56 Ridley, N. 1: 719; 2: 199 Riechman, A. 2: 637, 639, 642, 643 Rieck, G.D. see Van Loo, F.J.J. and Rieck, G.D. Riedi, P.C. see Lord, J.S. ef a/. Riegler, G. see Manghi, F et al. Riesterer, T. 2: 477,478,482,483; see also Osterwalder, J. et al.; Schlapbach, L. et al. Rietf'eld, H.M. 1: 440 Riethe, P. 2 649 Rieu, J. see Guedo, J.Y et al. Rieux, 3. and Goux, C. 5: 283 Rife, J.C. 2: 409 Rifkin, J.A. see Clapp, P.C. et al.; Clapp, R.C. et al.; Moncevicz, A. et al. Rifkin, J.A. et al. 1: 196, 208 Rigney, D.A. 2: 599; see al.so Neilmann, P. et U / . Rigney, J.D. see Lewandowski, J.J. et al. Rigney, J.D. and Lewandowski, J.J. 3: 550 Rigney, J.D. et al. 2: 64, 295 Rigsbee, J.M. see Shin, S.M. et al. Rikukdwa, S. see Takasugi, T. et al. Riley, F.L. 3 664 Rimini, E. see Baeri, P. et al. Ringel, S.A. see Rohatgi, A. et al. Rinne, F et al. 1: 309
Riseborough, P. 1: 218; see also Lawrence, J.M. et al., Severing, A. et al. Rishel, L. see De Graef, M. et al. Rishel, L.L* see Mura~eedharaii,K. er d. Rishel, L.L. et al. 3: 592, 594, 595, 597 Ritchie, R.O. see Badrinarayanan, K. et al.; Bloyer, D.R. et al.; Campbell, J.P et al.; Gilbert, C.J. et al.; Hong, M.-H. et al.; McKelvey, A.L. et al.; Murugesh, L. et id., Soboyejo, W,O. et al.; Venkateswara Rao, K.T. et al. Ritschel, A.J. see Baldi, R.W. et al. Ritter, A.M. 1: 640 Ritter, C. see Schobinger-Papa~an~ellos, P. et al. Ritter, D.L. et al. 3: 56 Ritter, J.J. see McMichael, R.D. et al. Rittner, M.N. and Abraham, T. 3: 755 Ritz, G. see Hebenstreit, W. et al. Ritz, K. see Pan, A.I. et al. Rivers, S.B. et al. 3: 222, 223 Rivet, J. see Ollitrault-~ichet,R. et al. Rivet, J. et al. 1: 352 Riviere, J.C. see Young, E.W.A. et al. RiviZre, J.P. see Beaufort, M.F et al.; Jaouen, C. et al. RiviZre, J.P et al. 1: 567, 568, 573, 574, 577 Riviere, R. see Whittetiberger, J.D. et al. Rivlin, V.G. 2 175, 177, 308 Rivlin, V.G. and Raynor, G.V. 3: 803 Rizzo Assuncao, F.C. see Xia, S.K. et al. Ro, T. see Yamabe-Mitarai, Y. et al. Ro, Y. see Gu, Y et al.; Yamabe, Y. et al., Yamabe-Mitarai, Y. et al., Yu, X.R. et al. Robadey, J. see Beeli, C. et ul. Robb, J. see Campisi, I.E. et al. Robbins, M. see Appelbaum, A. et al. Robbms, M.O. 1: 24, 32; see also Hawkins, R.J. et al. Robenson, C.R. see Andrews, P.V et al. Roberge, R. 2: 360 Roberts, B.E. 1: 16 Roberts, B.W. 1: 952; 3: 806 Roberts, C.S. 2: 407, 421, 424, 428 Roberts, R.J. 3: 592 Roberts, S.G. see Booth, A.S. and Roberts, S.G. Roberts, S.R. 2: 623 Roberts, W.C. 1: 5 Roberts, W.L. et nl. 1: 635 Roberts-Austen, W.C. 1: 3, 6 Robertson. 1.M. see Bond, G.M. et al. Robertson, J.L. I: 463,482; ,see also Chen, L.C. et al.; Pashley, D.W. et al.; Reinhard, L. et al. Robertson, J.L. et al. 1: 457 Robertson, J.M. 1: 161 Robertson, W 1: 618; see also Sundaram, v et al. Robertson, W.D. see ~undarani,V.S. and Robertson, W.D., Sundaram, V.S. et al. Robins, K.R. see McClusky, R. et al. Robinson, F.N.N. see Kurti, N. et al. Robinson, G,T. 2 507 Robinson, K. I: 473; 2 177, 178, 182, 186 Robinson, P 2: 640, 642 Robinson, P.M. 1: 109;
Author hzdex Robinson, P.M. and Bever, M.B. 3: 800 Robinson, R.C. 1: 991, 992, 993 Rocha, H.J. 1: 759 Rocker, W. see Hirschler, W and Rocker, W Rocker, W aiid Kohlhaas, R. 3: 527 Rockett, A. see Mullan, C.A. et al., Park, D.G. et al., Ramanath, G. et al. Rockett, A. et al. 3: 663, 672 Rodbel~,K.P. see Svilan, V et al. Rodgers, J.R. 1: 7, 255, 258, 259, 260 Rodriguez, C. 1: 843 Rodriguez, C.O. see Methfesse~,M. et al. Rodriguez, M. 2: 180, 182 Rodriguez, M.L. see Oglesby, P.L. et al. Rodriguez-Carvajal, J. see Blatico, J .A. er" al., Pannier, J. et al. Rodriguez-Fernandez, J. see Blanco, J.A. e f al. Rodway, G.H. I: 640 Roelots, L.D. 3: 222 Roessler, J,R, see Chatterjee, A. et al. Roetti, C. see Clementi, E. and Roetti, C. Rogachev, A.S. see Mukasyan, A.S. et al.; Varma, A. et al. Rogachev, A.S. et al. 3: 732, 733 Rogl, P 1: 417; 2: 314; see also Coey, J.M.D. et al.; Leithe-Jasper, A. el al.; Weitzer, F.et al. Rogl, P and Effenberg, G. 3: 803 Rohatgi, A. et al. 2 423 Rohleder, I. see Blau, B. et al. Rohrer, H. *seeBinnig, G. and Rohrer, H.; Binnig, G. et al. Rollett, A.D. see Castro, R.G. et al.; Kocks, U.I; et al. Rollett, A.D. and Kocks, U.F 3: 361, 362, 363 Romagnoii, R.J. 2: 446 Roman, Y.G. and Adriaansen, A.P.M. 3: 669 Romanov, A.E. and Vladirnirov, V.1. 3: 408 Romanow. W. see Pope, D.P. et al. Romanow, W.J. see Neumann, D.A. et al.; Wernick, J.N. et al., Zhu, Q. et al.
Romeau, L).see Yacaman, M.J. et al. Romego, M. see Sainfort, G. et al. Rornpa, H.W.A.M. see Verbeek, B.H. et al. Ron, M. 2: 486; see also Koster, U. et al. Ronay, M. 2: 606 Ronchetti, M. 1: 481; see also Stemhardt, P.J. et al. Rong, T.S. el al. 3: 303 Ronning, C. et al. 3: 664 Roof, R.B. see Cromer, D,T. et al. Rooks, M.J. see Erdogan, T. et al. Rooksby, N.P 3: 49 Roosild, S.A. see Kosonocky, W.F et al. Roozeboorn, F, 3: 664 Roozeboorn, H. 1: 7 Roques, B. see Francoiz;, M. et al.; Lemoine, V. et al.; Veiiturm, G. et al. Roniik, G. see Ulvensoen, J,H. et aE. Rosa, C.J. et al. 3: 34 Rosa, R.A. 2: 202 Rosakis, A.J. see Bmck, H.A. cf al., Conner, R.D. et al. Rosanov, N.N. see Mak, A.A. et al.
Rosato, V. 1: 817, 820 Rosato, V et al. 1: 80 Rose, J.H. et al. 1: 83 Rose, R.M. see Cogan, S.F. et al., Flukiger, R. er al.; Schwall, R.E. et al. Rosei, R. 1: 136 Rosen, A. see Bendcrsky, L. et al., LOUbriel, G.M. et al. Rosen, M. 1: 875; see also Klimker, H. et al. Rosen, M, et al. 1: 889 Rosen, S. 2: 54, 55 Rosenbaum, H S . 3: 417 Rosenberg, H.M. 1: 1018 Rosenberg, M. see Kuncser, V et al. Rosenberg, R. 2 605 Rosenberger. A.H. see Larsen, J.M. et al.; Worth, B.D. et al. Rosenbert, M. see Zhou, R.J. et al. Rosenbluth, A. see Metropolis, N. et al. Rosenb~uth,M.N. see Metropolis, N. et az. Rosencher, E. see Thomas, 0. et al. Rosencher, E. et al. 3: 787 Rosenfeld, B. 1: 564 Rosenfeld, R. et al. 3: 384, 385, 398, 399 Rosenkranz, R. see Fromrneyer, G. et al.; Frommeyer, G.R. et al., Kroll, S. et al. Rosenthal, D.G. see Ward, C.H. el al. Rosiiiger, H.E. 2 137, 143, 144, 145; see alsa Cameron, D.J. et al.; Causey, A.R. et af. Rosler, J. see Schneibel, J.H. et al.; Valencia, 9.J. et al. Rosler, J. et al. 3: 317 Rosler, U. see Schubert, K. et al. Rosner, C.H. 2: 360, 377, 378; see also Markiewicz, W.D. et al. Rosner, H. et al. 3: 454 Ross, E.N. see Jayashankar, J.S. et al. ROSS,E.W. see Walston, W.S. et al. Ross, E. W. and O'Hara, K.S. 3 306, 320 Ross, P. 1: 616 Ross, P.N. 1: 616 Rossat-Mignod, J. 3: 165 Rossat-Mignod, J. et al. 3: 180, 181 Rosseindky, M.J. see Christides, C . et al. Rossetti, M. 1: 998, 1000; see also Berkowitz-Mattuck, J. et al. ROSSI,C. 1: 11 Rossiter, P.L. 1: 777, 781, 945, 946, 947 Rostova, T.D. see Elagm, V.I. et al. Roth, R.W see Dasbach, D. et al. Roth, S. see Murani, A.P. et al. Roth, W.C. 1: 172 Rothenflue. L. see Szaruga, A. et al. Rothman, B. see Setton, M. et al. Rothman, S. see Hoshino, K. et al. Rothman, S.J. see Hoshino, K. et al. Rothmann, A. 1: 115 Rottrnan, C. 3: 222 Rouault, A. see Fruchart, R. et al. Rouleau, C.M. see Park, R.M. et al. Roult, G. see Bertaut, E.F. et al. ROUS,P.J. see MacLareii, J.M. et al. Rousseau, R. see Tillard-Charbonnel, M. et al. Roux, J.P 1: 987, 989; see also Viefhaus, H. et al. Roviadi, G. see Atrei, A. et al.
935 Rowe, D.M. 2 453 Rowe, G.M. see Striitt, P.R. et al. Rowe, J.M. see Brun, T.O. et al. Rowe, J.M. et al. 1: 1021 Rowe, J.R. et al. 2: 380 Rowe, R.G. 2: 91, 93, 95, 101, 104, 119, 290, 291, 298; see also Banerjee, D. et al.; Jackson, M.R. et al.; Mozer, B. et al. Rowe, R.G. and Hall, E.L. 3: 303 115; 3: 545 Roy, J.A. 2: 136, 137, 138, 139, 142, 143 Roy, R.A. see Svilan, V. et al. Roy, S. see Sen, S. et al. Roy, S.B. et al. 1: 221 Royal, T. see Thadhani, N.N. et al. Royer, W.A. see Testardi, L.R. et al. Roytburd, A. see Bendersky, L.A. et al. Rozendaal, E. et al. 2: 479 Rozgonyi, G.A. 1: 17811 Rozgonyi, G.A. et al. 2: 617 Rozgus, A. 3: 858 Rozner, A.G. 1: 544; 2 61 Rozouoer, L.I. see Izerman, S A . et al. Rozum, S.N. see Markiv, V.Ya. et al. Ruano, O.A. see Jiniinez, J.A. et al.; Wolfenstine, J. et al. Ruault, M.O. see Jaoueii, C. et al. Ruba, E.R. see Daulton, T.L. et al. Ruberto, M.N. see Willner, A.E. et al. Rubini, S. and Ballone, P. 3: 769, 770 Rubins, M.J. see Clapp. P.C. et al. Rubinstein, M. et al. 1: 479 Ruckle, D.L. 2 494 Rudberg, F 1: 5 Rudenko, V.K. see Shashkov, O.D. et al. Ruderman. M.A. 1: 440 Rudigier, H. see Ott, H.R. et al. Rudman, D.A. 2 383 Rudman, P.S. 1: 504, 507, 580, 773, 775 Rudy, M. 1: 506,916, 917,919, 923; 2: 61, 63; set also Jung, 1. et al. Rufer, H. see Sajovec, I;. et al. Ruge, H. 1: 107, 108 Rugg, D. 3: 483 Ruggiero, S.T. 2: 383 Ruhenbeck, A. 1: 16 Ruhl, R.C. et al. 1: 735 Ruble, M. 1: 927, 982, 986, 987, 989. 990; see also Schumann, E. et al. Rulison, A.J. see Kim, U .J. et al. Rullier-A~benque.F. 1: 579, 791, 792, 795, 797 Rumpf, T. see Wolff, I.M. et al. Ruadle, R.E. 1: 51; see also Florio, J.V et al. Ruoff, A.L. 1: 184; see also Ho, P.S. and Ruoff, A.L.; Vohra, Y.K. et al.; Weir, S.T. rt al. e Rosa, C.J. et al. 90 1: 663, 666, 674 Rusakova, 1. see Sun, Y.Y. et al. Rusetskaya, N. see Seinenova, E.L. et d. Rusetskaya, N.Yu. see Semenova, E.L. et al.
Rushford, M.C. see Wemberger, D.A. et at.
936 Rusing, J. and Herzig, Chr. 3: 290 Rusovic, N. 2: 55, 56, 57 Russ, S.M. 2: 294 Russel, S.W. see Li, J. et al. Russell, K. see Froes, F.H. et al. Russell, K.C. 1: 693, 706, 866, 867 Russell, S. see Miracle, D.B. et al. Russell, S.M. et al. 2: 60, 65, 66 Rustichelli, F. see CaciufTo, R. et al. Ruterana, P. see Braisaz, T. et al. Ryabinin, M.A. see Ryadchenko, V.M. et al. Ryabov, V.P see Markiv, V.Ya. et al. Ryabov, V.R. see Fal’chenko, V.M. et al.; Vivchar, 0.1. et al., Zarechnyuk, O.S. et al. Ryadchenko, V.M. et al. 3: 864 Ryan, C.E. see Marshall, R.C. et al. Ryan, D.H. 3: 181; see also Liao, L.X. et al., Stroin-Olsen, J.O. et al.; Thomson, J.R. et al. Ryan, D.H. et al. 3: 182, 183 Ryan, R.D. see Eberhardt, J.E. et al. Ryan, R.G. 2: 623 Ryback, G. 1: 635 Rybicki, G.C. 1: 983, 987, 990 Rybin, V.V. see Perevezentsev, V.N. et al. Ryge, G. 2: 576 Ryge, G. et al. 2 576 Ryoo, H.S. see Kang, B. et al. Rytz-Froidevaux, Y. ,see SalathC, R.P. et al. Ryunosuke, N. see Koyama, K. et al. Rzeznik, M.A. see Kwon, Y.U. et al. Rzynian, Ch. see Kek, S. et al. Saada, G. 1: 525, 543, 544, 548, 915; see also Shi, X. et al. Saam, W.F. see Ho, T.L. et al.; Jaszczak, J.A. et al. Saarinen, K. see Alatalo, M. et al.; Corbel, C. et al. Saarinen, K. et al. 3: 288 Sabariz, A.L.R. and Taylor, G. 3: 56, 60, 61 Sabes, P.N. see Kelton, K.F et al. Sabes, P.N. et al. 1: 454 Sabochick, M.J. 1: 88, 792, 817; see also Devanathan, R. et al.; Lam, N.Q. et al.; Lutton, R.T. et al.; Shoemaker, J.R. et al.; Zhu, H. et al. Saboungi, M.L. 1: 66, 70; see also Fortner, J. et al.; Johnson, G.K. and Saboungi, M.L.; Price, D.L. and Saboungi, M.L.; Price, D.L. et al., Reijers, H.T.S. et al.; Richardson, J.W et al. Saboungi, M.L. et al. 1: 662, 666, 667, 680, 686; 3: 246, 254. 255, 256 Sabrie, J.L. see Tixador, P. et al. Saburi, T. 1: 831, 837; see also Pak, H.R. et al. Saburi, T. et al. 1: 832, 845 Sachdev, J. 2: 506, 511 Sachdev, S. 1: 479, 480 Sachetti, N. see Blau, B. et al. Sachs, G. 1: 828, 913 Sachtler, W.M.H. 3: 224 Sadananda, K. see Feng, C.R. and Sadauanda K.
A ~ t ~ Index or Sadananda, M. and Feng, C.R. 3: 486 Sadananda, K. et al. 2: 223; 3: 275, 317, 318, 319, 320, 334, 345 Sadler, R.A. 2: 621; see also Geissberger, A.E. et at. Sadoc, A. see Belin-Ferre, E. et al. Trambly de Laissardiere, G. et al. Sadoc, A et al. 1: 479: 3: 146 Sadowitz, J.D. see Cheong, S.W et al. Sadwick, L.P. see Kirn, Y.K. et al. Sadwick, L.P et al. 2 621 Saeger, K.E. 2: 653 Saettas, L. see Kalogirou, 0. et al. Sagar, P.K. et al. 2 125, 126 Sagawa, M. et al. 1: 748; 2: 314, 315 Sahashi, M. see Hashimoto, T. et al.; Tokai, Y. et al.; Tomokiyo, A. et al. Sahay, B.B. see Mishra, A.K. et al. Sahin, 0. et al. 1: 1007 Sahoo, M. et al. 2 599 Sahu, P.C. et al. 3: 161, 731 Said, M.R. et al. 1: 939 Samfort, G. see Bollmann, W et ail. Sainfort, G. et al. 2: 204, 206 Sainfort, P see Sadoc, A. et al. St John, J. 1: 242, 243, 419, 420 Saito, H. see Horikoshi, Y et al. Saito, 1. see Osaka, T. ef al. Saito, J. set. Monnaga, M. et al. Saito, J.-I. see Harada, Y. et al, Saito, K. 2: 345 Saito, T. 3: 841; see also Nasako, K. et al.; Ogawa, T. et al. Saito, T. see Nishino, K. et al. Saito, Y et al. 1: 454 Saitovitch, H. see Xia, S.K. et al. Sajovec, F. et al. 1: 352 Saka, H. 1: 536, 542 Saka, H. et al. 1: 498, 506 Sakagami, T. see Umakoshi, Y et al. Sakagarni, Y. see Umakoshi, Y. et al. Sakaguchi, J. see Yamaguchi, A. et al. Sakai, N. see Minoniura, S. et al. Sakai, T. 2 509; see also Iwakura, C. et al. Sakai, T. et al. 2: 486, 509, 510; 3: 107 Sakaki, H. 2: 425 Sakakibara, H. sec. Tanaka, M. et al. Sakamoto, K. see Zhu, S.M. et al. Sakamoto, Y. see Flanagan, T.B. et al. Sakata, K. see Tokushima, T. et al. Sakata, M. see Kaibe, H.T. et al.; Kawasumi, I. et al. Sakata, M. et al. 2: 610 Sakata, T. see Tokushima, T. et al. Sakhanskaya, I. see Kuranov, A. et al. Sakidja, R. see Goken, M. et al.; Nunes, C.A. et al. Sakidja, R. et al. 3: 657 Sakinada, K. see Tachikawa, K.Y. et al. Sakuma, T. 1: 835 Sakurai, J. see Takabatake, T. et al.; Yoshimoto, N. et al. Sakurai, T. see Yamakoshi, S. et al. Sakuri, T. see Takasugi, T. et al. Sala, Ph. see Ohl, M. et al. Salamon, M. 3 281,283; see also Mehrer, H. et al.; Mishin, Y. and Farkas, D. Salamon, M.B. 1: 1020; see also Satija, S.K. e1 al. SalathC, R.P et al. 2: 414, 425 Saldin, P.V. see MacLaren, J.M. er al.
Salehi, M. and Hossei, R. 3: 671 Salerno, J.P see Vojak, B.A. et al. Sales, B.C. 3 106 Sales, B.C. see Sharp, J.VV et al. Sales, B.C. et al. 3 106 Salesse, M. see Sainfort, G. et al. Salih, A.S.M. see Rozgonyi, G.A. et ul. Salishchev, G. see Imayev, R. et al. Salishchev, G. et al. 3: 651 Salishchev, G.A. see Tmayev, R.M. et al.; Kaybyshev, O.A. et al. Salishchev, G.S. et al. 3: 639 Salokatve, A. see Saannen, K. ef al. Salomon, D. see Potzel, U. et al. Salpietro, E. see Bruzzone, P. et czl, Salzberger, U. see Schumann, E. et al. Samarth, N. see Ding, J. et al.; Jeon, H. et al. Sambongi, T. et al. 3: 232 Sarnhoun, K. and David, F 3: 30 Sammells, A.F. 2: 510 Sampath, S. see Tiwari, R. et al. Sampathkumaran, E.V. et al. 3: 527 Sampson, W.B. 2: 361; see also McClusky, R. et al. Samson, S, 1: 11, 325, 41 1 Samson, Y. see Durr, H.A. et al. Samsonov, G.V. 2: 630, 648, 649; 3: 36, 42, 808 Samsonov, G.V. and Bondarev, V.N. 3: 808 Sarnsonov, G.V. and Perminov, V.P 3: 808 Samuel, R.L. 2: 490 Samwer, K. 2 479; see also Yeh, X.L. et al. Samwer, K. et al. 3: 809 Sanchez, C . see Naka, S. et al. Sanchez, J.L. et al. 2: 311 Sanchez, J.M. 1: 41, 613, 619; see also Becker, J.D. et al.; Cadeville, M.C. et al.; Hnwhns, R.J. et al.; Kikuchi, R, et al.; Kosugi, M. et al.; McRae, E.G. et al.; Nohri, T. et al.; Tso. N.C. et al.; Wu, Y.P. et al. Sanchez, J.M. and Morian-Lopez, J.L. 3: 228 Sanchez, J,M. et al. 1: 30, 31, 38, 39, 41, 496, 502 Sandananda, K. and Vasudevan, A.K. 3: 336, 347 Sanders, D.M. 3: 665 Sanders, J.B. see Winterbon, K.B. et al. Sanders, R.E. Jr, see Griffith, W.M. et al. Sanders, T. see Welpman, K. et al. Sanderson, R.T. 3: 101 Sandlund, L. 2: 401, 402 Sandrock, G.D. 2: 484; see also Goodell, P.D. et al. Sandrock, G.D. et al. 2: 477, 478, 482, 484, 485,486 Sands, T. 2: 621; 3: 783, 784; see also Harbison, J.P. et al.; Tabatabaie, N. et al. Sands, T. et al. 2: 621; 3: 565, 783, 787 Sanganeria, M. 1: 788 Sanger, P. et al. 2: 373 Sankar, S.G. see Craig, R.S. et al.; h a n g , M.Q. et al. Sankaran, S.N. see Wallace, T.A. et al.; Wiedemann, K.E. et al.
A ~ t Index ~ o ~ Sankey, O.F. see Ramachdndran, G.K. et al. San Martin, A. see Quyen, N.H. et id. Sano, H. et al. 3: 672 Sano, T. see Bolt, P.J. et al.; Tian, W.H. et al. Santella, M.L. 3: 510; see also Sikka, V.K. and Santella, M.L. Santella, M.L. et al. 1: 655; Santiago, J.J. see Setton, M. et al. Santoro, G.J. 1: 986, 987, 989,990 Santoro, G.J. et al. 1: 983, 986, 987, 989 Sapelkin, A.V. see B r a ~ k i nV.V. , et al. Sapozhkova, T.P. see Vlasova, E.N. et al. Saqib, M. see Szaruga, A. et al. Sar, R. see Wang, H. et al. Sargeiit, C.M. 1: 845; see also Arunachalam, V.S. and Sargent, C.M. Sargent, G.S. see Allen, C.W. et al. Sargent, K. and Huffman, M. 3: 483 Sargent, P.M. 1: 913 Sargent, S. 2: 25 Saris, F.W. 2: 610, 611, 619, 625; see aIso Barbour, J.C. et al.; de Reus, R.et al.; Denier van der Con, A.W. et al., Hung, L.-S. et al.; Nastasi, M. et al.; Pretorius, R. et al.; Sinke, W. et al. Saris, F W. et al. 2: 618, 625, 626 Sarkar, B.K. see Sen, S. et al. Sarma, D.D. et al. 1: 137 Sarrao, J.L. see Migliori, A. et al. Sarrazin, T. 1: 845; 2: 150; see also Vanderschaeve, G. and Sarraain, T.; Vanderschaeve, G. ef al. Sasada, T. 2: 591 Sasaki, F. see Mis~ii~ia, T. et al. Sasaki, G. et al. 1: 586, 599, 600 Sasaki, K. 1: 651; see also Takeyama, M. et al. Sasaki, T. see Yanagihara, M. et al. Sasaki, W. see Kurosawa, K. et al. Sasaki, Y. see Asaiiabe, S. et al. Sasamori. K. see ~ a w a m u r aY. , et al. Sass, S.L. 1: 586, 588, 590, 903; see also Kung, H. et al.; Yoo, M.H. et aE. Sastry, D.H. see Sundar, R.S. et al. Sastry, G.V.S. 1: 475 Sastry, S.M.L. 1: 504, 505, 528, 535, 654; 2: 20. 22, 83, 98; see also Soboyejo, W.O. et al. Sastry, S.M.L. and Lipsitt, H.A. 3: 328 Satdarova, F. 1: 782 Satija, S.K. et al. 1: 164 Sato, A, see Jumonji, IS. et al. Sato, H. 1: 50, 775, 786; 2: 568; see also Gschwend, K. et al.; Ikeda, M. et al., McCoy, J. et al.; Maeda, H. et al.; Toth, R.S. et al. Sato, H. et al. 1: 775, 786 Sato, IS.2: 366; see also KakdO, M. et al.; Sano, H. et al. Sato, R.et al. 2: 394, 448 Sato, M. see Takahashi, T. et al. Sato, S. I: 836, 837, 838, 839, 841; see also Yanagihara, M. et al. Sato, T. see Wanada, S. et al.; Watanabe, s. et al. Satoh, K. see Nasako, IS.et al. Satoh, Y. see Aoki, K. et al. Satya Murthy, N.S. et al. 2: 309
Sauer, R.W. 2: 216 Saunders, G.A. see Gunton, D.J. and Saunders, G.A.; Miller, A.J. et al. Saunders, N. 1: 696, 698; 2: 192, 193; 3: 478 Sauiiders, N. and Miodownik, A.P. Sauiiders, N. et al. 2: 193 Saunders, R.C. see Yao, N.P. et al. Sauthoff, G. 1: 506, 911, 913, 914, 915, 916,917,918,919,921,922,923,925. 926, 927, 928; 2: 61, 63, 240, 650; 3: 231,275,288,297,312,313, 361,451, 574, 621, 809; see aDo Engell, H.-J. et al., Jung, I. et al.; Liu, C.T. et al., Neumann, P and Sauthoff, G.; Palm, M. et al., Von Keitz, A. et al.; Wolff, I.M. et al., Wuiiderlich, W. et al. Sauvage, M. et al. 1: 305 Savage, H.T. 2: 399; see also Clark, A.E. et al. Savage, H.T. et al. 2: 399 Savage, M.F. see Srinivasan, R. et al. Savalii, P. Yu. see Il’nitskaya, O.N. et aE. Savikova, L.A. see Molchanova, L.V. et al. Savino, E. see Pasianot, R. et al. Savino, E.J. 1: 616, 617; see also Farkas, D. et al.; Pasianot, R. et al. Savitskii, E.M. 1: 270, 271; 2: 240, 246 Savitskii, E.M. and Gribulya, V.B. 3: 823, 824 Savitskii, E.M. and Kiselyovn, N.N. 3: 812,813,814,815,823,824,825,827 Savitskii, E.M. et al. 1: 270; 3 812, 823, 824, 825, 827, 832, 833 Savtchenko, I.B. see Klakov, M.P et al. Savvinov, A S . see Tokarev, V.N. et al. Savvinov, AS. et al. 1: 715, 716, 717 Saw, C.K. see Schwarz, R.B. et al. Sawada, H. et al. 2: 207 Sawada, T. see Hasegawa, H.et al. Sawai, T. see Hishinuma, A. et al. Sawamura, 1. 2: 158 Sawatzky, E. 2: 444, 446, 448 Saxl, 0. see Smith, G.D.W. et al. Saxton, W.O. see Rnowles, K.M. et al. Sayashi, M. see Maki, IS, et al. Sayers, C.M. 1: 682 Scanlon, J.C. see Ayer, R. et al. Scanlon, R. see DeIl’Orco, D. et al.; Van Oort, J.M. et al. Scarbrough, J.O. see Koch, C.C. et al. Scarr, G.K. 2 86; see also Baes~ack,W.A. 111et al.; Marquardt, B.J. et al.; Shih, D.S. et al. Scattergood, R.O. see Bacon, D.J. et al.; Barnett, D.M. et al. Schaaf, G. see Mikulla, R. et al. Schaake, H.F. see Reed, M.A. et trl. Schad, R.G. see Paunovic, M. et al. Schaedler, R.M. see Gregory, E. et al. Schaefer, €3.-E. 3: 280, 282; see also Brossmann, U. et al.; Kurnnierle, E. et al.; Shirai, Y. et al.; Valeeva, A.A. et al.; Wurschum, R. et al., Zhang, X.Y. et al. Schaefer, H.-E. and Badura-Gergeii, K. 3: 276, 277, 279, 280, 281, 282 Schaefer, H.-E. and Schmid, G. 3: 277
937 Schaefer, H.-E. et al. 1: 564; 3: 277, 278, 279, 280, 28 1 282, 285, 286, 29 1 Schaefer, R.J. 1: 454, 645; see also Bendersky, L. er al., McAlister, A.J. et al. Schaefer, R.J. et al. 1: 454, 465; 2: 185 SchaefYer, J. et al. 1: 987, 996 Schafer, H. see Steglich, F. et al. Schzfer, H. and Eisenmann, B. 3: 11.5, 128 Schafer, W 1: 356 Schaffer, G.B. and McCorrnick, P.G. 3: 730. Schafrik, R.E. 1: 883, 885, 888; see also Lipsitt, H.A. et al. Schall, P. 3: 399, 401 Schall, P. see Yang, W.G. et al. Schanzer, S. 2: 275, 284 Schapmk, F.W 1: 587, 598, 600; see also Anto~opoulous,J.G. et al.; Buis, A. et al.; Pestman, B.J. ef al.; Tichelnar, F.D. et al. Schatteman, G.C. 1: 242, 243 Schaublin, R. see Baluc, N. et al. Scheers, P.V.T. see WolfT, I.M. et al. SchefTel, R. see Kohl, W. et ul. Scheffer, H.T. 2: 523 ScheBer, M. 1: 580 Scheil, E. I: 831; 2: 85 Schellenberg, L. see Jorda, J.L. et LIE. Scheltens, F.J. see Evans, D.J. et al. Schepp, P see Reppich, B. et al. Scherrer, H.2: 453 Scherrer, P. 1: 10 Scherrer, S. 2: 453 Schetzina, J.F. see Ren, J. et al. Scheunemann-Frerke~,G, 1: 528 Scheurs, J. see Baron, M. et al. Schianchi, G. see Bencl, S. et al. Schick, C. 3: 222 Schicktanz, S. et al. 1: 162, 163 Scbiffmacher, G. see Tominez, E. ef al. Scliilke, P W. see Wood, J.H. et al. Schiller, T.L. see Tsutsumi, S. et al. Schilling, J.S. see Cornelius, A.L. et al. Gangopadhyay, A.K. and Schilling, J.S. Schillinger, W, et al. 3: 635 Schiltz, R.J. Jr. 1: 87.5, 887, 889, 920 Schiltz, R.J. Jr. et al. 1: 875 Schirnanski, F.-P. w e Gerling, R. et al. Schimonura, 0. see Imai, M. et al. Schindler, C. see Moon, N.C. et al. Schirber, J.E. I: 1022; see also List, R S , et al.; White, G.K. et al. Schlapbach, L. 2: 476, 477, 479, 481, 482, 483; 3: 48,810; see also Aebi, P. et al.; Burger, J.P et al.; Chartoum, D. et al., Meli, F. et al.; Osterwalder, J. et al.; Pilo, Th. et al., Sandrock, G.D. et al.; Stuck, A. et al.; Zuttel, A, et al. Schlapbach, L, et al. 2: 479, 483, 484; 3: 107 Schlatter, H. see Van Ommen, A.H. et al. Schlemzer, C . see Kestler, H, et al. Schlemper, K. and Thomas, L.K. 3: 232, 233, 237, 239 Schlesinger, M.E. Jr. 1: 64 Schlesinger, ME. et al. 3: 547 Schlichting, J. 1: 998, 999, 1003; 2: 294, 295; see also Fitzer, E. et al.
938 Schlier, R.E. 1: 183 Schlottmann, P. 1: 211, 212 Schluckebier, G. 1: 104 Schl~ckebier,G. and Predel, B. 3: 687 Schmalzried, H, 1: 95; see also Pelton, A.D. and Schmalzried W. Schmauder, S. see Schneibel, J.H. et al. Sclimeding, A. see von Schneriiig, H.G. et al. Schmerling, M.A. et al. 3: 56 Schmid, H. 2: 308 Schmid, M. see Gauthrer, Y et al., Hebenstreit, E.L.D. et al.; Hebenstreit, W et al.; Varga, P and Schmid, M. Schmid, M. et al. 3: 213, 214, 223 Schmidt, B. et al. 1: 988; 2: 204 Schmidt, F.K. 2 295 Schmidt, P.C. 1: 67; 3: 238 Schmidt, R.D. 2: 597 Schmidt, V.H see Boa, 2;. et al. Schmit, J.L. see Bowers, J.E. et al. Schmit, J.N. 1: 72 Schniitt, D. 1: 441; see also Besnus, M.J. et al., Blanco, J.A. et al.; Gignoux, D. and Schmitt, D. Schmitt, D. see Bauer, E. et al.; Blanco, J.A. et al. Schniolz, P. see Mitterer, C . et al. Schniutzler, R.W. et al. 1: 669 Schneibel, J.H. 1: 433, 907, 917; 2: 42, 139, 157, 166; see also George, E.P. et al.; Ltu, C.T. et al., Someski, B. et al., Sparks. C.J. et (d. Schneibel, J.H. et al. 1: 918, 922; 2: 64, 158, 161, 166, 168, 169, 170; 3: 489, 655 Schneider, A. 1: 110 Schneider, D. et al. 3: 653 Schneider, E. see Schicktaraz, S. et al. Schneider, G. see Landgraf, F.J.G. et al.; Liu, N.C. et nl.; Moreau, J.M. et lzl.; Sclinclle, W. see Giguere, A. et al.; S t ~ ~ d e l ~ a H.W. i e r , et al. Schneider, H.J.S. see Hoenig, W.E. et al. Schneider, J.F 2: 652 Schneider, I(.see Bauer, R. et al., Singheiser, L. et al. Schneider, M, 1: 676 Schneider, W. see Campisi, I.E. et al. Schneider-Muiitau, H.J. sep 0111, M. e r al. S c h n e ~ a n nH. , see Schuster, H.U. et al. Schnittgrund, G.D. see Fritzemeier, L.G. et al. Schnittny, Th. see Kuhrt, Ch. el al. Schnitzke, K. see Schultz, L. et al., Wecker, J. et al. Schnitzke, K. et al. 2 314 Schnotz, G, see Schumann, E. et al. Scliob, 0. and ParthC, E. 3: 89 Schobar, T. 1: 598 Schdbel, J.-D. 2: 309 Schober, H.R. 1: 2019 Schober, T. and Bellufi, R.W 1: 598 Schober, T. see Balluffi, R.W. et al. Schobinger-Papaman~ellos,P. et al. 3: 100 Schoeck, G. 1: 496, 503, 540; 3: 437, 463; st'e also Korner, A. and Schoeck, G. Schoenberg, T. 2: 291
Author Index Schoenes, J. 2: 448; see also Brandle, H. et al.; Habermeier, H.-U. et al. Schoeiifcld, S.E. et al. 3: 371 Schoenflies 1: 309 Schofield, J.H. 3: 576 Schofield, M. 2 516n Schogl, S.M. and Fisher, F.D. 3: 426 Schoijet, M.J. 1: 849 Scholl, R. et al. 3: 657 Scholz, 11. see Blum, M. er al. Scholz, U.D. see Evers, C.B.H. et al. Schonberger, U. see Eriist, F. et al. Schonfeld, B. 3: 73; see also Biihrer, W. et al.; Reinhard, L, et al. Schorr, M. see Becker, S. et al. Schrag, G,1: 110 Schrank, J, et al. 1: 958, 959 Schreiber, E. et al. 1: 195, 202, 208, 873 Schretter, P. see Clemens, H. et al.; Koeppe, C. et al. Schrey, F. see Appelbaum, A. et al. Schrobi~geii,G. see Campbell, J. and Schrobilgen, G. Schrock, H.W. see Clemens, H. et al. Schroder, A. see Gerniann, A. et al. Schroder, K. 3: 808 Schroeder, D, see Rockett, A. et al. Schroeder, P.A. see Blatt, F.J. et al. Schroeder, T.A. 1: 833, 834 Schroer, W. see Wallow, F et al. Schroll, R. see Gumbsch, P. and Schroll, R. Schroll, R. et al. 3: 437, 452, 772 Schropf, H. et al. 3: 731, 759 Schroter, W. see Ceorge, A. et al. Schryvers, D. see Tanner, L.E. et al. Schubert, E. 1: 733; see rrZso Clemens, H. et al. Scliubert, E.F see Hunt, N.E.J. el al.; Kopf, R.F. et al. Schubert, K. 1: 112, 113, 115, 116, 119, 232, 279, 282, 305, 40311; 2 156; see also Bahn, S. and Scliubert, K., El Boragy, M. et al.; Pfeifer, H.U. et al. Schubert, K. et al. 1: 114 Schublin, R. see Baluc, N. and Schublin, R. Schublin, R. and Stadelniann, P.A. 3: 462 Schuh, C. and Dunand, D.C. 3 298 Schule, W. see Maier, K. et al. Schuler, P 2: 584 Schulson, E.J. 2: 133 Schulson, E.M. 1: 522, 567,591, 594,655, 791,793,803,887,897,902,903,915, 925, 997; 2: 18, 58, 60, 61, 133, 134, 135, 136, 137, 138, 139, 142, 143, 144, 145; 3: 509, 510; see also Baker, I. et al., Causey, A.R. et al.; Fang, J. et al.; Howe, L.M. et al.; Turner, R.B. et ul.; Weihs, T.P. et al. Schulson, E.M. et al. 1: 791, 806, 809, 898, 902,908; 2 30, 134, 137, 143 Schultz, H. see Eibner, J.E. et al. Schultz, W.H.see Steeves, M.M. et al. Schultz, L. 1: 734; 2 3 15; see also Eckert, A. et al.; Eckert, J. et al.; Katter, M. et al.; Kubis, M. et al., Rembges, M. et al.; Sclinitzke, K. et al.; Schropf, H. et al.; Wecker, J. et al. Schultz, L. and Katter, M. 3: 804 Schultz, L. et al. 2: 314, 315, 316, 360
Schultz, N.see Chu, T.L. et al. Schultz, P.A. 1: 156, 921 Schulz, M. see Reppich, B. et al. Schulz, P. 1: 908 Schulz, R. see Blouin, M. et al. Schulze, G. 1: 107 Schulze, G.E.R. 2: 238; 3: 214; see also Pautler, P. et al. Schulze, M. 1: 782 Schulze, M.P. 2 395, 399, 400, 401 see also Galloway, N . et: al. Schulze, M.P et al. 2: 400 Schulze, 0. 1: 917 Schumacher, C. see Klaumunzer, S. et al. Schumann, E. et al. 1: 982 Schumann, G. see Reppich, B. et al. Schiinemann, U. see Koster, U. et al. Schwer, P.J. et al. 1: 480 Schurmann, H. see Kumpfert, J. et al. Schusr, M. see Tomaschko, Ch. et al. Schuster, H.U. 3: 238; see also Czybulka, A. et al.; Drews, J. et al.; Kistrup, C.J. and Schuster, H.U., Petersenn, A.V. and Schuster, H.U. Schuster, H.U. et al. 3 238 Schuster, H.V. see Eberz, U. et al. Schuster, H.W. see Drews, J. et al. Schuster, J.C. 1: 113 Schuster, W. 3: 29 Schutz, R.J. 2: 624 Schutze, M. see Becker, S. et al. Schwab, G.-M. 1: 684; 2 647 Schwab, R. 1: 565, 574 Schwall, R.E. et al. 2: 361 Schwaller, P. see Aebi, P et al. Schwander, P. see Kozubski, R. et al., Kral, F. et af. Schwnrtz, A.J. see Nsmng, L.M. et al.; Wang, J.N. et al. Schwartz, A.J. and Tanner, L.E. 3: 56 Schwartz, B.B. see Fltikiger, R. et al. Schwartz, D.M. 2 278 Schwartz, D.S. 2: 224, 295; see also Shih, D.S. et al.; Soboyejo, W.O. et al. Schwartz, L.H. 1: 478 Schwarz, K. see Blaha, P et al.; Finstad, T.C. et al., Moruzzi, V.L. et al. Schwarz, R. see Hellner, E. et al. Schwarz, R.B. 1: 696, 700, 701, 733, 736; 2: 260, 263, 268, 273, 274, 281, 611; see also Alamgir, F.M. et al.; Clemens, B.M. et al.; Desch, P.B. et al.; He, U.et al., Jin, 0. et al., Miller, M.K. et al.; Shen, T.D. et al.; Srinivasan, S. et al.; Srinivasan, S.R. et al.; Wright, W.J. et al.; Yvon, P.J. and Schwarz R.B. Schwarz, R.B. and Funk, L. 3: 695 Schwarz, R.B. and He, Y. 3: 685,688,690 Schwarz, R.B. and Jin, 0.3: 702 Schwarz, R.B. and Johnson, W.L. 3: 681 Schwarz, R.B. and Koch, C.C. 3: 681 Schwarz, R.B. and Mitchell, J.W 3: 695 Schwarz, R.B. et al. 1: 706, 733; 3: 657, 685, 738 Schwarz, S.A. see Sands, T. et al. Schwarz, V see Wendt, H. et al. Schwarzenbach, D. 2: 237, 364 Schweika, W. 1: 40, 292, 304 Schweika, W. et al. 1: 614, 615 Schwellinger, P. see Leamy, H.J. et al.
Azlthor Index Schwer, R.E. see Harris, M.et al. Schwerrnann, W 1: 105 Schwitzgabel, G. see Moses, Z. et al. Scilla, G. see Thomas, 0. et al. Scobey, I.H*see Todd, A.@. et al. Scorey, C.R. see Deevi, S.C. et al.; HajdigOl, M.R. et al., Mistler, R.E. et al.; Strauss, J.T. et al. Scott, D.M. see Marshall, E.D. et al.; Tseng, W.F. et al. Scott, E.R.D. 1: 3, 631, 633; 2: 306 Scott, G.D. I: 479, 679 Scott, J. et al. 1: 973 Scott, M.G. see Gregan, G.P.J. et al. Scott, R.E. 1: 726 Scott, S.W. 1: 640 Scott, T.L. see Kirk, M.A. et al. Scotti, S. 2: 412 Seager, C.H. 1: 958 Seager, G.C. 1: 565 Seah, M.P see Briggs, D. and Seah, M.P. Seaman, C.L. see Lacerda, A. et al. Seainan, C.L. et al. 1: 221 Sears, J.W et al. 1: 640, 642 Secoue, M. see Guivar’c, A. et al. Seebauer, E.G. see Mendicino, M.A. et al. Seeger, A. see Dais, S. et al.; Schaefer, H.-E. et al.; Shirai, Y. et al.; Wurschum, R. et al. Seeger, A. et al. 3: 361 Seeger, J. 2: 81; see also Dahms, M. et al., Koeppe, C . et al. Seelantag, W 3: 238; see Eberz, U. et al. Seelmann-Eggebert, M. see Richter, H.J. et al. Seetharaman, V see Boehler, C.J. et al.; Semiatin, S.L. et al. Seetharaman, V and Semiatin, S.L. 3: 623, 624 Seetharaman, V. et al. 1: 653 Segal, B. see Amador, C. et al. Segal, E. see Craig, R.S. et al. Segal, V.M. et ul. 3: 628 Segnan, R. see Koon, N.C. et al. Segtrop, I(.see Schneider, D. et al. Seibold, A. 2: 156 Seidel, T.E. see Pal, C.S. et al. Serdnian, D.N. 1: 797 Seifert, H. see Hoenig, H.E. et al. Seigle, L.L. 1: 923; 2 490 Seiler, A. see Sehlapbach, L. et al. Seino, Y. 2: 523 Seitz, F. 1: 15, 496, 764, 784, 803 Seitzman, L.E. see Perepezko, J.H. et al. Sekhar, J.A. 1: 454; see also Dey, G.K. and Sekhar J.A., Ho, C.T. er al. Sekhar, J.A. et al. 3: 604, 609 Seki, I(.see Takahashi, T. et al. Seki, Y. 1: 700, 701 Selaliom, N. 1: 64 Sellars, C.M. SEC Prakash, U. et al. Sellers, C.M. see Blackford, J.R. et al. Sellmyer, D.J. 1: 1018; see also Oh, J.E. et al.; Wang, Y et al. Sellmyer, D.J. et al. 2: 314, 315 Selman, G.L. see Corti, G.W. et al.; McGill, I.R. and Selman, G.L. Selman, G.L. and Midgley, R.J. 3: 65 Selvam, P. see Bonhomme, F. et al. Semenenko, V.Ye. et al. 2: 653 Semenov, V. see Rabkin, E. et al.
Semenova, A.D. et al. 3: 78 Seinenova, E.I. 2: 518, 523, 524 Semenova, E.L. and Kudryavtsev, Y.V 3: 56 Semenova, E.L. et al. 3: 56, 60 Semenovskaya, S. see Khachaturyan, A. et d. Semenovskaya, S.V see Khachaturyan, A.G. et al. Semiatin, S.L. 1: 644; 2 86; 3: 623, 624, 626, 649; see also Dutton, R.E. et al.; Lombard, C.M. et al.; McQuay, P.A. et al., Seetharaman, V. et al. Semiatin, S.L. and Seetharaman, V. 3: 619, 627, 630, 631 Semiatin, S.L. et al. 1: 639, 644, 653, 655; 2: 125,127,298; 3: 623,626,631,632, 650, 651 Seminozhenko, V.P. see Dumas, J. et al. Semrad, E.E. see Motrya, S.F et al. Sen, N. see Krunimacher, S. et al. Sen, S. 1: 912; see also Patten, E.A. et al. Sen, S. et al. 3: 756 Sinateur, J.P. 1: 579, 791, 792, 795, 797; see also Thomas, 0. et al. Senba, H. see Umakoshi, Y. et al. Senba, Hiroyuki et al. 3: 316 Sence, M. see Bigot, J.Y et al. Senchenko, A.A. et al. 1: 948 Senkov, O.N. see Imayev, R.M. et al.; Salishchev, G.S. et al. Seno, T. see Nakano, T. et al. Seo, D. et al. 3: 297, 302 Seo, D.Y. see McQuay, P.A. et al. Seo, D.Y et al. 3: 419 Sepiol, B. see Eggersmairn, M. et al., Kummerle, E. et al. Sepiol, B. and Vogl, G. 3: 290 Sepold, G. see Clemens, H. et al. Sequeda, F.O. see McGahan, W.A. et al. Sereni, J. see Bauer, E. et al. Sergev, S.S. et al. 2: 646 Seri, H. see Moriwaki, Y et al. Serikova, V.P. see Arkharov, V.1. et al. Serin, B. 1: 944 Serizawa, H. see Ohnaka, K. et al. Seto, T. 1: 619 Setton, M. and Van der Spiegel, J. 3: 664 Setton, M. et al. 2: 616 Severin, L. see Eriksson, 0. et al. Severing, A. et al. 1: 218 Severson, M.C. see Zimm, C.B. et al. Sevillano, J.G. 3: 361 Sevov, S.C. 1: 303; 3: 119; see also Bobev, S. and Sevov, S.C.; Gascoin, F. and Sevov, S.C.; Queneau, V. and Sevov, S.C.; Todorov, E. and Sevov, S.C., Xu, L. and Sevov, S.C.; )
939 Shagiev, M.R. see Imayev, R.M. et al.; Salishchev, G.S. e t al. Shah, D.M. 2: 25, 41, 42, 225, 241, 244, 245, 246,247,249,294,295,298,650; see also Anton, D.L. and Shah D.M.; Anton, D.L. et al.; Pope, D.P. et al. Shah, D.M. and Anton, D.L. 3: 311 Shah, D.M. and Cetel, A, 3 302,303,304, 305, 306 Shah, D.M. and Duhl, D.N. 3: 299, 31 1, 31 5 Shali, D.M. et al. 1: 997, 999, 1005 Shah, M. and Pettifor, D.G. 3: 81 1 Shahwan, C.J. see El-Masry, N.A. et al. Shaheen, S.A. see Mendoza, W.A. and Shaheen, S.A. Shaklee, J.B. see Foster, K. et al. Shalimov, A.S. see Kablov, E.N. et al. Shallcross, F.V. see Kosonocky, W.F et al. Sham, L.J. 1: 24, 26, 59, 129, 196; see also Kohn, W. and Sham, L.J. ~hamasundar,S. see Dutton. R.E. et al. Shamin, S.N. see Galakov, V.R. e f al. Shan, A. 1: 928, 989 Shanabarger, M.R. 1: 987, 992 Shang, P. see Perez, J.F. et al. Shang, P. e f al. 3: 410 Shang, S.S. 1: 649; see irlso Meyers, M.A. et al. Shani, Y. et al. 2: 416 Shank, C.V 2: 339 Shank, P see Fraas, K. et al. Sliaakar, S. 1: 923; see also Dust, M. et aE. ~hannette,G.W 1: 875, 887, 920 Shannon, R.D. 1: 249 Shao, J. 1: 604; see Zhou B. et al. Sliao, Y and Spaepen, F. 3: 683 Shapiro, S.M. see Grier, D.H. ez al.; Satija, S.K. et al. Shapiro, S.M. et al. 1: 163 Shappino, J.R. 2: 623 Sharipov, S1r.M. see Mukimov, K.M. et al. Sharma, R.C. see Lin, Y -C. et al. Sharp, J.W. et al. 3: 106 Shashkov, D.P. 1: 925, 956, 957 Shashkov, O.D. see Adrianovskii, B.P. et al., Razuvayera, B.D. et al. Shashkov, O.D. et al. 2: 567 Shatzke, M. 1: 956 Shavers, C.L. see Hyde, B.G. e f al. Shaw, B.J. see Randall, R. et al. Shaw, D.T. see Chen, D.Y et al. Shaw, M.G.. see Alman, D.E. et al. Shaw, K.G. and Reinshagerr, J.H. 3: 504 Shay, J.L. 2: 330 Shcherbakov, A.S. et al. 1: 947 Shcherbakova, Ye.V see Ivanova, G.V. et al. Shea, M. see Stoloff, N.S. et al. Shea, NI. et al. 1: 973 Sheasby, J.S. 1: 996 Shechtman, D. 1: 453, 482, 531, 916; 2: 147, 185; see also Bendersky, L. et al., Cahn, J.W. et al.; Lipsitt, W.A. et al.; Portier, R. et al., Schaefer, R.J. et al. Shechtman, D. and Jacobson, L.A. 3: 414 Shechtnzan, D. et al. 1: 453,454, 504, 534; 2 75, 77, 87, 185 Sheetman, R. et al. 3 648
940 Sheikon, I. see Thomasson, Y. et al. Shein, A.B. 2: 502 Shekhar, J.A. see Fu, M. and Shekhar J.A. Sheldon, W.A. see Nicholson, D.M.C. et al. Shelton, R.N. see Chan, J.Y. et al.; Moodenbaugh, A.R. et al. Shelyakev, A.V. see Matveeva, N.M. et al. Shemet, V. see Singheiser, L. et al. Shemet, V.J. see Kisly, P.S. et al. Sheinyakin, V.S. et al. 3: 29 Shen, B.G. see Cheng, Z.H. et al., Fung, K.K. et al. Shen, B.G. et al. 3: 98 Shen, J. see Chen, S, E t al. Shen, S . see Boone, D.H. et al. Shen, S.S. see Bruzzone, P et al., Dresner, 2. et al. Shen, T.D. see He, Y. et al. Shen, T.D. and Schwarz, R.B. 3: 682,683, 698, 701, 702 Shen, T.D. et al. 3: 682, 683, 691, 692 Shen, Y. see Cassada, W.C. et al. Shen, Z. et al. 3: 216 Shen, Z.X. et al. 1: 136 Sheng, L. et al. 2: 314 Sheng, Q.G. see Lim, S.P et al. Shepherd, F.D. see Kosonocky, W.F et al. Sherby, O.D. 1: 656, 916, 925; see also Cheng, G.H. et al.; Cheng, S.C. et al., Wolfenstine, J. et al. Sheman, A.M. see Dowling, W.E. Jr. et al. Shero, J.S. see Watako, Y. et al. Sherwood, R.C. see Nesbitt, E.A, et al. Sherwood, R.C. et al. 2: 309, 448 Sheshadri, B.S. see Sr~matki,S.N. et ul. Shevakin, A. 1: 924 Shevchenko, V.G. see Kanonenko, V.L. et al. Shi, %.-P see Galakov, V.R. et al. Shi, X. et al. 3: 356, 366, 367 Shiba, H. 1: 34 Shibasaki, I. see Okamoto, A. et al. Shibata, 9. see Nishizawa, J, et al. Shibata, N. see Ohki, A. et al. Shibata, N.et al. 2: 326 Shibata, T. 1: 927, 983, 986, 989; see also Taniguchi, S. and Shibata, T.; Taniguchi, S. et al. Shibuya, A. et al. 3: 671 Shibuya, M. 1: 777 Shibuya, T. see Takeuchi, S. et al. Shibuya, T. et al. 1: 484 Shida, Y see Maeds, T. et al. Shidlovsky, I. see Mayer, I. et al. Shieh, J.H. see Ku, H.C. et al. Shield, J.E. 1: 50 Shield, J.E. and Krarner, M.J. 3: 400 Shield, J.E. et al. 1: 477; 3: 78, 400 Shiflet, G.J. see Cassada, W.C. et al., Chen, H. et al.; He, Y. et al. Shiga, M. see Wada, H. et al. Shigematsu, T. 1: 842 Shigeo, H. see Morita, A. et al. Shigeoka, T. see Iwata, M. et al. Shih, D.S. 2: 77, 79, 86; see also Deve, H.E. et al.; Huang, S.C. et al.
Shih, D.S. et al. 1: 640, 973; 2 83, 85; 3: 368 Shih, H. see Jepsen, D. et al. Shima, S. and Oyane, M. 3 648 Shimada, E. see Jin, M. et al. Shimida, K. see Uchitomi, N. et al. Shimizu, H. see Tsuji,H. et al. Shimizu, K. see Bnami, K. et al., Fuchino, S. et al.; Kobajashi, T. el al.; Nakata, Y. et al., Ohara, T. et al.; Tadaki, T. et al. Shimm, M. 1: 682; 2: 392 Shimizu, T. et al. 3: 577, 581 Shtmizu, Y. see Tadaki, T. et al. Shimoji, M. 1: 668; see also Kitajima, M. et al. Shimokawa, T. et al. 1: 916; 2: 153 Shimomura, 0. see Asaumi, K. et al. Minomura, S. et al. Shimotake, H. see Hesson, J.C. et al. Shiinotomai, M. see Doyama, M. et al.; Wang, T.M. et al. Shirnuzu, T. see Maki, R.et al. Shimuzu, T. et al. 1: 993 Shin, S.M. et al. 2: 610 Shindo, D. see %saki, G. et al.; Takdsugi, T. et al. Shingu, H.P. see Nishitani, S.R. et al. Shingu, P.H. see Dela Torre, S.D. et aE.; Taguchi, K. et al. Shingu, P.H. et al. 3 685 Shiniaev, A.J. 1: 768 Shinn, N.D. see List, R.S. et al. Shino, M. see Hashimoto, T. ed al. Shinoda, D. see Asanabe, S. et al. Shinoda, T. see Hayashi, Tohru et al. Shmohara, K.. see Horikoshi, Y. et al. Shinohara, K. et al. 2: 329 Shinohara, M, see Ishinioto, H. et al. Shinoura, 0. et al. 3 671 Shinoyama, S. see Kamada, H. et al. Shinozaki, S. see Toth, R.S. et al. Shiobara, R. see Ohara, T. et al. Shioda, M. see Maki, K. et al. Shiotani, N. see Kaiser, J.H. et al. Shiozawa, L.R. see Zare, T. et al. Shirai, T. et al. 2: 342, 420 Shirai, Y. 1: 564; 3: 280, 41 I, 414; see also Fujiwara, T. et al.; Ikebuchi, M. et al.; Inui, H. et al.; Ito, K, et al.; Nishitani, S.R. et al.; Yawaguchi, M. et al. Shirai, Y et al. 3: 281 Shiraishi, T. see Hisatune, K. et al.; Ohta, M. et al. Shirane, G. 1: 162, 163, 442, 443; 3: 166; see also Shapiro, S.M. et al.; Yamada, T. et al. Skirane, G. et al. 1: 444 Shires, P.J. et al. 2: 584 Shirley, D.A. 3: 576 Shirokura, T. see Ochiai, S. et al. Shirokura, Y. see Ochiai, S. et al. Shklyarevskii, I.N. 1: 682 Shocktey, W. 2 335; see also Eshelby, J.D. et al. Shoemaker, C.B. 1: 252, 411; 2: 478; see also Bendersky, L.A. et al. Shoemaker, D.P. 1: 252, 411; 2: 478; see also Marsh, R.E. and Shoemaker, D.P
Shoemaker, D.P et al. 1: 397 Shoemaker, H.E. see Brentnall, W.D. et ul. Shoemaker, J.R. et al. 1: 88, 570, 575, 577, 799 Shoenberg, D. see Lipson, H. et al. Shollock, B.A. see Basoalto, H.C. et al. Shornojo, F and Okazaki, H. 3: 251,255 Short, M.N. 1: 625 Showaki, K. er al. 1: 958 Shpyrko, 0.G. see DiMasi, E. et al. Shteinberg, H. see Sehwarz, R.B. et al. Shtremel, M. 1: 782 Shtrikman, S. 1: 202, 208; see also Biham, 0. et al, Shugaev, V.A. see Rogachev, A.S. et al. Shuh, D.K. see Kirn, Y.K. et al.; Sadwick, L.P. et al. Shull, C.G. 1: 11, 441, 442; see aim Wilhnson. M.K. et al. Shull, C.G, and Smart, J.S. 3: 166, 167 Shull, R.D. see Jones, S.A. et al.; Kaufman, M.J. et al.; Ma, E. et al.; McMichael, R.D. et al. Shunk, F.A. 3: 786 Shyue, J. et al. 3: 453, 774 Sibirtsev, D.S. see Sknpov, A V . et al. Sichevich, O.M. et al. 3: 100 Siddall, M. see Smathers, D. et al. Sidorko, V.R. see Lukashenko, G.M. et al. Sidorov, V.A. see Brazhkin, V.V et al.; Gavriliuk, A.G. et al. Sieber, H. see Sakidja, R. et al. Siegal, M. see Setton, M. et al. Siegbahn, K. et al. 3 136 Siegel, R.W 3: 753; see also Das Gupta, A. el al. Siegel, R.W. and Fougere, C.E. 3: 762 Siegel, S. 1: 11, 784, 803, 875, 888 Siegl, R. see Vitek, V. et al. Siegl, R. et al. 3: 407, 408, 772 S i e ~ a n nH.C. , see Sch~apbach,L. et al. Siegrist, T. see Cava, R.J. et al. Siegrist, T. et al. 1: 388; 3: 103, 104 Sieloff, D.D. see Brenner, S.S. et al. Sieloff, D.D. et al. 1: 590, 591, 593,594; 2: 30 Siemers, P.A. 2 75, 76, 83 Sievering, A. et al. 1: IS9 Sigalas, M. ef al. 1: 195, 203 Sigl, L.S. et at. 1: 862, 927 S i ~ u n dP. , 1: 692, 796; see also Winterbon, K.B. et al. Sigurd, D. see Miienpiiii, M. et al. Sih, G.C. 1: 882 Sikka, V.R. 1: 639, 6S5; 2 30, 39, 41, 43, 44, 199, 201, 203, 205, 207; 3: 492, 493, 502, 504, 505, 506, 51 3,604,609, 61 1, 612, 613, 639, 656; see also Choudhury, A. et al.; Deevi, S.C. and Sikka, V.K., Deevi, S.C. et al.; Hajaligol, M.R. et al.; Knibloe, J.R. et al.; Liu, C.T. and Sikka, V.K., Liu, C.T. et al.; McKamey, C.G. et al.; McKarney, C.C. et al.; McQuay, P.A. and Sikka, V.K.; Mistler, R.E. et al., Sahoo, M. et al.; Wang, J.N. et al.; Wright, R.N. and Sikka, V.K. Sikka, V.K. and Baldwin, R.H. 3: 502
Author hdex Sikka, V.K. and Deevi, S.C. 3 503, 604, 61 1 Sikka, V.K. and Santella, M.L. 3: 606 Sikka, V.K. et al. 2: 45, 202, 204, 205; 3: 361,368, 502, 503, 504, 572,607,609, 611, 612, 613, 614, 639, 654, 656 Sikka, V.W. see Deevi, S.C. el al. Silberberg, Y. see Sands, T. et al. Silcock, J,M. 3: 475; 2: 272,276, 283,284; see also Kear, B.H. et al. Silva, P.R.J. see Xia, S.K. et al. Sirnakov, G.V. see Batsanov, S.S. et al. Simcoe, C.R. 2: 91 Simensen, C.J. 2: 177, 184 Simizu, S. see Huang, M.Q. et al. Simkiiis, R. see McQuay, P.A. PT al. Simkowtch, G. see Zelanki P.D. Simmons, G. 1: 203, 204 Simmons, J. see Rao, S . et al.; Rao, S.I. et al. Siminons, J.P see Subramanian, P.R. et al. Simmons, J.P. et al. 3: 437, 439, 445 Simmons, R.Q. 1: 561 Simmons, R.Q. and Balluffi, R.W 1: 561; 3: 277 Simon, Ch. see Provost, J. et al. Simon, F.E. see Kurti, N. et al. Simon, G. 1: 319; sec. also Chen, C.H. ef al. Simon, J.P. see Papon, A.M. et al. Simonen, E.P. 1: 839, 841 Simons, D.S. 1: 1020 Simonsen, 1,K. et al. 1: 649 Simozar, S. 1: 247 Simpkins, R. see McQuay, P.A. et al. Sirns, C.T. see Kear, B.H. et al. Sims, C.T. et d.2: 3, 9 Sinis, D.M. et al. 2 44 Sirns, J. see Fowler, H.A. et al. Sinclair, J.E. 1: 80, 523, 611; 2 153; see also Finnis, M.W. and Sinclair, J.E. Sinclair, R. see Goo, E. et al., Moberly, W.J. et al. Sinfelt, J.H. 2: 482 Singer, I.L. see Hubler, G.K. et al. Singer, R.F. see Fitzgerald, T.J. and Singer R.F.; Gromann, J. et al., Hugo, F. et al. Singh, A. see Ranganathan, S. et al.; Van Tendeloo, G. et al. Singh, A.K. et al. 3: 460 Singh, D.J. 1: 134, 137, 138, 195, 199; see also Perdew, J.P et al.; Pickett, W.E. et al. Singh, D.J. et al. 1: 132, 133, 137, 197 Singh, H.P. see Misra, S. et al. Singh, J. see Suryanarayana, C. et al. Singh, J.P. et al. 3: 623 Singh, N.K.P. et al. 1: 677 Singh, P.P. I: 955; see also Asta, M. et al. Singh, R. see Tiwari, P. et al. Singh, R.N. 1: 665, 667, 668, 669, 671, 673, 674, 675, 676, 677, 678, 679; see also Jha, T.S. et al., Mishra, A.K. et al., Ramachandra Rao, P. et al.; Singh, N.M.P. et al. Singh, R.N. and March, N.H. 3: 26, 245, 246 Singh, S.K. 2: 382
Singh, S.R. and Howe, J.M. 3: 408, 409, 411,413, 422 Singhal, L.K. 2: 283 Singheiser, L. see Restall, J.E. et al.; Wagner, R. et al. Singheiser, L. et al. 1: 998, 999; 2: 227; 3: 479 Singleton, M.F see Nash, P. et al. Sinha, A.K. 1: 473; 2: 231 Sinha, A.K. et al. 1: 704 Sinhil, S. 1: 677 Sinharoy, S. et al. 1: 619; 3: 216, 219 Sinke, W et al. 2: 605, 623, 625 Sinkler, W. et al. 1: 706 Sinnerna, S. et al. 1: 643 Sinvhal, R.C. 1: 108 Sinyakova, S.I. see Vedernikov, M.V. et al. Sinyushko, V.G. see Levin, E.M. et al. Sipilg, V see Fugleberg, S . et al. Sisson, M.J. see Kelly, M.J. et al.; Wickenden, D.K. et al. Sistare, G.H. 2: 559, 560, 561, 562, 563, 568, 570, 571 Sitaud, B. 1: 782; see also Dimitrov, C. et al. Sitaud, B. et al. 1: 577 Sivil, C.S. 2: 652 Sivokha, V.P see Khachin. V.N. et al.; Matveeva, N.M. et al.; Savvinov, A S . et at. Six, W et al. 2: 306 Sizek, H.W. and Gray, G.T. 111 3: 368, 371, 373 Sizmann, R. 1: 785 Sjoberg, J. et al. 1: 360 Sjolander, A. see Bengtzelius, U. et al. Skelly, D.W. 2 290, 291, 298; see also Jackson, M.R. et al.; Rowe, R.G. et al. Skelton, E.F. see Yu, S.C. et al. Skillman, S. 1: 231, 243 Skinner, A.J. 3: 841; see also Lill, J.V et al.; Pettifor, D.G. et al. Skinner, D.J. 1: 891; 2: 178, 180, 182, 185, 194; see also Brown, A.M. et al.; Bye, R.L. et al.; Kiin, N.J. et al.; Zedalis, M.S. et al. Skinner, D.J. et al. 2 178, 180, 182, 184 Skinner, D.P 2 414, 415, 417, 432 Skjerpe, P. 2: 175, 176 Skobelkin, V.N. see Balagurov, A.Y. et al. Skog, IS.see Saarinen, K. et al. Skold, K. see Brun, T.O. et al. Skold, K. see Price, D.L. and Skold, K. Skold, K and Nelin, G. 3: 252 Skolnik, L.H. see Kosonocky, W.F. et al. Skolozdra, R.V 2 226; see also Karla, I. et al.; Vert, R. et al. Skolozdra, R.V. et al. 2: 596 Skomski, R. see Leithe-Jasper, A. et al. Skorov, D.M. see Ananyn, V.M. et al. Skowronek, C.J. et al. 3: 584 Skripov, A.V. et al. 3: 253 Skriver, H.L. 1: 58, 83; 2: 440 Skrotzki, W. see Ledig, L. et al. Skyllas-Kazaeos, M. 2 511 Slack, D.S. see Bruzzone, P. et al., Markiewicz, W.D. et al. Slack, G.A. 1: 1025, 1027; see also Tritt, T.M. et al.
941 Slack, S.S. et al. 2: 374 Slansky, E. see Johan, E. et al. Slastnikova, L.F. see Hertzriken, S.D. et al. Slater, J.C. I: 33, 129, 229 Slifkin, L.M. 1: 559, 761, 765; 2: 134 Sloman, H.A. 1: 101 Slonczewski, J.C. 2: 392 Sloyzov, v v. 2 10 Sluiter, M. 1: 34,37,38,40,41; 3: 800; see also Asta, M. et al., Asta, M.D. et al.; Gonis, A. et al.; Johnson, D.D. et al.; Mikalopas, J. et al.; Turchi, P et al.; Turchi, P.E.A. er al. Sluiter, M. et al. 3: 41, 44, 47, 67, 68, 70 Sluiter, M.H.F. see Turchi, P.E.A. et aE. Sluiter, M.H.F. et al. 3: 204, 205 Slyozov, V.V 1: 866; 2: 263 Small, C.J. 1: 745 Smallman, R.E. 1: 503,504,505, 542, 574; 2: 58, 61; see also Campany, R.G. et al.; Fraser, H.L. et al., Hameed, M.Z. et al.; Hume-Rothery, W. et al.; Ngan, H.W. et al.; Rong, T.S. et al.; Yu, H.F. et al.; Zhu, W, et al. Smarsly, W. see Dahms, M. et al.; Kestier, H. et al.; Wagner, R. et al. Smathers, D. 2: 353 Smathers, D. et al. 2: 361, 363 Smedskjaer, L.C. see Das Gupta, A. et al. Smeltzer, W.W. 1: 990 Smialek, J.L. 1: 982, 983, 987, 990, 991, 992,993,994, 1009; 2: 54, 55, 57, 167, 168; 3: 570, 577; see also Brady, M.P. et al., Bnndley, W.J. et al.; Doychak, J. et al.; Hebsur, M.G. et al., Lowell, C.E. et al.; Parfitt, L.J. et al. Smialek, J.L. and Humphrey, D.L. 3: 575 Smialek, J.L. et al. 1: 987, 988, 994, 995, 996; 2: 124, 168; 3: 577 Smigelskas, A.D. 2: 361 Smirnov, A.A. 1: 28, 760, 761, 767 Smirnov, L.V. see Greenberg, B.A. et al. Smirnov, V.A. 3: 30, 31 Smirnov, V.I. 1: 7, 8 Smimova, N.L. 1: 403 Smirnova, N.L. et al. 1: 261 Smit, H.H.A. et al. 2: 312 Smith, A.A. 2: 594 Smith. A.B. see Neumann, D.A. Smith, A.B. III see Zhu, Q. et al. Smith, C . 2: 649 Smith, C.H. 3: 698 Smith, D.A. I: 585, 586, 598; see also Paunovic, M. et al. Smith, D.K. et al. 1: 181 Smith, D.L. 2: 648 Smith, D.Q. 2 438 Smith, E.G. and Lang, C.I. 3: 55 Smith, G.C. 2: 258 Smith, G.D.W. see Kim, S. et al. Smith, G.S. et al. 3: 531 Smith, H.G. see Pintschovius, L. et al. Smith, 1.0. 2: 277, 278 Smith, J.A. see Menzies, R.G. et al. Smith, J.F. 1: 66,70,875, 887, 889,920; 3: 802; see also Aitken, E.A. and Smith, J.F.; Schiltz, R.J. Jr. et al. Smith, J.F. and Hansen, D.A. 3: 115 Smith, J.F. et al. 1: 69
942 Smith, J.L. SCE Demczyk, B. et al.; Fisk, Z . et al., Fowler, C.M. et al.; Giorgi, A.L. el al.; Matthias, B.T. et al.; Ott, H.R. et al., Stewart, G.R. et al. Smith, J.R. see Rose, J.H. et al. Smith, J.S. 1: 733 Smith, L.S. et al. 3: 461 Smith, M.P. 2 80, 82, 83, 91, 93, 107 Smith, P. see Hsia, S.L. et al. Smith, P.L. 1: 184; see also Owen, N.B. et al. Smith, P.M. 2: 629; see ulso Chao, P.C. et al. Smith, P.R. 1: 861, 993; 2 289, 292; see also Graves, J.A. et al. Smith, P.R. et al. 2: 289, 292, 294, 298; 3: 495, 584 Smith, P.W see Chemla, D.S. et al. Smith, R.A. see Soboyejo, W.O. et al. Smith, R.R. 2: 402 Smith, R.S. see Richter, H.J. et al. Smith, R.W. 1: 646; see also Castro, R.G. et al. Smith, S.D. see Bigot, J.Y. et al. Smith, S.D. et al. 2: 430, 411, 412 Smith, T.L. see Cheng, H. et al. Smith, T.R. 3: 739; see also Kallingal, C.G. et al.; Matsugi, K. Smith, T.R. et al. 3: 326 Smithells, C.J. 1: 875, 883, 886; 2: 148 Smyth, P. see Fayyad, U . N . et al. Snoek, J.L. 2: 306; see also Six, W. et al. Snow, A.I. see Florio, J.V et al. Snow, D.B. see Anton, D.L. et al.; Breinan, E.M. et al. So, F.C.T. see Molawa, E. et al., Nieh, C.W. et al.; Zhu, M.F. et al. So, F.C.T. et al. 2: 618, 625, 626, 628 Sob, M. see Paidar, V et al.; Schaefer, H.-E. et al. Sob, M. et al. 3: 200 Sobczak, E. and Auleytner, J. 3: 149 Sobolev, A.N. see Siclievich, O.M. et al. Soboyejo, W. see DiPasquale, J. et al. Soboyejo, W.O. 3: 337; see also Aswath, P.B. et al.; Lu, G.-Y. et al.; Srivatsan, T.S. et al., Ye, E". et al. Soboyejo, W.0. et al. 334, 336, 337, 342, 344, 345 Socolar, J.E.S. 1: 461, 480, 482; see also Lubensky, T.C. et al., Onoda, G.Y et ccl. Sodaiii, Y. 1: 547, 548,915; see also Vitek, v et al. Soderland, P. see Wills, J.M. et al. Sodervall, U. see Diviiiski, S.V. et al. Soffa, W. see Zhang, B. er al. Sof€a, W.A. I: 852, 853, 854, 855; 2: 282; see also Rao, M. and Soffa, W.A.: Shang, B. and Soffa, W.A.;Strychor, R. et al. Soga, N. see Schreiber, E. et al. Soisson, F. see arti in, G. et al. Sokolov, A.V. 2: 437,438 Sokolova, G.K. see hrkharov, V.I. et al. Sokolovskaya, E.M. see Guzey, L.S. et al.; Tsurikov, V.F. et al. Soldner, L. see Klau~iinzer,S. et al. Solenthaler, C. see Mllner, P. et al. Soler, J.M. 1: 208 Soltys, J. see Kozubski, R. et al.
Author hidex Solv'yev, S. see Zinsser, W.A. et al. Sonianathan, C.S. see Satya Murthy, N.S. et al. Somekh, R.E. see Highmore. F.J. et al.; Kieschke, R.R. et al.; Wangts, C.H. et al. Somer, M. see von Schnering, H.G. et al. Somer, M, et al. 3: 115 Somerfeld, A. 1: 440 Somers, M.A.J. see Reader, A.H. et al. Soiniesh, B. et al. 3: 288 Sommelet, P. 1: 118 Sommer, A.H. 2 418 Sommer, IF. 1: 666, 667, 668, 669, 671, 673, 674, 678, 679; see also Kek, S. et al.; Moser, Z. et al.; Rosa, C.J. et al. Sommer, F. et al. 1: 95, 735, 739 Sommerfield, A. see Grimm, H.O. and Sommerfield, A. Sommers, C.B. see Kiibler, J. et al. Somorjai, G. 1: 609; see also Baird, R. et al.; MacLaren, J.M. et al. Somov, A.I. see Semenenko, V.Ye. ez al. Somoza, J.A. see Gallego, L.J. et al. Son, J.-Y. et al. 3: 150 Sondericker, D. et al. 1: 615, 616; 3: 219 Sondheimer, E.H. 1: 944 Sondhi, S. et al. 3: 44 Song, H.Y. and Wang, X. 3: 736 Song, S. see Daulton, T.L. et al. Song, S.G. et al. 3 423 Song, Z.Y et al. 1: 536 Songina, O.A. 2: 518, 521, 523, 524, 646, 647 Sonomura, H. see Horinaka, H. et al. Sood, D.K. see ~ a t t a g ~ iG. n , et al. Soper, A.K. see Caciuffo, R. et al. Sorantin, P. see Blaha, P. et al. Sorby, H.C. 1: 7 Sosrosoedirjo, B.I. see Hisatsune, K. et al. Sotier, S. see Huckstein, K. et al. Soubeyroux, J.L. see Ayres de Campos, 9. et al., Obbade, S. et al. Soudani, S.E. see Semiatin, S.L et al. Southwell, G. et al. 2 227, 228 Southwell, R.P. see Mendicino, M.A. et al. Spaczer, M. see Diaz de la Rubia, T. et al. Spada, F.E. see Berkowitz, A.E. et al. Spadon, M. see Blau, B. et al. Spaepeii, F. 1: 454, 457, 463, 473, 474, 478,479,480,537,690,696,733, 742, 743, 744; see also Chen, L.C. et al. Spaepen, F. and Taub, A.I. 3: 695 Spaepen, F. and Turnbull, D. 3: 683, 693 Spaepen, F. et al. 1: 477 Spaetig, P. see Baluc, N. et al. Spahr, M.E. see Winter, M, et al. Spain, I.L. see Yu, S.C. et al. Spanner, G.E. 2: 637 Spanner, J. 2: 564, 573 Spano, M.L. see Clark, A.E. et al. Sparka, U. see Appel, F. et al. Sparks, C.J. see Porter, W. ct al., Porter, W.D. et al.; Reinhard, L. et al. Sparks, C.J. et al. 1: 500; 2: 157, 165 Spatz, P. see Ziittel, A. et al. Specht, E.P. see Sparks, C.J. et al. Spedding, F.H. et al. 2: 518 Speich, G.R. 1: 835
Speicher, W. 1: 666, 674 Speidel, M.O. see Miillner, P. el al. Speier, W. see Hoekstra, H.J.W.M. et al., Sarrna, D.D. et al. Speier, W. et al. 3: 149 Spencer, P.J. 1: 992, 999; 3: 800 Spencer, P.J. see Kubaschewski, 0. et al. Spencer, R.M. et al. 1: 34311 Sperl, W. see Lamparter, P. et al. Sperling, A. see Hartfield-Wunsch, S.E. et al. Sperry, E.S. 1: 16 Sperschneider, C.J. see Bowers, J.E. et al. Spicer, W.E. see Shen, Z.X. et al. Spichkin, Yu.1. 3: 527; see Dan'kov, S.Yu. et al.; Nikitin, S.A. et al. Spiers, G.D. see Callegari, A. et al. Spingarii, J.R. I: 925 Spitzl, R. see Niehus, H. et al. Spohr, P.A. see Kurti, N. et al. Spohr, R. see Audouard, A. et al., Trautmann, C. et al. Spooner, F.J. 1: 252; see also Wilson, C.G. et al. Spozhnikova, L.V. see Bannov, S.M. et al. Sprague, R.W. see Jin, R. et al. Sprengel, W. see Valeeva, A,A. et al.; Zhang, X.Y et al. Sprengel, W. and Schaefer, H.-E. 3: 288 Sprenger, H. et al. 2 296 Springer, T. see Lottner, V. et al. Springer, T. and Richter, C. 3: 252, 253 Spnngford, M. see Rcinders, P.H.P. et al. Sprosser-Prou, J. see vom Felde, A. et al. Sproul, W.D. 3: 665 Spruiell, J.E. see Brooks, C.R. et al. Sprunger, P.T. et al. 3: 213 Srimathi, S.N. et al. 3: 670 Srinivas, V. see Dunlap, R.A. et al. Srinivasan, R. see Singh, J.P et al., Szaruga, A. et al.; Weiss, I. et al. Srmivasan, R. et al. 3: 366, 367 Srinivasan, S. see Schwarz, R.B. et al. Srinivasan, S. et al. 1: 647; 2: 153, 157; 3: 756, 758 Srinivasan, S.R. see Schwarz, R.B. et al. Srinivasan, S.R. et al. 3: 486 Sriram, S. 1: 989; see also Mentzer, M.A. et al. Sriram, S. et al. 1: 533, 534; 2: 82; 3: 413, 419, 424 S r i r a i ~ a ~ u r t hA.M. y , 2 296 Sritharan, T. see Greenwood, G.W. et al. Srivastava, P.K. et al. 1: 735 Srivastava, P.L. ,we Jha, I.S. et al. Srivatsan, T.S. 1: 989; see also Soboyejo, W.O. et al. Srivatsan, T.S. et al. 3: 326, 332 Srolovitz, D. see Vitek, V. et al. Srolovitz, D.J. see Chen, S.P et d., Farkas, D. et al.; Gibala, R. et al., Najafabadi, R. et al.; Waiig, H. et al. Srolovitz, J. 1: 820 Stabile, P.J. see Loubriel, G.M. et al. Stiiblein, H. 2: 318 314, 315, 316; 3: 96; see also El-Masry, N.A. et al.; Liu, N.C. et al. Stadelmaier, H.H. and Reinsch, B. 3: 97 Stadelmaier, H.H. et al. 2 310, 314, 316
Author Indgx Stadelmann, P.A. see Schublin, R. and Stadelmann, P.A. Stadler, H. ,we Schmid, M. et al. Stadler, R. et al. 3: 240, 241 Stadnik, Z.M. 1: 484; see also Stroink, G. et al. Stafeev, V.T. see Khryapov, V.T. et al. Stafford, G.R. 1: 454; see also Tierney, B.J. et al. Stafford, K.N. et al. 2: 492 Stahl, D. 2: 648 Stahl, R. see Jacobi, H. and Stahl, R. Staiger, W. see Yu, D. et al. Stanek, P.W. SCE Castro, R.G. et al. Stankiewicz, J. see Morellon, L. et al. Stanley, C.J. 1: 625, 626 Stanley, J.T. see Balanzat, E. et al. Stansbury, E.E. see Brooks, C.R. et al. Stark, E.A. see Chakraborty, S.B. and Stark, E.A. Stark, J.P. see Sanchez, J.M. et al. Starke, E.A. 2: 178 Starke, E.A. Jr. 1: 845 Stassis, C. 1: 874, 876; see also Goldman, A.I. et al. Stassis, C. et al. 1: 156, 1020 Staton-Bevan, A. 2: 35 Staton-Bevan. A.E. 1: 548; 2 24; see afso Holdway, P. and Slaton-Bevan, A.E. Staubli, M. see Lupinc, V. et al. Staunton, J.B. see Turchi, P.E.A. et al. Staunton, J.B. et al. 1: 41 Stauss, G.H. see Rubinstein, M. et al. Staveley, L.A.K. see Parsonage, N. and Staveley, L.A.K. Steeb, S. 1: 792; see also Lamparter, P et al. Steeds. J.W. 1: 524, 8130, 881 Steele, B.C.H. 1: 95 Steeves, M.M. 2: 363 Steeves, M.M. et al. 2: 374 Stefanescu, D.M. 1: 912 Stefaniay, V. et al. 2 175 Stefanon, M. see Caciuffo, R. et al. Stefanou, N. see Koch, J.M. et al.; Koenig, C. et al. Stefanou, N. et al. 1: 569 Steglich, F. 1: 211, 215; see also Fraas, K. et al. Steglich, F. et al. 2: 229 Steiger, N.H. 3: 267 Steigmeier, E.F. see Dismukes, J.P. et al. Stein, D. see Palmer, R. et al. Stein, D.F. 2 27 Stem, D.W. 1: 104, 105 Steinemann, S . 3 231, 241 Steinemann, S.G. 3 74 Steinemann, S.G. et al. 3: 231, 233, 237, 24 I Stemer, A. 1: 95 Steiner, W. see Weitzer, F. et al. Steinfink, H. 1: 243; see also Iglesias, J.E. and S t e i ~ n kH. , Steinhardt, P.J. 1: 457, 480, 482, 491; see also Bancel, P.A. et al.; Levine, D. et al., Lubensky, T.C. et al.; Onoda, G.Y et al. Steinhardt, P.J. et al. 1: 479 Steinhorst, M. 1: 994, 995, 1005; see also Grabke, H.J. et al.
Stein~ann,S.G. see Anongba, P.N.B. and Steinemann, S.G. Steinmetz, P see Lemoine, V. et al. Stekly, Z.J.J. 2 364 Stekly, Z.J.J. and Gregory, E. 3: 103 Stekly, Z.J.J. et al. 2: 381 Stemple, N.R. see Suchow, L. et al. Stepanov, G.N. see Gavsiliuk, A.G. et al. Stephens, J.R. 1: 994; 2: 199, 200, 201, 206, 207, 228; see also Hebsur, M.G. et al.
Stephens, P.W 1: 457, 482; see also Bancel, P.A. et al.: Goldman, A.I. et al., Guryan, C.A. et al. Stephens, P.W. et al. 3: 257 Stephenson, D.J. 2: 496; see also Hancock, P. et al.; Nicholls, J.R. er al. Stephenson, G.B. see Koster, U. et al. Stern, E.A. 2: 439, 440, 441; see also Das, B.K. et al. Stern, E.A. et al. 1: 479 Stern, R. 1: 875, 887 Sternberg, A. see Zirnm, C.B. et al. Sternberg, D.D. see Hemker, K.J. et al. Sterne, P.A. see Mikalopas, J. ei al.; Nicholson, D.M. et al.; Nicholson, D.M.C. et al.; Pei, S. et al.; Stocks, G.M. et al. Sterner-Rainer, L. 2: 564 Stetson, A.R. see Brentnall, W.D. et al. Steuer, J. 1: 773 Steurer, W. 1: 470, 484; 3: 387 Stevens, G.T. et al. 1: 721, 726 Stevens, J. see Green, G. et al. Stevens, K.W.H. 3: 171, 172 Stevens, R. see Howson, M.A. et al. Stevens, R.N. 1: 918 Stevenson, A. see Housh, S. et al. Steves, R.N. see Davies, C.K.L. et al. Steward, C.J. 1: 110 Steward, S. see Felter, T. et al. Stewart, A.M. see FitzGerald, J.D. et al. Stewart, D.C. 3: 601 Stewart, G.R. 1: 21 1; see also Giorgi, A.L. et al., Roy, S.B. et ul. Stewart, G.R. et al. 3: 46 Stewart, M.J. 2: 136 Steyaert, S. et al. 3 98 Steyert, W.A. 3: 522 Steyn, H. DeV. see Wolff, I.M. et al. Sticht, 5. see Maurer, T. et al.; Oppencer, P.M. et al. Stiegler, J.O. 1: 914, 924; 2: 18; see also Johnson, L.A. et al.; LIU,C.T. et al., Whang, S.H. et al. Stievenard, D. see Bourgoin, J.C. et al. Stinchcombe, R.B. see AshraE, J.A. et al. Stinton, D.P. et al. 3: 669 Stirling, W.G. et al. 1: 154 Stobbs, W.M. 2: 188, 190; see a1,w Knowles, K.M. et al. Stockbarger, D.C. 3: 542 Stocker, H.J. see Forrest, S,R. et al. Stocks, G.M. 1: 24, 27; see also Ceder, G. et al.; Gonis, A. et al.; Gyorffy, B.L. et al.; Johason, D.D. et al.; Nicholson, D.M. et al., Nicholson, D.M.C. et al., Pei, S. et al.; Sluiter, M. et al., Turchi, P.E.A. et al. Stocks, G.M. et al. 1: 498, 505 Sloeckel, D. see Dueng, T. et al.
943
Stoloff, N.S. 1: 521, 534, 546, 896, 913, 914,915,920,921,925,927,970; 2: 3, 13, 18,24,28, 158, 164, 165,203,298, 598, 651; 3: 309, 325, 334, 346, 451, 613; #3# 654, 657; see also Almarm, D.E. et al.; Boettner, R.C. et al.; Cainus, G.M. et al.; Castagna, A. et al.; Fox, T. et al.; German, R.M. et al.; Gordon, D.E. et al., Kallingal, C.G. et al.; Kear, B.H. et al., Koch, C.C. et al.; Korinko, P.S. et al.; Kuruvifla, A.K. et al.; Liu, C.T. et al.; Matuszyk, W. et al.; Scott, J. et al.; Shea, M. et al., Smith, T.R. et al. Stoloff, N.S. and Alven, D.A. 3: 334, 339 Stoloff, N.S. and Davies, R.G. 3: 74, 364, 501, 758 Stoloff, N.S. and Liu, C.T. 3: 333, 339, 346, 347 Stoloff, N.S. and McKamey, C.G. 3: 613 Stoloff, N.S. and Sims. C. Stoloff, N.S. et al. 1: 993; 204, 207; 3: 503 Stolwijk, N.A. 1: 577; see a l ~ oBakker, H. et al.; Kroll, S. er al. Stolwijk, N.A. et al. 1: 566, 575; 3: 290 Stone, D.S. see Chang. Y.A. et al. Stone, W.E. 1: 628 Stonehouse, A.J. see Booker, J. et al., Pame, R.M. et al. Stoner, S.L. et al. 1: 925 Storer, F.H. 1: 3 Stori, H. see Mitterer, C. et al. Storm, A.R. see Testardi, L.R. et al. Stormer, H.L. see Dingle, R. et al. Storozhenko, A.I. 1: 718, 720 Stott, M.J. 1: 25, 78 Stoudt, M.R. see Ricker, R.E. et al. Stout, M.G. scc Kocks, U.F. et al. Stoutt, J.J. and Crimp, M.A. 3: 655 Stover, E.T. 2: 61, 296 Stowell, M.J. 2 178; see also Jacobs, M.N. et al., Pashley, D.W et al. Stower, D. see Ishiyama, S. et al. Stoyev, P.I. see Papirov, 1.1. et al. Strafford, K.N. see Datta, P.K. et al., Du, H.L. et al. Strafford, K.N. and Hanzpton, A.F. 3: 708 Strakana, R.E. see Clark, A.E. et al. Strandburg, K.J. see Widom, M. et al. Strane, J.W. see Li, J. et al. Strangwood, M. see Hippsley, C.A. et al., Srivatsan, T.S. et al. Strangwood, M. et al. 1: 925 Stranski, I.N. 1: 167 Strater, K. see Hanak. J. et al. Strauch, D. 1: 152, 153, 157 Strauss, J.T. et al. 3: 504, 655 Street, G.B. 2 448 Street, R. 1: 935 Streetman, B.G. see Block, T.R. et al. StreiE, R. I: 987, 991, 994, 996; 3: 566, 567, 577 Streiff, R. and Poize, S. 3: 570, 577 Strid, J. 2 184 Strife, J.R. see Nardone, V.C. et al.
944 Stringer, J. 1: 980; see also Goebel, J.A. et al.; Irving, G.N. et al.; Liu, C.T. et al. Stringer, J.F. 1: 987, 988, 994 Strizker, B. 2: 481; see also d’Heurle, F.M. et al.; Samwer, K. et al. Strnat, K.J. 2: 312, 313 Strnat, R.M.W. 2: 312, 313 Stroh, A.N. 3: 392, 393; see also Frank, F.C. and Stroh, A.N. Stroink, G. 1: 484 Stroink, G. et al. 2: 310 Strom-Olsen, J.O. see Altounian, Z. et al. Strom-Olsen, J.Q. et al. 1: 743, 745 Strong, H.M. see Bovenkerk, H.P. et al. Strongin, R.M. see Zhu, Q. et al. Strossner, K. see Werner, A. et al. Strothers, S.D. 2: 201 Strudel, J.L. see Gogia, A.K. et al.; Guimier, A. and Strudel, J.L. Struebing, V.O. see Matthias, B.T. et al. Struik, D.J. 1: 457 Strum, M.J. see m ens hall, G.A. et al. Strutt, P.R. 1: 921, 927; see also Bevk, J. et al., Hocking, L.A. et al., Kear, B.H. and Strutt, P.R.; Polvani, R.S. et al. Strutt, P.R. et al. 1: 537 Strychor, R. 2 95 Strychor, R. et al. 1: 857; 2: 96, 97, 98 Stucheli, M. see Mason, T.E. et al. Stuck, A. et al. 3: 140 Stucke, M. see Semiatin, S.L. et al. Stucke, M.A. see Larsen, J.M. et al.; Martin, P.L. et al.; Subramaniaii, P.R. et al., Vasudevan, V.K. et al. Stucke, M.A. et al. 3 445 Stucki, F. see Schlapbach, L. et al. Stucky, M. see Corbel, C. et al. Studer, F see Provost, J. et al., Toulemonde, M. et al. Stuijts, A.L. 2: 390 Stupart, G.V. see Lipson, H. et al. Stiiwe, H.P. 1: 252 Su, C. see Semiatin, S.L et al. Su, C.H. et al. 3: 28 su, L.M. 2: 335 SU,L.M. et al. 2: 335 Suarez, N. see Sanchez, J.L. et al. Subbanna, G.N. see Chattopadhyay, K. et al. Su~rahi~anyam, J. 1: 987, 994; 2: 124, 167, 168; see also Ranganath, S. et al. Subra~anian,P.R. see Bewlay, B.P. et al., Dimiduk, D.M. et al.; Hensha~l, G.A. et al.; Massalski, T.B. et al.; Mazdiyasni, S. et al.; Mendiratta, M.G. et al.; Parthsarthy, T.A. et al. Subra~anian,P.R. et al. 1: 433, 434; 2 157, 296; 3: 490, 491, 541, 545, 548, 552, 554, 555, 557, 558, 658, 802 Suchow, L. et al. 1: 351 Sucksmith, W. and Thompson, J.E. 3: 171 Suda, S . 2 486; see also Sandrock, G.D. et U / , Sudarshan, T.S. see Chatterjee, S. et al. Sud~arsanan,R. see Rohatgi, A. et al. Suenaga, H. see Ta~dsugi,T. et al. Suenaga, N.1: 808; %E 358, 361, 388; see also Bussiere, J.F. et al. Suenaga, M. and Clark, A.F 3: 806
Author Index Suffczynski, M. 1: 232 Suga, S. see Ogawa, S. et al. Sugai, T. see Hanamura, T. et al. Sugawara, H. see Ishikawa, M. et al. Sugawara, T. see Chiba, A. et al. Sugeta, T. 2: 335 Sugii, K. 2 328 Sugimoto, M. see Ando, T. et al.; Komatsu, K. et al. Sugita, T. see Inoue, A. et al. Sugiyama, K. 2: 336 Sugiyama, K. et al. 2: 330 Sui, H.X. et al. 3: 758 Sullenger, D.B. and Kennard, C.H.L. 3 9 Sullivan, C.P. et al. 2 502, 511 Sumer, A. 1: 875 Sumida, M. see Nakano, T. et al. Sumiyama, K. see Xu, Y. et al. Sumiyama, K. et al. 1: 773 Summers, C.J. see Chiaiig, K.H. et al. Summers, L.T. see Gregory, E. et al. Summers, L.T. et al. 2 361 Summers, S. see Kuhn, W.E. et al. Sumski, S. see Hayashi, 1. et al. Sun, H. 2: 314; see also Coye, J.M.D. et al.; Fujii, H. et al., Hu, B.P. et al., Otani, Y. et al. Sun, H. et al. 2: 314 Sun, J. et al. 3: 443 Sun, R.C. et al. 1: 956 Sun, T. see Heilmann, P. et al. Sun, W. see Hiraga, K. et al. Sun, Y. see Couret, A. et al. Sun, Y.Q. 1: 496, 512, 527, 528, 545, 548, 549; 2: 137; 3: 367,438,442,453,461; see also Ezz, S.S. et al.; Hazzledine, P.M. et al., Korner, A. et al. Sun, Y.Q. et al. 1: 509, 512, 534, 545, 551; 3: 411, 413 Sun, Y.X. see Lin, C. et al. Sun, Y.Y. et al. 3: 103 Sun, Z. SPP Chen, G. et al.; Hashimoto, H. et al. Sun, Z.Q. et al. 3: 639 Sunagawa, I. 1: 186, 192 Sundar, R.S. et al. 3: 612, 654 Sundaram, V. 1: 618 Sundaram, V et al. 1: 618, 619 Sundaram, V.S. and Robertson, W.D. 3: 218 Sundarmi, V.S. et al. 3: 218 Sundaresan, R. see Suryanarayana, C. et al. Sundgren, J.-E. see Adibi, F. et al.; Greene, J.E. er al., Hultman, L. et al., Karlsson, L. et al.; Petrov, I. et al. Sundman, B. 1: 98, 99; see also Anderson, et al.; Andersson, J.-0. et al. Sung, C.M. et al. 2 191; 3: 154 Sum, 1. see Finetti, M. et al., Miienpaa, M. et al.; Pal, C.S. et al.; Zhu, M.F. et al. Suni, 1. et al. 2: 618, 623, 625, 626 Sunko, J. see Hu, Y.Z. et al. Sup~~manian, P.R. see Massalski, et al. Suresh, S. see Aswath, P.B. et al.; Christman, T. et al., Ramamurthy, U. et al. Suri, S.K. see Bhalla, AS. et al. Suri, S.K. et al. 1: 181
Sumach, S. see Baro, M. et al., Yavari, A. et al. Surinch, S. see Baro, M.D. Suryanarayana, C. 1: 467, 734; 3: 724, 727, 749, 750, 751, 753, 755, 762; see also Frefer, A. et al.; Froes, F.G. and Suryaiiarayana, C.;Froes, F.H. et al., Li, W. et al., Mukhopadhyay, D.K. et al. Suryana~dyana,C. and Jones, H. 3: 755 Suryanarayana, G . and Koch, C.C. 3: 750, 751, 753, 756 Suryanarayana, C. and Norton, M.G. 3: 760 Suryanarayana, C. et al. 3: 753, 756, 758 Suryanarayana, S.V. see Mahan Rao, P.V. et al. Susini, P. see Magini, M. et al. Suski, W. 3: 99 Suski, W and Troc, R. 3: 804 Susman, S. ,see Brun, T.O. et al.; Price, D.L. et al. Susta, J. et aE. Camptsi, I.E. et al. Susz, C.P. 2 560, 571, 572 Sutcliffe, C.H. 1: 498 Sutherland, W. 1: 1023 SutliR, J.A. see Bewlay, B.P. and Sutliff J.A.; Bewlay, B.P. et al., Henshall, G.A. et al. Sutton, A.P. 1: 601; see also Wang, G.-J. et al. Sutton, M. see Koster, U. et al. Suwa, M. see Yasuda, K. et al. SuZUki, A. 2: 415; see ~ I S OKOmdtSU, K. et al. Suzuki, E. see Tachikawa, K. et al. Suzuki, F. see Kiyoshi, T. et al. Suzuh, H. 3: 352, 461 Suzuki, H. et al. 2 175, 176 Suzuki, K. 1: 1024; see also Edagawa, K. et al.; Kimura, K. et al.; Kobayashi, T. et al.; Nakamura, N. et al.: Nenibach, E. et al., Xu, Y et al. Suzuki, K. et al. 1: 512, 522, 525, 529; 2: 22; 3: 702 Suzuki, K.A. see Makmo, A. et al. Suzuki, N. see Rido, Y. er al. Suzuki, R. see Nishizawa, J. et al. Suzuki, R.O. et al. 1: 650 Suzuki, S . 1: 591, 593, 594; see also Koyama, K. et al.; Takasugi, T. et al. Suzuki, S. and Takeuchi, S. 3: 458 Suzuki, T. 1: 499, 530; 2: 214; see also Fraas, K. et al.; Hashimoto, K. et al., Hayashi, Tohru. et al., I-iirano, T. et al.; Kobayashi, K. et al., Matsumara, T. et al.; Mishima, Y. et al.; Miura, S . et al., Noguchi, 0. et al.; Ochiai, S. et al., Otsuka, K. et al., Takabatake, T. et al.; Takizawa, H. et al.; Tounsi, B. et al.; Wee, D.M. et al,, Yadagawa, M. et al.; Yodogawa, Y et a/., Yoshihara, M. et al. Suzuki, T. and Mutoh, 1. 3: 31 Suzuki, T. et al. 1: 498, 505, 527, 539, 914, 915; 2 18, 23, 24; 3: 361 Suzuki, Y. see Matsuoka, T. et al. S U Z L IY. ~ ~and , Niihara, K. 3: 486 Svane, A. 1: 138 Svanne, A. 1: 130
945 Svechnikov, V.N. et al. 1: 388 Svedberg, R. 1: 994 Svensson, E.C. et al. 1: 160 Svezhova, S.I. see Bulycheva, Z.N. et al. Svilan, V et al. 3: 674, 675 Svitashev, ELK. 2: 418, 419, 432 Swainson, I.P. see Wirris, M.J. et al. S w a b , R.A. 2 10 Swam, P.R. see Flower, H.M. et al. Swam, P.R. et al. 2 199, 205 Swanson, M.L. see Schulson, EM. et al. Swanson, M.L. et al. 1: 818 Swartz, P.S. see Bean, C.P. et aE. Swartzeiidruber, L.J. see McM~c~iael, R.D. et aE. Sweedler, A.R. 1: 803 Sweeney, W.T. see Oglesby, P.L. et al. Sweetman, D. see Conn, R.W. et al. Swendsen, R.H. see Widoni, M. et al. Swenson, C.A. me Cetas, T.C. et al. Swihart, J.C. see Nicholson, D.M.C. et al. Switendick, A.C. 2 478, 479; see also Sighs, M. et al. Sy, C.C. 2: 510 Sykes, C. 1: 941; 2: 203 Symes, E.N. see Fleischer, R.L. et al. Syoiio, Y, see Fujinaga, Y et al.; Yamasaki, T. et al. S~utkina,V.I. see Adrianovskii, B.P. et al.; Razuvayera, B.D. et al.; Shashkov, O.D. er al. Syutkina, V.I. and Yakovleva, E.S. 3: 413 Szabo, P. 2 318 Szagvari, A. see Kuhn, W.E. et at. Szasz, A and Kojnok, J. 3 142 Sze, S.M. see Crowell, C.R. and Sze, S.M. Sze, S.M. and Gummel, H.K. 3: 787 Szenas, G. 1: 821 Szofran, F.R. and Lehoczky, J. 3: 28 Szynka, D. .we Cox, G. et al. Szytula, A. 3: 179, 180; see also Duraj, M. et al. Szytula, A. and Eeciejewicz, J. 3: 117, 178, 807 Ta~atabaie,N. see Harbison, J.P. et al.; Palmstrom, C.J.et al.; Sands, T. et al. Tabatabaie, N. et al. 3: 787 Tabbernor, M.A. 1: 541; 2: 54, 58 Tabory, C . see Compaan, A. et al. Tachi, M. see Nishi, Y. er al. Tachikawa, K.Y. 2: 361; see also Marluewicz, W.D. ei al.; Takeuchi, T. et al., Tanaka, Y . et al. Tachikawa, K.Y. et al. 2: 353, 360, 361, 367 Tacita, K. see Tetsui, T. er al. Tada, N. see Kiyoshi, T. et al.; Yamaguchi, K. et al. Tadaki, T. see Nakata, Y. et al. Tadaki, T. et al. 1: 836, 840, 841; 3: 56 Tagaki, Y. see Harada, U. et al. Tag~izadeh,M.R. see Smith, S.D. et al. Taglauer, E. see Voges, 2). et al. Taguchi, K. see Nishida, K. et al. Taguchi, K. et al. 3: 653 Taguchi, 0. see Hashimoto, U. et al.
Taguchi, T. 2: 346 Taillard, R. 2: 277, 278 Taillard, R. and Pinneau, A. 3; 65 Taillard, R. et al. 2 277 Tairov, Y.M. 2: 327 Takabatake, T. see Kyogaku, M. et al. Takabbsttake, T. et al. 1: 217, 218 Takagahara, T. 2 424,425 Takagi, H. see Cava, R.J. et al.; Ikushirna, K. et al. Takahashi, A. see Tokai, Y. et al. Takahashi, H. see Kagayama, T. et al.; Matsumara, T, et al., Watako, U. et al. Takakashi, J. see Nobuki, M. et al. Takahashi, K. see Kamisada, Y . et al.; Ogawa, T. et al. Takahashi, S. see Chiba, A. et al., Koguchi, N. et al., Nishikiori, S. et al. Takahashi, S. er al. 2: 332 Takahashi, T. see Kido, G. et al.; Liu, Y et al. Takahashi, T. and Oikawa, H. 3: 320 Takahashi, T. et al. 1: 136; 2: 79 Takahashi, Y. 1: 682; see also Ando, T. et al. Takahura, T. see Miyamoto, Y. et al. Takai, M.et al. 3 671 Takamura, J. see Ikematsu, Y . et al., Narita, N. and Takamura, J. Takamura, M. see Showaki, K. et al. Takamura, S. 1: 785 Takano, Y. see Ishimoto, H. et al. Takao, T. see Tsuji, H. et al. Takasaki, A. and Furuya, Y. 3: 653 Takasugi, S. see Yoo, M.H. et al. Takasugi, T. 1: 498, 505, 528, 598, 599, 603, 604,897,898,907,908,924,927, 928, 970,913; 2: 27, 28, 29, 38, 39, 214, 215, 216; 3: 347; see also Hasegawa, S. et al.; ~ a s a h a s hN. ~, et al.; Sasaki, G. et al., Yasuda, N. et al.; Yoo, M.H. et al., Yoshida, M. and Takasugi, T.; Yoshida, M. et al. Takasugi, T. et al. 1: 91, 512, 544, 591, 592, 593,595, 596,655,656,896,897, 898, 899,900, 904, 925; 2: 27, 29, 33, 214, 216; 3: 49, 356, 366, 370 Takasuki, T. see Liu, Y. et al. Takayama, M. 1: 986, 989, 993, 995 Takayama, S. 1: 734 Takayama, T. see Rao, S.M. et al. Takayasu, M. see Steeves, M.M. et al. Takebe, H. see Fujita, S. et al. Takeda, S. see Tamaki, S. et al. Takeda, Y. see Kim, S.M. et al.; Murata, U. et al. Takehara, H. see Mura~ami,Y. et al.; Yamamoto, T. et al. Takei, A. and Ishida, A. 3: 577, 580 Takei, H. see Oota, A. et al. Takei, M. see Pintschovius, L. et al. Takei, W.J. 1: 442, 443 Takenaka, T. see Yamamoto, T. et al. Takeo, T. see Tsuji, H. et al. Takeshi, K. see Koyama, K. et al. Takeshita, N. see Kobajashi, T. et al. Takeshita, T. 3: 101, 102; see also Nakayania, R. and Takeshita, T. Takeuchr, A. see Inoue, A. et al. Takeuchi, H. et al. 2: 343
Takeuchi, S. 1: 470, 485, 498, 528, 529, 534, 537, 547, 925; 2: 17, 19, 22, 24, 25; 3: 451; see also Edagawa, K. et al.; Hashimoto, T. et al.; Kimura, IS. et al.; Nernbach, E. et al.; Shibuya, T. et al.; Suzuki, K. et al.; Suzuki, S. and Takeuchi, S. Takeuchi, S. and Kuramoto, E. 3 450 T. et al.; Tachikawa, K.Y. et al. Takeuchi, T. et a/. 2: 360 Takeuchi, Y see Hashi~oto,T. et al. Takeya, H. et al. 3: 104, 519 Takeyarna, M. 1: 927; see also Cahn, R.W et al. Takeyama, M. et al. 2: 87, 623; 3: 648 Takezawa, K. 1: 836, 831, 838, 839, 841 Takigawa, M. see Kasaya, M. et al. Takigawa, Y. ,we Kurosawa, K. et al. Takiztawa, H. et al. 3: 107 Takizawa, S. see Mohri, T. et al. Takizawa, S. et al. 1: 67 Takizawa, T. see Kiyoshi, T, et al. Talalayeva, E.V. see Nikitin, S.A. et al. Talanov, V.M. and Frolova, L.A. 3: 812, 824 Talboom, F.T. et al. 2: 492 Taluts, G.G. see Greenberg, B.A. et al. Talvacchio, J. see Sinharoy, S. et al. Tarnaki, S. 1: 667; see also Matsunaga, S. et al. Tamaki, S. et al. 3: 26 Tamamura, T. see Nakao, M. et al. Tammanii, G. 1: 6, 7, 8, 10, 12, 15, 16, 115, 759; 2: 604, 628 Tammann, G. and Jaiider, W. 3: 27 Tamnann, G. and Koliman, K. 3: 30 T a ~ a n G. ~ and , Schafmeister, P. 3: 33 Tamura, M. see Zhu, S.M. et aE. Tarnura, N. see Yang, W.G. et al. Tan, T.Y. see IIrIsia, S.L. et al. Tan, T.Y. et al. 2 327 Tan, Y.N. see Han, Y.F. et al. Tanabe, H. see Suzuki, H. et al. Taiiabe, T. 1: 733 Tanaka, H. 2 379 Tanaka, H. et al. 2 336 Tanaka, 1. see Miyamoto, Y. et al. Tanaka, K. 1: 957, 958, 960; see also Minonishi, Y . et ul.; Nishiiio, M. et al.; Nonaka, K. et al. Tanaka, M. see Hashimoto, Y. et al., Terauchi, M. et al. Tanaka, M. et al. 3: 783, 784, 785 Tanaka, R. see Yoshihara, M. et al. Tanaka, T, see Ikeda, M. et al.; Ishizawa, Y. et al.; Nishikiori, S. et al., Otani, S. et LIJ. Tanaka, Y. see Ogawa, T. et al.; Tachikawa, K. et al.; Tsuji, H. et al.; Yamaguchi, A. et al. Tanaka, Y. et al. 1: 957 Tandon, J.L. see Kattelus, H.P. et at., So, F.C.T. et al.; Zhu, M.F. et al. Tandy, P.C. 1: 635 Tang, G. see Valeeva, A.A. et al. Tang, H.C. see Bennett, L.H. et al. Tang, J. see Matsumara, T. et al. Tang, J. et al. 3: 162
Author Index
946 Tang, N. see Han, X.F
et
al., Liu, J.P
et al.
Tang, N.et al. 3: 99 Tang, W. et al. 3: 100 Tang, W.H. see Rao, G.H. et al. Tangri, K. 2: 75 Tani, T. see Hisatsunc, K. et al. Tani, T. et al. 2: 566 Taniguchi, H. see Ohara, T. et al. Taniguchi, H. et al. Ogawa, T. et al. Taniguchs, S. 1: 927,983, 986, 989; 3: 577, 580, 581 Taniguchi, S. and Shibata, T. 3: 578 Taniguchi, S. et al. 1: 983, 986, 987, 988, 989, 995; 3: 578, 581 Tanihata, I(.see Miynmoto, Y. et al. Tanirnura, S. see Higashi, K. et al. Tanino, M. see Hanamura, T, et al.; Ikematsu, Y. et al. Tanner, L. and Okamoto, H. 3: 802 Tanner, L.E. 1: 509, 510, 511, 735, 746; see also Askenazy, P. et al.; Gronsky, R. et al.; Krishnan, K.M. et al.; Shapiro, S.M. et al. Tanner, L.E. et al. 2: 54 Tanovic, B. see Green, A.J. et al. Tao, L. see Chen, N.-Y. et al. Tao, M. see Park, D.G. et aE. Taoka, T. see Takeuchi, S. et NI. Taplin, D.M.R. see Ashby, M.F. et al. Taranenko, I.A. see Papirov, 1.1. et al. Taranenko, V.T. see Makyta, M. et al. Tardif, H.P. see Bouchard, M. et al. Tardy, F.J. see Gas, P et al. Tarfa, T. see Dirnitrov, C. et al. Tarnacki, J. 2 159, 163, 170 Tarnoczi, T. et al. 1: 743 Tarnopol, L. 2: 564, 565 Tarschisch, L. 2: 310 Tassin, N.see Bronoel, G. et al. Tatami, K. see Fujii, H. et al. Tatarkina, A.L. see Tsurikov, V.F. et al. Tatebayashi, T. et al. 2: 392 Tatishvilli, D.G. see Aleksandrov, B.N. et al. Tatro, R.E. see Baldi, R.W. et al. Tatsuo, T. .we Ha~iada,S. et al. Taub, A.I. 1: 591, 593, 594, 898, 899, 900, 902,904,908; 2: 27,214,245,254; see also Briant, C.L. et al.: Chang, K.M. et al.; Fleischer, R.L. and Taub, A.L.; Huang, S.C. et al.; Liu, C.T. et al. Taub, A.I. and Spaepen, F 3 696 Taub, A.I. et al. 1: 899, 904, 907, 908; 2: 27, 29, 30, 32, 38, 39 Taubkin, 1.1. see Khryapov, V.T. et al. Taunier, P. see Fruchart, R. et al. Taunt, R.J. 2: 19, 22, 23, 270, 271 Tavendal, A.J. see Eberhardt, J.E. et al. Tawancy, W.M. 3 418 Tawara, Y. see Fidler, J. et al.; Ohashi, K. et al. Tax, R.B. see Van der Ladn, M.T.G. et al. Taxi], P. and Mahenc, J. 3 671 Taylor, A. 1: 565, 566, 568; 2: 18, 54, 55, 58, 199, 200; see also Bradley, A.J. and Taylor, A. Taylor, C.E. see Dell’Orco, D. et al. Taylor, D.F. 2 565; see also Leinfelder, K.F. et al.
Taylor, G.I. 2: 37; 3 361, 370 Taylor, J.B. 1: 252 Taylor, J.R. see Gosnes, A.S.L. et al. Taylor, M.A. 1: 477 Taylor, R. see Basinsky, Z.S. et al.; David, W.I.F. et al. Taylor, R.E. see Touloukian, Y.S. et al. Taylor, R.L. see Goela, J.S. et al. Taylor, R.W. see Kosonocky, W.F. et al. Taylor, S, see Hill, P.J. et al. Taylor, S.S. and Biggs, T. 3: 75 Taylor, T.A. see Tucker, R.C. et d . Taylor, T.A. et al. 2: 492, 493 Taylor, T.N. I: 987, 993; see also Hanrahan, R.J. Jr. et al. Taylor, W.H. 1: 477; 2: 177 Teaturn, E. et al. 1: 109,249,252,366,382 Tebbe, K.F. see von Schnering, H.G. et al. Tedenac, J.C. see Chapon, L. et al. Teed, K. see Zribi, A. et al. Tegze, M. see Stephens, P W . et al. Teh, H. 1: 777, 778, 779 Teillet, J. see Steyaert, S. et al. Tejada, J. see Yu, R.H. ef al. Tellcr, A. see Metropolis, N. et al. Teller, E. see Metropolis, N. et al. Ternkin, H. et al. 2: 421, 425 Temmerman, W.M, see Nicholson, D.M. et al., Pei, S. et al.; Stocks, G.M. et al. ten Kate, H.H.J. see Van der Laan, M.T.G. et al.; Van Oort, J.M. et al. ten Kate, H.H.J. et al. 2: 377 Tepesch, P.D. et al. 3: 196, 205 Terakura, K. 1: 70; 3: 142; see a00 Mohn, T. et al., Takizawa, S. et al. Terakura, K. et al. 1: 41, 67, 68, 197; 3: 205 Teraoka, Y. 1: 610, 613, 614, 615, 619; 3: 22 1 Terasaki, T. see Nishizawa, J. et al. Terauchi, H. see Kamigaki, K. et al. Temuchi, H. et al. 3 275 Terauchi, M. et al. 3: 148 ter Avest, D. see ten Kate, H.H.J. et al. Terda, M. see Tachikawa, K.Y. et al. Terekhov, G.I. see Vederiiikov, M.V. et al. Terekhova, V.F sec Kripyakevich, P.I. et al. Terepka, F see Brady, M.P. et aE. Terkelsen, B.E. see Piearcey, B.J. and Terkelseii B.E. Terlinde, G. 2: 108 Ternes, J.K. et al. 3: 356 Ternes, K. see Farkas, D. and Ternes, K., Farkas, D. et al. Terr, D.G. 2: 493 TersofK J. see Finstad, T.G. et al. Teshima, F. see Kyogaku, M. et al.; Takabatake, T. et al. Teslyuk, M.J. 1: 410 Tessereau, A. see Claeyssen, F. et al. Testa, A. see Baroni, S. et al. Testardi, L.R. et al. 1: 875, 890; 2: 408; 3: 78 Teter, J.P. see Clark, A.E. et a2.; Moffett, M.B. et al. Tetsui, T. 3: 598, 600 Tetsui, T. et al. 3: 635 Teukolsky, S.A. see Press, W.H. et al. Teuscher, H. see Kramer, P et al. Tewari, S.N. 2: 296
Teytel, U. see Kuranov, A. et al. Thadhani, N.N. I: 912; 3: 737, 740, 743; see also Ferreira, A. et al. Thadhani, N.N. see Chen, T. et al.; Counihan, P,J. et al., Dunbar, E. et al.. Glade, S.C. and Thadhani N.N.. Grakam, R.A and Thadhaiii N.N., Grebe, H.A. and Thadhani N.N.; Lee, J.H. and Thadhani N.N., Namjoshi, S.N. and Thadhani N.N.; Vandersall, IS.and Thadhani N.N.; Vreeland, T. Jr. et al.; Xue, H. et al. Thadham, N.N. and Aizawa, T. 3: 723, 741, 742, 743 Thadhani, N.N. et al. 3: 744 Thakker, A. see Larsen, S.E. et al. Thakker, A.B. see Postans, P.J. et al. Thalrneier, P. 1: 159 Thanailakis, A. 2: 344 Thangaraj, N. see Chattopadhyay, K. et al. Tharp, A.G. see Smith, G.S. et al. Theron, C.C. see Li, J. et al.; Pretorius, R. et al. Thessicu, C. et u2. 3: 162 Thevenin, J. see Epelboin, 1. et al. Thibault, N.W. 1: 183 Thiedemann, D. see Schuster, H.U. et al. Thiel, F.A. see Hong, M. et al.; Kortan, A.R. et al. Thiel, P.A. see Shen, Z. et al. Thiel, R.C. see Smit, H.H.A. et al. Thiele, J.-U. see Ronning, C. et al. Thiele, W. see Singheiser, L. et al. Thiessen, K.P 3: 743, 744 Thirukkonda, M. see Weiss, 1. et al. Thirumalai, D. 1: 679, 680 Thoelke, J.B. 2: 400; see aBa Galloway, N. et al. Thorn, A.J. see Akinc, M. et al. Thoma, D.J. see Chu, F. et al.; Hanrahan, R.J. Jr., Perepezko, J.H. et at. Thoma, R.A. 2: 380 Thomas, B.J. 3: 858; see also Taillard, R. et al. Thomas, B.W.J. see Baines, G.S. et al. Thornas, C.R. see Dalal, R.P. et al. Thomas, D.K. see Wilson, C.G. et al. Thomas, F. et al. 3: 162 Thomas, G. see Berkowitz, A.E. et al., Radmilovic, V. et al. Thomas, J.P. see Baudin, Thomas, L.E. see Charlot, L.A. et al. Thomas, L.K. see Schlemper, K. and Thomas, L.K. Thomas, M. see Douin, J. et al.; Naka, S. et al. Thomas, M. et al. 1: 535, 543, 544, 545; 2: 103 Thomas, M.C. see Baiii, K.R. et al.; Broomfield, R.W. et al. Thomas, M.P. see Palmer, I.G. et al. Thornas, 0. see Charai, A. et al.; Guenin, G. et al.; Harper, J.M.E. et al.; Hbrnstroni, S E . et al. Thomas, 0. et al. 1: 386; 2: 611, 622 Thomas, R.E. see Doyle, B.L. et al. Thomas, R.E. et al. 2: 626 Thomas, R.J. see Denner, S.G. et al. Thomas, R.N. see Hobgood, H. et aE. Thomas, W.J.O. 1: 101, 110,245,246,252
Author Index Thomasson, J. see Thomas, F. et al. Thomasson, Y. et al. 3: 162 Thome, L. see Audouard, A. et al.; Garrido, F. et al. Thome, R. 2: 375 Thome, R.J. et al. 2 380 Thompson, A.W. 1: 925, 928, 993, 994, 997, 1000; 2: 112, 114, 115, 282; see also Cho, W et al.; Chu, W.Y et al., Dimiduk, D.M. et al., Ward, C.H. et al. Thompson, C. see Wright, C.R.A. et al. Thompson, C.V see Ma, E. et al.; Miura, H. et al. Thompson, E.R. see Breinan, E.M. et al. Thompson, J.C. ez al. 1: 665 Thompson, J,D. 1: 221; see also Canfield, P.C. et al., Fisk, Z. et al.; Hundley, M.F. et al.; Lacerda, A. et al.; List, R.S. et al.; Movshovich, R. et al.; Severing, A. et al.; Shen, T.D. et al., Watako, Y. et al. T h o ~ p s o nJ.R. , 2 610; see Sales, B.C. et al. Thompson, L.J. see Meng, W.J. et al. Thompson, M. de K. 2 652 Thompson, M.O. see Poate, J.M. et al. Thompson, R. see Rice, J.R. and Thompson, R. Thompson, R.D. see Eizenberg, M. et al.; Ottaviani, G. et al. Thompson, R.D. et al. 2 622 Thornpson, W.A. see Guntherodt, G. et al. Thompson, W.T. see Pelton, A.D. and Thompson W.T. Thomson, C.V. 2: 654 Thomson, J.R. et al. 3: 182 Thomson, R. 2 166 Thornton, D.E. 1: 663, 665 Thornton, J.A. 3: 665, 666 Thornton, P.H. 2 213 Thornton, P.H. et al. 1: 548, 882, 914, 915; 2 17, 30 Thorvaldsson, T. see Liu, P. et al. Threadgill, P.L. 1: 656; 2: 128 Tliiimmler, F. see Jesse, A. et al. Thursfield, G. 2: 178 Tiainen, T. see Lagerboni, J. et al.; Schwarz, R.B. et al. Tian, F. see Bergstrom, D.B. et al. Tian, W.H. et al. 1: 860 Tibballs, J.E. see March, N.H. et al. Tibballs, J.E. et al. 2: 186 Tibeno, R.C. see Chao, P.C. et al. Tice, W.K. see Bretiim, E.M. et al. Tichelaar, D.D. 1: 587 Tichelaar, F.D. 1: 598, 600; see also An~onopou~ous, J.G. et al.; Buis, A, er al.; Pestman, B.J. et al. Tichelaar. F.D. et al. 1: 496, 511, 590 Tichonova, E.A. 1: 763 Tichy, G. see Heredia, F.E. et al. Tichy, G. et al. 1: 512, 525, 550; 2: 23, 24, 164, 165 Ticknor, L. 1: 97 Tiearney, T.C. Jr. and Natesan, K. 3: 707 Tien, J.K. 2 11, 12; see also Becker, J.D. et d.; Sanchez, J.M. et al., Wu, Y.P. et al. Tien, J.K. et al. I: 928; 2: 282; 3: 49
Tierney, B.J. et al. 3: 671 Tietze, H. et al. 1: 164 Tilford, C.R. see Cetas, T.C. et al. Tillard-Charbonnel, M. see Belin, C. and Tillard-~harbonnel~ M. Tillard-Charbonnel, M. et al. 3: 123 Tiller, W.A. see Bever, M.B. et al. Tilley, R.J.D. 3: 232 Tilly, G.3 298 Timbie, J.P 1: 682 Timme, R.W. 2: 402 Timmons, C.F et al. 2: 584 Ting, C.Y. 2 231, 605, 623, 624 see also Krusin-Elbaum, L. et al. Tishiii, A.M. 3: 527; see also Dan’kov, S,Yu. et al.; Nikitin, S.A. and Tishin, A.M.; Nikitin, S.A. et al.; Pecharsky, V.K. et al. Tishkevich, V.M. see Larikov, L.N. et al. Tisoner, T.C. see Sun, R.C. et al. Tissink, H.C. see de Reus, R. el al. Tissen, V.G. et al. 3: 161 Titchener, A.L. see Bever, M.B. et al. Titran, R.H. 1: 987,988,991,995; see also Vedula, K, et al. Titran, R.M. et al. 2: 206 Tiwari, P et al. 3: 235 Tiwari, R. ef al. 1: 642 Tixador, P et al. 2 382 Tobisch, J. see Poschmann, I. et al. Todd, A.G. see Kelly, M.J. et al.; Wickenden, D.K. er al. Todd, A.G. et al. 2 626 Todo, S. see Hashimoto, T. et al. Todorov, E. and Sevov, S.C. 3: 115, 116, 122, 126, 127 Toennies, J.P see Harten, U. et al. Togaiio, K. see Takeuchi, T. et al. Togawa, N. see Sagawa, M. et al. Tokai, Y et al. 3: 527 Tokarev, V.N. et al. 1: 717 Tokci, Zs. et al. 3: 290 Tokizane, M. see hieyanam, K. et al. Tokizane, M. et al. 1: 925 Tokuhara, H. see Sagawa, M. et al. Tokumitu, K. see Aizawa, T. and Tokumitu K. Tokushima, T. et al. 2 329 Tolochko, M.I. see Bulycheva, Z.N. et al. Toloui, B. et al. 1: 743, 745 Tom, C.M. see Lebensohn, RSA.and Tom, C.N. Tomanek, D. 1: 80 Tomaschko, Ch. et al. 3: 267 Tomasett, L.R. 2 328 Tomashik, V.N. and Grytsiv, V.I. 3: 27, 31, 33 Tomashik, V.N. and Grytsiv, V.I. 3 803 Tomaszewicz, P. 1: 995 Toniehawa, S. SPE Wada, H. et al. Tominez, E. et al. 3: 103 Tornioka, H. see Inouye, A. et al. To~iyoshi,S. 1: 445 Tomizuka, C.T. 1: 761, 765; see alsa Kuper, A.B. et al.; Kupper, A.B. et al. Tomka, G.J. see Lord, J.S. et al. Tomkiewicz, M, et al. 2 511 Tomkowicz, Z. see Duraj, M. et al. Tomlinson, P. see Pecharsky, V.K. et al. Tomokiyo, A. see Hashimoto, T. et al.; Tokai, Y. et al.
947 Tomokiyo, A. et al. 3: 527 Tornokiyo, Y. see Eguchi, T. et al. Tomzig, E. ,see Sajovec, F. et al. Tonejc, A. 2: 178 Toner, J. see Horn, P.M. et al., Levine, D. et al.; Lubensky, T.C. et al. Tong, C.H. see Hsu. S.E. e f al. Tong, S. see Wuttig, M. et al. Tong, W. and Ravichandran, G. 3: 738 Tonkov, E.Yu 3: 154, 160, 161, 162 Tonneau, A. et al. 3: 339. 340 Topler, J. 2: 485 Topor, L. I: 64 Topper, T.H. and Y q M.T. 3: 341 Torikachvili, M.S. see Meissner, G.W. et al.; Seaman, C.L. et al. Torndahl, L.-E. see Fridberg, J. et al. Tornquist, W. see Prieskorn, J.N. et al. Torok, E. see Zogg, H. et al. Torrecillas, R. ,see Liang, K.M. et al. Torres, M. 1: 491 Torrini, M. see Atrei, A. et al. Torrisi, A. see Duchateau, J.P.W.B. et al. Tortorelli, P. 3: 339 Tortorelli, P.F. see McKarney, C.G. et al.; Natesan, K. and Tortorelli, P.F. Tortorelli. P.F. and Be Van, J.H. 3: 573 Tortorelli, P.F and Natesan, K. 3: 503, 613 Tortorelli, P.F. et al. 1: 1007, 1008 Toruckner, W.G. see Angers, L.M. et al. Tosatti, E. see Ercolessi, F. et al. Toshio, T. see Matsuno, S. et al. Tosi, M.P. 1: 764 Tostmann, NI. see DiMasi, E. et al. Toth, L.E. 1: 297, 299, 300, 303, 580 Toth, R.S. 1: 50; 2 568 Toth, R.S. et al. 1: 727 Totolici, J.E. see Manaila, R. et al. Toulemonde, M. see Defour, C. et al., Dufour, C , et al.; Provost, 9. et al.; Trautmann, C. et al.; Wang, Z.G. et al. Toulemonde, M. et al. 1: 821; 3: 269, 270 Touloukian, Y.S. 1: 1019, 1020, 1022 Touloukian, Y.S. et al. 1: 1023, 1024, 1025, 1026 Tounsi, B. 1: 498, 505, 528, 551 Tounsi, B. et al. 1: 545, 546, 548; 2: 214 Tourand, G. see Funnel-~el~ise~i~, M.C. et al. Tova, J. see Zribi, A. et al. Towle, N. see Lang, @.I.et al. Towle, N.R. 3: 74 Towle, N.R. et al. 3: 74 Towner, J.M. 1: 836, 838, 841 Townsend, H.E. see Zoccola, J.C. et al. Toyada, N. see Uchitomi, N. et al. Toyota, N.see Pintschovius, L. et al. Toyyota, N. see Iwasaki, H. et al. Trajkovic, D. see Blau, I). et al. Tralmer, J.M. see Hesson, J.C. et al. Trambly de Laissard~ere,G. et al. 3: 143, 146 Tran, J. see Nicholsoii, D.M.C. et al. Tran, L. see Koiawa, E. et al. Trankle, G. see Brunner, IS. et al. Trattner, D. see Pfeiler, W et al. Trautmann, C. et al. 3: 268 Travaglini, G. 1: 217; see also Siegrist, T. et al.
Author Index
948 Traverse, A. et al. 3: 151 Travina, N.T. 2 271 Treat, D.W. see Epler, J.E. et al. Trebin, H.R. see Mikulla, R. et al. Trkglia, G. see Landesman, J.P et al.; Meunier, I. et al.; Turchi, P. et al. TrCglia, G. et al. 1: 28, 29 TrCheux, D, 1: 757; see also Labarge, J.-J. et al. Tretyakov Yu, D. see Dumas, J. et al. Triaiitafillou, J. see Beddoes, J. et al.; Chen, W R. et al. Triboulet, R. 2: 326 Triboulet, et al. 3: 806 Trickey, S.B. see Blaha, P et al. Tricb, L. see Schrank, J. et al.; Veith, G. et al. Triotskii, B.S. see Zakharova, A.M. et al. Tritt, T.M. et al. 3: 106 Trivedi, R. 1: 839, 841 Troev, T. see Wurschuni, R. et al, TroRer, M.B. see Park, R.M. et al. Troian, S.M. 1: 480 Tromp, R.M. 3: 213 Trottier, T.P 1: 997; 2: 133, 144 Trunin, R.F see Altshuler, L.B. et al. Tsai, A.P. see Inoue, A. et al., Kortan, A.R. et al., Masumoto, Y. et al.; Matsubara, E. et al., Yerauchi, M. et al. Tsai, A.P. et al. 1: 457, 484 Tsai, M.Y. see d’Heurle, F.M. et al. Tsang, T.S. and Cho, A.Y. 3: 783 Tsang, T.S. and Ilegems, M. 3: 783
Ottaviani, G . et al. Tsaur, B.Y et al. 1: 705; 2: 609 Tschinkel, J.G. see Gianiei, A.F. and Tschinkel J.G. Tshaikovski, E.F. 1: 766 Tshornaia, L.F see Larikov, L.N. et al. Tsien, L.-H. see Compaan, A. et al. Tsiok, O.B. see Brazhkin, V V. et al. Tsiovkin, Yu.N. see Kourov, N.I. et al. Tsipas, D.N. 1: 983, 986 Tso, N.C. see Kosugi, M. et al. Tso, N.C. et al. 1: 41 TsokoY, A.O. see ~ a r u § i nE.P. , et al. Tsu, R. see Esaki, L. and Tsu, R. Tsucktya, K. see Calderon, H.A. et al. Tsuchiya, M, see Oota, A. et al. Tsuchiya, Y see Tarnaki, S. et al. Tsuda, H . see Mabuchi, H. et al. Tsuci, C.C. 1: 703, 704, 748; see also Agye~~ K.net~ al. Tsuge, A, 1: 651 Tsui, P see Gaydosh, D.J. et al. Tsuji, H. see Ando, T. ef al. Tsuji, €3. et al. ;Z: 374 : 86; see also Hashimoto, K. et al.; Kasahara, I<. et al., Nobuki, M. et al.; Takeyarna, M. et al.
Tsukamoto, 0. 2: 382 Tsunashima, S. see Di, G.Q. et al. Tsurikov, V.F. et al. 1: 719; 3: 60 Tsurisaki, K. see Takasugi, T. et aZ. Tsuruoka, H. see Taniguchi, S, et al. Tsutsumi, S . et al. 2: 584 Tsuyama, S. see Mitao, S. et al. Tsuyama, S. et al. 2: 82 Tsvetkov, V.F. 2: 327 Tu, K.N. 1: 696; 2: 605, 614, 619; see also Eizenberg, M. et al.; Mayer, J.W. et al.; Olowolafe, J.O. et al.; Ottaviani, G. et al.; Poate, J.M. et al., Thompson, R.D. et al. Tu, K.N. et al. 1: 696 Tucker, R.C. see Taylor, T.A. et al. Tucker, R.C. et al. 2: 492, 493 Tucker, R.P. see Yang, W.J.S. er aE, Tun, Z. see Ryan, D.H. et al. Tunca, N. 1: 646 Tung, J. see Meschter, P.J. et al. Tung, R.T. see Hensel, J.C. et al. Tung, R.T. et al. 2: 231; 3 808 Tung, T. see Vanyukov, A.V. et al. Tuominen, S.M. 2: 282 Turchi, P.E.A. 1: 23,24, 34,36,37,38,40, 41; 3: 185; see also Gonis, A. et al.; Johnson, D.D. et al.; Landesman, J.P. et al.; Mikalopas, .T. et a[.; Sluiter, M.H.F. et al., Stocks, G.M. et al. Turchi, P.E.A. et al. 1: 34, 35, 36, 37, 38, 41, 44, 45, 46, 50, 56; 3: 205 Turco, G. 1: 186 Turilli, G. see Ibarra, M.R. et al. Turkalo, A.M. 2 54, 64 Turmezey, T. see Stefantay, V et al. Turnbull, D. 1: 465, 479, 480, 679, 688, 690, 692, 733, 739; 3: 683; see also Drehman, A.J. et al.; Kui, H.-W et al.; Seybolt, A.U. et al. Turnbull, D. and Hogman, R. 3: 320 Turner, C.D. et al. 2: 159, 161, 162, 166, I67 Turner, R.B. et al. 2: 134 Tustison, R.W 1: 440 Tutov, V.I. see Semenenko, V Ye. ei al. Tuttle, 3.R. see Albm, D.S. et al., Mooney, G.D. et al. Tuttle, J.R. et al. 2: 423 Tuval, E. see Singh, J.P. et al. Tylecote, R.F. 1: 3 Tyren, C. 2: 402, 405 Tyrkiel, E. 1: 15, 97 Tyurin, A L . see Annaorazov, M.P. et al., Nikilin, S.A. et al. Tzeng, W.S. see Polvani, R.S. et al. Ubbelohde, A.R. 3: 245 Uchida, H. see Nishino, K. et al. Uchida, S. see Cava, R.J. et al. Uchika~a,F. see M a t s ~ n oS. , et al, Uchikawa, Y. see Tsuji, H. et al. Uchitomi, N. et al. 2: 621 Uchiyama, S. see Di, G.Q. et al. Udoh, K. see Tani, T. et al. Udoh, K.-I. see Hisatsune, I(.et al. Udoh, K.-I. et al. 2: 566 Ueda, A. see Ohara, T. er al. Uehara, I. see Sakai, T. et al. Uehara, M. see Maeda, H. et al.
Ueki, T. see Golberg. D. et al.; Mori, M. et al.; Suzuki, R.O. et af. Uemori, R. see Hanarnura, T. et al. Uemori, R. et al. 2: 78 Uemura, K. 2 329 Uenishi, K. see Shingu, P.H. et al. Ueno, N. see Isbida, K. et al. Ueno, Y. see Otsuka, K. et al. Ueta, S . see Sumonji, K. et al. Wfintsev, V.B. see Zenlskov, V S . et al. Uher, C. see Chen, B. et al. Uhl, D. see Woenig, H.E. et aE. Uhlenbrock, St. see Galakov, V.R. et al. Uhlenhut, W, see Lebensohn, R.A. et al. Uhlfert, W. see Worvath, 3. et al. Uhlig, H.H. 2: 217 Ukai, T. see Tatebayashi, T. et al. Ulvensoea, J.H. et al. 3: 503 Umakoshi, Y. 1: 496, 498, 504, 506, 507, 510, 521, 536, 653, 913, 914; 2: 18, 105, 148, 149, 154, 202, 240; 3: 328, 414; see also Hagihara, K. er Hisano, T. et al.; Nekano, T. Yaniaguclii, M. and Umakoshi, Y., Yamaguchi, M. et al.; Yasuda, H. Y. et al. Umakoshi, Y. and Nakashima, T. 3: 418 Umakoshi, Y. and Yamaguchi, M. 3: 4S0, 460 Umakoshi, Y. et al. 1: 504, 507, 537, 539, 992,993,994; 2: 24,26,155,167,219, 220, 221,222, 223, 240; 3: 326, 327, 414, 456, 459, 621 U m e ~ a k J. ~ ,see Hiller, M.W. et al. Umemoto, M. see Calderon, H.A. et al. Unal, 0. et al. 1: 539 Underhill, E.M. 2: 307, 308, 309 Underwild, F.C. see Hensel, J.G. et al. Unertl, W. 3: 222 Unertl, W. see Campuzano, J.C. et al. Unertl, W.N. see Rivers, S.B. et al. Unnani, J. see Brewer, W.D. and Unnam, J. Unni, C.K. see Cordon, D.E. et al. Uno, H. see Ameyanam, K. et at. Uniiih, ICM. ,see Atzmon, M. et al. Ur, U.C. see Nash, P. et al. Urabe, K. see Nishida, T. and Urabe, K. Uray, L. see Gaal, 1. et al. Urba, J. 2: 383 Urban, B. see Eckert, A. et al. Urban, K. 1: 785, 786, 800, 802, 803; see also Banerjee, S . et al.; Cox, G. et al.; Csanady, A. et al., Eckert, 3 . et al., Feuerbacher, M. et al.; Grushko, B. et al., Holland-Moritz, D.M. et al., Jia, C.L. et al., Ranganathan, S. et al.; Rosenfeld, R. et al., Wang, R.W. et al.; Wollgarten, M. et al.; Wurschum, R. et al.; Yang, W.G. et al.; Zhang, Z. et al. Urban, K. et al. 1: 461, 463, 537; 3: 384, 385, 397, 398,400,401 Urban, R. 1: 567; 573 Urban-Erbil, B. 1: 777, 778 Urbanic, V.F. see Causey, A.R. et al. Uribe, F see Felter, T. et al. Urlin*V.D. S L J Altshuler, ~ L.B. et al. Ury, I. see Lee, C.P. et al. Uryanarayana, C. et al. 3: 73.9.
US Bureau of Mines 1: 635 US National Defense Research Council 2 390 Ushida, K. see Hao, S.M. et al. Usuda, M. see Osaka, T. et al. Uthurusamy, R. see Fayyad, U.M. et al. Utkin, A.V. see Kanel, G.L. et al. Uusimaa, P. see Saarinen, K. et al. Uy, L.H. see Meyers, M.A. et al. Uyeda, K. see Ogawa, T. et al. Uyeda, T. see Tadaki, T. et al. Uz, M. 1: 999 Uzuka, T. et al. 2: 565 Vahldiek, F.W. 1: 1007; 2: 124 Vaidya, W. see Wagner, R. et al. Vailht, C. see Caillard, D. et al.; Farkas, D. and Vailhe, C.; Farkas, D. et al. Vailhe, C. and Farkas, D. 3: 452, 766 Vaillant, F. see Osterwalder, J. et al. Vainshte~n,B.K. 3: 444 Vakshoori, D. 2: 431; see also Hong. M. et al. Valeeva, A.A. et al. 3: 275 Valencia, J.J. see Heredia, F.E. et al.; Hyman, M.E. ei al.; McCullough, C. et al.; Rosler, J. et al. Valencia, J.J. et al. 1: 640; 3: 311, 317, 318, 320, 648 Valetta, R.M. 1; 406 Valiev, R.Z. see McFadden, S.X. et al. Valiev, R.Z. et al. 1: 656, 925; 3: 754 Vallier, J.C. see Ohl, M. et al. Van Aken, D.C. see Gibala, R. et al., Mason, D.C. and Van Aken D.C. van Bruggen, C.F. see Otto, M.J. et al. Van Buren, H.G. 1: 764 Laan, M.T.G. et al. Van de Waal, B.W. 1: 207 van den Berg, J. see Palstra, T.M.M. et al. van den Berg, R. see von Lohneysen, H, et al. van den Beukel, A. 1: 743 van den Beukel, A. see de Vries, J. et ul.; Mulder, A.L. et al. van den Broek, J.J. see ~ i n e ~ u rT. a, et al. van der Aart, S.A. et atl. 3: 129 van der Broek, J.J. see van der Kolk, G,J. et al. van der Worst, F. see Xu, R. et al. van der Lingen, E. see Cretu, C. and van der Ltngen, E. van der Lugt, W. 3: 113, 246; see also Reijers, H.T.J. et al.; van der Aart, S.A. et al.; Verkerk, P. et al.; Xu, R. et al. Vandermeer, R.A. see Fonda, R.W. et al. Vander~a~l, I(.see Xue, H. et al. also Calka, A. et al.; Giessen, B.C. et al. Va~~erschaeve, G. 1: 531,543,845; 2: 150 Vanders~haeve,G. and Escaig, B. 3: 414, 446, 447,464 Vanderschaeve, G. and S Vanderschaeve, G, et al.
Vandervoort, R. see Campuzano, J.C. et al. Vandervoort, R.R. et al. 1: 918, 923; 2: 61, 63, 64 van der Groat, A.S. 2: 310 van der Groot, A.S. and Buschow, K.H.J. 2: 394 Van der Heide, R,G. 2: 201 van der Kolk, G.J. 2 610, 612; see also de Reus, R. et al., Duchateau, J.P.W.B. et al.; Minemura, T. et al. van der Kolk, G.J. et al. 2: 610, 613, 617, 618, 619, 628 Van der Laan, M.T.G. et al. 2: 379 Van der Linde, P.R. see Mondieig, D. et al. van der Lugt, W. see Reijers, H.T.J. et al., van der Marel, C. et al. van der Marel, C. et al. 1: 671 Van Der Meij, L. see Bakker, H. et al. van der Spiegel, J. see Setton, M. et al. van der Steeg, M.C. see Koch, A.J.J. et al. van der Valk, P.J. see Otto, M.J. et al. Van Der Wegen, G.J.L. et al. 3: 444 van der Woude, F. see Schurer, P.J. et al. van der Zwaag, S. see Mulder, A.L. et al. van Dijk, C. 1: 161; see also Reijers, H.T.J. et al. van Dover, R.B. see Cava, R.J. et al. Van Duzer, T. 2 383 Van Dyke, J. . ~ Campisi, e I.E. et al. Van Dyke, J.P 1: 1022 van Emden, V. see ten Kate, H.H.J. et al. van Engelen, P.P.J. 2: 448 van Engelen, P.P.J. et al. 2: 448 van Engen, P.G. 2: 308; see also Buschow, K.H.J. et al. van Engen, P.G. et al. 2: 448, 449 van Eysden, J. see Mahler, D.B. et al. Van Feggelen, M. see Van Ommen, A. et al.; Van Ommen, A.H. et al. Van Gend, M. see Stolwijk, N.A. et al. Vangenlielm, S.D. see Arkharov, V.S. et al.; Panko, T.S. et al. Van Gent, N. see Stolwijk, N.A. et al. van Gurp, G.J. 1: 956; 2: 605 Vanherzeele, H. 2: 414 van Wove, G. see de Potter, M. et al. Van Hove, M.A. 3: 211, 212 Van Hove, M.A. see Chen, W, et al.; MacLaren, J.M. et al.; Shen, Z. et al. Van Keuren, R.M. see Markiewicz, W.D. et al. Vanko. D'A. 1: 627, 628 van Konynenberg, R.A. see Guinan, M.W. et al. van Landuyt, J. see Tan], T. et al.; Udoh, K.4. et al. Van Leuken, E. see Sarma, D.D. et al. Van Lierop, J. see Ryan, D.H. et al. Van Loo, F.J.J. and Rieck, C.D. 3: 570 Van Loo, F.J.J. et al. 1: 727 Van Ommen, A.H. 1: 572, 577, 775,777; see also Bakker, H. et al. Van Ommen, A.H. and de Miraiida, J. 3: 277 Van Ornmen, A.H. ei al. 1: 561, 563, 565, 566, 573, 575, 577,177, 778 Van Oort, J.M. et al. 2: 377 van Rossum, M. see de Potter, M. et al.; Johnson, W.L. et al.
Van Sceiver, S.W see Miller, J.R. et al. van Schi~fgaarde,M. see Asta, M.D. et al.; Gall, K. et al.; Pasturel, A. et al, Van Tendeloo, G. see Leroux, C. e f al.; Tan], T. et al., Tanner, L.E. et al.; Udoh, K.-I. et al. Van Tendeloo, G. et al. 1: 477 Van Vechten, J.A. 1: 357, 358, 419 van Veen, A. see van der Kolk, G.J. et al. Van Vliet, J.P.M. see Young, E.W.A. et al. vanVucht, J.H.N. see Daams, J.L.C. et al. van Vucht, R.J.M. 2: 646 Van Winkel, A. 1: 777; see also Bakker, H. et al. van wont erg he^, J. 1: 733 Varekha, L.A. ,WE Rezukhina, T.N. et al. Varga, P. see Gauthier, Y. et al.; Hebenstreit, E.L.D. et al.; Hebenstreit, W. et al., Schmid, M. et al. Varga, P. and Scliniid, M. 3: 213 Vargas, J.R. see Goebel, J.A. et al. Varin, R.A. 2: 157, 159, 163, 164, 165, 166; see also Virk, I.S. et al. Variii, R.A. and Zbrotiiec, L. 3: 495 Varley, J.H.O. 3: 465 Varma, A. 3: 725 Varma, A. see Lebrat, J.-P et al., Mukasyan, et al.; Rogachev, A S , et al. Varma, A. et al. 3: 725, 726, 734 Vaima, C.M. 1: 156, 158, 164 Varnn, R.D. Jr. see ~ h a ~ a r m aW.N. n, Pt d. Varsanyi, M.L. see Leidheiser, H. et al. Varuhin, V.A. see ~ ~ m ~ d c hC.A. e v , et al. Vashchenko, N.D. see Kiselyova, N.N. et al. Vashukov, LA. see Kutolin, S.A. et al. Vasiliev, M. et al. 1: 618 Vasina, S.Ya. see Petrii, O.A. et al. Vassel, A. see Thomas, M. et al. Vassiliou, M.S. et al. 1: 649 Vasudtvan, A.K. 1: 391, 997, 998; see also Petrovic, J.J. and Vas~idevan,A.K., Petrovic, J.J. et al., Sadananda, K. et al. Vasudevan, A.R. and Petrovrc, J.J. 3: 574, 583, 657 Vasudevan, V.K. 1: 175; see also Court, S.A. et al.; Kumar, M. and Vasudevan, V.K.; Srmm, S. et al.; Stucke, M.A. et al.; Wang, P. et al.; Wheeler, R. e f al. Vasudevan, V.K. et al. 1: 531, 532, 533, 534, 543; 2: 78, 19, 82, 150, 151, 153, 159; 3: 414, 424, 446 Vaughan, D.J. 1: 635 Vaughn, D.A. see Uhlig, H.H. et al. Veal, B. see Campuzano, J.C. et ak. Veal, B.W. 1: 1020; see also Lst, R.S. et al.; Natesan, K. et al., Olson, C.G. et al.; Renusch, D. et al. Vecchio, K.S. see Kad, B.K. et al., Meyers, M.A. et al. Vecsey, G. see Blau, B. et al. Veda, A. See Fuchiiio, S. et al.
950 Vedam, K. see Li, J. et al. Vedel, I. see Burger, J.P. et al. Vedcrnikov, M.V 3: 106 Vcdernikov, M.V. et al. 2: 469; 3: 78 Vednne, P see Tixador, P. et al. Vedula, K. 1: 905, 924, 926; 2: 58, 59, 60, 61,64, 199,201,204,206,208; 3: 451. 496, 654, 655; Crimp, M.A. et al.; Nahn, K.H. and Vedula, K.; Khadkikar, P.S. et al., Rigncy, J.D. et al.; Titran, R.H. et al. Vedula, K. et al. 1: 904, 925; 2: 58 Vedula, R.M. see Crimp, M.A. et al. Vedyayev, A.V 1: 944 Veer, F.A. et al. 1: 768 Vehoff, H. 1: 918, 919; 2: 64, 65, 85, 224; see also Gken, M. et al.; Kempf, M. et al.; Weber, T. et al. Veith, G. et al. 1: 782 Vejisis, V see Nash, P.G. et al. Velge, W.A.J.J. and Buschow, K.H.J. 3 I00 Velickji, B. et al. 1: 25 Vellasamy, R. 2: 177 Venables, J.A. 3: 422 Venetsky, S. 3: 31 Venezia, A.M. see Haasch, R.T. et al. Venkateswara Rao, K.T. see Badnnarayanan, K. et al., Bloyer, D.R. et al.; Campbell, J.P et al.; McKelvey, A.L. et al.; Murugesh, L. et al., Soboyejo, W.0. et al. Venkateswara Rao, K.T. P t al. 3: 334, 336, 344, 347 Venkatraman, M, see Garg, S.P. et al.; Raghavan, V. et al. Venkat~iraman,S . 3: 338 Venke, G. see Ryadchenko, V.M. et al. VennGgues, P et al. 1: 777, 782 Ventkatesan, S . see Fetcenko, M.A. et al. Venturini, 6. see Francois, M. et al. Ventunni, G. et al. 3: 91 Verbeck, B.H. et al. 1: 138 Vereshchagin, Yu.A. see Senchenko, A.A. et al.; Shcherbakov, A.S. et al. Vergand, F. see Fargues, D. et al. Vergara, J. see Madurga, V. et al. Vergasova, L.L. see Zakharova, A.M. et al. Verger-Gaugry, J.L. see Brechet, Y . et al. Vcrhaeghe, M.F see Ghavmeau, J.P. et al. Verhoef, R. see Sinnema, S. et al.; Wolfers, P et ad. Verhoeven, J.D. see McMasters, 0.D. et al. Verhoeven, J.D. et al. 2: 395, 400 Verhoeven, V.W.J. see van der Aart, S.A. et al. Verink, E.d. Jr. see Brady, M.P. et al. Verkerk, P see van der Aart, S.A. et al.; Xu, R. et al. Vermaak, J.S. see Mays, C.W. et al. Vermilyea, M.E. 2: 380 Verrall, R.A. 1: 925; see also Arzt, E. et al. VerSnyder, F.L. see Piearcey, B.J. and VerSnyder F.L. VerSnyder, F.L. and Guard, R.W. 3: 542 Vert, R. et a!. 3: 99 VCrtes, A. see Lesdheiser, H. et al.
Author Index Vertogradskii, V.A. see Kovalev, A.I. et al. Vettcr, J. see Audouard, A. et al. Vetterling, W.T. see Press, W.H. et al. Vettier, C. see Blanco, J.A. et al.; Kwo, J. et al. Veyssi$re, P 1: 496, 503, 521, 522, 523, 527, 528, 636,540, 543,544,545,547, 548, 549, 551. 915; 2: 58, 80, 278; see also Beauchamp, P. et al., Caron, P. et al., Dirras, G. et al., Douin, J. and Veyssikre, P.; Douin, J. et al., Frangois, A. et al.; Hug, G. and VeyssiGre, P.; Hug, G . et al.; Khan, T. et al.; Shi, X. et al.; Thomas, M. et al.; Tounsi, B. et al. Veyssikre, P. and Douin, J. 3: 361, 621 VeyssiGre, P. and Noebe, R. 3: 440, 453 Veyssikre, P. et al. 1: 498, 505, 529, 540, 546; 2: 19, 20, 22, 271 Vianco, P.T. see Ray, U. et al. Viau, G. see Stroink, G. et al. Victoria, M. see Caro, A. et al.; Diaz de la Rubia, T. et al. Vidoz, A.E. 1: 512, 514 Vidoz, A.E. and Brown, L.M. 3: 367 Vidoz, A.E. et al. 1: 897, 923; 3: 364 Viefiaus, H. see Grabke, H.J. et al. Viefhaus, N. et al. 3: 64 Vieland, L.J. 1: 676 Viens, D.V. see Baker, I, et al.; Schulson, E.M. et al., Weihs, T.P et al. Vigneron, F. see Bellissent, R. et al. Vignoul, G.E. see Zen, J.K. et al. Viguier, B. see Kruml, T. et al.; Hemker, K.J. et al. Viguier. B. and Hemker, K.J. 3: 367 Viguier, B. et al. 3: 368 Vijayaraghavan, R. see Mazumdar, C. et al.; Nagarajan, R.et al. Vijh, A.K. 2: 502, 503, 504, 507, 51 1, 646 Vijh, A.K. et al. 1: 969; 2 503, 505, 506, 508, 509, 511 Viljoen, M. 3: 80 Villain, J. see Rossat-M~gnod,J. et al. Villars, P. see Wafner, J. et al. Villars, P 1: 7, 120, 138, 227, 228, 230, 234,236,237,241,242,248,252,255, 256,258,259,260,261,264,265,268, 269,272,277,278,280,281,289,294, 298, 303,305,306, 320, 328,357,358, 363, 364, 365,367, 370,377, 378,381, 382, 385,386,388, 390,391,393,394, 396, 397, 398, 403n, 419, 424,425, 431,433,434,437,625,632,876,878, 881, 885, 887; 2: 147; 3 26,87,88,94, 100, 104, 106, 797, 812; see also Chen, N.-Y. et al., Daams. J.L.C. et al.; Rabe, K.M. et al. Villars, P and Calvert, L.D. 3: 66, 238, 241, 784, 785, 797 Villars, P, et al. 1: 36, 227, 228, 234, 244, 249,250,261,265,277,358,382,423, 427,428,433, 434, 485; 3: 801 Villas-Boas, V. see Landgraf, F.J.G. et al., Moreau, J.M. et al. Viltange, M. 1: 787 Vimercati, G. see Lupinc, V. et al. Vinai, F. see Caciuffo, R. et al. Vinarcik, E.J. 1: 1009; see also Nesbitt, J.A. et al.
Vincent, G. see Rosencher, E. et al. Vincenzini, P. 3: 806 Vincze, I.J. see Balogh. J. et al. Vines, R. 2: 564 Vineyard, C. 1: 775, 785 Vining, C. 2: 453 Vinogradov, S.I. 1: 413; 2: 321 Vintaikin, E.Z. 2: 272 Virdis, P 2: 156 Virk, IS. 2: 157 Virk, IS. et al. 2: 157, 163 Visani, P. see Lopez de la Torre, A. et al. Vishnubatla, S.S. see Jan, J.P. and Vishnubatla, S.S. Visnbvsky, S . 2 438 Visscher, W.M. siv Migliori, A. et al. Viswanadha~,R.K. 2: 61; see also Kumar, K.S. et al. Viswanathan, G.B. see Sriram, S. et al.; Wang, P et al. Viswanathan, R. see Moodenbamgh, A.R. et al. Viswanathan, S. see Maziasz, P.J. et al.; Sikka, V.K. et al. Viswanathan, S . et al. 3 604, 607, 610, 611, 613 Vitek, V. 1: 88, 523, 541, 544, 547, 548, 598, 600, 601, 915; 2: 19, 20, 28; 3: 437, 438, 439, 451; see also Ackland, G.J. and Vitek, V.; Cockayne, D.J.H. and Vitek, V., Cserti, J. et al.; Duesbery, M.S. and Vitek, V.; Ezz, S S . et al., Girshick, A. and Vitek, V.; Heredia, F.E. et al.; Inuj, H. et al.; Ito, K. and Vitek, V.; Khantlia, M, et al.; Kruisman, J.J. et aE.; Kulp, D.T. et al., Mahapatra, R. et al.; Paidar, P eb al., Paidar, V. et al.; Pestman, 13.5. et al., Schroll, R. et al., Siegl, R. et al.; Sob, M. et al.; Tichy, G. et al.; Umakoshi, Y. et al.; Wang, G.-J. et al.; Wee, D.M. et al.; Wu, Z.L. et al.; Yamaguchi, M. et al.; Ydn. M. et al.; Znam, S. et al. Vitek, V et al. 1: 82, 88, 523, 598, 600, 915; 2: 28; 3 73, 437, 440, 693, 769, 772 Vitrishchak, 1.B. see Mak, A.A. et al. Vitta, S . et al. 1: 746 Vivchar, 0.1. et al. 2: 178 Vladimrov, V.I. see Romanov, A.E. and Vladimirov, V.I. Vlasova, E.N. et al. 3: 404 Vogel, H. 1: 690 Vogel, M. see Blau, B. et al. Vogel, R. 2 310; 3: 155 Voges, D. et al. 3: 225 Vogl, G. see Eggersmann, M. et al.; Klaumunzer, S. et al. Vogl, G. and Sepiol, B. 3: 282 Vogl, P 1: 56; see also Hafner, J. et al. Vogt, 0. see Rossat-Mignod, J. et al.; Stirling, W.G. et al. Vogtenhuber, D. and Podloucky, R. 3: 24 1 Vohra, Y.K. see Weir, S.T. et al. Vohra, Y.K. et al. 1: 184 Voigt, w 1: 202 Voit, H. see Tomaschko, Ch. et al. Vojak, B.A. et al. 2: 335 Vol, A.E. 1: 726
Author Index Volchenkova, R.A. 1: 232 Voleti, S.R. see Markiewicz, W.D. et al. Voline, K.J. see Reijers, H.T.J. et al. Volk, R. see Miiller, W. and Volk, K. Volkert, C.A. 1: 743, 744 : 476, 484; see also Potzel, U. et al. Volkov, A.Yu. see Greenberg, B.A. et al. Volmer, D.C. see Semiatin, S.L et al. Volmer. M. 1: 167 Volynskaya, N.V. see Alisova, S.P et al. vom Felde, A. et al. 1: 319 von Allnien, M. 2: 610; see also Samwer, K. et al. Von Bardeleben, H.J. see Bourgoin, J.C. et al. von Barth, U. 1: 132 von Hei~eiidahl,M. 1: 747 Von Kanel, H. 2 231 Von Keitz, A. see Engell, H.-J. et al. Von Kertz, A. et al. 3: 454 Vonken, E.-J. see Weegels, L.M. et al. von Liihneysen, H. et al. 1: 480 von Mises, R. 1: 921; 2: 243 von Nesda, A.R. see Jordan, A.S. et al. von Philipsborn, H. see Wen, C.P. et al. von Ranke, P.J. et al. 3 519, 527, 530 von Schnering, H.G. 1: 343n; 3: 128; see also Somer, M. et al. von Schnenng, H.G. and Honk, W. 3: 128 von Schnenng, H.G. et al. 3: 115, IS6 von Seefeld, H. see Cheung, N. et al. von Seefeld, H. et al. 2: 624 Von Stebut, J. see Dubois, J.M. et al. von Stetten, E.C. see Singh, D. et al. von Vegesack, A. 1: 9 von Welsbach, C.A. 2: 647 Voorhees. P.W. 1: 792 Voorrips, A.C. see de Reus, R. et al. Vorberger, J. Jr. 2: 584 Vorob’yova, V.P. see Kiselyova, N.N. et al. Voroshilov, Yu.V. see Motrya, S.F. et al. Voroshilov, Yu.V et al. 2: 309 Vosko, S.H. 1: 138; see also Perdew, J.P. et al. Vosko, S.H. et al. 1: 132 Voter, A.F. 1: 82, 84, 85; 3: 196; see also Chen, S. et al., Chen, S.P. et al., Clapp, P.C. et al.; Farkas, D, et al., Vitek, V. et al. Voter, A.F and Chen, S.P. 3: 769 Voter, A.F et al. 1: 82 Vouglit, J.D. SPP Sikka, V.K. et al. Vovchenko, G.D. see Semenova, A.D. et al. Voyer, J. see Boily, S. et al. Vozdvizhenskii, V.M. 3: 812 Vozdvizhenskii, V.M. and Falevich, V. 3: 812, 824 Vredenberg, A.M. see de Reus, R. et al.; Pretorrus, R. et al. Vredensky, D.D. see MacLaren, J.M. et al. Vreeland, T. Jr. 3: 738, 742; see also Schwarz, R.B. et al. Vreeland, T. Jr. et al. Vrolijk, J.W.G.A. see Van Los, F.J.J. et al. Vul, D. and de Fontasne, D. 3: 193
Vulf, B.K. 2 254 Vvedenski, D.D. 1: 604, 605 Waber, J.T. see Teatum, E. et al. Waber, J.T. et al. 1: 246, 253 Wachtel, E. 2: 309; see also Berner, D. et al.; Willman, N. et al. Wachtel, E. et al. 1: 565 Wachter, P. 1: 217; see also Allen, J.W. et al. Wachtman, J.B. Jr. 1: 875 Wada, H. see Iiioue, K. et al.; Kiyoshs, T. et al. Wada, H. et al. 3: 525 Wada, 0. see Yamakoshi, S. et al. Wada, S. see Hasegawa, S. et al. Wada, T. 2: 335 Wade, K. 3: 113, 114 Wadley, H.N.G. 2: 98 Wadsworth, J. 1: 656, 925; see also Chou, T.C. et al.; Nieh, T.G. and Wadsworth, J.; Nieli, T.G. et al.; Pharr, G.M. et al.; Wolfenstine, J. et al. Waegeinaekers, A.A.H.J. see Bakker, H, et al.; Van Ornmen, A.H. et al. Wagemann, B, see Grabke, H.J. et al. Waghniare, U.V. et al. 3: 439, 455, 456, 457,458 Wagman, D. see Hultgren, R. et al. Wagner, C. 1: 97, 768, 980; 2: 263, 502, 51 S Wagner, C.N.J. 1: 663, 739, 792; see also Cheng, J. et al. Wagner, F.E. see Weitzer, F. et al. Wagner, G.A. and Van den Haute, P. 3 263 Wagner, H.-G. see Jink, J. et al. Wagner, R. 3: 250 Wagner, R. 2: 10; see also Appel, F. and Wagner, R.; Appel, F et al.; Gerling, R. et al.; Rim, Y.-W. et al.; Oehring, M. et al.; Paul, J.D.H. et al. Wagner, S. 1: 93; see also Coutts, T.J. et al. Wahl, C. see Singheiser, L. et al. Wahl, J.B. see Broomfield, R.W et al. Wakabayashi, H. see Tomokiyo, A, et al. Wakabayash~,N. 1: 777; see also Pintschovius, L. et al. Wakayama, S. see Hashimoto, M. et al. Wakayama, S. et crl. 1: 604 W a k e ~ e ~G. d , see Spedding, F.H. et al. Wakita, K. see Tanaka, H. et al. Wakita, lS.et al. 2: 342 Wakoh, K. see Xu, Y. et al. Wakugawa, J.M. et al. 2: 409 Wald, F.V. 2: 346 Wald, M.S. see Brandon, D.G. et al. Waldman, J. 1: 991, 992, 993; see also Cook, J. et al.; Lee, E.W. et al. Walford, L.K. 2: 177, 178 Walker, A.C. see Bigot, J.Y. et al.; Smsth, S.D. et al. Walker, D.G. 1: 815 Walker, E. see Sinharoy, S. et al. Walker, J.G. 2: 390 Walker, L.R. see Jaccanno, V et al.; Liu, C.T. et al. Walker, M.J. see Howson, M.A. et al. Walker, M.S. see Motowidlo, L.R. et al.
95 1 Walker, M.S. et al. 2: 366 Walker, N. see Larsen, S.E. et al. Walkcr, R.M. see Fleischer, R.L. et al. Wall, M.E. see Tanner, L.E. et al. Wallace, D.C. 1: 1023 Wallace, T.A. see Wiedemann, K.E. et al. Wallace, T A . er al. 1: 987, 993 Wailace, W. see McRae, E. et al. Wallace, W.E. 1: 88, 109; see also Cheng, S.F. et al.; Craig, R.S. et al.; Huang, M.Q. et al., McRae, E.G. et al.; Zhang, L.Y. et al. Wallace. W.E. and Ackland, G.J. 3: 221, 226 Wallace, W.E. et al. 2: 482 Wallach, E.R. see Strangwood, M. et al. Wallbaum, H. 1: 112 Wallen, P see Hedenqvist, P et al. Wallow, F. et al. 1: 205, 206 Wallwork, G.R. 1: 995 Walmer, M.H. see Chen, C.H. et al. Walmer, M.S. see Chen, C.H. et al. Walser, B. see Wittenaue~,J. et al. Walser, R.M. 2: 606, 608. 609 Walsh, D.S. see Castro, R.G. et al. Walston, W.S. 2: 57, 58; see also Darolia, R. et al., Noebe, R.D. and Wa~ston, W.S.; Yu, KO. et al. Waiston, W.S. and Darolia, R. 3 494, 503, 604 Walston, W.S. et al. 3: 66, 494 Walter, J.L. 1: 690; 2: 296; see also Jackson, M.R et al. Walter, J.M. 1: 681 Walter, K.C. see He, X.-M. et al. Walters, G. see Jaimson, lS. et al. Walther, M. see Brunner, K. et aE. Walton, D.R.M. see David, W.1.F et al. Walzer, W. 1: 243 Wan, X.J. et al. 1: 927 Wang, C.S. 1: 197; see also Singh, D. et al. Wang, C.S. et al. 1: 130 Wang, E. 1: 252 Wang, F. see Liu, 2. et al. Wang, F.W. see Cheng, Z.H. et al. Wang, G. see Cockeram, B.V. et al.; Mueller, A. et al. Wang, G.-J. et al. 1: 601 Wang, G.X. and Dahms, M. 3: 653 Wang, H. 1: 203, 204; see also Gao, C . et al. Wang, H.Y see Najafabadi, R. et al. Wang, H.Y. et al. 1: 611 Wang, I. 2: 624 Wang, 3. see Chen, G. et al.; Hit, J. et al. Wang, J.G. see Zhang, L.C. et ul. Wang, J.L. see Wan, X.F. et al.; Tang, N. et al., Yang, C.P et al.; Yang, F.M. et al. Wang, J.N. et al. 3: 652 Wang, J.Y. see Chu, J.P. et al.; Shen, E.G. et al. Wang, K.L. see Ern5Y.K. et al.; Sadwick, L,P et al. Wang, L. 2: 290; see also Jin, R. et ul. Wang, L.B. see Lu, L. et al. Wang, L.G. see Paidar, V et al., Sob, M. et al. Wang, P. et al. 1: 844 Wasig, P.C. and Yang, J.-M. 3: 337 Wang, P.C. et al. 3: 328, 337
952 Wang, R. 2: 610; see also Yang, W.G. et al.; Yu, S. et al. Wang, R.H. 3: 381,382,384,398,400; see also Ding, D.H. et al.; Feng, J.L. et al.; Qin, Y.L. et al.; Wang, Z.G. ef al.; Wen, J.Q. et al.; Yang, W.G. et al.; Yao, D.Z. et al. Wang, R.H. and Dai, M.X. 3: 282, 384, 399, 400 Wang, R.H. et al. 3: 383, 384, 398, 400 Wang, S.Q. 2: 618, 625, 626; see also Hung, L.-S. et al. Wang, T.M. see Doyama, M. et al. Wang, T.M. et al. 1: 564, 570, 574 Wang, T.S. see Rossat-Mignod, J. et al. Wang, W. see He, C. et al. Wang, W.K. see Iwasakl, H. et al. Wang, W.L. see Wu, W.T. et al. Wang, X. see Song, H.Y. aiid Wang X. Wang, X.-M. see Zhang, D,-L. et al. Wang~U. 1: 132: see also Chen, G. et al., Chen, L. et al., Chen, L.-Q. et al.; Nicholson, D.M.C. et al. Wang, Y. et al. 2: 315 Wang, Y.G. see Han, X.F. et al. Wang, Y.-P. see Zhang, D.-L. et al. Wang, Y.Z. 2: 316; see also Yang, C.P. et al.; Yang, F.M. et al. Wang, Y.Z. et al. 3: 99 Wang, Z. see Park, D.G. et al. Wang, Z,G. et al. 3: 270, 383 Wangts, C.H. et al. 3: 672 Wanner, A. see Clemens, H. et al. Ward, C.H. 2: 93, 112, 118; see also Dirniduk, D.M. et al.; Kestner-Weykamp, H.T. et al.;
Ward-Close, C.M. 2: 298 Wardle, S. et al. 1: 534; 3: 411, 413 Wardle, S.T. 1: 540 Ware, R.M. 2: 329 Warekois, E.P. 1: 172, 184 Warckois, E.P. et al. 1: 172, 180 W a r p , M.E. see Busby, P.E. et al. Warin, D, see Epelboin, I. et al. Warlimont, W. I: 3, 792, 837, 839; 2: 55, 56, 57; see also Leainy, H.J. et al.; Zogg, H. et al. W a r ~ i ~ o nH. t , and Delaey, L. 3: 55, 56 Warlii~ont~Mei~r, B. et al. 1: 960 Warren, B, 3 54 Warren, B.E. 1: 461 Warren, D. see Libsch, J.F et al. Warren, M, see Jin, R. et al. Warren, W.W. 1: 684 Warren, W.W Jr. et al. 1: 479 Warrington, D.H. see Grimmer, H. et al. Waseda, V 1: 663, 820; 2: 502; see also Matsubara, E. et al.; Tamaki, S. et al. Wasielewskr, G.E, see Shih, D.5, et al. Wasilewski, R.J. 1: 512, 536, 5 875; 2: 54, 56, 58, 61 Wasilewski, R.J. et al. 1: 565, 904; 2 58, 59, 61; 3: 276, 280 Waszczdk, J.V see Hauser, J.J. et al. Watako, Y et al. 3: 162
Author Index Watanabe, D. 1: 49; 2: 564; see also Iwasaki, H. et al.; Ogawa, S . et al.; Yaniaguchi, S. et al. Watanabe, H. 1: 445; 2 327; see Ogawa, S. et al. Watanabe, K. see Mohri, T. et al., Terakura, K . et al. Watanabe, M. see Hiraga, K. et al. Watanabe, N, see Ikeda, M. et al.; Miura, Y. and Watanabe, N. Watanabe, S. see Chiba, A. et al.; Hanada, S. et al., Hosoda, H. et al.; Takasugi, T. et al. Watanabe, S. et al. 2 218, 219 Watanabe, T. 1: 733, 957, 958, 960 Watanabe, T. et al. I: 955 Watanabe, Y. see Sluiter, M.H.F. et al. Waterstrat, R.M. 2: 580, 589; 3: 78 Waterstrat, R.M. and Okabe, T. 3: 31 Waterstrat, R.M. et al. 2: 251 Watson, J.D. 1: 835 Watson, R.D. see Castro, R.G. et al. Watson, R.E. 1: 241, 242, 243, 420; see also Bennett, L.H. et al.; Davenport, J.W. et al.; Fernando, G.W. et al.; Goodman, D.A. er al. Watson, R.E. et al. 1: 62, 63, 64, 67, 68 Watson-Yang, T.J. see Hong, T. et al. Watts, B.R. 1: 959, 960 Waugh, J.L.T. 3: 841; see also Bergman, G. et al. Wawnes, F.E. 3: 584 Wawrousek, H. et al. 3: 810 Wayman, C.M. 1: 716, 830, 831, 833, 834, 836, 837, 838; 2: 54, 55; see also Duerig, T. et al., Hamada, Y. et al.; Inoue, H.R. et al., Inoue, H.R.P. et al.; Saburi, T. et al., Satija, S.K. et ai., Wu, M.H. et al. Wayman, C.M. and Harrison, J.D. 3: 54, 55 Weaire, D, see Heine, C. et al. Weaire, D.L. 1: 31, 33 Weast, R.C. et al. 1: 109 Weatherby, M.H. see Tucker, R.C. et al. Weatherly, G.C. see Mukhopadhyay, N.M. et al. Weathers, M.S. see Bird, J.M. et al. Weaver, F.J. see Williams, R.K. et al. Webb, D.J. see Chan, J.Y. et al. Webb, D.S. see Viswanathan, S. et al. Webb, G. see De Bussac, A. et al. Webb, G.W. see Pintschovius, L. et al. Weber, A. and Klages, C.-P. 3: 669 Weber, D. et al. 1: 574 Weber, E.T. 2: 648 Weber, H.P. see SalathC, R.P. et al. Weber, H.W. see Kirk, M.A. et al. Weber, J. see Roberts, W.L. et al. Weber, J.H. 1: 912 Weber, T. et al. 3: 225 Weber, W. 1: 67, 70, 152, 153, 156, 158, 162, 163, 164; see also Harmon, B,N. et al.; Keil, M. et al.; Pintschovius, L. et al. Webster, G.A, et al. 1: 917 Wechsler, M.S. 1: 574 Wechsler, M.S. et al. 1: 828 Wecker, J. see Katter, M. et al.; Schnitzke, E;. et al.; Schultz, L. et al. Wecker, J. et al. 2: 316
Weddington, V.L. see Dimiduk, D.M. et al. Wee, D.M. 1: 499; see also Suzuki, T. et al.; Yadagawa, M, et al.; Yodogawa, Y. et al. Wee, D.M. et al. 1: 549; Weeber, A.W. 1: 700; 2: LoefT, P.I. et at, Weegels, L.M. et al. 2: 413 Weekes, M E , 3 21 Weertman, J. see Cohen, J.B. and Weertman, J. Weertman, J, and ~ e e r t m a nJ.R. , 3: 300, 423 Weertman. J.R. see Angers, L. et al., Calderon, H.A. et al.; Weestina~,J. and Weertman, J.R. ~ e e r t s J., 1; 913 Weger, M. et al. 1: 953 Wehner, G. see Reppich, B. et al. Wehner, G,K. 3: 665 Wehrmann, R. 2: 294 Wei, J. see Sheng, L. et al. Wei, J.C.S. see Kolawa, E. et al. Wei, R.P. see Cao, M. et ul. Wei, S.-H. 1: 138; see also Lu, Z.W et al., Zunger, A. et al. Wei, S,-H. et al. 1: 41, 67, 68, 138 Weibel, E. see Binnig, 6. et al. Weiberg, W,H. see Van Have, M.A. et al. Weihs, T.P. see Schulson, E M . et al. Weihs, T.P. et al. 1: 543, 902; 2: 139, 143 Weijers, H.W. see Van Oort, J.M. et al. Wcil, R. 1: 875 Wein, W. et al. I: 759 Weinberger, D.A. see Ojima, M. et al. Weinber~es~ D,A. et al. Weinert, M. ~ e Davenport, e J.W. et al.; Fernando, G.W. et al.; Watson, R.E. et al.; Wei, S.-H, et al.; W i ~ m e sE. , et al. Weinstein, M. see Wolff, G.A. et al. Weinstock, H. 2: 383 Weir, S.T. see Vohra, Y,K. et al. Weir, S.T. et al. 1: 184 Weisberg, L. 1: 778, 782 Weiss, A. see Pady, H, et al. Weiss, H. 2: 654 Weiss, I. see Singh, J,P. et al. Weiss, I. et al. 3: 545 Weiss, J. see Semiatin, S.L. et al. We&, J.A. see Gelatt, C.D. Jr. et al. Weiss, L. 2: 572, 573 Weiss, P. I: 439 Weissenbesg, 1: 309 Weissman, S. 2: 282; see also Hirabayashi, H. and Weissman, S. Weissmann, S. 2: 564 Werzer, V.G. 1: 874; 2: 623 Weitzer, F see Leithe-Jasper, A. et al. Weitzer, F. et al. 3: 100 Welch, D.O et al. 1: 569, 579 Welch, L.5. see Young, E.W.A. et al. Welker, H. 1: 16 Weller, D. 2; 450 Weller, M. see Schaefer, H.-E, et al. Wells, A.F. 1: 400, 422, 428 Wells, C. see Busby, P.E. et al. Wells, C.H. see Paslay, P.R. et al. Wells, G . see Boily, S. et al. Wells, P. 1: 781
A ~ ~ h Index or Welpnian, K. et al. 2: 186 Welsch, G. 1: 991, 992, 993; 2: 124 Welsch, G. and Desai, P.D. 3: 809 Weissh, G. and Kaveci, A.I. 3: 575 Welsch, G. et al. 1: 992; 3: 575 W e l s h , B. 1: 215 Wen, C.J. 1: 95 Wen, C.J. et al. 1: 66, 70 Wen, C.P. et al. 2: 332 Wen, J.Q. et al. 3: 882 Wendelken, T.F. see Overbury, S. et al. Weiidhausei~,P.A.F. see Miiller, K.H. et al. Wendt, H. et al. 2: 507, 511 Wenger, G.M. ,we Ray, U. et al. Wenger, L.E. see Putatunda, S.K. et al. Wenger, S. et al. 2: 377 Wengert, P.R. 1: 100, 116; 2: 518 Wengert, S. et al. 3: 252 Wennekers, P see Richter, H.J. et al. Weiitorf, R.H. Jr. 1: 181; see also Bovenkerk, H.P. et al. Wenw~ng,€3. see Slieng, L. et al. Wenzl, H. see Sajovec, F. eB al. Weppner, W. 2: 502, 51 1 Werner, A. et al. 1: 184 Werner, A.G. 1: I86 Werner, F.E. 2: 317 Werner, R. 2: 486
Chin, G.Y. et al.; Jaccarino, V. et af.; Kunzler, J.E. et U/.; Nesbitt, E.A. et al.; Sherwood, R.C. et al., Testardi, L.R. et al. Wernick, J.H. et al. 1: 109; 2 309 Wert, J.A. 2: 157, 160, 162; see also Mysko, D.D. et a€.; Turner, C.D. ed al. Wert, J.J. 2: 599 Wertheim, C. see Jaccarino, V et al. Wesicken, R. see Nissen, H.U. et al. Wesley, D. see Shoemaker, J.R. et al. Wessel, S. see ten Kate, H.H.J. et al.; Van Oort, J.M. et al. Wessels, J.F et al. 3: 338, 347 West, A.W 1: 957 West, D.R.F. see Chakravorty, S. et al.; Flower, H.M. et al.; Khatee, A. et al. West, J. 1: 772, 773 West, J. et al. 1: 773 West, M.B. see Andrews, P V. et al. West, R.N. see Kaiser, J.H. et al., Lock, D.G, et al. Westbrook, J.H. 1: 3, 14n, 228, 385, 565, 575, 580, 586,595, 596, 597,639,657, 712,887,892,897,898,912,913,915, 918,921,923,925,927,935,956,957, 1000, 1003, 1006; 2: 7, 17, 18, 24, 30, 34, 35, 36,61,2Q2,237,254,408, 506, 521, 64%; 3: 22, 31, 54, 231, 275, 307, 352, 440, 501, 591, 810, 842; see also Aust, K.T. and Westbrook, J.H.; Guard, R.W. and Westbrook, J.H.; Hazrjl, W.C. and Westbrook, J.H., Jena, A. et al., Kear, B.H. et al.; Seybolt, A.U. and Westbrook, J.H.; Seybolt, A.U. et al.; Wawrousek, H. et al., Wood, D.L. and Westbrook, J.H.
Westbrook, J.H. and Fleischer, R.L. 3: 12, 13, 185, 231, 275, 617, 810 Westbrook, J.H. and Grattidge, W. 3: 818 Westbrook, J.H. and Wood D.L. 3: 227 Westbrook, J.H. et al. 1: 258, 261 Westengen, H. 2: 175, 176 Westengen, H. et al. 2: 186 Westermann, U. see Lugscheider, E. et al. Westerveld, J.P.A. 1: 574, 849 Westgren, A. 1: 10, 11, 13 Westgren, A.F. and Phragmen, G. 3: 114 Westlake, D.G. 1: 298; 2; 478; see cdro Kocks, U.F. and Westlake, D.G. Weston, E. 2: 652 Weston, W.F. see Ledbetter, H.M. et al. Westwood, W.D. 3: 665 Wever, H. 1: 580; see also Bose, A. et al.; Hahn, H. et al.; Hunecke, J. et al. Wever, H. et al. 1: 921 Wey, E. see Brooks, J.W. et al. Weyrich, K.H. 1: 133 Whalley, E. see Klug, D,D. et al.; Mishima, 0. et al. Whang, S.H. see Gao, Y.Q. and Whang, S.H. Whang, S.H. and Hahn, Y.D. 3: 621 Whang, S.H. et al. 1: 532; 2 17. 18; 3: 356, 366, 808 Wharton, W.R. see Shoemaker, J.R. et al. Wheatley, G. see Buck, T. et al. Wheatley, G.H. see Buck, T.M. et al. Wheeler, R. 1: 532; see also Boehler, C.J. et al.; Vasudevan, V.K. et al. Wheeler, R. et al. 1: 530, 531; 2: 154, 155 Whelan, M.J. see Cockdyne, D.J.H. et al.; Hirsch, P et al. Wherrett, B.S. see Smith, S.D. et al. Whetstone, J.R. see Mills, J.J. et al.; Mills, M.J. et al. White, A.E. 1: 650, 792; 2: 324, 330 White C., E.T. and Okamoto, H. 3 802 White, C.L. 1: 591,914,989; 2: 27,28, 31, 32, 33; see also Choudhury, A. et al.; George, E.P. et al.; Liu, C.T. et al., Santella, M.L. et a€. White, C.L. et al. 1: 86,591,593, 594. 620; 2: 30 White, E.W. see Bhalla, AS. et al. White, G.K. 1: 1024, 1026; see also Barron, T.H.K. et al.; Collocott, S.J. et al. White, G.K. et al. 1: 1025 White, H.J. 3: 813 White, J.C. 1: 172 White, J.F see Howard, J.K. et al. White, M.G. 2: 277, 278 White, R. see Price, D.L. et al. White, R.M. 1: 682, 936 Whitehead, B. see Saris, F.W. et al. Whiternan, J.A. 2: 277, 278, 279, 282 Whitfield, H.J. 1: 345 Whitmire, L.D. 1: 960 Whiting, P. see Bewlay, B.P. et al. Whitten, W.B. et al. 1: 875 Whittenberger, J.D. 1: 864, 925, 1024; 2: 57, 58,61,63,64,206; 3; 724; see also Aikm, B.J.M. et al.; Brown, S.A. et al.; DiPietro, M.S. et al.; Carg, A. et al.; Hebsur, M.G. et al., Kurnar, K.S. et al.; Locci, I.E. et al.; Lowell, C.E. et al.
953 Whittenberger, J.D. et al. 1: 925: 2: 62,63, 64, 170, 293, 294, 296; 3: 495 Whittle, D.P. 1: 912, 923, 980, 982; see also Hindam, H. and Whittle D.P., Irving, G.N. ef al.; Stafford, K.N. et al. Wichner, R.P 2: 642 Wicke, W. 1: 765 Wickenden, D.K. see Kelly, M.J. et al. Wickenden, D.K. et al. Wickramasekara, L. see Cadieu, F.J. et al. Wicks, G.W. see Erdogan, T. et al. Widom, M. 1: 484,491 Widorn, M. et al. 1: 482 Wieber, R.H. et al. 1: 768 Wieczorek, C. 3: 669 Wiederhorn, S.M. 1: 927; see also French, J.D. et al. Wieder~ann,K.E. see Wallace, T.A. et al. Wiedermann, K.E. et al. 2E 124 Wiedersich, H. I: 693 W i e ~ a n n W. , see Chemla, D.S. et al.; Dingle, R. et al.; ~ i l ~ eD.A.B. r, et al. Wiehs, T.P. see Schulson, E.M. et al. Wieiner, D. see Crabke, H,J. et al. Wier, B.E. see Hong, M. et al. Wiesmger, G . 2: 481; see also Weitzer, F. et al. Wiesler, D.G. see Youngquist, S.E. et at. Wiezorek, J.M.K. et al. 3: 421, 463, 623 Wigner, E. 1: 132 Wijn, H.P.J. 3: 803, 804 Wijngaard, J. see Otto, M.J. et al, Wik~und,P. see Nygren, S. et al.; Ostliiig, M. et al. Wilbrandt, P. 3: 410 Wilcox, R.E. see Markiewicz, W.D. et al. Wilcox, W.R. see Triboulet, et al. Wilde, G. see Sakidja, R. et al. Wilde, G. et al. 3 683 Wilder, D.R. see Wirkus, C.D. and Wilder, D.R. Wilder, M.P. see Espe, W. et al. Wildhagen, B. see Dahms, M. et' al. Wiley, J.D. see Doyle, B.L. et al.; Thomas, R.E. et al. Wiley, J.D. et d.2: 626 Wiley, 5.0. 3: 805 Wiley, R.C. see Buehler, W.J. et al. Wilfert, G.L. 2: 637 Wilk, K. see Vosko, S.H. et al. Wilkening, D. see Sikka, V.K. et al. Wilkens, B.J. see Palmstrom, C.J. et al. Wilkens, M. I: 806; see also Banerjee, S. et al., Jenkms, M.L. et al.; Urban, K. et al. Wilkes, P. 1: 694, 706, 784, 785, 800, 801, 802 Wilkins, B.J.S. 2 137, 143, 144, 145; see also Cameron, D.J. et al. Wilkins, B.R. see Tseng, W.F. et al. Wilkins, S.W. 1: 28, 50; see also March, N.H. et al. Wilkinson, A.J. see Jcnner, A.G.I. et al.; Parviiimehr, A. et al. Wilkinson, M.K. et al. 1: 935 Willardson, R.K. 2: 327 Willardson, R.K. and Beer, A.C. 3 805 Wille, L.T. see DreyssB, H. et d. Wille, L.T. and de Fontaine, D. 3: 205 Willems, J.J. 2: 509, 511
954 Willemsen, M. see van der Kolk, G.J. et al. Willemsen, M.F.C. see Duchateau, J.P W.B. et al. Willens, R.H. 1: 703, 704; see also Duwez, P. et al.; Klement, W.J. et al.; Nesbitt, E.A. et al. Willett, D.R. see Mitchell, 1C.W et al. Willett, R.E. see Ho~lingsworth,E.H. et al. Willey, D.B. et al. 3 108 Williains, A.R. 1: 24: see also Gelatt, C.D. Jr. et al.; Kiibler, J. et al.; Moruzzi, V.L. et al.; Terakura, K. et al. Willia~s,A.R. et al. 1: 60, 62, 63, 69; 2: 610 Williams, C. see Koon, N.G. et al. Williarns, D. 2: 590; see also Davies, G.J. and Williams, D. Williams, D.B. see Alarngir, F.M. et al.; Sung, C.M. et al. Wil~iams,D.B. et al. 1: 586 Williams, D.J. 1: 844 Williams, D.S. 2: 610, 617, 618 Williams, E. 1: 775 Williams, E.J. 1: 502, 799 Willia~s,E.K. see Brenner, A. e l al. Williarns, F. 1: 610 Williains, J.C. 1: 137; 2: 95, 282; 3: 473, 474, 477, 503, 591; see also Cho, W. et al.; Chu, W Y. et al.; Martin, P.L. et al.; Strychor, R. et al., Ward, G,H. et al. Williams, J.E.C. see Zhao, Z.P. et ul., Zhukovshy, A.Y. et al. Williams, J.E.C. et al. Williams, K.A. see La Williams, K.J. 2: 214 Williams, K.R. see Davies, P.W. et al.
Sales, B.C. et al., Sharp, J.W ct al. Williams, R.K. et al. 1: 1024, 1026, 1027 Wil~iams,R.S. %: 621; see aDo Kim, Y .K. et al., Lince, J.R. and Williams, R.S.; Sadwick, L.P. et al. Williams, W.S. 1: 580 Williamson, R.L see Korth, G.E. and Williamson, R.L., Rabin, B.H. et al.; Wnght, R.N. ef al. Willis, B.T.M. 1: 441 Willis, 3.0. see Fisk, Z. et al., Stewart, G.R. ~t al. Willis, R.F. see Campuzano, J.C. et al. Willman, N. et al. 3: 687 Willnecker, R. see Wilde, G. et al. Willner, A.E. et al. 2: 413 Wills, J.M. see Asta, M.D. et al. Wills, J.M. et al. 1: 195, 207 Wil~sher,A. see Lamarche, A.M. et aL. Wilm, A. 2: 564 Wilsdorf, N.G.F. 1: 498, 525; 2 13, 25, 137, 585; see aba Hawk, J.A. et al. Wilsdorf, M.G.F. et al. 2: 187 Wilshire, B. see Davies, P W et al. Wilson, C.G. 1: 252 Wilson, C.G. et al. 1: 409 Wilson, E.Y 1: 957
Author Index Wilson, G.C. 2: 507 Wilson, G.T. see Dresner, L. et al. Wilson, J.R. 1: 661 Wilson, L.W see Talbowm, F.T. et al. Wilson, R.H. 2: 646 Wimmer, E. et al. 1: 133; 3: 234 Win, W see Freeman, A.J. et al. Winburg, J.D. see Mills, M.J. et al. Windle, P.L. et al. 3: 22, 32 Wing, R.C. 2: 489 Winkler, P.J. et al. 1: 653; 2: 88 Wiiinicka, M.B. 2 159, 163, 164, 165, 166; see also Virk, IS. et al. Winter, E. see Hazzledirte, P.M. et al. Winter, H. 2: 91 Winter, M. 3 809 Winter, M. et al. 3: 250 Winterbon, K.B. et al. 1: 806, 811 Wipf, H. see Potzel, U. et al. Wire, M. 1: 214 Wire, M.S. see Stewart, G.R. et al. Wirkus, C.D. and Wilder, D.R. 3: 574 Wise, E.M. 2: 562, 564, 570 Wise, E.M. et al. 2 564 Wiseinan, M. see Campisi, I.E. at al. Wiser, N. 3: 691; see also Hickey, 13.5. et al. Wiss, T. and Matzke. H.J. 3: 267 Wissmnnn, P. 1: 956 Witcomb, M.J. see Compton, D.N. et al. Hill, P.J. et al., Hohls, J. et al.; Horner, I.J. ef al. Witcombe, M.J. see Harte, A.S. et al. Withers, R.L. see FitzGerald, J.D. et al. Witmer, C. see de Reus, R. et al. Witte, H. 1: 14, 107, 108, 322, 40511, 409; see also Pauly, H. et al. Wittenauer, J. see Bassi, C. et al. Wittenauer, J. et al. 1: 655 Wittenauer, 3.P. 1: 999 Wittig, F.E. 1: 97 Wittrner, M. 2 605, 622, 623, 624, 630; 3: 663; see d s o Krusin-Elbaum, L. et al. Witzen, M.B. see Suchow, L. et al. Wlodawski, M.S. see Mentzer, M.A. et al. Wohlbier, F.H. 1: 757; 3: 805 Wohlfarth, E.P. and Buschow, K.H.J. 3: 804 Wohlleben, D. see Sievering, A. et al. Wold, A. see Guen, L. et al. Woldt, E. 1: 743 Wolf, G.K. 3: 664 Wolf, R. see Sajovec, F et al. Wolf, U. see Ernst, F. et al. Wolf, V. see Giessen, B.C. et al. Wolf, W. see Stadfer, R. et al. Wolfe, R. see Sherwood, R.C. et al. Wolfenden, A. 1: 885, 886, 920; 2: 55, 56. 207 Wolfenstine, J. see Cheng, G. H. et al.; Cheng, S.C. et al.; Jeng, Y.-L. et al. Wolfenstine, J. et al. 1: 918 Wolfers, P see Obbade, S. et al. Wolfers, P et al. 3: 98 Wolff, G.A. 1: 17111, 172, 174, 178, 17811, 180, 181, 183, 186, 191, 192 Wolff, G.A. et al. 1: 168, 169, 170, 172, 173, 17811, 181, 192 Wolff, I.M. 3: 61, 75, 80; see also Cortie, M.B. et al., Hill, P.J. et al. Wolff, I.M. and Cortie, M.B. 3: 56, 77
Wolff, 1.M. and Hill, P.J. 3 59, 68 Wolff, X.M. and Pretorius, V.R. 3: 74 Wolff, X.M. and Sauthoff, G. 3: 55,60,61, 67 Wolff, I.M. et al. 3: 60, 61, 62, 67 WolE, J. see Franz, M. et al.; Kerl, R. et al. WolB, J. et al. 3: 281, 284, 286, 287 WolE, s. 2: 377 Wolfgarten, G. see Kruger, P et al. Wolgemuth, G.A. see Turner, R.B. et al. Wolk, G.D. see BIanpam, B. ef a/. Wall, C. see Harten, U. et al. Wollam, J.S. 1: 109 Wollan, E.O. 1: 447; see also Cable, J.W. P f al. Wol~enberger,H.J. 1: 559, 560, 599 Wollgarten, M. see Feuerbacher, M. et al., Rosenfeld, R. et al.; Urban, K. P t al.; Wang, R.H. et al., Zhang, Z . et al. Wollgarten, M. et al. 1: 463; 3: 399 Wolter, J.H. see Weegels, L.M. et al. Wolters, R.A.M. see Dirks, A.G. et al. Wottersdorf, G. 3: 113, 114 Woltersdorf, G. see Zintl, E. and Woltersdorf, G. Wolverton, C. see Asta, M.D. et al. Wolverton, C. and Zunger, A. 3: 194, 195 Wolverton, C. et al. 1: 32; 3 193, 194 Wong, J. see Randall, R. et al. Wong, J.Y. 3; 28 Wong, K.M. see Koflat, D.D. et al. Wong-%an, M. et al. 3: 62 Wonka, J. see Lugscheider, E. et al. Wonn, H. see Muller, Ch. et al.; Ziesche, P et al. Wood, C. 2: 453 Wood, D.L. 1: 596, 597, 897, 898,918, 923, 927, 1000, 1003; 2: 30, 34; see also Westbrook, J.H. and Wood D.L. Wood, D.L. and Westbrook, J.H. 3: 353, 354 Wood, D.M. 1: 41 Wood, E.A. 3 216 Wood, J.H. 1: 989 Wood, J.H. et al. 2: 492 Wood, J.R. see Foreman, A.J.E. et al. Wood, M.I. 2: 492; see also Nicho~~s, J.R. et al. Wood, T.H. see Miller, D.A.B. et al. Woodfield, A.P see Rowe, R.G. et al. Woodford, D.A. 3 318 Woodford, D.A. see Gigliotti, M.F.X. et al. Woodhouse, J.B. see Evans, D.J. et al. Woodson, H.N. see Stekly, Z.J.J. et al. Woodward, G. see Benedek, R. et al., Dimiduk, D.M. et al., Parthasarathy, T.A. et al.; Rao, S. et al.; Rao, S.I. et al. W o o d ~ a r d C. , et al. 1: 506, 540, 541; 3: 41 1 Woodward, J. see Gill, S.S. et al. Woolgarten, M. see Feuerbacher, M. et al. Woollam, J.A. 2: 438, 439; see also McGahan, W.A. et al., Oh, J.E. et al. Woolley, J.C. 1: 352; see also Hughes, O.H. et al.; Lamarche, A.M. et al. Wopersnow, W, and Raub, Ch.J. 3: 55
A ~ t h Index o~ Work, S. see Thadhani, N.N. et al. Worrell, W.L. 1: 64, 95, 988, 996 Worth, B.D. see Larsen, J.M. et al. Worth, B.D. et al. 3: 326, 484 Wortman, G. see Potzel, U. et al. Wdrwag, G. 1: 721 Wosnitza, J. see von Ldhneysen, H. et al. Wowchak, A.M. et al. 3: 783, 784 Wride, V. see Honey, F.J. et al. Wright, A.C. 1: 792 Wright, A.F. see Baskes, M.I. et al. Wright, A.J. see Qwen, N.B. et al. Wright, C.R.A. et al. 1: 5 Wright, C.S. see Ansara, I. et al. Wright, D.A. see Spencer, R.M. et al. Wnght, D.C. 1: 461 Wright, J. see Liu, C.T. et al. Wright, J.E. 1: 741 Wright, J.K. and Wright, R.N. 3: 654 Wright, J.L. see Alexander, D.J. et al.; Liu, C.T. et al.; Maziasz, P.J. et al. Wright, P.K. 3: 480; see also Jang, H. et al. Wnght, P.K. et al. 2: 65; 3: 331, 332 : 223; see also Knibloe, J.R. et al., Sikka, V.K. et d. Wright, R.N. and Sikka, V.K. 3: 504 Wright, R.N. et al. 1: 645; 3: 656 Wrsght, W.J. et al. 3: 694 Wrobel, J.S. 2 328 Wronski, A S . see Ansara, I. et al. Wu, C.Q. see Chu, T.L. et al. Wu, C.S. see Marshall, E.D. et al. Wu, D.H. see Inui, H, et al. Wu, G.H. see Yang, G.P. et al. Wu, J.J. see Chu, J.P. et al. WLI,J.S. see Sun, J. et al. Wu, M.H. 1: 836; see aZso Hamada, Y. et ad. Wn, M.H. et al. 1: 836, 841 Wu. S.K. see Hon, W.P. et al., Yang, Y.S. and Wu, S.R. Wu, S.K. and Lin, R.Y. 3: 668 Wu, S.K. and Wayman, C.M. 3: 56 Wu, T.W. 1: 453, 454 Wu, W. see Taniguchi, S. et al. Wu, W.T. et al. 2: 490 Wu, X. see Yu, S. et al. Wu, X.H. see Chen, X.F et al.; Wu, G.X. et ul. Wu, Y.-H. see Kawakami, Y. et al. Wu, Y.-H. et al. 2: 426 Wu, Y.K. see He, L.X. et al. Wu, Y.P. see Sanchez, J.M. et al. Wu, Y.P et al. 1: 496, 504 Wu, Z.L. et al. 1: 924; 2: 23. 159, 161, 162, 163, 164, 165, 166 Wubbenhorst, R. 1: 94 Wuensch, B.J. see Cava, R.J. et al. Wukusick, C.S. 1: 1000, 1002 WulE, G. 1: 167 Wun-Fogle, M. see CIark, A.E. et al.; MofTett, M.B. et al. Wunderlich, F. see Hartmann, €3. et al. Wunderlich, W. et al. 1: 927; 2: 78 Wuiinike-Sanders, W 1: 9 15 Wurschum, R. see Brossniann, U. et al.; Schaefer, H.-E. et al. Wurschum. R. and Schaefer, H.-E. 3: 283 Wurschuin, R. et al. 3: 277, 278,279, 280, 281, 282, 283, 284, 285, 291 W u r ~ a l l n e rIS. , see Wagner, R. et al.
Wuttig, M. see Gauthier, Y. et al. Wuttig, M. et al. 1: 612, 616, 617 Wyckoff, R.W.G. 1: 277,278,281,309; 3: 10 Wynblatt, P.P 1: 611, 612, 761, 762; see also Cheng, C.Y. et al. Wyim, J.D. see Nong, M. et al. Xi, M. see Hu, J. et al. Xia. S.K. et al. 3: 732 Xia, W. see Hong, (2.2.et al. Xia, W. et al. 2: 609 Xiang, Z.D. et al. 3: 579 Xiao, H.Z. see Ramanath, G. et al.; Rockett. A. et al. Xiao, L. and Abbaschian, R. 3: 583 Xie, W. see Jeon, H. et al. Xiedi, P see Sheng, L. et al. Xing, Z.P. see Han, Y.F. and Xing, Z.P.; Han, Y.F. et al. Xu, G.-B. et al. 1: 812, 818, 819, 820 Xu, J. see Guo, X.-Q. et al.; Hong, T. et al.; Lin, X. et al. Xu, J.-H. 1: 65, 66, 69; 2: 156, 157; see also Chen, B. et al.; Freeman, A.J. et al.; Lin, W et al. Xu, J.-H. et al. 1: 65, 66, 68, 69, 71 Xu, J.H. see Freeman, A.J. et al. Xu, J.M. see Li, H.S. et al. Xu, L. and Sevov, S.C. 3: 115 Xu, L. et al. 3 130 Xu, M. see Wuttig, M. et al. Xu, Q. see Zhang, Y.G. et al. Xu, R. see Verkerk, P. et al. Xu, R, et al. 3: 252, 255 Xu, W. and Moriarty, J.A. 3: 439 Xu, X.F see Lin, C. et al. x u , Y 2: 137, 143; see also Golberg, D. et al., Schulson, E.M. et al. Xu, Y . et al. 3: 61 Xu, Y.H. see Jia, C.L. et al. Xue, W. et al. 3: 734, 735 Yacaman, M.J. 1: 491 Yacaman. M.J. et al. 1: 491 Yadagawa, M. et al. 2: 23 Yafet, Y. see Majkrzak, C.F. et al. Yagn, K. see Qsakabe, N, et al. Yagi, Y see Yamaguchi, K. et al. Yajima, E. see Nishi, Y. et al. Yakel, H.L. 1: 396 Yakovenkova, L.I. see Gornostyrev, Yu.N. et al., Greenberg, B.A. et al. Yakovleva, E.S. see Adrianovskii, B.P et al., Syutkma, V.I. and Yakovleva, E.S. Yalisove, S. 1: 616 Yalisove, S.M. see White, C.L. et al. Yamabe, Y. 2: 75 Yainabe, Y . et al. 2: 76; 3: 67 Yamabe-Mitarai, Y. see Gu, Y et al.; Murakami, A. et al.; Yu, X.H. et al. Yamabe-Mitarai, Y. et af. 3: 67, 68, 69, 71, 843 Yamada, H. see Nakamura, H. et al. Yamada, J. and Demukar, N. 592 Yamada, T. see ~ s h i d aT. , et al. Yamada, T. et al. 1: 443; 2 507, 511 Yamada, Y . 1: 441,442 Yamadaya, T. see Sambongi, T. et al.
955 Yamagata,' T. 2: 202, 204; see also Murakami, A. et (tl. Yaniagishi, H. 2: 621 Yamagiwa, Y , see Ikeda, M. et al. Yamaguchi, A. et al. 2: 309 Ya~aguchi,K. et al. 2: 383 Yamaguchi, M. 1: 496,498,504,506,507, 510, 521, 536, 564, 653, 913, 914; 2: 18, 19, 88, 105, 148, 149, 154, 202, 240; 3 41 1, 414, 424; see also Fujiwara, T. et al.; Guder, S. et al.; Weatherly, L. ef al.; Tkebuchi, M. et al.; Inui, H. et al., Ishikawa, K. et al.; Ito, K. et al.; Jin, Z. et al.; Kini, Y,-W. et al.; Kishida, K. et al., Moriwaki, M, et al., Nishitani, S.R. et al.; Oh, M.H. et al.; Paidar, V. et al., Shimokawa, T. et al.; Siegl, R. et al.; Umakoshi, V et al.; Vitek, V. et al.; Yokoshima, S. and Yamaguehi, M.; Yoo, M.H. and Yaniaguchi, M. Yamaguchi, M. and hi,H. 3: 363, 446, 447, 477,495 Yamaguchi, M. and Shirai, Y. 3: 280 Yamaguchi, M. and Umakoshi, Y. 3: 275, 325,332,361,368,406,414,477,621 Yamaguchi, M. et al. 1: 496, 497, 500, 501, 506, 512, 529, 536, 537, 541, 547, 916; 2: 20,22, 147, 148, 150, 151, 153, 154, 155, 164, 165; 3: 241, 410, 414, 441, 442, 648 Yamaguch, S . see Hirabayashi, M. et al. Yamaguchi, S. et al. 1: 726 Yamaguchi, Y see Takabatake, T. et al. Yamaka, E. see Kanaya, H. et al. Yamakawa, K. see Iiiaba, M. et at. Yamakoshi. S, et al. 2: 336, 426, 427 Yamamoto, I-I. 2: 309; see also Sagawa, M. et al. Yamamota, M. 1: 618; see also Ishikawa, M. et al., Nakano, T. et al. Yaniamoto, N. see Ashida, A. et al.; Hortnaka, H. et al. Yamamoto, R. see Hashimoto, M. et al.; Wakayama, S. et al. Yamamoto, T. see Aoki, K. et al.; Takeuchi, S . et al. Yamamoto, T. et al. 2: 410, 411; 3: 671 Yamane, M, see Hisatune, K. et al.; Ohta, M. et 01. Yamane, T. see Umakoshi, Y. et al.; Yama~uchi,M. et al. Yamane, Y. see Umakoshl, Y. et al. Yamaoka, S. see Kagamda, M. et al. Yamaoka, T. et al. 2: 336 Yamasaki, T. et al. 3: 740 Yamauchi, H. 1: 775, 781, 855 Yamauchi, K. see Yoshizawa, Y. et aE. Yamauchi, K.J. see Yoshizawa, Y. et al. Yamauchi, Y. see Kikuchi, R. et al. Yamazaki, IS.see Shirai, T. et al. Yamazaki, S. see Shirai, T. et al. Yamoaka, S. see Mishiiia, 0. et al. Yaniuguchi, M. see Umakoshi, Y. and Yamaguchi, M. Yan, L.M. et al. 3: 812, 824 Yan, M. see Chen, D. et al.; Sinkler, W. et al. Yan, M. et al. 3: 770, 771, 772 Van, Y.F see Wang, R.H. et al.
956 Yan, Y.F. and Wang, R. 3: 400 Yan, Z. et al. 1: 751 Yaaagihara, M. et al. 2: 409 Yanagimoto, K. see Yamaguchl, A. et al, Yanagisawa, E. see Umakoshi, Y et al. Yanagitani, A. see Aoki, K. et al. Yaney, D.L. 2: 63 Yang, A.B. see Olson, C.G. et al. Yang, B. see Jaszczak, J.A. et al. Yang, B.X. see Shapiro, S.M. et al. Yang, C.N. 1: 507 Yang, C.P. see Wang, Y.Z. et al. Yang, C.P. et al. 3: 94 Yang, C.Y. see Fung, K.K. et al. Yang, D.Z. see Sui, H.X. et d. Yang, F. see Hu, J. et al.; Tang, W. et al. Yang, F.M. see Hadjipanayis, G.C. et al.; Han, X.F. et al.; Tang, N. et al. Yang, F.M. et al. 3: 94, 99 Yang, G.Q. X: 641 Yang, H. see Bakker, H. et al. Yang, H.S. see Valiev, R.Z. et al. Yang, H.S. et al. X: 657, 925; 2 128 Yang, 3. see Mackin, T. and Yang, J. Yang, J. et al. 3 99 Yang, J.-I. see Yang, Y.-c. et al. Yang, J.L. see Lm, C. et al.; Zhang, D. et al. Yang, J.M. see Boss, D.E. and Yang, J.M,; Nieh, T.G. et al. Yang, J,-M. 2: 290; see also Wang, P.C. et al. Yang, K.N. see Meissner, G.W. et al. Ymg, L.C. see Mullan, C.A. et al.; Rockett, A. et al. Yang, L.C. and Rockett, A. 3: 664 Yang, L.H. see Benedek, R. et al. Yang, M.S. 2: 621 Yang, Q.B. 1: 483 Yang, R. et al. 1: 860 Yang) S.-C. see Lee, T.-S. et al. Yang, S.S. see Hou, D.H. et al. Yang, S.W. see Gigliotti, M.F.X. et al.; Jackson, M,R. et al. Yang, W. see Gibala, R. et al. Yang, W.J. 2: 63, 64 Yang, W.J.S. 1: 791, 813 Yang, W.J.S. et al. 1: 791, 813 Hu, C,Z. et al.; Wang, R.H. et al. Yang, W.G. et al. 3: 379, 383, 384, 397, 400, 899 Yang, W.Y. see Sun, Z.Q. et al. Yang, Y. see Yang, J. et al. Yang, Y. et al. 3: 667 Yang, Y,-C. et al. 2: 315 Yang, Y.-M. see Bsennan, P.C. et al. Yang, Y.S. and Wu, S.K. 3: 407 Yanir, E. see Mayer, I. et al. Yankovskii, Kh.1. see Semenova, A D . et al. Yano, N. see Inouye, A. et al. Yano, T. see Ito, K. et al. Yanson, I.R. 1: 682 Yanson, T.I. et al. 1: 725 Yao, D,Z. et al. 3: 395 Yao, L.-X. see Chen, N.-Y. et al. Yao, N.P. et al. 2: 510; 3: 251 Yao, Y.D. see Pan, Y.C. et al. Yarbrough, W.A. 1: 180 Yariv, A. see Lee, C.P. et ~ l .
Yarmolysek, Ya.D. see Skolozdra, R.V. et al. YannoIyuk, Ya.P. see Grin', Yu.N. et al.; Kalychak, Ya.M. et al.; Sichevich, O.M. et al. Yarmoshenko, Yu.M. see Galakov, V.R. et al. Yarovetz, V.I. see Gladyshevskii, E.I. et al. Yartys, V.A. et al. 3: 100 Yasaka, H. see Takeuchi, H. et al. Yasuda, H. et al. 1: 875, 887 Yasuda, H.Y. see Nekano, T. et al., 'Umakoshi, Y. et al. Yasuda, H.Y et: al. 3: 326, 332, 334 Yasuda, M. 1: 774; 2: 282, 562, 565, 566; see also Kanzawa, Y. et al., Ohta, M. et aE.; Shirai, T. et al., Tani, T. et al., Udoh, K.-I. et al.; Uzuka, T. et al. Yasuda, K. et al. 1: 640 Yasuda, T. et al. 2: 326 Yasuda, Y. see Hisatsune, K. et al. Yasuoka, H. see Ikushima, K. et al. Yata, M. et al. 1: 651 Yates, J.T. Jr. I: 183 Yatsenko, S.P. 3 32, 33, 808 Yatskar, A. see Canfield, P.C. et a t Yavasi, A.R. 1: 773, 809; see also Bordeaux, F. and Yavari A.R.; Gialanella, S. et al. Yavari, A.R. et al. 1: 788 Yavari, P. 1: 918 Yayama, H. see Hashimoto, T. et al.; Tokw Y. et al.; Tomokiyo, A. et al. Yazbeck, J. see Rebbah, A. et al. Ye, C.T. see LID,C. et al. Ye, C.-T. see Yang, Y.-C. et al. Ye, F see DiPasquale, J. et al.; Soboyejo, W.O. et al. Ye, F, et al. 3: 342, 343, 345 Ye, H.Q. see Zhang, L.C. et al. Ye, Y.Y. see Kong, X.J. et al. Yealoz, R. see Heinz, D.L. and Yealoz R. Yeh, C.-Y. et al. 1: 422, 424 Yeh, J.J. 3: 137 Yeh, J.J. and Lindau, I. 3: 137 Yeh, L.S.R. et al. 2 511 Yeh, X.L. et al. 1: 695, 734; 3: 681 Yelon, W.B. see Hu, 2;. et al., Luo, H. et al.; Sauvage, M. et al. Yen, B.M. et al. 3 758 Yencha, A.J. see Daniell, A.J. et a/. Yermolenko, A.S. see Ivanova, G.V. et al. Yi, H.C. 1: 646 Yi, S.H. see Perepezko, J.H. et al. Yi, S.S. et al. 3; 56 Yixn, W.M. 2: 464 Yin, B.1. see Benedek, R. et al. Yin, W. see Lupinc, V. et al. Yin, Y, see Yu, S. et al. Yip, S. 1: 817; see also Limoge, Y. et al. Yodagawa, Y.M. see Mishima, Y. et al. Yodogawa, M. see Mishima, Y. et al.; Ochini, S. et al. Yodogawa, Y. et al. 1: 498, 499 Yogurtcu, Y.K. see Miller, A.J. et al. Yokokawa, T. 1: 71; see also Gu, Y. et aE,; Murakami, A. et al., Yamnbe-Mitarai, Y. et al. Yokoshima, S. see Kishida, K. et al.; Nakagawa, Y.G. et al.
Yokoshima, S and Yamaguchi, M. 3: 484 Yokola, R. see Komatsu, 3'. et al. Yokoyama. N. et al. 2: 621 Yokoyama, T. see Ohashi, K. et al. Yokoyama, Y. see Inoue, A. et al. Yoiton, C.F. 2: 87; see also Clemens, H. Yolton, C.F et al. 3 634, 649, 651 Yonesaki, T. see Nasako, I(.et al. Yonezu, I. see Nasako, K. et ul.
902, 915, 920, 922; 2: 24, 25, 28, 58, 148, 150, 167, 202; 3 407, 411, 413, 417, 418, 422; see also Fu, C.L. and Yoo, M.N.; Hazzledine, P.M. et al.; Lee, J.K. and Yoo, M.H.; Lee, J.W. rb al.; Minonisbi, Y. and Yoo, M.H.; Yoshimi, K. et al.; VeyssGre, P et al. Yoo, M.H. and Fu, C.L. 3: 405,407,422, 626 Yoo, M.H. and Hishinuma, A. 3: 410, 422, 423 Yoo, M.H. and King, A.H. 3 419 Yoo, M.H. and Lee, J.K. 3: 414 Yoo, M.H. and Loh, B.T.M. 3 405 Yoo, M.H. and Yamaguchi, M. 3: 411, 424 Yoo, M.H. et al. 1: 88, 535, 552,915, 916; 2 37, 155; 3: 36l,406,411,418,419, 424, 621, 626 Yoo, Y.S. see Henry, M.F. et al. Yoon, D.Y see Henry, M.F. et al. Y o o s ~ h a r a M. , and Kim, Y -W. 3: 479, 480,481 York, Y. 2: 573 Yoshida, H. 2: 202, 204 Yoshida, J. see Kato, K. et al. Yoshida, K. I: 945; see also K~rosawa,K. et al.; Oku, S. et al. Yoshida, M. 1: 498, 505, 528; 2: 214, 215, 216; see aZscr Bonhomme, F. et al.; Takasugi, T. et al. Yoshida, M. and Takasugi, T. 3: 443 Yoshida, M. et al. 3: 461 Yoshida, T. see Enami, IS,et al., Okamoto, A. et al. Yoshihara, M. and Kim, Y.-W 3: 479, 48 1 Yoslnihara, M. et al. 3: 577 Yoshimi, K. see Hosoda, H. et al. Yoshimi, M.and Hanada, S. 3: 368, 372, 373 Yoshimi, K. et ul. 3: 368 Yoshimoto, W. et al. 2: 310 Yoshiiiaga, H. see Sakai, T. et al. Yoshitorni, T. see Kido, G. et al. Yoshizaki, K, see Hashimoto, Y. et al.; Kishida, T. et al. Yoshizawa, Y. et al. 1: 749; 3: 702, 754 Yosouf, K. see Sahu, P.C. et al. Yost, F.G. 2: 654 You, Y.B. see Ku, H.C. et al. Youchison, D.L. see Castro, R.G. et al. Young, ASP.see Berkowitz, A.E. et al. Young, D.J. see Sou~hwel~, C.et al. Young, E.W.A. Ir 989, 990 Young, E.W.A. et al. 11987, 989 Young, F.A. 2: 585
Young, J.L. see Heyne, C.J. et al. Young, J.M. see Dowson, A.L. et al. Young, L. 2: 503 Young, M. see Adam, E. et al. Young, M.G. see Sham, Y. et al. Young, P. 3 32 Young, P.L. 1: 889 Young, R.M.K. 2: 176 Youngqmst, S.E. et al. 1: 484 Yu, see Semenova, E.L. et al. Yu, A. see Palmstrom, C.J. er al. Yu, D, see Hartwig, K.T. et al. Yu, D. et al. 1: 463 Yu, D.P et al. 3: 379 Yu, H.F. et al. 3: 443 Yu, K O . see Oti, J.A. and Yu., K.O. Yu, K.O. et al. 3: 602, 603 YU, L. 3: 742 Yu, L.H. 1: 649 Yu, R. 2: 481 Yu, R.H. et al. 3: 691 Yu, S. see Tan, T.Y. et al. Yu, S.C. see Kiln, K.S. et al. Yu, S.C. et al. 1: 184 Yu, W.Z. see Schaefer, H.E. et al. Yu, X. see Gu, Y. et al. Yu, X.H. et al. 3: 69 Yu, Y.V 2: 231 Yu, Z.Z. see Ciapp, P.C. et al. Yu-Zhang, K. et al. 1: 457, 465 Yuan, M. see Cheng, J. et al. Yuasa, S. et al. 3: 56 Yuda, S. see Kamigaki, K. et al. Yudina, N.V. et al. 3: 813, 814, 816 Yue, W. 1: 130 Yukawa, N. see Morinaga, M. et al.; Murata, Y. et al. Yung, E.K. and Turlik, I. 3: 670 Yung, H.F. see Gottstein, G. et al. Yupko, L.M. see Svechnikov, V.N. et al. Yurchikov, E.E. see Elsukov, E.P et al.; Ermakov, A.E. et al. Yuzhin, A.I. see Muslimov, I.S. et al. Yvon, K. 1: 302, 303, 304, 393, 394; 2: 478; see also Belin, et al.; Bonhomme, F. et al.; Dunand, A. et al.; Gueramian, M. et al.; Parthti., E. et al. Yvon, P.J. and Schwarz, R.B. 3: 731 Zabala, R.J. 1: 883, 885, 887; 2 240, 244, 245, 246, 247, 248, 2.52; see also Fleischer, R.L. et al. Zabel, H. see Youngquist, S.E. et al. Zabel, W,et al. 3: 257 Zacharia, T. 2: 202 Zackay, V.F. 2: 203 Zador, S, 1: 95 Zagoruiko, N.G. 3: 818, 819, 822 Zahavi, J. and Fehrsson, P.E. 3: 670 Zaikovskaya, L.V, see Panko, T.1. et al. Zaitsev, A.I. see Goman'kov, V.1. et ul. Zak, T. see Schaefer, H.E. et al. Zakharov, V.V. see Elagin, V.1. et al. Zakharova, A.M. et al. 2 95 Zamanza~eh,M. see Bavarian, B. et al, Zanazzi, E. see Atrei, A. et al. Zandbergen, H.W. see Cava, R.J. et al.; Siegrist, T. et al. Zand~ergen,H.W et al. 3: 104 Zsng, Y. see Ding, Y. et al.
Zang, Z, and Couture, A. 3: 298, 31 1 Zangwill, A. 1: 88 Zanio, K. see Rowe, J.M. et al. Zare, R. et al. 1: 172 Zarechnyuk, 0,s. 1: 413; see also Kripyakev~ch,P.I. et al.; Vivchar, 0.1. et al,; Yanson, T.I. et al. Zarechnyuk, O.S. et al. 2: 178, 179 Zdremba, E. 1: 78 Zasedatelev, M. see Kurnakov, N.S. et al. Zavada, J.M. see Mentzer, M.A. et al. Zavalij, L. see Zribi, A. et al. Zawadawski, F. 1: 213 Zbroniec, L. see Varin, R.A. and Zbroniec, L. Zebreva, A.I. see Kozlovskii, M.T. Zec, S. see Durovic, D. et al. Zedalis, M.S. 1: 891; 2: 184, 185; see also Angers, L. et aE.; Das, S.K. et al.; Skinner, D.J. et al. Zedalis, M.S. et al. 2: 180, 193 Zee, R. 1: 800, 801, 802 Zee, R.H. 1: 694,706 Zehetbauer, M. see Schrank, J. et al. Zehner, D.M. see Graupner, H. et al.; Hammer, L. et al.; Lui, S.-C. et al. Zeijlexnaker, H. see de Reus, R. et al. Zeitlin, B.A. 2 361; see also Gregory, E. et al. Zelanki, P.D. and Simkovitch, G. Zelaya-Angel, 0. see Huerta-Ruelas, J, et al. Zelenin, L.P see Balina, Ve.A. et al.; Senchenko, A.A. et al. Zeller, R. see Hoekstra. H.J.W.M. et al.; Sariiia, D.D. et al.; Speier, W. et al. Zembutsu, S . see Ohki, A. et al.; Shibata, N. et al. Zemczuzny, S.F see Kurnakov, N.S. et al. Zemek, J. see D'hiia, E. et al. Zemskov, V.S. et al. 3: 812, 814, 816 Zeper, W.B. 2: 442, 450 Zerobin, F see Wenger, S. et al. Zeumer, B. and Sauthoff, G. 3: 316 Zghal, S. 3: 422 Zhan, Q.B. see Yan, L.M. et al. Zhan, W.S. see Cheng, Z.H. et al.; Fung, K.K. et at.; Shen, B.G. et al. Zhan, W.Z. see Fung, K.K. et al. Zhang, B. see Tseng, W,F et al. B. and SoEa, W.A. 3: 404 Zhang, B. et al. 1: 774 Zhang, B.4. see Yang, Y.-C. et al. Zhang B. 1: 247, 248 Zhang, B.S. see Lin, C. et al.; Zhang, D. et al. Zhang, D. et al. 3: 664 Zhang, D.-L. et al. 1: 484 Zhang, H. see Lacerda, A. et al. Zhang, H. et al. 3: 671 Zhang, H.W see Shen, B.G. et al. Zhang, L.C. et al. 3: 414 Zhang, L.Y. et al. 3: 179, 180 Zhang, M. see Yu, S. et al. Zbang, S. see Berkowitz, A.E. et al.; Fang, R.-Y. et al.; Lu, L. et al.; Mikkola, D.E. er al., Nic, J.P. et d. Zhang, S. et al. 2: 147, 156, 158, 160, 161, 163, 164, 165, 166, 168,426 Zhang, S.B. er al. 1: 252 Zhang, S.J. et al. 2 139
Zhang, S.Y. see Shen, B.G. et al. Zhang, T. see Inoue, A. et al. Zhang, T, et al. 1: 690; Zhang, T,X. see Lou, L Zhang, W. see Chen, S. et al.; Deevi, S.C. and Zhang, W. Zhang, W.J. see Chen, G.L. et al. Zhang, X. 1: 454,457,461,463,475,477, 478, 482; see also Ali, W.et al.; Dirnitrov, C. et al, Zhang, X.-D. see Yang, Y.-C. et al. Zhang, X . G . 1: 133; see also Conis, A. et al.; Stocks, G.M. et al. Zhang, X.J. see Zhang, M,et al. Zhang, X.-X. see Fang, R.-Y et al.; Yu, R.H. et al. Zhang, X.Y. et al. 3: 283, 285, 286 Zhang, Y. see Cken, S. et al. Zhang, Y.G. et al. 1: 532, 534 Zhang, Y.G. and Cha~urvedi,M.C, 3: 413 Zhang, Y.-H,, see LI, T. Zliang, Y.L. see Rao, G.H. et al. Zhang, Z. 1: 463; see also He, L.X. et al.; Urban, K. et al.; Wollgarten~M. et al. Zhang, Z. et al. 1: 463, 465, 484, 537 Zhang, Z.-D. see Zhang, D. et al. Zhang, Z.X. see Cheng, Z.H. ef al. Zhang, T. see Inoue, A, et al. Zhang, Y.K. see Zhang, H, et al. Zhao, G.L. 1: 137, 143 Zhao, J . 4 . 3: 841, 848; see also Bewlay, B.P. et al. Zhao, J.G. see Fung, K.K. et al. Zhao, J,-T. et al. 1: 292 Zhao, L. 2: 75; see also Beddoes, J. et al.; @hen, W.R. et al.; Dudzinski, D, et al.
Zhao, R.W, see Han, X.F. et al.; Yang, F,M. et al. Zhao, X.A. see So, F.C.T. et al. Zhao, X.L. see Campbell, S.J. et QE. Zhao, Y. see Tang, W. et al. Zhao, Y.L. see Pao, Y.H. et al. Zhao, Z.G. see Tang, N. et al. Zhao, Z.P. see Zhukovsky, A.Y. et al. Zhao, Z.P. et al. 2 371 Zhemchuzhnii, S.F, 1: 14 Zheng, L.R. see Blanpain, B. et d. Zheng, L.R. et al. 2: 628 Zheng, M. et al. 3; 572 Zheng, Y.H. see Hadjipanayis, C.C. et al. Zheng, Z. see Zhang, T. et al. Zherdev, A.A. see Nikitin, S.A. et al. Zhihul, L. see Sheng, L. et al. Zhilyaev, A.P. see ~ c ~ a d d eS.X. n , et al. Zho, J. see Hu, Z.W. et al. Zhou, B. et al. 1: 270; 3: 812 Zhou, E. see Suryanarayana, C. et al. Zhou, G.F. see Bakker, H. et al. Zhou, G.F aiid Bakker, H. 3: 731, 761 Zhou, J.G. see Fung, K.K. et al. Zhou, J.M. see Yang, Y. et al. Zhou, L.see Chen, S. et al. Zhou, 0. see Zhu, Q. et al. Zhou, P 2: 430,431,432 Zhou, R.J. et al. 3: 168 Zhou, S. see Yu, S. et al. Zhou, X. see Chen, C.et al. Zhou, Y. see He, C. et al. Zhou, Y.Q. see Fung, Y. et al.
958 Zhou, Y.Z. see Fung, K.K. et al. Zhu, H. see Liu, C.T. et at. Zhu, H. et al. 1: 798, 799 Zhu, H.X. and Abbaschian, R. 3: 654,657 Zhu, J. see Yang, C.P. et al.; Yu, R.H. et al. Zhu, J.G. see Hong, Q.Z. et al.; Palmstrom, C.J. et al. Zhu, J.H. see Wan, X.J. et al. Zhu, M. see Sui, H.X. et al. Zhu, M.F. see So, F.C.T. et al. Zhu, M.F et al. 2: 610, 618, 619, 624, 626 Zhu, M.J. et al. 1: 67 Zhu, Q. et ul. 3: 257 Zhu, S.M. and Tjong, S.C. 3: 65 Zhu, S.M. et al. 3: 654 Zhu, W. et al. 3: 302 Zhu, Y see Lou, L. et al.; Zhang, T. et al. Zhu, Y.M. 1: 536, 542 Zhuang, Y. 1: 463 Zhukov, S.G. see Bodak, 0.1. et al. Zhukovsky, A.Y. see Zkdo, Z.P et al. Zhukovsky, A.Y. et al. 2: 371 Ziesche, P. see Muller, Ch. et al. Ziesche, P. et al. 1: 137 Ziman, J.M. 1: 663, 944, 1018 Zimm, C. et al. 3: 520, 523
Author Index Zimm, C.B. see Pecharsky, V.K. et al. Zimm, C.B. et al. 3: 527, 529 Zinoviev, V. see Weihs, T.P. et al. Zinsser, W.A. and Lewandowski, J.J. 3: 345, 491 Zinsser, W.A. et al. 3: 491 Zintl, E. 1: 13, 319; 3: 113 Zintl, E. and Brouer, G. 3: 113 Zintl, E. and Dullenkopf, W. 3: 113, 126 Zintl, E. and Woltersdorf, G. 3: 11 3, 114 Zintl, E. et al. 3: 238 Zinnsky, S . 1: 875 Znam, S. 3: 445 Znam, S. et al. 3: 445 Zocco, T.G. see Hanrahan R.J. Jr. et al. Zoccola, J.C. et al. 2: 519 Zogg, H. et al. 2: 184 Zok, I;. 1: 655 Zolliker, P. see Belin, et d. Zolotorevskii, V.S. see Khalim, A.A.R. et al. Zorc, T.B. 2 367 Zotov, V.P. see Ananyn, V.M. et al. Zou, G. see de Potter, M, et al. Zou, J. see Yoo, M.H. et al. Zou, X.D. et al. 1: 477 Zougmork, F P t al. 1: 743 Index compiled by GeoJTey C. Jones
Zribi, A. et al. 3: 663 Zschack, P. see Reinhard, L. et al.; Sparks, C.J. et al. Zuber, J.R. see Forrest, S.R. et al. Zuckermann, M.J. 1: 1028; sec also Thomsoii, J.R. et al. Zuercher, M.-H. 1: 747 Ziirnkley, T. see Ruminel, G. et at. Zumkley, Th. see Mehrer, E€. et al. Zunger, A. 1: 41, 130, 132, 229, 230, 232, 233,236,242,243,252,253,260,261, 319, 357,358,419,420,421,422,424; 3: 192, 198, 199, 202; see also Ferreira, L.G. et al.; Lu, Z.W et al.; Wei, S.-H, et al.; Wolverton, C. and Zunger, A.; Yeh, C.-Y. et al. Zunger, A. et al. 1: 41, 67, 68 Zupan, M. see Hemker, K. et al. Zutavern, F.J. see Loubriel, G.M. et al. Zuttel, A. see Chartouni, D. et al., Meli, F. et al.; Schlapbach, L. et al. Zuttel, A. et al. 2 486; 3: 108 Zuurendonk, T.J. see Bakker, H. et al. Zwicker, U. 2: 156 Zydzik, G.J. see Hunt, M.E.J. et al.
ote: sort order is letter-by- letter^ spaces being ignored. Preliminary pages are indicated by their roman numbers, Figures and Tables by italic page numbem, footnotes by suffix ‘n’. “IMC” means “intermetallic compound”. II-VI compounds, 3: xlii data sources, 3: 806 growth by MBE, 3: 780 III-V compounds, xlii, 3: 663, 732 data sources, 3: 800, 805, 806 growth by MBE, 3: 780, 781 IV-VI compounds, 3: xlii V-VI compounds, 3: xlii A2 structure, 1: 317, 765 see also I lattice complex A15 compounds ab znitio calculations, 1: 64, 68 disordering by milling, 3: 759 extrinsic properties, 2: 355-359 fabrication methods, 2: 360-364 ferroelastic transitions in, 3: 258 intrinsic properties, 2: 352-355 irradiation effects, 1: 792, 794-795, 808 mechanical properties, 2: 358-359 Pearson symbol for, 3: xxxi, xxxiii, 13 phase diagrams, 2: 353-354, 355 phase transitions, 1: 162-163 phonon dispersion effects, 1: 157, 158 point defects in, 1: 578-579 silicides, 2: 223-224 strain effects, 2: 354-355 space group for, 3: xxxi, xxxiii, 13 su~erconductors,1: 16, 748, 792, 808, 951, 1017-1018, 2: 223, 351-384 applications, 2: 364, 366, 367-384 characteristics, 2: 351-352 compared with other superconductors, 2: 366-367 transition temperatures listed, 2 352 surface reconstruction, 1: 619 surface structure 2 16 vacancy migration, 1: 579 see also Cr,Si structure type A286 [superalloy], 2: 3, 7, 42, 270, 272, 276 ABB alloys, 3 481, 482, 595, 596, 597 ABC notation, 1: 277, 278, 345, 358 ABDz Heusler alloys, prediction of, 3: 831-832 Ab initio calculations, 1: 31, 55-73 a~tiph~se-boundary energies, 1: 497-498, 514; 2: 58 for faults, 3: 457-458 for phase diagrams, 3: 185, 192, 206 and surface structures, 3: 215, 227
methodology, 1: 58-61 results, 1: 61--72 heats of formation, 1: 61-65, 67-71 phase diagrams, 1: 73, 114, 115 phonon dispersion curves, 1: 156-157 plaaar-defect energies, I: 71-72 str~~ctural energies, 1: 65-67 vacancy formation enthalpies, 3: 280 see also First-principles calculations Abundance of elements and cost/price considerations, 2 254 semicond~ct~rs, 2: 347 ABX, compounds, prediction of, 3: 825, 826 AB,X, compounds, with ThCr,Si, structure, prediction of, 3: 828-829 AB,& compounds, prediction of, 3: 825827, 828 Academic staff. 011 Internet, 3: 862 Accelerated oxidation see Pest phenomenon at intermediate temperatures, 1: 1000-1006 Accuracy of materials prediction 835 see also Pest degradation. . . Acoustic phonons, 1: 149 Acoustoelectric devices, 2: 345-346 Acousto-optic spectrum analyzer, 2: 413 Acousto-optical properties, database on, 3: 816 Acronyms listed and defined, 1: xix; 2 xvii, 3: xxiii-xxix Actinides amalgams, 3 23 beryllides, 3: 37, 46 dataa sources, 3 807 Activation area, dislocations in quasicrystals, 3: 398 Activation energy for creep, 3: 299, 321 Activation volume, dislocations in quasicrystals, 3: 398 Active concentration range principle, compound-formati~npredicted using, 1: 257, 264 Active corrosion, 1: 966 Active magnetic regenerator (AMR) magnetic refrigerator, 3: 522-524 Active oxidation, 1: 997 Active vibration controllers, 2: 403 Adamantane-structure compounds, 3: xxxv
additional experimental rules, 1: 356358 classification of, 1: 343-344 compos~t~ons, 1: 348-352 binary compounds, 1: 349 homogeneity-range diagram used, 1: 350-351 methods of calculation, 1: 348-352 inulticoinponent compounds, 1: 349-352 ternary compounds, 1: 349 defect compounds, 1: 344 normal compounds, 1: 344 ordered types, I: 344, 352-356 valence-electron rules, 1: 345-348, 35 1 tolerated deviations, 1: 352 Adamantane-structur~equation, 1: 346-347 Adamantane structures, defimtion and classification, 1: 343-344 Adamantine structures, 1: 343 [footnote] Adiabatic approxima~~on, 1: 151 Adiabatic bond-charge model, 1: 152-1 53 Adiabatic demagnetization, 3: 5 19-520, 521-522 Adiabatic elastic constants, 1: 873 compared with isothermal elastic constants, 1: 876 Advanced sheet rolling process (ASRP), 3 631,632 Aeroengine applications, 2: 46, 55, 87, 88, 237, 287,294 Aerospace systems, business issues, 3: 473 Aftordability, high-te~peraturestructural materials, 3: 476-477, 482-483 Ag-Al-Mn system, 2 308 Ag-A1 system, 1: 102 Ag-Al-Ti system, 1: 720; 2: 160, 164 Ag-Au-Cu-Pd system, 2: 651 649 Ag-Au-Cu-Zn system, 2 567-569 Ag-Au-Fd system, 2: 560, 570-572, 649 Ag-Au system, 2: 561 Ag-Bi system, 1: 102, 103 Ag-Cd system, 1: 836; 2 538 Ag-Cu-Ni-Pd system, 2: 651 Ag-Cu-Pd system, 2: 651 Ag-Cu-Pt system, 2: 651 Ag-Cu system, 1: 703, 705; 2: 561 Ag-Cu-Ti system, 1: 720 Ag-Cu-Zn system, 1: 837
Subject Index Age ~ a r d e n ~ ngold g , jeweiry alloys, 2 560, 562-569 Ag-Ge system, 1: 102, 103, 104, 703, 704; 3: 161 Ag-Hg naturally occurring compounds, 1: 632 Ag-In system, 1: 102; 2: 521, 522 Ag-Mg system, 1: 8, 596, 597, 956,3: 312 Ag-Mg system, creep behavior, 3; 31 1, 312 Ag-Pb system, 1: 102 Ag-Sb natural~yoccurring ~ompounds,1: 632 Ag- Sn system, 1: 102; 2: 521, 522, 576, 577 Ag-Ti system, 1: 102 Ag-Zn system, 1: 836, 846 Ag-Zr system, 1: 740 Air-conditioning applications, Aircraft propulsion systems, 3: 473, 478479, 541 AI-Au layered coatings, 3: 76 AI-AU system, 1: 116, 662, 673; 2: 572-573 AlAu, structure type, 1: 338 Al-AU-Cu alloys, 3: 77 AlB, structure type, 1: 388 atomic envkron~nents,1: 250, 25 1, 367-370, 373, 374, 375 atomic ~adii-~nteratomicdistance relationship, 1: 378, 379 atoms mixing on one position, 1: 376378 crystallographic data, 1: 364 hexagonal structures derived from, 3: 93,94 rare-earth compounds, 3: 88,89,91,93, 93 separation ia structure maps, 1: 389, 429 AI,Ba structnre type, 1: 364, 365, 373, 374, 375 AI-Ba system, 3; 156 AI-Be-Ni system, 1: 718 AI-Bi system, 1: 662 AI-B-Ni system, EAM potential, 1: 8245 Al,CeMn, structure type, 1: 413 AlCo alloys, solid solution hardening in, 3: 354, 355, 356 Al-Co-Cr-Mo-Ni-Ti alloys, 1: 867 Al-Co--Cr-Ni-Y coatings, 2: 493 AI-CO-Cr-Y coatings, 2: 492, 493 Al-Co-Fe system, 1: 717, 718, 719 A1-CO-Md system, 1: 719 AI-CO-Ni quasicrystals dislocations in, 3 385 plastic deformation of, 3: 399, 400 AI-Co-Ru system, I: 718, 719 AI-CO system, 1: 115, 116, 985, 995 Al-CO-Ti system, 2: 159 AI-Cr-Fe alloys, alloying addition of ruthe~ium,3 65 AI-Cr-Fe-Ni alloys, 3 65 AI-Cr-M-X-based coatings, 2: 492, 494, 495,496 AI-Cr-Nb-Si-Y system, 2: 169
Al-Cr-Nb-Ti system, 2: 83, 86 AI-Cr-Nb-Ti alloys consolidatioii of, 3: 650 creep resistance, 3 428, 430 strain hardening in, 3: 368, 369, 370, 371, 372, 372 synthesis of, 3: 649 twinning in, 3: 421 Al-Cr-Ni alloys, 3: 70 AI-Cr-Ni system, 2 13, 14 AI-Cr-Ru system, oxidation-resistant coatings, 3: 65 Al-Cr-Si system, 2 229 A1--Cr system, 1: 94, 985, 996; 2: 186187 AI-Cr-Ti alloys, twinning in, 3: 413, 428 AI-Cr-Ti coatings, 3: 579-580, 581 AI-Cr-Ti system, 2 158, 163, 164, 165, 168, 169, 170 AI-Cr-Ti system, phase diagram, 3: 581 AI-Cr-X intermetallics, 2: 186-187 AI-Cr-Y system, 2: 187 A1-Cr-Zr system, 2 185, 187 Al-Cu-Fe quasicrystals dislocations in, 3 385 mechanical properties, 3: 397 plastic deformation of, 3: 400 Ai-Cu-Li-Mg-Xr system, 2: 189, 190, 192 Al-Cu-Mg system, 1: 107-108, 410 Al-Cu-Mn-Zn system, 2: 541 AI-Cu-Ni system, 1: 832,846; 2 536, 538, 546, 556 AI-Cu-Pd system, 1: 712, 713, 719 Al-Cu-Si system, 2: 643 AI-Cu system, 1: 116, 632, 662, 1020; 2: 537, 594-595 Ai-Cu-Ti system, 1: 720; 2 159, 164 AI,CLIstructure type, 1: 385, 388, 389, 391-393, 396,428, 429 Al-Cu-Zn system, 1: 832, 836, 837, 838, 839-840, 841, 842; 2: 536, 538, 555, 556 AI-Fe-Mn-Si system, 2: 186 AI-Fe-Mo-V system, 2: 185 AI-Fe-Nb system, 1: 861; 2: 206 AI-Fe-Ni system, 1: 719, 725, 726, 846, 861, 923, 926; 2: 307, 308 Alfenol [magnetic material], 2: 306-307 AlFe alloys, 3: 496 applications, 3 514 casting of, 3: 611-614 extruded, 3: 505 fatigue crack growth in, 3: 339-341,346 fatigue properties, 3: 332 forming of, 3 639 future needs, 3: 516 high-temperature fatigue strength, 3: 332,334 joining of, 3: 510 powder metallurgy applications, 3: 654455 processing of sheet, 3: 506 suppliers, 3: 5 1 1-5 13 synthesis of, 3: 654-655, 759 AI-Fe-Ru system, 1: 719 Al--Fe-Si system, 2 176-178, 217-219, 307 Al-Fe-Si-V system, 2 178, 180 coarsening rates, 2: 186 phase relationships, 2: 183, 193 physical properties of alloys, 2: 184-185
structure of alloys, 2: 180-184 AI-Fe-Si-X intermetallic compounds, 2: 180-185 AI-Fe system, 1: 114, 662, 766 corrosion behavior, 1: 971-973 crystallographic transformation, 1: 845 crystal structure, 2: 199, 200 dispersoids in aluniinum alloys, 2: 176-178 magnetic materials, 2: 307 microstruct~ira~ stability, 1: 865, 866 oxidation behavior, 1: 985, 995 phase stability, 2 199, 200 see also Iron aiuminides AIFe, alloys, 3 496 applications, 13: 513 casting of, 3: 611-614 extruded, 3: 505 fatigue crack growth in, 3: 338-339 fatigue properties, 3: 331-332 forming of, 3: 639 joining of, 3: 510 oxidation of, 3: 572 powder inetal~ur~y applications, 3: 654 processing of sheet, 3: 506 reinforced with Al,O, fibers, 3: 583-584 suppliers, 3: 51 1 synthesis of, 3: 654 tw~nningin, 3: 41-17 Al-Fe system surface segregation in, 3: 224, 228 thermal vacancy f o ~ a t i o nin, 3: 282 see also Iron aluminides AlFe, films, 3: 786 Al-Fe-Mo-Ni alloys, 3: 65 Al-Fe-Ti system, 2: 159, 164, 165 AI-Fe-V system, 2 184 Al--Ga-La system, 1: 378 AI-Ca-Ni system, 1: 725 AI-GeNi system, 1: 725 AI-Ge system, 2: 522 Al-Hf-Ni system, 1: 727 AI-Elf system, 2: 187, 188 Al-Ir system, 1: 115, 116, 985, 996; 3: 5859 phase diagram, 3: 57 Al-Ir-Ni-Ta alloys, 3: 66 Al-Ir-Ru system, 3: 58 Alkali-ion-impact collisron ion scattering spectroscopy (ALICISS), 3: 214 Alkali-metal-antimony systems, 3: 129 Alkaii-metal-arseiiic systems, 3: 129 Alkali-metal-bismuth systems, 3: 129-1 30 Alkali-me~al~arbon compounds, 3: 250, 257-258 Alkali-metal-gallium systems, 3: 123 Alkali-nietal-i~diumsystems, 3: 123-125 Alkali metals amalgams, 3: 23 extraction of, 3: 29 Alkaline-earth metals, anialganis of, 3: 23 AI-La system, 1: 662, 664, 673, 680, 735, 738 Al,LaNi, structure type, 1: 413 AI-Li system, 1: 47, 48, 70; 2: 188-192, 510 AI-Li system, phase ~ i a g r a mca~culations, 3: 202-205 Al-Li-Zr system, 2: 188-192 Alloy design, 2: 40-42, 80, 192, 282-284 Alloy identification, 3: 502
96 I
Subject Index Alloying and brittleness, 3: 842 to stabilize crystal structure, 3: 842 Alloying additions ductility affected by, 1: 898-900; 38-40, 59-60, 80-81, 215-21 oxidation beliavior affected Alloys design of, 3: 811-855 combinatorial approach to, 3: 841 iron-based, 3: 496 nickel-based, 3: 492-494, 501-502 nio~iuni-based,3: 844-845 titanium-bas~~, 3: 477-485, 49 503 see cllso AlFe . . .; &Fe, . . .; AlNb, . . .; AlNbTi, . . .; AlNi . .; A N , . . , Alloying effects
in selection o f high-te~perature intermetallics, 2 243 superalloys, 2: 8-9 titanium aluminides, 2: 80-83 Alloy stability, 1; 38-41 applications, 1: 41-51 Alloying strategy, molybd~num-silicon alloys, 3: 488 Al-Mg system, 3 156 429 Al-Mn-Pd quasicrystals dislocations in, 3 379, 380, 382-383, 385 mechanical properties, 3: 397-398 plastic deformation of, 3: 399-400 Al-Nb system, 1: 994-995 system, 1: 982-989, 990-991; 57 system, I: 992, 993, 994; 2: 8 1-82 Mo-Si system, 1: 998-999 Al-Mg system, 1: 664, 959 AI-Mn-Nb-Ti system, 2: 86
Al-Mn-Si system,
AI-Mo-Nb-Ta-Ti system, 2: 91, 93 A1-Ma-Nb-Ti system, 2 93, 125, 126 Al-Ma-Nb-V-Ti system, 2: 91, 93 Al-Mo-Ni system, 2: 43, 46 A1 MO systeiii, 1: 696, 996-997 AlNb, alloys fatigue crack growth in, 3: 342, 346 fatigue properties, 3: 332-333 A1,Nb-based compounds, 2: 169 Al-Nb-Ni system, 2 650, 651 Al-Nb system, 2: 187, 188, 353 oxidation of, 1: 984-985, 992, 994-995 Al-Nb system, inechanical alloying of, 3: 725 Al-NbTi alloys creep behavror, 3: 303 ductility mechanisms, 3: 774 extrusion of, 3: 628, 629 fatigue crack growth in, 3: 337, 342, 343, 346
lamellar microstructure, 3: 619 mechanically alloyed, 3: 739 strain hardeniiig in, 3: 368, 369, 370, 371, 372, 372 twinning in, 3: 417 AI-Nb-Ti system, 1: 844, 857, 858, 859, 860, 862, 863, 864, 973; 2: 91-128 crystal structures, 2 94 oxidation-resistant coatings, 2: 124, 168 phase diagram, 1: 863; 2 93 SiC-fiber-reinforced coniposites, 2: 291-292 293, 294 AlNb, structure type, 1: 422 AlNbTi, alloys orthorhombic, 3: 494495 creep behamor, 3: 303 fatigue crack growth in, 3: 337 faults in, 3: 460-461 strain hardening in, 3: 369, 370, 370 A1-Ni alloys grain boundary properties, 3: 771-772 relaxations of surface layers, 3: 220-221 synthesis of by coil?bustion/reaction synthesis, 3: 737 by mechanical alloying, 3: 656657, 728-730, 739 by shock-induced reactions, 3: 741, 744 AlNi alloys, J: 494, 502 casting of, 3: 602-604 creep behavior, 3: 311-312, 494 dislocations in, 3 772 fatigue crack growth in, 3: 341-342 fatigue properties, 3: 330-331 mechanical properties, 3: 737 oxide-~ispersioii-streiig~hened~ 3: 656657 powder metallurgy applications, 3: 656-657 precipitation-hardened, 3: 3 16--317 shock consolidation of powders, 3: 739 suppliers, 3: 513 thermal conductivity, 3: 603 AlNi-based alloys, 2: 53-69 antiphase-boundary energies, 2: 270-27 1 in composites, 2: 290, 293-294 displacive phase transitlons, 1: 163 ductility, 1: 846; 2 58-60 fabrication and processing of, 2: 67-69 fatigue behavior, 2 65 fracture properties, 2 64-45 oxidation behawor, 1: 984, 989-991 oxidation resistance, 2: 57 polyerystalline, 2: 58, 60, 61 precipitation strengthening in, 2: 271-273 shape-memory alloys, 2: 538, 556 strength, 2 60-64 toughness improvement, 1: 860 in turbine airfoil design, 2 65-67 AlNi-based coatings, 3: 309 Al-Ni system ab z ~ i ~calculations io of properties, 1: 68, 69 atomistic calculations, 3: 768-772 atomic volume as function of atomic fraction, 1: 115
oxidation behavior, 1: 981-984, 989-991 efyects of alloying additions, 1: 982-984, 990-991 phase diagrams, 1: 114, 982; 2: 5, 55, 557, 607 phase diagrtm calculations, 3: 205 phase-formation sequence prediction, 2: 607, 608-609 surface rippling phenomena, 1: 6 17 synthesis in, 1: 649450 vacancy forination, 1: 566 AN,-based alloys, 2: 17-47 AlNi, alloys, 3: 310, 31 1 492-493, 501502 alloy design, 2: 40-42 polycrystalline alloys, 2: 40-41 alloying elements in, 3: 611
creep behavior, 3: 309-31 3 dislocations in, 1: 527, 528, 539; ductility improveme~t,1: 899-903 environmental embritt~ement,2: 28-29> 46 effect of alumnuin cmentration, 2: 30, 31, 32
fatigue properties, 3: 329-330, 493 forming of, 3: 639 for s e l ~ - d ~ ~ u s3:~ o300 n, fracture of, 2: 27-30 future needs, 3: 516 future research, 2: 46-47 joining of, 3: 509-510 mechanical properties, oxidatioii behavior, 1: polyerystalline, 2 40, 41 powder metal~urgyapplications, 3: 655-656 processing of sheet, 3: 506 properties, 3: 493, 515 structural applicat~o~s, sul~dationof, 3: 711-7 superplastically formed, 3: 505-506 suppliers, 3: 5 11 synthesis of, 3: 656, 736 twinning in, 3 41 1 welding of, 2: 4 4 4 5 yield strength, 2: 598 Alnico [magnetic material], 2: 307, 308, 318 Al-Ni-Ru system, 3 60 Al-Ni-Ti system, 1: 718, 727; 2: 159 Al,Ni,-type compounds, thermodynamlc properties, 1: 112-113 Alonso [solubility] plots, 1: 247 Al-Pd system, 1: 115 A1-Pd-Ti system, 2: 160 Alpha-brass polycrystals, twinning in, 3: 425 Al-Pt system, 1: 115, 696, 985, 995-996; 3: 70, 72-73 Al-Rh system, 1: 115, 116
962 Al-Ru system, 1: 985, 996; 2: 251; 3: 5556, 58, 59 phase diagram. 3: 57 Al-Ru-Ti, system, 3: 62 Al-Ta system, 1: 986, 997 AltaVista search engine, 3: 858, 859, 870 AlTi alloys, 3: 477-485, 503 casting of, 3: 591-602 dislocations in, 3: 367 extrusion of, 3: 627-630 fatigue crack growth ia, 3: 334-337, 337-338, 346, 347 fatigue properties, 3: 328, 328-329 forging of, 3: 626-627 forming of, 3: 617-638 hot-working limits, 3: 626 joining of, 3: 510 strain hardening in, 3: 368, 369, 370, 371, 372, 372 suppliers, 3: 513 toughness, 3: 426427 twjnning in, 3: 411, 413-414, 418, 424 see also Titani~maluminide
massive-like transformations, 1 : 844 oxidation resistance, 2: 8 1-82, 124 processing of, 2: 83-87 w r o ~ g h alloys, t 2: 83,85-86 Al-Ti system ab initio calculations of properties, 1: 68, 70 atomistic calculations, 3: 772-774 hydrogen-embrittlement susceptibility, 1: 973-974 neighborhood diagrams, 1: 432, 433
see aDo Titanium aluminide; . . trialuminide creep behavior, 2: 114-120 crystal structures, 2 94-95 current status of development, 2 91-93 deformation behavior, 2: 101-106 fatigue crack growth, 2: 122, 123 fracture toughness, 2: 120-122 micromechanisms of fracture, 2: 108 oxidation bebavior, 2: 124 phase equilibria, 2: 95 phase tr~~nsfoi~iations, 2: 96-101 processing of 2: 125-128 superplastic forming of, 1: 657 synthesis of, 3: 736 tensile behavior, 2: 1 0 6 114 effect of heat treatment, 2 107-1 12 effect of hydrogen, 2: 113 effect of strain rate, 2: 113 effect of temperature, 2: 112-1 13 welding of, 1: 655, 656 AI,Ti alloys, 3: 495-496 A1,Ti-based compounds DO~*-structure,2: 148-155, 167-168 dislocntions in, 1: 530-531 L1,-structure, 2 155-167, 168-169 A1,Ti structure type, 1: 430, 431
Subject Index see DO,, type compounds Al-Ti-V system, 2 124 Altmarkite [mineral name], 1: 629; 3: 21n a-Alumina scales, 3 708 Alummide coatings, 3: 309,563, 564, 565574 growth mechanisms, 3: 566 platinum-modified, 3: 64, 65, 72, 74, 564,565, 567, 568 diffusional transport in, 3: 568-570 Aluminides atomic volume as function of atomic fraction, 1: 115-1 17 B2-phase antiphase boundaries in, 3 850 characterization by TEM, 3: 848-850 composition, 3 847 decomposition of, 3: 850-851 degree of order, 3: 849 field of cxistence, 3: 847-848 stoichiometry, 3: 849-850 coatings based on, 2 57, 124, 168, 170, 491, 495 constitutionai defects, 3: in 358 creep resistance, 3: 31 1-312, 313 crystal structures, 1: 91 1 CsC1-structure aluminides, 1: 137-144 data sources, 3: 808 electrochemical studies, 1: 967, 969 formation temperatures, 3: 676 fracture behavior, 3: 226-227 heat capacity, 1: 1020-1021 heats of formation, 3: 225 mechanically alloyed, 3: 757 optical properties, 3: 232-233 oxidation behavior, 1: 981-997, 1000-~003,1005 oxide protective scales, 1: 981 prediction of formation, 3: 829-830, 831 single crystal, 2: 41, 42, 67, 153, 202 strain hardening in polycrystalline intermetallics, 3: 368373 single crystals, 3: 365-367 surface reconstructions, 3 225 surface structure, 3: 2 19-220 thin films, 3: 675-676 thermal properties, 1: 1024, 1026-1027 see aZso Cobalt . . .; Iridium Iron . . .; Nickel . . .; Ru
-
aluminide A l u m i n u ~alloys, phase diagram calculations, 3: 202-205 Alumiirium-based glasses, 1: 735, 742 Aluminium-based quasicrystals, 1: 454-457, 465,485, 742 Aluminized steel, 2: 519 Aluminum alloys, strengthen in^ of, 2 282 Aluminum-based bearing materials, 2: 595, 596 Aluminum bronzes, 2: 594 Aluminum-coated steel, 2 519 Aluminum-gold alloys, 2: 572-573 Aluminum-lithium alloys, 2: 188-1 92 phase diagram calculations, 3: 202-205 precipitation strengthening of, 3: 202 see also Al-Li system Aluminum-rich intermetallic dispersoids in alum in^ alloys, 2: 175-195, 282
Alu~inum-zinc-coated steel, 2 519 Al-V-Zr system, 2: 185, 188 A1,X compounds, synthesis by mechanical alloying, 1: 647 Al-Zn naturally occurring compounds, 1: 632 Al-Zr system, 1: 740,997; 2 133-145, 185 see also Zirconium aluminide; . t~ialu~in~de A1,Zr4 structure type, 1: 409, 41 1 Amalgamation coatings, 2: 519-521 Amalgamation reactions dental amalgam, 2: 580-582 metal-recovery processes, 1: 4; 2: 515-516 Amalgams, 3: xxxv, 21-35 with actinides, 3: 23 with alkali metals, 3: 23 with alkaline-earth metals, 3: 23 analogies of low-melting alloys to, 3: 33-34 applications, 3 29-33 barium amalgam, 2: 518 chemical analysis, 3: 30 chemical preparations, 3: 30-3 1 copper amalgam, 2: 518, 519-520, 649 dental applications, 2 575-590 gold amalgam, 2: 4; 2: 515, 519-520 isotope separation, 3: 30 lead a m ~ ~ g a m 2:,522 non-chemical applications, 3: 3 1-32 on mirrors, 2: 408, 520-521 powder metallurgy, 3: 29-30 preparation of, 3: 23, 24, 25-27 by cementation 25-26 by chemical reactions, 3: 26 by direct solution, 3: 25 by electrochemical reduction, 3: 25 by mechanochemical reaction, 3: 26 by powder meta~lurgy,3 29-30, 643 process metallurgy, 3: 29 in chemical analysis, 3 30 classification of, 3: 22-25 complex, 3: 27-29 definition, 3: 22 diffusion in, 3: 26-27, 30 IMCs formed in, 3: 27-29 isotope separation for, 3: 30 with d-metals, 3: 24 with hgh-melting nietals, 3: 23-24 with lanthanides, 3: 23 with low-melting metals, 3: 23 with nonmetals, 3: 25 silver, 3 21, 795 silver amalgam, 2: 515 silver-tin amalgam, 2 575-590 sodium amalgam, 2: 518 solubility of elements in, 3: 22 structure, 3: 26-27 thorium amalgam, 2: 5 18 toxic properties, 3: 31 use 111 analytical chemistry, 2 646, 647 zinc aIlIdlgam, 2: 518 see also Dental amalgams American Society for Metals, ~ ~ ~ a Alloy Phase ~ i a ~3: ~187~ ~ s , Amman decoration [on Penrose tiles], 1: 480,481,482 Amorphization
963
Subject Index and diffusion~1: 696, 739 hydrogen-induced, 1: 695, 734; 2: 479 interfacial-reaction-induced, 1: 695-699, 733-734 irradiation-induced, 1: 692-695, 733, 791, 808-822 dose required, 1: 8 1 1, 8 12, 815 experimental observations, 1: 808-8 17 mechanisms, 1: 817-822 mechanical~deformation-induced~1: 699-701, 733, 734 models for, 1: 820-821 pressure-induced, 1: 701-703 quasibinary/q~~asiternary systems, 1 : 721, 722, 723, 724 thermal-quench-induced, 1: 689-692 see also Order changes Amorphous alloys, 3 68 1-705 controlled crystallization of, 3: 702, 740, 754-755 cooling rates, 3: 681 crystallization of, 2: 609 crystallization temperatures, 2: 610, 612, 6l8, 625 devitrification, 3: 740, 751, 754-755 as diffusion barriers, 2: 605, 609-614, 618,625-627 examples, 1: 735 future prospects, 3: 701-703 inagneto-optical applications, 2: 449-450 model for structure, 1: 739-742 pressure effects 161 stability of, 2: 609-614 synthesis of IMCs by devitrification of, 3: 740, 7.51, 754-755 see uZso Metallic glasses Amorphous ferromagnets, 1: 439, 449-450 Amorphous phases, 1: 731-751 annealing effects, 1: 747 atomic-size difference criteria, 1: 739 coordination numbers in, 1: 739, 742 criteria for formation and stability, 1: 737-739 and crystalline compounds, 1: 749-75 1 crystallization of, 1: 745-749 and liquid alloys, 1: 679-681 production/s~thesisof. 1: 689-703, 733-734 by chemically drrven systems, 1: 696-699, 733-734 by ~rradiati~n, 1: 692-695, 733, 791, 792 by mechanically driven systems, 1: 699-701, 733, 734 by pressure, 1: 701-703 by quenching from liquid/vapor, 1: 689-692, 733 structural relaxation of, 1: 742-745 properties affected, 1: 743 structure, 1: 739-745 superplasticity, 1: 728 systems showing formation tendency, 1: 734-736 t ~ e ~ o d y n a m i c1: s , 736-737; 2: 613 ~ o ~ p h o powders us consolidation of, 3: 685 synthesis of, 3: 645, 652453, 654, 685 Amorphous silicon, 1: 732
Amphoteric elements, amalgams, 3 23-24 Analytical-chemistry applications, 2: 646-4547 Anderson Hamiltonian, 1: 212, 213, 218 Anderson lattices, 1: 213, 214 Andrade-fi creep, 3 302 Angle-resolved inverse-photoemission spectroscopy (ARIPES), 1: 135, 136 compared with ACAR, 1: 137 Angle-resolved photoe~ission,3 137 Angle-resolved photoemission spectroscopy (ARPES), 1: 41, 135, 136 compared with ACAR, 1: 137 Angle-resolved ultraviolet photoemission spectroscopy (ARIJPS), 3: 137, 139 Angular-correlati~n-of-aiinihil~~tioiiradiation (ACAR) methods, 1: 135, 136-137 compared with other methods, 1: 137 Angular momentum theory, for quasicrystals, 3: 386 Angular valence-orbital factor, 1: 238, 254, 256, 257, 264, 270, 390,420 see also Valence-electron factor Anisotropic elasticity, dislocation cores affected by, 1: 540, 541 Anisotropic elasticity theory of islocations, 3: 393 Anisotropic magnetostriction, 2: 389-390 Anisotropy magnetocrystalline, 1: 937, 939; 2: 303, 304, 305, 306, 390, 392,478 magneto-optical effects, 2 439, 442 of inechaiiical properties, 2: 66-67, 288 Anisotropy coefficients [for permanent magnets], 2: 305 Anisotropy factor, 1: 202 calculated values, 1: 206 Annealing, amorphous alloys affected by, 1: 747 Annealing twins, 3 410, 430 Anodic behavior, 2: 502 Anodic dissolution, 2: 502, 503 Anodic oxidation, 2: 503 Anomalous yield-stress behavior, 2: 7, 8, 24, 165, 243 see also Flow-stress anomaly Antibonding states, 1: 63, 64 Anti-CaF, structure, 1: 111 [footnote] Antiferromagnetic model [of magnetic structures], 1: 443 Antiferromagnetism, 1: 439, 935, 938 m chromium chalcogenides, 2: 331 co-existence with superconductivity, 2 230 in heavy-fermi on compounds, 1 : 2 1 1, 215 Antiferromagnetic phases, 3: 177, 178 Antiferroi~agnetic-to-ferromagnetic transition, 3: 98, 177, I80 Anti-fluorite structures, 3: xxxv Antimonides, 1: 626 Antimony, %atom cluster, in Ziiitl phases, 3: 116 Anti-PbC1, structure, 1: 11In Antiphase boundaries (APBs), 1: 40, 495-514; 3: 438 in Al,Ti, 2 149 in €32 structure, 3 451-452, 452, 850 in C11, structure, 3: 455. 456-457, 457, 458
in C54 structure, 3: 460
in L1, structure 444, 445, 445 in L1, structure, 3: 440, 441, 441, 442 in L2, structure, 3: 453, 454 and thermal vacancies, 3: 291 a ~ o r p ~ i z a t i oat,n 1: 816 chemical effects, 1: 496, 502-504 1: 496 c~assi~catioii, in DO,,/DO,,/1, relation, 1: 499; diffuse interface format, 1: 865 and dislocations, 1: 511-514, 543 aiid disorder-order transformations, 1: 496, 500, 780 and domain growth, 1: 864 dragging of APBs, 1: 546 electrical resistivity influenced, 1: 946-947 and grain boundaries, 1: 587, 589, 605 ideal APBs, 1: 496, 497, 514 deviations due to chemical effects, 1: 496, 502-504 deviations due to structural effects, 1: 496, 500-502 2: 201 in iron a~ui~inides, in L1, compounds, 2 19, 20, 139 nearest-neighbor models, 1: 500 nucleation at, 1: 780 residual contrast, 1: 501, 502 scattering by, 1: 946 segregation of alloying elements, 1: 504 segregation of vacancies, 1: 503-504 structural effects, 1: 496, 500-502, 925 surface energy, 1: 55. 71, 504-509 tubes see A n t i ~ h a s e - b o u n ~.~.r.ytubes in two~dimensionalAB alloy, 1: 495-496 Antiphase-b~~ndary ~AP~)-coup~ed dipoles, 1: 551, 552 Antipliase-boundary (APB) energies, 1: 5.5, 71, 504-509 ah mltict calculations, 1: 497-498, 514; 2: 58 for AINi, and alloys, 2: 23, 270-271, 283 anisotropy, 1: 497-499 B2 compounds, 3 452453,453 calculatioiis,*I: 507-508 for dispersoids/precipitates, 2: 269, 270-27 1 and dissociation of dislocations, 1: 525, 526, 531-532 effect of alloying additions, 2: 59 effect of long-range order parameter, 1: 507 experimental dete~indtion,1: 508-509 listed, 1: 505-507 €32 compounds, 1: 506 DO, compounds, 2: 153 DO, compounds, 1: 507; 3: 454,455 DO,, compound, 3: 450-451, 451 DO, compounds, 1: 506; 3 447-448, 449 L1, compounds, 1: 506; 3: 445-446, 447 Llz compounds, 1: 505; 2: 23; 3; 442, 443-443 L2, compounds, 1: 507
964 plot vs disorder-order transition temperature, 1: 508 and structural stdbility, 1: 499-500 for superalloys, 2: 284 temperature dependence, 1: 503, 540 use in fitting of Ni-A1 cross potential, 1: 83, 85 Antiphase”b0undary (APB) tubes, 1: 495-497, 512-514, 550 Antiphase boundary (APB) tubes, 3: 367 Antiphase domaiiis (APDs), 1: 495-514 in aniorphous phases, 1: 700 in Au-Cu alloys, 2: 564 in bainite phases, 1: 841 coalescence of, 1: 773-774, 865, 866 crystallography, 1: 509--510 disorder-order transition, 1: 509, 510-511 effect of composition on size, 2: 568 isostructural transformation, 1: 509-5 10 magnetic structures modelled, 1: 442, 443, 450 morphology, 1: 509-5 10 nucleation at walls, %: 567 ‘Swiss-cheese’ structure, 1: 509, 510 in two-dimensional AB alloy, 1: 495-496 Antireflection coatings, 2: 410 Antisite defects in AlNi,-based alloys, 2: 18 f o r ~ a t i o nof, 1: 564-566, 692, 795-796 phonon dispersion affected by, 1: 160 in trialuminides, 2: 157 Anti-site defects, and vacancy formation, 3: 280, 710 Anti-site disorder, 3: 759 Anti-structural bridge (ASB) mechanism, 3: 771, 773 Antis~rLictureatoms, 1: 922 Anti- winning, 3: 405, 406, 419, 431 APBs see Antiphase boundaries APDs see Antiphase domains Applications of IMCs, 3: 469-588 amalgams, 3: 29-33 beryllides, 3: 48-49 FeAl alloys, 3: 514 Fe,A1 alloys, 3: 513 historical, 1: 3 precious-metal compounds, 3: 61- 77 rare-earth compounds, 3: 97 structural general-use, 3: 501-518 hi~h-t~mperatur~, 3: 471-499 TiAl alloys, 3: 515 see also Structural applications Approximants [in quasicrystals], 1: 459, 465, 470; 3: xxxv see also Crystal . . .; Rational see also Prediction Aqueous corrosion i n ~ e ~ e t a l lcompounds, ic 1: 967-970 reactions occurring, 1: 965 Arnchno clusters, Zinti phases, 3: 114, 115, 120, 124, 125 Arc evaporation technique, 3: 665 Arc melting, beryllides prepared by, 3: 47
Subject Index Arc spraying, 1: 642 Archetypes see Prototype structures Aristotypes sec Prototype structures Arrhenius plots, 1: 979, 983, 992 Artificial intel~igence and data analysis, 3: 821-823 meaning of term, 3: 835 problems and perspectives, 3: 834-835 use of, 3: 811-839 Arsenides, 1: 626; 2: 621 Arsenopalladinite [mineral name], 1: 627, 632 As-cast Al-Ti alloys, toughness, 3: 426 As-Cd system, 3: 156 As-CO system, 3 156 As-Fe compounds, 1: 632 AsFeS structure type, 1: 388 AsZFestructure type, 1: 388 As-K system, 3 156 Aslanov’s crystal chemical model, 1: 366, 38 1 As-Ni compounds, 1: 632 AsNi structure type, 1: 250,251,261,422; 2: 309 AsNi-type compounds in structure maps, 1: 262,421,422,424, 426, 431 thermodynamic properties, 1: 112 As-Pd CompoLIilds, 1: 632 As-Pd system, 3: 156 As-Pt system, 2: 524 As-Sb compounds, 1: 632 As-Si system, 3: 156 As,Ti structure type, 1: 388 ASTM specifications, 3: 513 Astroloy, fatigue behavior, 3: 330, 331, 338 Astroloy [superalloy], 2: 6, 42 Astronautics Corporation of America, magnetic refrigerator, 3: 523 As-Zn system, 3: 156 Atokite [mineral name], 1: 627, 630, 632 Atomic diffusion, and poiiit defects, 1: 574-577, 580 Atomic displacement waves, 1: 716 Atomic environment approach, 1: 237-241, 364-366, 381 combined with predictive models, 1: 38 1 Atomic environments (AEs), 1: 237, 273, 363-382 in AIB, structure, 1: 367-370 111 CaIn, structure, 1: 370-372 in close-packed structures, 1: 280, 283, 294 irregular, 1: 367 in space filling, 1: 251 Atomic environment types (AETs), 1: 237-241, 273, 364 and coordination numbers, 1:239,262, 366, 373, 374, 375 labelling, 1: 239, 365, 366 listed for range of cja ratios, 1: 373, 375 most frequently occurring AETs, 1: 238, 239, 259, 264 number per crystal structure, 1: 255, 256, 373 see also Single-environment types Atomic force microscopy (AFM), 3: 215, 225 image of fracture surface. 3: 226, 227
Atomic images, grain boundaries, 1: 586, 581 Atomic jumps activated, 1: 786 forced, 1: 786 Atomic-layer epitaxy (ALE), 2: 422-423 Atomic magnetic moment, 3: 166 Atomic number, 1: 231, 232 atomic properties predicted using, 1: 229, 230 listed for various elements, 1: 233 Atoniic-number €actor, l:, 232, 270 in AP-AN plots, 1: 230, 233 atomic properties grouped under, 1: 23 1 Atomic packing diffusivity affected by, 1: 921-922 see also Close-packed . . .; Cubic closepacked . . .; ~ e x a g o n aclose~ packed . . . Atomic potentials, 3: 769 Atomic-probe field-ion microscopy (APF~M),3: 687 Atomic properties, 1: 228 grouped uiider factors, 1: 231 see also Atomic-number . . .; Cohesiveenergy .; E ~ ~ c t r o c ~ e m. i.c ~ l Size . .; Valence-electron factor Atomic property expressions (APES), 1: 228 structure maps using, 1: 242-243 Atomic radii close-packed crystal structures, 1: 279, 282, 298 and interatomic distances, 1: 253, 378-381 and space filling, 1: 249, 252 Atomic radii. listed, 3 214 Atomic-me di~erences/ratio Laves phases, 1: 197, 409, 763 liquid alloys, 1: 662, 677-678 MoSi,-type compounds, 1: 386 structural stability affected by, 1: 102-104, 420,422 for a m o ~ h ~ phases, us 1: 739 see also Size factor Atomic shuffling, 3: 405, 417 Atomic-sphere approximation (ASA), 1: 133 Atomic structure types, 3: 16 see also CIose-packed structures Atomic volume dependence on atomic fraction, 1: 115-117 measurement, 1: 114 see also Partial atomic volunie Atomistic models grain boundaries, 1: 600-603, 605 properties predicted uslng, 1: 77 surfaces, 1: 610-411, 620 see also Embedded-atom method Atomistic models, platinum group aluminides, 3: 56 Atomistic simulations, 3 452, 765-778 applications Mo-Si system, 3: 774-775 Ni-A1 system, 3 768-772 Ti-A1 system, 3 772-774 for free surfaces 215, 220 future developments 775-776 see also ~mbedded.atom method .$
965
Subject Index Atomization, 1: 733, 749, 773 synthesis by, 3: 644 see also Centrifugal . ., Gas . . ., Water atomization Atom location by channeling enhanced microanalysis (ALCHEMI), 3: 288, 359, 850 Atom-probe (AP) analysis, grainboundary chemistry studied using, 1: 591, 595, 596 Atom-probe field-ion microscopy (APFIM), 3: 64,66 Atarns, coordination of, 3: 7 AuBe, structure type, 1: 410 Au-Bi naturally occurring compounds, 1: 632 AuCd structure type, 1: 261, 422 Au-Cd system, 1: 102, 662, 830, 831; 2: 538 Au-CO system, 1: 705 Au(Cr,Mn,V,Ti) systems, 1: 727 Au-Cs system, 1: 669-670 Au-Cu alloys surface energies, 3: 226 surface properties, 3: 2 17-21 8 Au-Cu-Fe system, 3: 65 Au-Cu-Ni system, 2 570, 571 Au-Cu-Ni-Zn system, 2: 560, 572 Au-Cu-Pd system, 1: 721, 726 Au-Cu-Pt system, 1: 726 AuCu structure type, 1: 119, 261, 422, 426; 2: 308; see also L1, type compounds AuCu, structure type, 3: 88, 89, 217 see also L1, type compounds Au-Cu system ab initio calculations of properties, 1: 67-68 de-alloying in, 1: 970 gold alloys, 2: 561-562, 563 liquid alloys, 1; 662 metastable phases, 1: 706 naturally occurring compounds, 1: 632 surface-induced ordering, 1: 614-615 thermal expansion behavior, 1: 1023 twinning in, 1: 844 vacancy formation in, 1: 565 AuCu, structure type, 1: 268, 408, 409, 430, 43 1 , 444 Au-Cu-Zn system, 1: 836, 840; 507, 538 Au--Gd system, 1: 703 Auger electron spectroscopy (AES), 1: 591-594,611,897,901,1001,1003; 2: 27, 28, 214; 3: 214, 780, 785 limitations, 1: 894 Au-Ge system, 1: 703,704 Auger effect, 3: 140 Augmented plane-wave (APW) method applications, 1: 133; 2: 219 see also F u ~ l - ~ o t e n tlinear~al augmented plane-wave (FLAPW) method. Linear augmented planewave (LAPW) method Augmented spherical-wave (ASW) method, 1: 31, 60 applications, 1: 41, 64, 65, 68, 69, 70 Au-Hg system, 1: 632; 2: 515-516, 518 Au-L~system, 1: 696, 735 Au-Nd system, 1: 703 Au-Ni system, 1: 69
Au-Pr system, 1: 703 Auricupride [mineral name], 1: 627, 632 Au-Sb system, 1: 703, 704 Au-Si system, 1: 691, 704 Au-Sm system, 1: 703 Au-Sn system, 1: 662, 703, 704, 959 ALI-T~system, 1: 703, 704 Au-Ti system, 2: 559, 569-570 Automotive applications, 3: 61-62, 472, 503, 514,515, 599-601 Automotive parts applications, 2: 45, 88, 543, 544 Au-Y system, 1: 696, 703 Au-Zn system, 1: 662, 673, 678; 3: 156 Au-Zr system, 1: 696, 740 Avalasiche photodetectors/photodiodes, 2: 341, 342,420-421,425 Awaruite [mineral name], 1: 627, 631, 632 Axial ratios A1,Cu-type compounds, 1: 392 enthalpy of formation affected AI,Ni,type compounds, 1: 113 AsNi-type compounds, 1: 112 Hume-Rothery phases, 1: 106 lnNi,-type compounds, 1: 112 Laves phases, 1: 109 MoSi,-type compounds, 1: 386, 387 stability of Hume-Ro~heryphases affected by, 1: 105-106 see also Lattice cla ratio B f type compounds colored compounds, 3: 238 metallic vs covalent/ionic bonding, 1: 183, 184 Pearson symbol for, 3: xxxi, xxxiii, 13 space group for, 3: xxxi, xxxiii, I3 see al,ro ClNa structure type, NaCl structure type 8 2 aluminides chemical composition, 3: 847 field of existence, 3: 847-848 B2 pliases characterization by TEM, 3: 848-850 decomposition of, 3: 850-851 degree of order, 3: 849 stoichiametry, 3: 849 B2 type coinpounds ab initio calculations, 1: 64 APB energies. 1: 506 cell occupancy, 3: 355 colored compounds, 3: 236, 237, 238 constitutional defects in, 3: 358 crystallographic elements and parameters of twin modes, 3: 407 crystal structure, 1: 62, 520, 856, 858, 896; 2: 54, 55, 62, 200 defects/dis~ocations in, 1: 535-536, 542, 798 diffusion in, 1: 765-766 disordering by milling, 3: 759 displacement vectors of stacking-faultlike defects, 3: 463 effect of boron, 3: 59 effect of third alloying element on mechanical properties, 3: 288 flow-stress anomaly, 1: 536 fracture of, 1: 904-906 grasn-boundary structure model, 1: 603 ion tracks in, 3: 272 iridium aluminides, 3: 58-59
Pearsoii symbol for, 3: xxxi, xxxiii, 13, 439 point defects in, 3: 354 quasibinary/q~dsiternarysystems, 1: 713-719 ruthenium alummides, 3: 55-58 segregation at surfaces, 3: 224-225 solution hardening of, 3: 354, 355. 356 space group for, 3: xxxi, xxxiii, 13, 439
temperature, 1: 614 surftice structure, 3: 219-220 surface terminations, 1: 6 15, 6 16 thermal vacancies in, 3: 275-276, 289290 twinning in, 3: 416 vacancy foimation in, 3: 275 vacancy migra~ion,1: 575, 576 see aZso AIFe; AlFe3 alloys; ClCs structure type, CsCl structure type B2/B32 competition, 1: 66-67, 856 B8, structure, 1: 858 B8, type compounds, disorder in^ by milling, 3: 759 B9 type compounds, metallic vs covalent/ ionic bonding, 1: 183, 184 BXl structure, 1: 62, 575, 577, 798 B19 structure, 1: 715, 716, 717 B l 9 structure, 1: 716, 717 B32 structure, 1: 62, 856
space group for, 3: xxxi, xxxiii, 13 see also NaT1-type
...
B-1900 ~s~peralloy], 2: 6 Babbitts, 2: 592-594 Babelfish translation program, 3: 870 Ba-Hg system, phase diagram, 3: 23 BaHg,, structure type, 1: 398, 433 Baikov Institute, 3: $24 Baiii deformation, 1: 827, 828 Bain transformation, 3: 199 ~ a i ~ i t1e: ,835 compositional alteration, 1: 840 crystal habit plane, 1: 838 crystal structure, 1: 836 growth kinetics, 1: 839-840 interfacial structure, 1: 841 microstructure, 1: 837-838 nucleation and growth, 1: 838-839 shape changes, 1: 836-837 Bainite transformations, 1: 835-842 and atomic disordering, 1: 841 in ceramics, 1: 842 in ferrous alloys, 1: 835 models describing, 1: 841-842 in ordered intermetallic alloys, 1: 836 Ball-milling, 3: 652. 723, 724, 727 amorphization by, 1: 699-700 disordering by, 1: 788 synthesis using, 1: 647 Ba-Mg system, 1: 667, 668 Bandpass [optical] filters, 2: 410--41I Band structure aluminides, 1: 137-144; 2 54
966 ~alcu~ation, 1: 127-144 basis sets, 1: 133-134 computational imethods, 1: 130-1 34 for hydrides, 2: 479, 480 other factors affecting^ 1: 134 shape approx~niationsused, 1: 133 thcory, 1: 127-130 experimental probes, 1: 135-137 interpretation, 1: 134-135 Banerje~Urban[irradiation~induced orde~ng/disorde~ngl model, 1: 802 BaPb, structure type, 1: 407408, 430, 431; 3: 88, 89 Barrier layers, 2: 410, 522 Barrier layers, sulfidation slowed by 710 Barium--iixmury phase diagram, 3: 23 Baskes-Angelo -Moody (BAM) interatomic potential, 3: 769, 770,771,772 Bask~tweavestruct~~re, AI-Nb-Ti alloy, fatigue crack growth in, 3: 337, 338 Bat tcries molten-salt, 3: 251 nickel-metal-hydrid~ batteries, 3: 107- 108 recycling of mercury from, 3: 33 silver chalcogenides in, 3: 250 Battery electrodes, 3: 107, 794 BaZn, structure type, 1: 413 B,BaNi, structurc type, 1: 415 B.c.c see Body-centred cubic . , B,CeCo, structure type, 1: 412, 413, 415, 416 BCr structure type, 1: 261, 364, 372, 373, 375,422 separation in structure maps, 1: 262, 421,424,426 Beam transport magnets [in particle accelerators], 2: 377 Bearing materials aluminum~based,2: 595, 596 cadmium-based, 2: 596 cobalt-based, 2: 596-597 copper-based, 2: 591, 594-596 lead-based, 2 , 592, 593, 594, 596 nickel-based, 2 596-597, 598-599 tin-based, 2: 592, 593, 594 Beer-Lambert law, 3: 145 Belov's classification [of structure types], I: 403 Bergnian, B. G., 3: xlii Bergmaii phase, 1: 475, 483; 3: xxxv Berthollet, Claude Louis, 3: xliii Berthollides, 3: xxxv Berthollide compounds, 1: 273 solid solubility, 1: 248 structure mapping, 1: 245 Beryllides, 3: 37-51 applications, 3: 48-49 hydrogen storage materials, 3: 48 oxidation resistant coatings, 3: 48 refractory materials, 3: 48 creep, 3: 43-44 crystal structure, 3: 41, 42 data sources, 3: 808 diffusion behavior, 3: 46 ductil~brittletransition temperature, 3: 43 elastic properties, 3: 42 hardness, 3: 43 h~avy-fermioncompounds, 3 46
Subject Index htgh-temperature strength, 3: 42-43 MBe-type compounds, 3: 40 MBe2-type compounds, 3: 40 MBe,-type compounds, 3: 40 MBe,-type compounds, 3: 40 MBe,,-type compounds, 3 40, 42, 42 MBe,,-type coiiipounds 39, 41, 42 MBe,,-type compounds, 3: 38 M,Be,,-type compounds, 3: 40, 42 magnetic properties, 3: 45-46 mechanically alloyed, 3: 757 mechanical properties, 3: 42-44 oxidation behavior, 1: 999-1000, 1001, 1005 oxidation of, 3: 44-45 oxide protective scales, 1: 1000; 3: 48 physical properties, 3 38, 39,40, 45-46 properties, 3: 42-46 as reinforcing phases, 3: 49 structure, 3: 38-42 synthesis methods, 3: 46-47 liquid reactions, 3: 47 liquid-solid reactions, 3: 47 mechanical alloying, 3: 47 powder reactions, 3: 46-47 thermal spraying and sputtering, 3: 47 temperature dependence of resistivity, 3: 45 ternary and higher-order, 3: 49 thermal properties, 3: 44 toughness, 3: 43 Bery~~um electronic structure, 3: 37 IMCs, 3: 800 toxicity, 3: 49 Beryllium oxide, reactions, 3: 44 Beta-brasses, 3: xxxv temperature effect on color, 3: 240 Beta-bronze, 3: 874, 875 Beta-manganese phases, 3: xxxv Beta-tungsten phases, 3: xxxv Bethe lattice, 1: 32 Be-Th-U system, magnetic ordering in, 1: 215, 216 B e Z r system, 1: 740, 742 B-Fe-Mo system, 3: 161 B-Fe-Nd magnets, 3: 97-98, 102-103 B-Fe-Nd system, 2: 314, 318 B-Fe-Si system, 1: 749 BFe structure type, 1: 261, 364, 372, 373, 375,422 separation in structure maps, 1: 262, 421, 424,426 &Fe system, 1: 735, 742 Bhat~a-Thorntoncorrelations, 1: 665 Bhatia-Thorntoii partial structure factors, 1: 663 Bi-Cd-Hg-W alloy, 3: 31 BiF, structure type, 1: 430,43 1,444,3: 88, 89 see also DO, type compounds Bi-In system, 3: 161 Bi-K system, 1: 662 Bi-Li system, 1: 662 Bi-Mgsystem, 1: 662, 664, 669, 670, 673, 676, 677 Binary defect adamantane-structure types, 1: 353-354 Binary normal adamaiitane-structure types, 1: 352-353 Bi-Na system, 1: 662, 673
Binder materials, 3: 512 Binding energy measurement of, 3: 136 in PES spectra, 3 137 Biographical sketches, 1: 3 [footnote], 6, 7, 8, 13 Biographical sketches (on Internet), 3: 860,361462 Bi-Pb system, 3: 161 Bi-Pd naturally occurring compounds, 1: 632 Birch equation, 1: 199, 208 Birefringence, 2 330 Bi-Sb system, 2 464 Bismanoi [magnetic material], 2: 309 Bismuth, ion tracks in, 3: 268 Bismuth, recovery of, 2: 517 BismLIth~des,1: 626 Bismuth strontium calcium copper oxide (BSCCO) superconductors, 2: 366 Bi-Te system, 2: 462-464 Bi-Zn system, 1: 662, 679 Blade cooling, 3: 541, 542 Bloch's theorem, 1: 59, 128, 129, 198, 319, 485, 942 B-Mo-Si system, 3: 487-490, 487 phase diagram, 3 487 B-Ni system, 1: 735, 741, 742 Body-centred cubic (b.c.c.) alloys dislocation in, 1: 535-537 ground-state diagram, 1: 851 ground-state properties, 1: 41-44 see also B2 compounds Body-centered cubic (b.c.c.1 alloys/ compounds deformation twinning in, 3: 416-417 lattice parameters, 3: 34.3 surface structure, 3: 219-220 thermal vacancy formation in, 3: 275, 279 see aEso €32. ., DO, type compounds Body-centred cubic (b.c,c.) denvative structures, 1: 309-341 I framework, 1: 3 17-325 nets, 1: 336-341 notations used, 1: 309-317 see also CsCI-structure; I franiework Body-centred cubic (b.c.c.) lattice ab initio calculations, 1: 62 combination with f.c.c, structure, 1: 169 ordering maps, 1: 35, 37 packing density, 1: 281 stability in transition metals, 1: 58 Body-centered cubic (b.c.c.) metals, strain hardening in, 3: 362, 363 Bolting, high temperature, 2: 5 19 Bond characteristics, 1: 11 Bond characterization from crystal morphology, 1: 167-192 from theimodynamic properties, 1: 91-122 Bond charge (BC) model, 1: 152-153 Bond directionality, 1: 238, 281, 282, 291 dislocation cores affected by, 1: 540-54 1 at grain boundaries, 1: 604 Bonded magnets, 3: 102 Bonding, 1: 19-222 Wusion affected by type, 1: 762-764 interaction between intermetallic compounds determined by, 1: 712
Subject Index plasticity affected by, 1: 919-921 Bonding-charge mode. 3:l 179 Bonding states, 1: 63, 64 Bond loosening, premelting phenomena explained by, 1: 110 Bond valence concept, 1: 354 Bookmarl~(on Internet), 3: 876 Booksellers on Inteinet, 3: 860 Boole, George, 3: 876 Boolean logic, 3: 876 Internet searching using, 3: 858 Borane cage clusters, 3: 114 Boride coatings. 2: 507 Bonde-reinforced sibeides, oxidation behamor. L: 1008 Boiides, 1: 413-414 Borishanskite [mineral nnme], 1: 627, 630. 632 Born-,Wayer potential, 1: 152, 154 Born-von Kdrinin model, 1: 152. 156 see also Force-constant models Boron creep behavior affected by, 2 206 ductility affected by, 1: 656. 899-904, 905,906,924; 2: 17,29-30, 59,204 adverse effect of sulfur, 1: 904 fracture affected by, 1: 896 interstitial boron in AIM,, 1: 85-86, 899- 900 .. prenous-metal compounds affected by, J: 59, 68 recrystallization ot AI-TI allogs affected by, 3: 625 segregation dt grain boundaries, 1: 593, 594, 638, 899. 901,904 solubility in Ni,X, 1: 593, 594 Boron-doped AI-Ni-based alloys ductility, 1: 656, 786 787, 899-904, 905-906, 924, 2: 17, 29-70 free-surface seaeeation. I: 620 gram-boundary stftlctures, 1: 587, 588, 602 603, 605 oxidation resistanE. 1: 9x9 Boron oxide, as flux for metdlhc glasses, 3: 687-684 Boundary scattering [in thermal conduction], I: 1028 Bragp formula, 3: 166 Riagg scattering, magnetic structures, 1: 440 Bragg-Williams appioximation, 1: 25, 26, 39, 502, 570,610,619. 785,945 Bragg-Wilhms model bee GorskyBragg-WilliamP (GBW) inethod/ model Bragg-Williams LRO model, I: 775, 777, 799 P-Biass [CuZnJ, 1: 3, 5, 101, 119, 319, 320 de-alfoying phenomenon, 1: 970 diffusion in, I: 762, 764-765 drslocatioiis in, 1: 536, 540 in electrochemrcd electrodes, 2 646 vacanucs in, 1: 562, 564 y-Brass [Cu,Zn,]. 1: 101, 113, 322. 325, 141, 627, 703 +Brass [CuZnJ, 1: 101 Brass inclusion [in pyrites], l:, 627, 632 Brasser, ree u1.10 Beta-brasses Bravais lattices, 1: 167, 169, 171, 179 splittings of I lattices. 1: 322
Bravais-type crystal Ttructures, number of atoms per unit cell, 1: 256 Brdzing, 3: 510 Blazing alloys, 1: 3, 2 522 difiusion barriers used, 2 604 Bieatlring deforni,ition, 1: 159 Breinsstrahlung isochromat spectrosiopy (BIS), 3: 148-150 constant initial state mode, 3 150, 150 isochromat mode, 3: 148, 149 spectral mode, 3: 148, 149 Brem's'strahhmg iadialion, 3: 138, 148 Brewer, Leo, 3: xliii Rridgman method, 3: 542, 546 547, 602 Bridgman [single-crystal groh th] method, 1: 643. 2: 67, hX, 395 Bridgman-Stockbarger method, 3: 542, 543 Brilliance, meaning of term, 3: 242 Bi tllouin scattering, amorphous phases studied usmg, 1: 695, 814 Brillouin Lone, 1: 100, 101, 106 notation used for symmetry points, 1: 138 Brilloum-zone sampling, I: 132-133 for phonon dispersion curtes, 1: 150 special-points method, 1: 132 tetrahedron method, 1: 132 Brittle fracture E2 compoundb. 1: 904 L1, compounds. 1: 897 modeis for, 1: 907-909 Brittleness, 1: 895 and alloying, 3: 842 of C15 Laves phases, 3: 415 of ceramics, 3: 341 criteria, 2: 166-167 dynamic effect, 2: 31 of ;>-TlAl,3: 421 at gidin boundaries, 2 27-28 in titaniuin alumirndes, 2 77-78, 147 in lrialuminide compoundc, 2: 148, 163, 166-167 see aiw Cleavage , Duchlity enhancement, Intergrmular fracture Brittle-to-ductile traimtion temncrature 1: 911. 918 919 high-temperature mtennetallics, 2: 244, 245, 246 iron alummides, 3: 61 1 molybdeiium disihcide, 3: 486 nickel alummidc, 3: 502 silicides. 2: 218 titmium alummide, 3: 503 Btoken-bond model, 3: 224. 226 Bronze, 1: 3 Bronze bearing alloys, 2: 594-596 Bronze process [for mpercoilductors], 2: 361 Browsing the Internet, 3: 857, X76 Broyden [mixing] method, 1: 197 Brunnei-Scbwarzenbxh atomicenvironment definition, i:237, 364 Buckyballs, 3: xxxv Bulk metallic glasses, 3: 681 682 alloy sy.stems, 3: 682 composition range.?, 3: 682, 686, 6S8, 691 synthesir of, 3: 682485 Bulk modulus, 3: 241
967 for AI,Ti, 2: 149 calculated values, I:206, 885 composition dependence, 1: 8x7 determinatioii by first-principle calculations, I: 200, 201 for polycrys~als,I: 202, 206, 209. 885 estimatioii from single-crystal constants, 1: 883 884, 885 Burgers circuit, 3: 381 Burgers vectors, 3: 364 Buried heterostructure (BH) lasers, 2 336, 338 Buried kayers [of silicides], 2: 330 Burner rig tcsts, 2: 495 'Burst' effect anomaly. 2: 399, 400 Buschow model, 2: 611 C l structure. 2: 212 C1 type compnuids colored compounds, 3: 236. 237, 238 Pearson symbol for, 3: xxxi, xxxiii, 13 space group for, 3 xxxi, xxxiii, 13 see nlro CaF, (lluorite) structure type C1I, structure. I: 62, 520. 539, 575; 2 212 displaceinent vectors of stacking-faultlike defects, 3: 463 elastic properties of compounds, 1: 877 Pearson symbol for. 3: xxxii, xxxiii, 14, 439 space group for, 3: xxxii, xxxiii, 14,439 stacking faults in: 3: 454458 unit cell, 3: 455, 456 see crisa MoSi, structure type C l lb/C40/C54competition, 1: 67 C14 type contpounds beryllides. 3: 49 Pearson symbol for. 3: xxxi, xxxiii, 14 space group for, 3: xmi, xxxiii, 14 see also MgZn, structurc type C15 type conipounds beryllides, 3: 49 Pearson syinbol for, 3 xxxi, xxxiii, 14. 439
point defects in. 3: 354 solution hardening of, 3 356 space group for, 3 xxxi, xxxiii, 14, 439 twinning in. 3: 415 see iihv Cu,,Mg structure type C15 structure, 2: 395 C15, type compotinds Pearson symbol for, 3: xxn, xxxiii, 14 space group for, 3: xxxi, xxxiii, 14 see also AuBe, structure type CI 5h structure see Cu,Mg structure type Cl6 type compounds Pearson symbol for, 3: xxxii, xxxiii, 14 space group for, 3: xxxii, xxxiii, I4 see ulso AI,Cu structure type C16 structure stabilizing factors, 1: 392 see also AlzCustructure type C3X type compounds Pearson symbol for. 3: xxxii, xxxiii, 14 space group for, 3: xxxii, xxxiii, 14 twinning in, 3: 418 C40 type compounds displacement vectors of stacking-faultlike defects, 3: 463 Pearson symbol for, 3: xxxi, xxxiii, 14. 439
968 space group for, 3: XXXI, xxxiii, 14, 439 stacking faults in, 3: 458-460 C40 structure, 2: 212 C54 type compounds displacement vectors of stacking~faultlike defects, 3: 463 Pearson symbol for, 3: xxxii, xxxiii, 14, 439 space group for, 3: xxxii, xxxiii, 14,439 compounds, 3: 832 CaAuSn structure type, 3: 95 CaBe,Ge, structure type, 3: 90, 92 Cable-in-conduit conductors, 2: 361, 373, 380 Cabriite [mineral name], 1: 627, 632 CaC, structure type, 3: 89, 89 Ca-Cu system, I: 95 CaCu,, structural formula for, 3: 6 CaCu, structure type, I: 41 1-434 rare-earth compounds, 3: 94, 95, 96 Cadmium amalgams, 3: 22 Cadmium-based beanng alloys, 2: 596 Cadniiuin chalcogeiiide thiii films, 2: 5 11 Cadmium-sharing clusters, in Zintl phases, 3: 122, 122 Cadmium telluride, solubility in mercury, 3: 28 CAESER test program, 3: 601 CaF, (fluorite) structure type, colored compounds, 3 74-77,238 CaF, structure type, 1: 117, 388 in structure maps, 1: 389, 428, 429 see also Anti-CaF, structure type Cage structures, 3: xxxvii, 8, 9, 126, 127, 254-255 Ca-Hg system, 3: I56 CaIn, structure type, 1: 364, 370-372, 373, 375 Calciothermic reduction processes, 1: 650 CaLiSi, structure type, 3: 92 Calorimetry, long-range-order changes studied, 1: 777, 778 Ca-Mg system, 1: 662; 3: 156 Canada, platinum group metal reserves, 3: 80 CANDU nuclear power reactor, 1: 791; 2: 133 Caiined extrusion, 3: 627-628, 639, 648, Ca-Pb system, 3: 156 ,.we Lattice c/a ratio Carbides, 1: 299-302, 304, 396 electrical applicatioiis, 2: 470 electrochemical behavior, 2: 506 see also CFe; . . .; CFe,W, structure type Carbriite [mineral name], 1: 627, 632 CaRh,B, structure type, 3: 92 Car-Parrinello method, 3: 197 Ca-Si system, 3: 157 Cardiovascular applications, 2: 554 Castability, i ~ e a n i ~ofgterm, 3: 597 Cast alloys iron aluinmides, 2: 201-202 superalloys, 2: 7,8 titanium aluminides, 2: 83-85 c/n ratio
Subject Index Casting, 3 475, 503-504, 591-616 alternative technologies, 3: 593-594 centrifugal, 3: 592, 608, 613 challenges, 3 591 and compositional control, 3: 606-607 equiaxed investment, 3: 609 Exo-Melt process, 3: 604-606 of F e 4 1 alloys, 3: 611-614 and grain size, 3 609 hydrogen evolution during, 3: 61 1 impurities in, 3: 606-607 of ingots, 3: 592 investment, 3 592-593, 608-609 and macroporosity, 3: 610 microstructure affected by, 3: 594-596, 609-610, 613 models for, 3 598 of NiAl alloys, 3: 602-604 of Ni,Al alloys, 3: 604-61 1 oxidation in, 3: 607 production readiness NiAl alloys, 3: 604 TiAl alloys, 3: 598-601 sand, 3: 607-608, 612-613 tape, 3: 507, 655 of TiAl alloys, 3 591-602 of turbine components, 3: 601, 609 Casting alloys, composition, 3: 596-598, 610-611, 613-614 Catalysts, precious metal compounds, 3: 78 Catalytic app~ications,1: 15; 2: 647 Catalytic converters, 3: 54, 514 Catastrophic degradation, 1: I000 see also Pest effect Catastrophic oxidation of molybdenum silicides, 3: 486 of niobium silicide composites, 3: 658 see also Pest phenomenon Cathodic charging [by hydrogen], 2: 477, 486 Cauchy pressure, 1: 79 Cauchy pressure criterion, 3 215, 241, 772, 774 typical values, 3: 241 Cauchy relationships, 1: 874 listed for various mtermetallics, 1: 875 Ca-Zn system, 3: 157 CdCl, structure type, 1: 389, 428, 429 Cd-Cu system, 2 502-503 Cd-Hg system, 2: 652 Cd-Hg-Te system, 2: 418-419; 3: 31-32 Cd-In system, 2: 538 Cdl, structure type, 1: 388, 389, 428, 429 Cd-Mg system, 3: 157 phase diagram calculations, 3: 205 Cd-Ni system, 2: 519, 520 Cd-Pd system, 3 157 Cd-Pt system, 3: 157 Cd-Sb system, 3: 157 P-T phase diagram, 3: 155, 161 P-T-X phase diagram, isobaric crosssections, 3 155, 160, 161 Cd-Te system, 1: 676 CeAlSi, structure type, 3 92 CeCd, structure type, 3: 88, 89 CeCo, structure type, 1: 416 CeCoAl structure type, 3: 91, 93, 93 CeCu, structure type, rare-earth compounds, 3: 88, 89,91, 93, 93, 95 CeCu, structure type, 1: 413
Ce-Fe system, 2 647, 648 CeFeSi structure type, 3: 91, 93, 93 Ce-Gd system, 1: 407 Ce-Ho system, 1: 407 Cell occupancy, B2 compounds, 3: 355 Cementation, amalgams prepared by, 3: 25-26 Cemented carbides, 3: 643 CeNi, structure type, 3: 88, 89 CeNiSi, structure type, 3: 90, 92 CeNi, $i4 structure type, 3 100 Centered mesh (in reconstruction of surface layers), 3: 217 Centrifugal atom~~dtion, 3: 644 Centrifugal casting, 3: 592, 608, 613 Ceramic matrix composites, compared with intermetallic composites, 2 288 Ceramics, and brittle inte~etallics,3 347 Ceramics, transfo~ationsin, 1: 835, 842 CeRhAl structure type, 3: 91, 91, 93 CeRhGe, structure type, 3: 92 CeRhSn, structure type, 3: 92 Ce-Sm system, 1: 407 Ce-Tb system, 1: 407 C-Fc--Ni system, 1: 830; 2 557 CFe, structure type, 1: 385, 398-399, 430 CFe,W, structure type, 1: 385, 394-396 Cha-Cha program, 3: 864 Chain structure types, 3: I7 Clialcogenides h~~h-pres$ure phases, 3: 756 meaning of term, 3: xxxvi mechanically alloyed, 3: 757 prediction of ABX, compounds, 3: 825,826 AB,X, compounds, 3: 8254327,828 see also Selenides; Sulfides; Tellurides Chalcopyrite [mineral name], 1: 353, 359 structure, 2: 324 Chalcopyrite-type compounds, 2: 324, 330 Chalcopyrites, 3: xxxvi, 672 chdlmers-Martius criterion, 1: 542 violations, 1: 535, 543 ~harged~particle tracks, 3: 263 see also Ion tracks Charge transfer, I: 117, 144 close-packed alloys, 1: 282 compou~ds,3: XXXVI embrittlement caused by, 1: 907 hydrides, 1: 298-299 Chelikowsky plots, 1: 246-247 Chemical analysis, amalgams used, 3: 30 Chemical bonding, interaction between inte~etalliccompounds determined by, 1: 712 Chemical composition dental amalgam, 2: 584 and dislocation cores, 1: 539-540 elastic properties affected by, 1: 886-8 87 electrical conductivity affected by, 1: 953-9 54 gold alloys, 2: 570, 571 plasticity affected by, I: 922-924 shape-mem~ryalloys, 2: 538, 557 Tribaloy alloys, 2: 597 Chemical disorder, 3: 245 Chemical disordering, 1: 693, 694, 701, 706, 860 and irrad~ationeffects, 1: 817, 818
Subject Index Chemical enthalpy excess, 1: 688 Chemically etched ion tracks, 3: 263, 264, 268 Chemical ordering, in quasicrystals, 1: 463 Chemical properties, 1: 15-1 6 Chemical reactions, amalgams in, 3: 3031 Chemical twinning model, 1: 741, 742 Chemical vapor deposition (CVD), 1: 652; 2 291,328,360,409,491,628; 3: 667-670 advantages/disadvantages, 1: 652; 3: 668 applications, 3: 669 disadvantages, 3: 669 examples of thin-film IMCs, 3: 669 Chem~sorption,2: 482 Chemistry, meta-sites on Enternet, 3: 865 ChemWeb search engine, 3: 858, 859 Chen-Yan-Liu (CYL) interatomic potential, 3: 772, 773 Chevrel, Roger, 3 xliii Chevrel phases, 3: xxxvi data on, 3: 833 Chi phase, 1: 322; 3: xxxvi C h ~ m n e ~ l a d dphases, er 3: xxxvi Chinese Academy of Science, Institute of M e ~ ~ ~ l3:~ 824 gy, Chirped [diffraction] gratings, 2: 415 Chisel toughness Cr-Ti alloys, 2 242 high-temperature mtermetallics, 2 245, 248 ruthenium aluminide, 3: 55 tools for meas~rement,2: 247, 248 Toughness see also Toughness Chomferide [mineral name], 1: 627, 632 Chro~ium ductility of A N , agected by, 2: 40 oxidation behavior affected by, 1: 984-986, 989, 993 see also Cr- . . . Chromium-based alloys, coatings used for, 2: 228 Chromium chalcogenides, as semiconductors, 2: 330-3 32 Chrysalis Technologies, Inc., 3: 507, 511 Chrysoberyl, 3: 45, 48 Chudley-Elliott model, 3: 251, 253, 256, 258 Ghuk Kam [gold] jewelry, 2: 569 CIELAB color scheme, 3: 232 Cinnabar [mineral], 3: 21 Cinnnbar-structure compounds metallic vs covalent/ionic bonding, 1: 183, 184 separation in structure maps, 1: 422, 426 see also HgS structure type Class, meaning of term, 3: 835 Class c o ~ p a c t n ~ sas s , hypothesis, 3: 834 Classical electrostatic interaction, 1: 78 Classification adamantane structures, 1: 343-344 of algorith~s,3: 822 of ainalgarns, 3: 22-25 antiphase boundaries, 1: 496 crystal faces, 1: 170
crystal habits, 1: 184-189 of crystal structures, 3: 3-4 gold jewelry alloys, 2: 559-560 meaning of term, 3: 835 of nanostructures, 3: 750 of phase diagrams, 3: 185 phases, 1: 11 stimture types, 1: 403 Classification rules, 3: 812 prediction by, 3: 8 12-8 13 Classification scheme, 3: 835 Clathrates meaning of term, 3: xxxvi Zintl phases, 3: 130-131 Clausius-Clapeyron equation, 2: 534 ClCs-structure, 1: 336, 422 Al,Ni, compounds, 1: 112, 113 aluminides~1: 116, 137-144 brittle vs ductile behavior, 1: 137 phase transitions, 1: 163 Hume-Rothery phases, 1: 101 Interatomic distance relationsh~p,1: 253 manganese atloys, 1: 444, 445 and order-disorder phenomena, 1: 119 and polymorphism, 1: 118, 261 separation in structure maps, 1: 420, 421,422,424,426 see also B2 structure; Body-centred cubic , . Cleavage fracture, 1: 895-909 examples, 1: 895 meaning of term, 1: 895 Cleavage process, and surface freeenergy, 1: 171 ClFPb structure type, 1: 388 Climb dissociation, 1: 546 ClNa-structure density-melting-point plots, 1: 268 interatoi~icdistance relationship, 1: 253 metallic vs covalent/ionic bonding, 1: 183, 184 polymorphism, 1: 261 separation in structure maps, 1: 262, 420, 42 1, 422,424, 426 ClNa-type Zintl phases, 1: 110 Clebsch-Gordan coefficients, 3: 167 Close-packed stacking planes, in martensite, 2: 531 Close-packed structures, 1: 277-306, 403-417; 3: 9-10, 12n ab initio calculations, 1: 61 complex CP alloys, 1: 291, 292 cubic CP interstitial alloys, 1: 299-302 cubic CP structures, 2 270, 280, 287-29 1 equally sized atoms, 1; 406-408 homologous series, 1: 293-294, 405 homometric structures, 1: 292-293 interstitial CP alloys, 1: 303-304 nearest-neighbors, 1: 280 notation used, 1: 306 point symmetry of ordered phases, 1: 294 stacking sequences, 1: 277-282 therma~vacancies in, 3: 275, 279, 289 unequally sized atoms, 1: 408-414 see also Cubic-packed . . .; Hexagonal close-packed structures, A1 5, l,, structure types
969 Closo clusters, Zintl phases, 3: 114, 115, 123, 124, 125, 127, 127 CI,Pb structure type, I: 388, 389, 429 Cluster algebra, 3: 193 Cluster Bethe lattice method (CBLM), 1: 24, 32 applications, I: 41 Cluster-expansion free energy, in calculation of phase diagrams, 3: 194196 Cluster formation, close-packed structures, 1: 284, 287 Cluster functions, 3: 193 Cluster glasses, 1: 440 Cluster variation method (CVN), 1: 30, 38, 51 I, 610,775 for construction of phase diagrams, 3: 191-194, 201, 845 compared with mean-field model, 3: 195 example of application, 3: 202-205 dynamic displacenients in, 3 201 as multi-site mean-field model, 3: 195 static displacements in, 3: 201 Clusters Mackay, 3: 398 in Ni-base alloys, 3 286 vacancy, 3: 286 in Zintl phases arachno, 3: 114, 115, 120, 124, 125 closo, 3: 114, 115, 123, 124, 125, 127 germanium, 3 115-1 16 lead, 3: 116 mercury, 3: 122 nido, 3: 114-115, 120, 125 zinc, 3: 120-121 C-Mo-Si system, 2 226-227 CMSX-2, creep behavior, 3: 302, 843 CMSX-3, creep behavior, 3: 302 CMSX- 10, high-temperature strength, 3: 70 Coal gasifier applicat~ons,2: 207-208 CoAs, structure type, 3: 106 Coating processes, 3: 507, 509 Coatings, 1: 646,990,995,996, 997, 1003, 1009; 2: 57, 124, 169; 3: 561-588 AlNi-based, 2: 57 A1,Ti-based, 2: 124, 168, 170 alummide, 2 491, 492, 495-497 amalga~ation,2 408, 519-521 anti-diffusion, 3: 563 antireflection, 2: 410 cobalt-tin 2: 507 for composites, 3: 582-585 concentration profiles, 3: 578 for conventional alloys, 3: 565--575 copper-tin, 2 507 creep behavior in, 3: 298, 309 design considerations, 3: 562-565 d8iisional transport in, 3: 568-570 diffusivities, 3: 569 electro deposited, 2: 521 failure modes, 3: 562-564 fiber, 3: 583 gold, 2: 519-520 graded, 2: 498-499 growth mechaiiisms, 3: 566 hot dip, 2 518-519 for IMCS, 3: 575-582 interdiffusion, 3: 563 iron a l ~ m i n ~ d3:e , 571-574
Subject Index
970 by magnetron sputtering, 3: 570, 571, 572, 581 of MCrAlY, 3: 580- 582 by inicro~elding,3: 572 nickel aluminide, 3: 309, 565-567 Ni-Cd, 2: 519 Ni-Zn, 2: 521 noble metal, 3: 584 overlay, 2 489-492 by pack process, 3 566, 570, 572, 577579 p e r f o ~ a n c echaracteristics, 2: 494-498 corrosioa resistance, 2: 494-496 erosioii resistance, 2: 496, 497 ranking order of coatings, 2 495 thermal stability, 2: 496, 498 platinum-~odified,3: 64, 65, 72, 74, 564,565, 567-568 precious metal modified, 3: 567-568 for reinforcement fibers, 2 293 requirements for oxidation/corrosion resistance, 2 490 silicide, 3: 574-575 silicide-based, 2: 21 1, 227-229 for steel, 2: 519, 520 for TiAl, 3: 575, 577-580 tinplate, 2: 507, 518, 519 titanium aluminide, 3: 570-571 for windows, 2: 41 1 see also Diffusion; Alurninides, Beryllides, oxide protective scales Coating techniques for producing, 2 227, 49 1, 492-494, 518-52 1 Cobalt aluminides entlialpy of formation, 3: 605 point defects in, 3: 354 Coba~t~based bearing alloys, 2: 596-597 Gobalt-based binary compounds, magneto-optical applications, 2: 444-446 Cobalt borides, as magnetic materials, 2: 307 Cobalt-lanthan~~e [magnetic] coinpowds, 2 312-314 Cobalt-tin coatings, 2: 507
272, 276-277, 649 Co-Cu-Ni-Ti system, 1: 722-724 Coercivity, 2:304 Co-evaporation, thin films deposited by, 3: 667, 473 Co-extrusion, 3: 504-505, 505, 509 Ce-Fe naturally occurring compounds, 1: 632 Co-Fe-Ni system, 1: 721 Co-Fe-Ni-Ti system, 1: 712, 715-717; 2: 557 Coherence strain, 1: 867 Coherency stren~t~ening, I&: 266, 283 Coherent-potential approximation (CPA), 1: 25-26 methods based on, 1: 25-30
Concentration-functional theory; Embedded-cluster method; Generalized perturbation method; Korringa-Kohn-Rostoker . . . Coherent precipitates, 2: 4, 5 Coherent scattering, 3: 248, 251-252 Coherent twiii-boundary (CTB) energy, 3: 407 and twinning, 3: 424, 428 Coherent twins, 3: 410 Cohesive energetics, 1: 5 6 5 8 Cohesive energy, 3 768 Cohesive-energy factor, 1: 232, 270 in AP-AN plots, 1: 230, 233 atomic properties grouped under, 1: 23 1 cohesive energy vs melting temperature, 1: 232 Co-Hf system, 2 445 Coil-spring behamor, shape-memory effects compared, 1: 833; 2: 535 Coincidence-site lattice (CSL) boundaries, 1: 586 Coincidence-site lattice (CSL) [grainboundary structure] model, 1: 598, 600, 954 Coincident Doppler broadening, 3: 280, 287-288 Cold-crucible [melting] techniques, 1: 640, 643 Cold rolling, 1: 655 Cold working, 3: 506-507 Collapse of cascades to dislocatioii loops, 1: 807-808 to vacancy loops, 1: 803 Collective electron theory, 1: 936, 945, 949, 1018 Collision cascade parameters, 1: 8 11, 813 Collision cascades, 1: 579, 692, 809 Colloidal dispersions, metals in mercury, 3: 24 Color, 3: 231-244 and complex dielectric function, 3: 233 and electronic properties, 3: 233-236 human perception, 3: 232 physicist’s view, 3: 232-233 temperature effects, 3: 240 Coloradoite [mineral], 3: 21 Color, composition dependence in gold alloys, 2 560, 571 Colored compounds, 3: 74-77, 231 applications, 3: 75-77, 231 and related properties, 3: 240-242 binary compounds, 3: 75, 238 elastic properties, 3: 240, 241 listed, 3: 75, 231, 238, 239, 240 quaternary compounds, 3: 75. 240 structure maps, 3: 237 structures, 3: 236-240 temperature effects, 3: 240 ternary coinpounds, 3: 75, 239 Colored karat gold alloys, 2 559 Colored metals, 3: 231, 232 Colossal-magnetoresistance (CMR) inatenals, 3: 104 Columnar-grain superalloys, creep behavior, 3: 306 Gombustion synthesis, 2 44; 3: 645, 723, 724, 732-737 chemical furnace method, 3: 736 and densification, 3: 735-736 see also
electric-~eld-activated,3: 723, 725, 734-735 limitations, 3: 733-734 and matenal properties, 3: 736-737 and milling, 3: 757-758 phase-formation mechanisms, 3: 732735 and porosity, 3: 733-734 thermal explosion method, 3: 732 timescales, 3: 724 wave-propagation method, 3: 732 Combustion velocity, 3: 734 Co-Mg system, 3: 157 Commercial alloys ABB-2, 3: 4815482, 486, 595, 596, 597, 600 Astroloy, 3, 330, 331, 338 CMSX-2,3:, 302, 843 CMSX-3,3, 302 CMSX-10, 3:, 70 FA-129, 3: 332, 333, 339, 341, 573, 613 FA-186, 3: 573 FAL, 3: 573, 613 FAP-Y, 3: 333, 339 FAS, 3 573, 613 FAX, 3: 573 FINEMET, 3: 702, 754 GE 1541,3 715 GE 48-2-2, 3 595, 597, 601 Hastelloy, 3: 364, 418 Honeywell WM, 3: 597 IC6, 3: 608-609, 611 IC50, 3: 219 IGZ18, 3: 329, 330, 505-506 IC221/221M, 3: 329,341,492,493,515, 609, 606-61 1 IC396LZr, 3: 611 IC438, 3: 611 IHI, 3: 597 IMI 834, 3: 630 IN 100,3: 298, 300,480 IN 713C, 3: 329,630 IN 713LC, 3: 600 IN 7 1 8 , 3 481, 482, 630 rN ~ B L C ,3: 306 Inconel, 3: 42 KS, 3: 481-482, 630 Mar M 002, 3: 564, 56.5, 567, 568, 569 Mar M 200, 3: 306 Mar M 247, 3: 70 N A N ~ P E R3: ~ 702 , Platigeiiil’, 3: 75-76 PWA 1484,3: 486, 489 Rene’ N6, 3: 489 Rene’ 80, 3 331, 332 Reiie’ 95, 3: 630 RV 8413, 3: 715 SC-INC-6-7, 3: 305 Spangol~TM, 3 76-77 TAB, 3: 597 TlMET 1100,3: 571 TMS-79, 3: 68 TMS-80, 3: 68 45XDTM,3:. 596, 597 47XDTM,3 481,482,490,594,597, 597 Commercial Internet sources, 3: 872-873 Commerci~lization factors affecting, 3: 4 7 1 ~ 7 3477, , 602 of Fe-A1 alloys, 3: 61 1 of Ni,Al alloys, 3: 493, 604 of TiAl alloys, 3: 602
Subject Index Com~~unications systems applications. 2: 336, 342, 346, 413,414,428 Co-MoSi system, 2: 596-597 Companion crystal-structure interrelatio~sliips,1: 261, 268 Compensation [magnetization] point, 1: 938 Compensation point writing, Complementary color, 3: 233 Com~lementarytwins, 3: 405, 406, 414, 43 I Complex amalgams, 3: 27-29 Coniplex close-packed alloys, 1: 291, 292 interstitial alloys, 1: 303-304 Complex formation in liquids, 1: 671-673 energy of forination, 1: 672-673 temperature dependence, 1: 673 partition function describing, 1: 676-4377 quasilattice model, 1: 673-675 Complex stacking fault (CSF) energies Ll, compounds, 3: 443 L1 compound^, 3: 445 Complex stacking faults (CSFs), 2: 20, 149 in C11, structure, 3 455, 4.57 in C40 structure, 3: 459 in C54 structure, 3: 460 in DO,, structure, 3: 448-449, 449 in DO,, structure, 3: 447, 448 in L1, structure, 3: 444, 445, 445 in L1, structure, 3: 440, 441, 441 Composite-electroplating process, 2: 492 Composite microstructures Allr-based materials, 3: 59, 61 AIRu-based materials, 3: 61, 67 Composites, 1: 926; 2 170, 208, 223, 233, 254-255, 287-298 application temperatures, 2 288 coatings for, 3: 582-585 co-continuous phases in, 2: 287, 288 continuous reinforcements used, 2 287, 290 creep resistance, 3: 3 17-3 19 damage tolerance, 2 294, 297 developments, 2: 296-297 diffusion barriers used, 2: 295, 604 di~continuousreinforcements used, 2: 289-290 with ductile reinforcements, 3: 336, 737 fatigue behavior, 3: 337, 338, 341, 342, 344, 345, 346, 491 hybrid composites, 2: 290 in-situ, 2: 650, 654 iron"alumin~de-based,2 208 m~gnetostrictivematerials, 2: 401 MoSi,-based, 2 223, 233, 294-295 nickel-aluminide-based, 2 293-294, 650 niobium silicide, 3: 490-492, 541-560 oxidation behavior, 1: 1006-1008 processing of, 2 297-298 synthesis of, 3: 724 systems used, 2: 290-296 titanium-alu~mide-based,2: 170, 291-293 Composition see Chemical composition Compositional short-range ordering, 1: 744-745 Co~positionalsuperlattices, 2 425 Compound-formation plots, 1: 12, 234, 235, 236, 237, 264, 265 see also Structure maps
Compound-formation tendency, 1: 232-236, 270 binary systems, 1: 12, 232-234, 257-258, 264 ternary systems, 1: 234-236, 237, 257, 258, 264 Compound twin, 3: 405,431 Compressibility, of copper, 3: 738 Compressibility, liquid alloys, 1: 665, 673 Compressional stress-strain curves, Al-Ru alloys, 2 251 Compression testing cobalt disilicide, 3: 241 and strain hardening, 3: 371 Compressive creep, 3: 299 Compressive strength, dental amalgam, 2: 585 Computer design of alloys, 3: 81 1-839 list of terms, 3: 835-836 Computer disk drive readlwrite heads, 2: 318,435 Computer learning, 3: 822, 835 Computer models amorphous alloys, 1: 742 diffusion, 1: 762 dislocations, 2: 23, 238 glass formation, 1: 679-680 irradiation-induced aiiiorphization, 1: 820 strengthening, 2 260 Computers, optically based, 2 346-347 Computers, and professional language, 3: 818 Concentration fluctuations, 1: 27 glass formation, 1: 680-681 m liquids, 1: 665-668, 675, 677, 678 Concentration-functio~altheory (CFT), 1: 24, 25, 26-28 applications, 1: 41, 44, 47 Concentration range rule, compound formation predicted using, 1: 257 Concentration waves method, 1: 26; 3: 191, 202 Concentric circle grating surface-emitting (CCGSE) lasers, 2: 431 Concentric-tube [crystal-growth] system, 1: 189 Concept, meaning of term, 3: 835 C o n c e p t - f o ~ a t ~ osystem, n 3 819, 822 Condal alloys, 2: 524, 525 Coiiditional spinodal decomposition, 1: 852 Conducting films, precious metal compounds, 3: 78 Conductors, 2: 469, 653 Configurational entropy, 1: 27, 38; 3 193 cluster expansion of, 3: 194, 201 Con~gurationmatrix, 3: 194 CONFOR computer learning system, 3 819, 822, 834, 835 Congruent ordering process, 1: 853 Co-Ni-Si system, 2: 216-217 Connolly-Williams method (CWM), 1: 24, 30-32, 51; 3: 194 applications, 1: 41, 47 Consolidation of amorphous powders, 3: 685 of TiAl alloys, 3: 647 Constant initial state (CIS) spectroscopy, 3: 150 Constitutional defects
97 1 identification of, 3: 358 in solution hardening, 3: 352-353 evidence for 353 Constitutional studies, 1: 5-9 Constitutional vacancies, 3: 275, 279 Consumption, meaning of term, 3: 793 Contact matersals, 2: 652-653 Continuous cooling t r a n s f o ~ a ~ o n s (CCTs), in AlTi,-based alloys, 2: 98, 99 Continuous Electron Beam Accelerator Facility (CEBAF), 2: 376-377 Continuous ordering, 1: 851 Continuous solid solutions, 1: 406 in quasibinary/quasiternary systems, 1: 712 Contrast experiment technique, and dislocations in quasicrystals, 3: 383 Controlled etching, 2: 502 Convergent-beam electron diffraction (CBED), 3: 381 and dislocations in quasicrystals, 3: 381-383 Conversational processor, 3: 818,819,820 Coiivex-hull method (for phase diagram), 3: 197-198, 202,204 Cooling power, relative, 3: 525 Cooperative ordering, 1: 1020 Coordination numbers, 1: 250, 262 amorphous phases, 1: 739, 742 and atomic environment types, 1: 239, 262, 366, 373, 374, 375, 382 close-packed structures, 1: 284, 403-405,414-415 notation, 1: 239 P,Th,-type phases, 1: 399 Coordination p o ~ y ~ e d r1 a:,238, 280-281, 309, 365; 3: 3 illustrated, 1: 239, 366, 404, 427 notation used, 1: 313, 317, 427 symbols, 3: 8 Coordination types, 1: 238, 243, 268, 273, 372, 381 listed, 1: 373, 375 in structure map pin^, 1: 244, 245 Copley-Kear model, 1: 546-547 Copper compressibility of, 3: 738 ductility, 3 763 single crystal, strain hardening in, 3 362 thermal conductivity data, 1: 1025-1026 see also Cu Copper a l ~ ~ m i ~ ~ed~et hs ,a ~ pofy formation, 3: 60.5 Copper amalgam, 3: 2111, 31 Copper-based bearing materials, 2: 591, 594-596 Copper-based shape-memory alloys, 2 536, 538, 543-544 Copper-gold alloys, 1: 970 surface properties, 3: 2 17-2 18, 226 see also Au-Cu alloys see also Au-Cu system Copper-tin coatings, 2: 507 Copper-zinc-based shape-memory alloys, 2 532-534 CO-Pt system surface-induced ordering, 3: 223-224 surface structure, 3: 218 Core losses [in electric motors], %: 317
972 Correlation variables, 3: 194 Corrosion, 1: 927, 965-974 active, 1: 966 dental amalgam, 2 586-587 by mercury, 3: 31 principles, 1: 965-967 see aiso Active . . .; Aqueous . .; Crevice; Intergranular . . .; Passive corrosion Corrosion loss, for FeAI, 3: 716 Corrosion resistance amalgams, 3: 31 effect of coatings, 2: 489, 494-496 iron aluminides, 2: 207-208; 3: 714 shape- memo^ alloys, 2: 538 stainless steels, 3: 62 in wet conditions, 2: 507 zirconium aluminide, 2 144 Corrosive wear, Tribaloys, 2: 597 Co-segregation effect, 1: 86 CO-Si system, 1: 650, 999 CO& structure type, 1: 364,373,375, 379, 380, 381 separation in structrrre maps, 1: 389, 429 Cosmic rays, recording of, 3: 271 CO-Sn system, 1: 696, 766; 2: 507 COSP [oxidative lifetime prediction] program, 1: 1008, 1009 CO-sputtering, 1: 651 Cost considerations, 2: 250, 252, 254 COST 50 program, 2: 495 Co-Th system, 1: 705 Coulomb explosion model, 1: 821-822 Coulombic interaction in close-packed structures, 1: 285, 295, 296 Hume-Rothery phases, 1: 101 interstitiai alloys, 1: 300, 302 Laves phases, 1: 109 Coulometric titration method, thermodynamic properties determined using, 1: 95, 97 Counterfeit gold coins, 3: 31 Counter-gravity casting, J: 592-593 Couplings, 2: 536, 538-540 CO-U system, 1: 735, 738 Covalency crystal habit affected by, 1: 176-179, 183, 184 detection in band structure, 1: 134-135 Covalent bonding, 1: 73 close-packed alloys, 1: 295, 296 diffusion mechanisms affected by, 1: 763-764 Grimm-Sommerfeld compounds, 1: 109 interstitial alloys, 1: 300, 302 transition-metal aluminides, 1: 144 Covalent compounds interstitial atoms in, 1: 301 lattice dynamics, 1: 152-1 53 Covalent impurities, EAM used in description of, 1: 88 CO-WC nanocry~tallinepowders, 3 754 CO-Y system, 2: 445 Go-Zn system, 3: 157 CO-Zr systern, 1: 696, 740 ..cP2(12i)framework, 1: 335, 336 Crack growth
and embritt~emen~ by hydrogen, 3: 340 environmental effects, 3: 336-337, 337, 339-341, 346, 347 and hydrogen, 3 337, 339,340, 347 see also Fatigue crack growth Crack initiation and growth mechanisms, 3: 337, 347 Crack propagation rates, 2: 42 Crack-tip twinning, 3 419 CrB structure type, 3 89, 89 Cr-Cu system, 1: 705 Creep 3: 297-324 and alloying, 3: 306-312 anisotropy, 3: 304-306, 317, 319, 321 of beryllides, 3: 43-44 of composites, 3 317-319 constitutive model equation, 3: 299,321 environmental e@ects,3: 303-304, 320321 and fatigue, 3: 298 and grain boundaries, 3: 319-32 1 and grain texture, 3: 306 incubation period for, 3: 299, 302-303 and interparticle spacing, 3: 315 irradiation conditions, 2: 134 and lamellar stren~thening,3: 3 17 measurement and analysts of, 3: 299304 modeling of minimuni creep rate, 3 299-302 of ~ o S i ~ - b a s ecomposites, d 3 317-3 18, 485,486 of NiAI, 3 494 of nickel-base superalloys, 3: 297, 300, 302, 309-312, 842, 843 of niobium~silicidecomposites, 3: 489, 553-555 and ordering, 3: 308-312 rupture failure, 3: 298 rupture life, factors affecting, 3: 314 and slip geometry, 3: 304-305 strengtheners, 3: 307 stress, density-normali~ed,for highp e r f o ~ a n c ealloys, 3 480 of titanium aluminides, 3: 297, 302303, 317 and twinning, 3 428, 430 of two-phase alloys, 3: 312-317 see also Coble . .; Diffusional . . .; Dislocation. . .; Harper-Dorn . . .; Inverse . .; Minimum creep rate; Nabarro-Herring creep; Plastic deformation; Primary creep Creep behavior Condal alloys, 2: 525 dental amalgam, 2: 585-586 high-temperature intermetallics, 2: 244 iron aluminides, 2: 206, 207 nickel aluminides, 2: 42, 43, 63-64 silicides, 2: 225 superalloys, 2: 14, 62-63, 74 titanium al~minides TiAl, 2: 74,78-79,82,83,92; 3: 302303, 317 Ti,AI, 2: 114-120 trialu~nides,2: 169 Creep-fatigue interaction, 3: 298 Creep rate, secondary, high-temperature materials, 3: 489 Creep resistance
alloying effects, 3: 306-312 and diffusion coefficient, 1: 923 effect of second phase, 1: 926; 2 82 factors affecting, 3: 300, 304-321 geometrical factors affecting, 3: 304306 and grain size, 3: 319-320 and minor"e~e~ent additions, 3: 32032 1 and prec~pitationh a r d e n ~ n ~ in AN-based alloys, 3; 316-317 in nickel-based superalloys, 3: 313-3 16 and solid-solution $trengthening, 3: 307-308 and stoichiometry, 3: 309-3 12, 315 temperature dependence, 1: 922 two-phase alloys, 3: 3 12-3 17 Crevice corrosion, 1: 966 Cr-Fe-Mn-Ni-Si system, 2: 557 C r - F e 4 system, 2: 225, 226 CrFe structure type, 1: 119, 249, 253 Cr-Fe system, 1: 98, 632 Critical current density [superconductors], 2: 356 effect of strain, 2: 358, 359 Critical resolved shear stress (c.r.s.s.) effect of ternary additions, 2 37 for twintling, 3: 403, 415 temperature dependence, 3: 414, 415 mckel aluminides, 2: 19, 26, 58 silicides, 2: 220 titanium aluminides, 2: 102 trtaluininide compounds, 2: 162 in underaged alloys, 2: 269-270 Cr-Mo-Nl alloys, deformation twinning of, 3: 418 Cr-Mo-Si system, 2: 220, 221 Cr-Ni-Si system, 2: 228 Cr-Ni system, clustering tendencies in, 3: 307 Cr-Pt system, 3 70 Cr-Si system, 3: 157 Cross-over experiments, SRO kinetics studied, 1: 781 Cross potentials fitting for Al-B-Ni system, 1: 83-85 properties used in fitting, 1: 83, 84 Cross-slip, 1: 525, 529, 547 models, 2: 24-25, 27, 37 Cross-substitution diagram, adatnantane-structure compounds, 1: 348, 349 Crowdions, 1: 570, 580, 797 Cr-Si system, 1: 999 Cr-Si-Ti system, 2: 229 Cr-Si-Ti-V system, 2: 229 CrSi, structure type, 1: 388, 389, 390,428, 429 Cr,Si structure type, 1: 409,411, 430, 431 see also A15 compounds Cr-Ti system, 1: 751; 2: 240, 242 Crucible m~terials,2: 524-525 Crucible melting, cold wall, 3 592, 636 Cryosurgery applications, 2: 460 Crystal approximants, 1: 471478 decagonal~p~ase quasic~stals,1: 466, 477478,483 ic~sahedral-ph~se quasicrystals, 1: 476-477 indexing of, 1: 461
Subject Index Crystal chemical model of atomic iiiteractions (CCMAI), 1: 366, 38 1 Ciystal cliemical parameters, 3 814 Crystal equilibrium forms, 1: 168, 169, 172, 173 Crystal faces, classification, 1: 170 Crystal-field interactions, 1: 937; 3: 170173 Crystal-field parameters, 3: 172 determination of, 3 173 Crystal growth Bridgman method, 1: 643; 2: 67,68, 395 Czochralski method, 1: 643; 2: 68, 395 faceted growth, 1: 856 fundamental principles first described, 1: 167 Crystal habits boron nitride, 1: 182-183 collective mapping, 1: 186-189 cubic face combinations, 1: 186-1 89 definition and analysis, 1: 170-171 derived habits compared with observed habits, 1: 180-183 diagram for derivation, 1: 185, 186 doubling of lattice spacing, 1: 183 effect of near-equi~ibrium~o~equilibrium conditions, 1: 189-19 1 experimentally obtained end-forms, 1: 189, 190 factors deterinin~ng,1: 174-183 geometric classification, 1: 184-1 89 influence of atomic substitution, 1: 168, 171, 173-174 influence of Bravais lattice combination, 1: 169, 171 silicon carbide, 1: 182, 183 tabulation and mapping, 1: 184-186 see also Negative . . ., Positive ciystals Crystalline electric-field (CEF) excitations; 1: 159 Crystallin~~to-amorphous transfo~ation irradiation-induced, 1: 692-695, 733, 791, 809, 811, 816 see also Amorphization Crystallization of amorphous alloys/phases, 1: 745-749 temperatures, 2: 610, 612, 618, 625 driving force for, 1: 690, 746 eutectic, 1: 747 kinetics of, 1: 747 polymorphic, 1: 746-747 primary, 1: 747 surface, 1: 748 Crystallographic point groups, 1: 309, 310, 311 see also Point group . . . C~stallographicstudies, historical, 1: 9-1 1 C~stallographictra~sformations,1: 827-846 Crystallography data sources, 3: 797-799 on Internet, 3: 873 Crystal morphology, bond characterization using, 1: 167-192 Crystal structure maps see Stixcture maps Crystal structur~s,1: 225-486 of Al-Fe-Si-V compounds, 2: 180, 182-1 83 of alurninides, 1: 911; 2: 17, 18, 54, 55, 94-95, 199, 200
of bainite phases, 1: 836 of body-centred cubic structures, 1: 62, 309-341 classification of, 3: 3-4 of close-packed structures, 1: 61, 277-306 companion interrelationships, 1: 261, 268 data availability, 1: 227, 228 of dental-amalgam compounds, 2 576 and entropy of fusion, 2: 639 factors governing, 1: 227-273, 364 strategy to find, 1: 228-232 glossary of terms, 1: 272-273 of high-temperature intermetallics, 2 240, 244,245, 249 of hydrides, 1: 297, 298, 299-302, 393; 2 475,478 illustrated, 1: 61, 62, 520, 858; 2 17, 55, 94, 148, 212, 324, 330 interaction between intermetallic compounds determ~nedby, 1: 712 nomenclature, xxiii-xxvi, 1: 273; 3: xxxi-xxxiv, 12, 13-15 and physical properties, 1: 264--268 pressure effects, 3: 161 for rare-earth compounds, 3: 87-96 regularities, 1: 236-264 active coiicentration range approach, 1: 257, 264, 270 atomic environment approach, 1: 237-241, 264, 270 chemistry principle, 1: 264, 270 information-prediction systems proposed, 1: 270, 271, 272 simplicity approach, 1: 255-256, 264, 270 solid-solubility map approach, 1: 244-248, 264 space-filling approach, 1: 248-254 stoichiometric restraint approach, 1: 257-260, 264, 270 structural relation approach, 1: 260-264 structure map approach, 1: 241-244, 264, 270 symmetry approach, 1: 254-255, 256, 264 of semiconductors, 2 323,326,329,330 solid solubility affected by, 1: 247-248 stability, 3: 841 structural-energy differences, 1: 65 symmetry, 3: 11-12 of trialuminides, 2 148 Crystal symmetry, plasticity affected by, 1: 921-922 Crystal-to-glass transformation, 1: 702 CSM X-2 superalloy, 2: 259 Cr-Zr system, 1: 740 CsCl structure type 3: 89, 89 growth of coinpounds by MBE, 3 784786 see aZso B2 type compounds; Bodycentered cubic. . ., ClCs-structure C-Si-Ti system, 2: 410 Cs-Na system, 1: 677 Cs-Pb system, 1: 686 Cs-Sb system, 3: 1.57 Cube [atomic environment], 1: 238, 239 frequency plot, 1: 240 Cube slip, 2 13; 3: 304, 305
973 Cubic alloys, 1: 419, 431 Cubic close-packed (c.c.P.) interstitial ~ ~ I o Y s1:, 299-302; 3: 8, 9 Cubic close-packed (c.c.P.) structures, 1: 278, 250, 287-291 see also A15 . ., Ll2 Compounds, etc. Cubic ordered structures see €32.. .; DO,. .; Ll,. . ., L2, type compounds Cubic system, point groups, 1: 312 Cubooctahedron, 1: 238, 239, 281, 403, 404 frequency plot, 1; 240 structure map, I: 245 Cu-Er system, 1: 696 Cu-Fe-Hg alloy, 3: 32 Cu-Fe-Ni system, 2: 307, 308 Cu-Fe system, 1: 706 Cu-Ga system, 1: 843, 846 Cu-Ga-Ti system, 1: 720 Cu-Hg system, 1: 632, 633 Cu,Mg-MgZn, quasibinary section, 1: 107 Cu-Mg-Ni system, 1: 108 Cu-Mg-Si system, 2: 643 Cu,,Mg,Si, structure type, 1: 385, 397 Cu,Mg structure type, 1: 250, 252, 389, 410,428,429 Cu-Mg system, 3: 157 Cu-Mg-Zn system, 2: 643 Cu-Mn system, 2 538 Cunico (magnetic material], 2: 307 Cunife [magnetic maternal], 2 307 Cu-Ni-Ti system, 1: 711, 721, 722-724, 728 Cupalite [mineral name], 1: 627, 628, 632, 633 Cu-Pd system, 1: 49-50, 632, 780 Cuprate superconductors, 1: 16, 136 Cuprite [mineral name], 1: 320 Cuprostibite [mineral name], 1: 627, 632 Cu-P-Si system, 2 643 Cu-Pt naturally occurring compounds, 1: 632 Curie p o i n t ~ t e ~ p e r a t ~ r e ferromagnetic semiconducting compounds, 2: 331, 332 magnetic materials, 2: 306, 308, 309, 312, 313, 314, 315, 316, 317, 391, 479 Curie point writing, 2: 441, 448 Curie temperature, 3: 519 listed for various materials, 3: 526-527 rare-earth compounds, 3: 98, 99, 100, 177, 526-527, 530 effect of interstitial modification, 3: 98, 178 Curie-Weiss law/regime, 1: 220, 682, 937 Cu-Sb naturally occurring compounds, 1: 632 Cu,Sb structure type, 1: 364, 373, 375, 379, 380, 381; 2: 309 separation in structure maps, 1: 389, 429 Cu-Si system, 2: 643 Cu-Sn-Ni system, 1: 843 Cu-Sn system, 1: 95,662; 2: 408,507, 538, 594-596 see also Bronze Cu,Tb structure type, 2: 310 Cu-Th system, 1: 703, 705
974 CuTi structure type, 1: 261, 422, 426 CuTi, structure type, 1: 430, 431 Cu,Ti structure type, 1: 430, 431 Cut-projection method, quasicrystals studied using, 1: 459 Cutting of IMCs, 3 509 Cutting tools, 2: 46 Cu-Y system, 1: 696 Cu-Zn system bainite phases/transformations, 1: 836-837, 838, 839, 841 beta brass (CuZn), see CsCl structure type ent~aipyof formation as function of atomic fraction, 1: 101 gamma brass, see Cu,Zn, structure type heat capacity, 1: 1019-1020 naturally occurring compounds, 1: 632 phase diagram calculations, 3: 205 phase diagrams, 1: 44, 45, 101; 2: 536 phase stability properties, 1: 44-47 regular solution, 1: 101 short-range order in, 1: 46 superelasticity, 1: 834 weal- performance, 2 591 see also b, g and e brass; Brass; Muntz metal Cu,Zn, structure type, 1: 322, 325, 326, 327 Cu-Zr system, 1: 735, 740, 742 CW structure type, 1: 364, 373, 375, 426 Cybernetical-statistical approach to design of materials, 3: 832-834 Cyberiietics, 3: 821 Cyclic hardening, 3: 326, 328 Cyclic oxidation, 1: 979,989,998, 1008; 2 124, 168, 169 Czochralski method, 3: 546, 546 Czochralski [single-crystal growth] technique, 1: 643; 2 68, 395 DO, structure, 1: 62, 520, 856 2: 62, 200, 212 APB energies, 1: 507 dislocations in alloys, 1: 536-537 fracture of compounds, 1: 907 see also BiF, structure type DOl9structure, 1: 61, 65-66, 520, 534 dislocations in alloys, 1: 534-535 elastic properties of compounds, I: 879-8 80 see also Ni,Sn structure type DO,, structure, 1: 61, 65, 520, 529-530; 2: 148, 156 A1,Ti-based compounds, 2: 148-155 APB energies, 1: 506 dis~ocationsin alloys, 1: 530-532, 542, 543 elastic properties of compounds, 1: 878-8 79 fracture of com~ounds,1: 907 relation to Ll, structure, 1: 499, 520, 530; 2 148, 156 see also AI,Ti structure type DO,, structure, 1: 61, 65; 2: 148, 156 fracture of co~pounds,1: 907 relation to L1, structure, 1: 499; 2: 148, 156 DO, type compounds, 3: 275 colored compounds, 3 238
Subject Index crystallographic elements and parameters of twin modes, 3: 407 displacement vectors of stacking-faultlike defects, 3: 463 Pearson symbol for, 3 xxxi, xxxiv, 14, 439 space group for, 3: xxxi, xxx~v,14, 439 stacking faults in, 3: 453-454 twinning in, 3: 416-417 DO, structure, 2 156 DO,vtype compounds, 3: 275 crystallographic elements and parameters of twin modes, 3: 408 isp placement vectors of stackin~~aultlike defects, 3 463 Pearson symbol for, 3: xxxi, xxxiv, 14, 439 space group for, 3: xxxi, xxxiv, 14, 439 stacking faults in, 3: 448-451 twinning in, 3: 417-418 DO,, type compounds crystallographic elements and parameters of twin modes, 3: 408 displacement vectors of stacking-faultlike defects, 3: 463 Pearson symbol for, 3: xxxii, xxxiv, 14, 439 space group for, 3: xxxii, xxxiv, 14, 439 stacking faults in, 3: 446-448 twinning in, 3: 414-415 DO, type compounds Pearson symbol for, 3: xxxii, xxxiii, 14 space group for, 3: xxxii, xxxiii, 14 twinning in 415 Dl, type compounds Pearson symbol for, 3: xxxii, xxx~v,14 space group for, 3: xxxii, xxxiv, 14 twinnmg in, 3: 418 D1, type phases, 1: 727-728 D2, structure see NaZn,, structure type D7, structure, see P4Th3structure type D8, structure, 2: 212 D8, structure see Mn,,Th, structure type D8, type compounds Pearson symbol for, 3: xxxii, xxxiv, 15 space group for, 3: xxxii, xxxiv, 15 twinning in, 3: 418 see also Ivfn,Si, structure type D(60) framework, 1: 325, 327, 328, 331, 332 d-group elements, amalgams, 3: 24 D-type Shockley partial dislocations, 3: 423 Dalton, John, 3: xliii Daltonide compounds, 1: 273 solid solubility, 1: 248 stoichiometric-ratio distribution binary compounds, 1: 258, 259 ternary compounds, 1: 258, 260 str~icturemapping, 1: 245 Daltonides, 3: xxxvi Damage tolerance, in composites, 2: 294, 297 Damping, 2: 535-536, 554-555, 649-650 Danbaite [mineral name], 1: 627, 632 Dangling bonds, 3: 216 Darken-Gurry diagrams, 3: 8 12 see also Generalized Darken Model Darken and Gurry solubility plots, 1: 245-246 Darken’s [viscosity] equation, 1: 668
Data, meaning of term, 3: 818, 835 Databases, 3: 8 13-8 16 on acousto-~e~~tro-/noni~~i~ar-optical properties, 3: 816, 817 application of AI to analysis of, 3: 816819 design principles for, 3: 814 on phase diagrams with semiconducting phases, 3 816 on ternary compo~~ids, 3: 8 15-8 16 Data centers, 3: 808-809 Data mining (on Internet), 3: 857, 876 Data sources, 3 797-810 aluminides, 3 808 beryllides, 3: 808 c~stallogra~hy, 3 797-799 gallides, 3: 808 germanides, 3: 808 indides, 3: 808 kinetics, 3: 805 magnesides, 3: 808 magnetic properties, 3: 803-804 miscellaneous, 3: 809-8 I0 phase diagrams, 3: 800-803 physical and other properties, 3: 803804 rare-earth compounds, 3 806-807 semiconductors, 3: 805-806 silicrdes, 3: 808 superco~ductors,3: 806 t h e r ~ o d y n a ~pi rc~ ~ e r t i e3s , 799-800 the~o-physicalproperties, 3: 807-808 Data storage applications, 2: 435, 438, 44 1 Deal-Grove model, 3 674 De-alloying, 1: 970 Debye-Gruneisen correction, 3 201 Debye model, 1: 203, 1018 Debye t~mperature,1: 885, 1018, 1022 Debye temperatures, metallic glasses, 3: 698 Debye-Waller correction, 3: 251, 253,259 Decagonal-phase quasicrystals, 1: 453, 465-471 atomic structures, 1: 483-484 crystal approximants, 1: 477-478 Decagonal quasicrystals (DQCs), 3: xxxvi calculation of elastic displacement field around isl location, 3: 395-397 characteristics, 3: 389 elastic constants, 3: 395 plastic deformation of, 3: 399 Decomposition modes, 1: 852, 853 Decomposition pathways reaction kiiietics, 1: 853-857 thermodynamics, 1: 851-853 Decomposition reaction [for dislocation], 1: 527 examples for various structures, 1: 527-528, 533, 535, 536, 537, 539, 543 Decorative coatings, 3: 76, 231 Decrepitation, of hydrides, 2: 478, 482 Deep eutectic features, 1: 689, 691, 737 Defect calculations for Al-Ni system, 3: 768-772 for Al-Ti system, 3: 772-774 angular contributions, 3: 773 for Mo-Si system, 3; 774-775 Defect cascades, collapse of, 1: 807
S ~ b j Index ~ ~ t Defect concentrations, req~Iirementfor amorphization, 1: 818 Defect derrvatives, 3: I2 Defect energies, 3: 770 Defect-enhanced diffusion, Defect icosahedron, 1: 404 Defect identification, 3: 358-359 Defect rhombododecahedron, 1: 404 Defect simulations, 3 765-778 Defect structures, 1: 112, 493-621 defect adaman~~ne-structure compounds, 1: 344, 347, 349, 353-3 54, 355-3 56 and diffusion, 3 275-293 in iron aluminides, 2: 199-201 tetrahedral structures, 1: 346, 348 see also Antisite . . .; Interstitial . . ., Point defects; Vacancies Defect tolerance, in aluminides, 2: 66 Defocused convergent-beam electron diffraction, and dislocations in quasicrystals, 3: 381-383 Deformation amorphization by mechanical means, 1: 699-701, 733, 734 of B2 phases, 3: 851 of dental amalgam, 2: 586 of nickel aluminide alloys, 2: 22-27,46, 63-64 role OC ear-Wilsdorf configuration, 1: 544 of silicides, 2: 218-219 structure t r a n s f o ~ a t i o nby, 1: 405 of superalloys, 2: 13-14 and temperature, 3: 621 of titanium aluminide alloys, 2 73-74, 101-106 of tnalurmnides, 2: 150-154, 162-167 see also Plastic deformation Deformation maps, 1: 916, 918-919 D e f o ~ a t i o n~ e c h a n ~ smaps m power law creep region, 3 299, 647 for pressure sintering, 3: 647 Deformation mechanisms, titanium aluminides, 3: 621-623 Deformation processing powder billets, 3: 646-648 of titanium aluminides, 3: 649 Defo~mationtwinning, 1: 844,9 15-9 16; 3: 403, 431 in bcc-based structures, 3: 415-417 and creep, 3: 428 in fcc-based structures, 3: 411-415 in ~ e x a g o n astructures, ~ J: 417-418 $trengthening effects, 3: 424425 see also Twinning Degenerate semiconductors, 2: 329 De Haass-van Alphen (dHvA) techniques, 1: 135, 136, 212 compared with ACAR, 1: 137 Delta phases, 3: xxxvi Demixing of liquid alloys, 1: 677-678 Demolition applications, 2 555 Dendritic precipitates, 1: 867 Dendritic segre~ation,3: 621 Dense-random-packing (DRP) model, 1: 473,479, 740-741 Densification, 1: 4 and combustion synthesis, 3: 735-736
pressure-induced, 3: 742 in transient liquid phase, 3 725 Deiisification mechanism map, for HIP consolidation, 3: 637 Density Al-Ce-Fe phases, 2: 179 alunlinides, 2 55, 73, 74, 134, 199, 293 Condal alloys, 2: 525 listed for various IMCs, 3: 502 pressure effects, 3: 161 semicond~~ctors/semtmetals, 2: 328 shape-memory alloys, 2: 538 silicides, 2: 213, 294 trialum~nides,2 147, 148, 170 see also Specific gravity Density-functional theory (DFT), 1: 58, 129-130, 196-197; 3: 196,439 basis sets chosen, 1: 198 and quasicrystals, 1: 461, 480 see also Ab initio . . .; First-principles calculations Density-melting point plots, 1: 266, 268 Density of states (DOS) and band structure, 1: 134, 135; 2: 479, 480 disilicides, 3: 234235 at grain boundaries, 1: 605 integrated density of states (IDOS), 1: 26 investigation of in occupied states, 3: 135-145 in unoccupied states, 3: 145-150 in liquid alloys, 1: 670 models used in tight-binding calculations of cohesive energetics, 1: 56 in phonon dispersion, 1: 161 in quasicrystals, 1: 484, 485 quasigap/pseudogap in distribution, 1: 58, 63, 671 and superconductivity, 1: 952 in transition-metal aluminides, 1: 138, 140, 141, 1020 Dental amalgams, 3 : 3; 2 575-590; 3: 31, 231, 795 amalgamation reactions, 2: 580-58 1 clinical performance, 2 587-588 compared with other restorative materials, 2: 575 compressive strength, 2: 585 crystal structure of compounds, 2: 576 dimensional changes during hardening, 2 582-583 dissolution of mercury into Auid, 2: 589 economic factors, 3: 794 electrocheinical properties, 2: 586-587 hardening rate, 2 584 inechan~calproperties, 2 584-586 microstructures, 2: 581-582, 583 phase diagrams of compounds, 2 576-578, 579 silver-tin alloy powder, 2: 578-580 microstructures, 2: 580 preparation and processing of, 2 5 79-5 80 solidus temperature for, 2 583-584 typical compositions, 2: 584 uniqueness, 2 575, 590 vaporization of mercury from, 2: 588-5 89 Dental gold alloys, 2 564, 565, 649
975 Dental materials, 2: 575, 649 Deoxidants, 2: 523, 524 Depletion-layer conversion devices, 2: 345 Derivative structures, 1: 405, 41 1 defect derivatives, 1: 412 deformed derivatives, 1: 406, 410, 412 Design of alloys, 3: 841-855 of inorganic substances, 3: 8 19-823 of processing paths, 3 841 Design considerations castings, 3: 598 coatings, 3 562-565 high~temperaturestructural materials, 3: 475-476 Designer coatings, 3: 585 Desorption behavior, hydrides, 2: 477 Detectors see Infrared . .; Optical . .; Photo . . ., Thermoelectric detectors Deutendes, 1: 299, 300; 2: 477; 3: xxxvi Devitrification, 1: 748-749 Dew-point method, vapor pressure d e t e ~ ~ n using, e d 1: 93, 94 ..d1,(60) framework, 1: 328, 332, 333 Diamond bonding, 1: 109 Diamond octahedra, 1: 180 Diamond, structural formula for, 3 5, 7 Diamond structure, 2: 323 partial substitution effects, 1: 168 Dielectric breakdown, 2 504 Dielectric constant, semiconductors, 2: 327, 345 Dielectric tensor, 2: 439 Diesel engine parts, 2 45, 46 Dies and molds, 2: 45 Differential cross-sect~on,3: 247 Differential scannin~calorimetry (DSC), metallic glasses, 3: 686, 699 Differential thermal expansion technique, vacancy concentration determined by, 3: 277, 281 Diffraction gratings, 2: 41 5-416 Diffraction-pattern indexing, application to quasicrystals, 1: 460-461, 470-471 Diffraction techniques, 3: 247-248 surfaces studied by, 3: 212 see ulso Neutron.. .; X-ray diffraction Diffuse scattering intensities, 1: 40 cu-Pd alloys, 1: 49 Cu-Zn alloys, 1: 46 PdRh,, I: 50 Diffusion, 1: 757-768 activation energy for boron in AINi,, 1: 594 in amalgams, 3: 30 applications of, amorphous phases formed by, 1: 696-699; 2 619-620 superconductors fabricated using, 2: 361-362 synthesis of intermetallics using, 1: 645-646, 757-760 in beryllides, 3: 46 in coatings, 3: 567, 568-570 coniputer simulation of, 1: 762 and covalent bonding, 1: 763-764 and creep resistance, 1: 921 defect-enhanced, 3: 724 and defect structures, 3: 275-293 of dental amalgam phases, 2 581, 584 experimental techniques, 3: 276-278
976 of hydrogen, 2: 483-484 and ionic bonding, 1: 764 in liquid alloys, 1: 668-669 in liqud mercury, 3: 27 and metallic bonding, 1: 763 multiple, 3: 841 reactive, 1: 768; 3 673-676 at surfaces, 3: 225 Diffusion activation energy, boron m A N , , 1: 594 Diffus~onalcreep, 1: 917-918; 2 117, 121 Diffusional interactions, 1: 861-868 Diffusion barriers, 1: 645-646, 967; 2: 498-499, 603-630; 3: 675 amorphous alloys as, 2: 605, 609414, 618, 625-627 applications, 2: 295, 357-358, 604, 629 in composites, 2: 295, 604 effect of impurities, 2: 609, 628-629 in ~etallizationschemes, 2: 622-628 preparation and use, 2: 628-629 principles, 2: 603 requirements, 2: 603 stacked, 2: 627 in superconductors, 2: 357-358 types, 2 604406
stuffed barrier, 2: 605, 625 Diffusion bonding, 1: 656; 2: 521; 3: 298299 with superplastic forming, 1: 657 Diffusion coefficients boron in AlNi,, 1: 594 and creep, 3: 302, 307, 313 experimental determination, 1: 764-768 and solid-state reaction synthesis, 3 724 Diffusion couples, 1: 696, 697, 86 1, 862 Diffusion mech~nisms,1: 760-762 divacancy mechanism, 1: 761-762 effect of bonding type* 1: 762-764 interstitialcy mechanism, 1: 760-761 jump mechanisms, 1: 576, 760, 762 ring mechanism, 1: 762 sublattice mechanism, 1: 761 vacancy mechanism, 1: 76 1 Diffusivity, and ordering, 3: 364 Diffusivity, transition metals in 2: , 176 a~umi~um Dimer model, 1: 183 Diode lasers, 2 325, 326, 328, 336, 338339, 346, 347,418,427-431 Diode sputter in^^ 3: 665 Diopside, structural formula for, 3: 5 Direct chill casting method, 2 175, 176 Direct configura~iona~ averaging (DCA) method, 1: 24, 32-33; 3: 194 Direct-current (DC) sputtering, 3 665 Directional bonding, 1: 281, 282, 291 at grain boundaries, 1: 604 Directional~ysolidified materials aluminides, 2: 42, 43, 651 in composites, 2: 654 conductors, 2: 653 magnetic materials, 2 317
Directional solidification, 3: 541-544 Bridgman method, 3 542,546547,546 Czochralski method, 3: 546, 546 of autectics, 3 544-545 float zone processing, 3: 545, 546 of NiAl alloys, 3: 602-604, 656 Directional solidification techniques, 1: 639, 643-644, 912 Dirichlet con~truction,1: 365 Disclination lines, 1: 473, 474-475, 479 Disilicides, 2: 211, 219-223, 615 bondiiig in, 3 235 electrical conductivity, 3: 235 optical properties, 3 232, 233, 235 prediction of formation, 3: 829 Schottky barner heights, 2: 230 structural-energy differences, 1: 67 structures, 3: 456 see also Silicides Dislocation climb, 1: 916; 2: 134, 586; 3: 314 factors affecting, 3 315 Dislocation-constrictionenergies, 2: 25 Dislocation cores and alloy composition, 1: 539-540 computer simulations, 2 19, 23 effect of anisotropic elasticity, 1: 540,541 effect of bond directionality, 1: 540-541 factors affecting, 1: 539-54 1 mechanical properties affected by, 1: 546-550 properties in various structures, 1: 527539 sessile, 1: 545 and slip systems, 1: 541-543 Dislocation creep, 1: 916-917,919; 2: 117, 121 Dislocation debris, 1: 550-552 Dislocation density, in deformed quasicrystals, 3: 399-400 Dislocation dynamic based model, 3: 305 Dislocation dynamics, superalloys, 2 13 Dislocation locks, 1: 525, 529, 543-544, 545 conservative locks, 1: 545 Dislocation mechanisms, antiphase boundaries formed, 1: 496-497 Dislocation- article interactions, 2: 258-263 Dislocation-based resistivity models, 1: 959,960 Dislocation mobility factors affecting, 3 622 strain hardening affected by, 3 365 Dislocation models for twinmng, 3 407408,421-422 Dislocation motion in metallic glasses, 3: 693 in plastic deformation of quasicrystals, 3: 399-401 Dislocation pileups, 3: 419, 421, 762 Dislocation reactions, in quasicrystals, 3: 398-399 Dislocations, 1: 15, 5 19-553 a < I10 > type in strain hardening, 3: 366-367 APB-caused, 1: 511-514 atomistic calculations, 3: for 772 in B2 alloys, 1: 535-536, 542 in b.c.c. alloys, 1: 535-537
in Cl 1, phases, 1: 539 choice of slip directions, 1: 542 in DO, alloys, 1: 536-537 in DO,, alloys, 1: 534-535 in DO,, allays, 1: 529-532, 542, 543 definitions, 1: 519, 521 dissociation ofs 1: 5 1 1-5 12, 524-527 in DO,, alloys, 1: 531-532 in L1, alloys, 1: 528-529; 2: 22, 139, I62 D-type Shockley partials, 3: 423 elastic properties af'fected by, 1: 880-882, 883, 884 e ~ e c t ~ i cconductivity a~ affected by, 1: 958-960 fine structure observed, 1: 521-523 fine structure simulated, 1: 523 Frank and Shockley partials, 3: 409, 423, 430 in iron alummides, 2: 201 in L1, alloys, 1: 532-534, 542 in L1, alloys, 1: 527-529 in L2, alloys, 1: 536-537 in Laves phases, 1: 537 line tension of, 2: 263-266, 272 long straight, in quasicrystals, 3: 392395 mobility, 1: 519, 522 experimental determination, 1: 523 mobility of, 2: 154 non-conservative dislocation, 1: 546 in ordered structures, 1: S19 pinning of, 2: 25 and point defects, 1: 580 in quas~crystals,1: 463, 537, 538; 3: 379-402 Burgers vectors for, 3 381-385, 385 elastic fields around, 3: 389-397 image of dislocation line, 3: 383 and plastic defor~ation,3: 397-401 Volterra process for, 3: 379-381 sessile, 1: 543-546 and surface structure, 1: 620 in Ti-48A1-2Cr-2Nb, 3: 367 in tri~luminidecompounds, 2: 150-153, 162 type-a superdislocations, 1: 534, 535 velocity-stress relationship, 2: 238 see aZso Antiphase-~oundary(APB) tubes; Superdislocations; Superlattice extrinsic stacking fault . , ., Superlattice stacking fault . . . Dislocation structure, A N 3 , 2: 27 Dislocation structu~es,and fatigue damage, 3: 326 Disordered close-packed alloys, 1: 304-305 Disordered materials, strain harden~ngin, 3: 362-364 Disordered metal hydrides, dynamic disorder in, 3: 252-253 Disordering, 3: 245 from mechanical alloying, 3: 73 1-732, 758, 759, 761 of nanos~ructures,3: 759-761 near surfaces, 3: 221, 222-223 see also Chemical disorder; Dynamic disorder; Lattice melting anisotropy; Orientational disorder; Traiislational disorder
Subject Index Disordered solid alloys, and liquid alloys, 1: 679, 681 Disorder effects and antiphase boundaries, 1: 502-503 on phonon dispersion, 1: 159-162 in quasicrystals, 1: 461 substitutional disorder, 1: 160-161 on thermal conductivity, 1: 1025 Disordering, 1: 771-788 by mechanical deforma~ion,1: 701, 786-788 irradiation-induced, 1: 692-695, 733, 791, 803-808 thermal, 1: 771, 779-781 Diso~d~r-order transition antiphase boundaries formed, 1: 496, 500 antiphase domains, 1: 509, 510-51 I APB energies vs transition temperature, 1: 508 see also Order-disorder transition Disorder parameter, 1: 38 Disorder quench in^, 1: 772-773 Disorder softening, in high-temperature plastic defoinxition of quasxrystals, 3: 401 Disorder trapping, 1: 857 Dispersion-strengthened aluminurn alloys, 2: 175 properties and applications, 193195 Dispersotds, 2: 175-195, 257-284 Al--Cr-X compounds, 2: 186-1 87 Al-Fe compo~nds,2: 276-185 Al-Ce-Fe system, 2 178-180, 194 Al-Fe-Si-X system, 2: 180-185, 194 Al,(Hf, Nb,Ti, V, Zr) compomds, 2: 187-1 88 AI,(Li,-xZrx) ~ompounds,2: 188-192 alummum-transition-metal composite Al,Li/Al,Zr, 2: 190, 191, 194-195 in composites, 2: 293 Displacement energies, 1: 567, 570, 794, 797 Disp~acementsper atom, 1: 797 Disp~acementwaves, 1: 858 Displacive energy, 3: 192, 199 3: 199-20 I Displacnve ~nt~ractions, Displactve ordering, 1: 860 Displacive phase transitions, 1: 162-164 Distortion derivatives, 3: 12 Distortion energy, of Hume-Rothery phases, 1: 103-104 Distr~butedfeedback (DFB) lasers, 2 339 Divacancies, 3: 281 Divacancy diffusion mechanism, 1: 761-762 Divergent flaps, with TiAl sheet, 3: 638 Domain coarsening, 1: 774, 865, 866-868 Domain wall pinning, 3: 442 Dom [dislocation creep] equation, 1: 916; 2 63 Dominant diffusing species (DDS), 3: 674 Doping, sulfidation slowed by, 3: 709 Dorn equation, 3: 302, 312 Double differential cross-section, 3: 248
Double heterojunct~onis~ructure, 2 328, 336, 338, 426,427 Double twinning, 3: 416, 431 DS IN738LC, creep behavior, 3: 306 Dual [crystal] habit, 1: 189 Ductile IMCs, 3: 851-854 Ductile-to-brittle transition temperature (DBTT) beryllides, 3: 43 and solid solution hardening, 3: 356 of TiAl sheets, 3: 633-634 see also Brittle-to-ductile transition tempe~ature Ductility compared to conventional alloys, 3: 474, 476 Condal alloys, 2: 525 effect of alloying, 2: 39, 59, 60, 80-81, 214, 215-216, 251 effect of grain size, 2: 33-34, 78 effect of microstructure, 2: 77, 78 effect of off-stoichiometry, 2 32 effect of test environment, 2 216 elevated-temperature AI-Fe-Si-V alloys, 2 194 aluminide alloys, 2: 39-40, 62, 113, 114 factors affecting, 1: 897, 898, 918-919, 923,924 metallic glasses, 3: 693-696, 702 nanocrystal~ine,3: IMCs 763
58-60, 74; 3: 494, 502 at room temperature, 3: 853-854 aluminide alloys, 2: 38-39, 69, 77-78, 80 ruthenium aluminide, 3: 55-56 silicides, 2: 214, 220 superalloys, 3: 842 and ternary add~t~ves, 3: 736-737 TiAl sheets, 3: 633, 651, 652 titanium aluminides, 2 77-78, 83, 92, 107, 111, 113, 114 trialuminides, 2 166, 167 and twinnin~,3 425426,427 wrconium aluminide, 2: 140, 142-143, 144 Ductility enhaiicement, 1: 620, 656, 786-787, 845-846, $60,924 iron alurninide alloys, 2: 204 nickel aluminides, 2: 29-30, 59 silicides, 2: 2 15-2 16 titanium a I u ~ n i d e s2: , 80-81, 91 Ductility mec~anisms,Al-Nb-Ti alloys, 3: 774 Ductiiization, 3: 842 by boron, 3: 843 trialuminides, 2: 154-1 55 Dulong-Petit limit, 1: 1019 Dumbbell structure [of iiiterstitials], 1: 567, 578, 785, 796, 798 Duplex m~crostructures,3 318, 478, 628 ~ickel- base^ superalloys, 2 1 1 titanium aluminides, 2: 76, 109-1 12 Dy-Fe-Tb system, 2: 392-403 Dy-La system, 1: 407 Dynamical structure factors, 3: 248 Dynamic disorder, 3: 245, 250-259 in disordered metal hydrides, 3: 252253 in fast-ion phases, 3: 250-252
977 in ferroelastic phases, 3: 258-259 in fullerides, 3: 257-258 in graphite intercalation compounds, 3: 257 in paddl~-wheelphases, 3: 255-257 in rotor phases, 3: 253-255 Dynamic embrittling mechanism, 2: 33 Dynamic pressure applications, 3: 160 Dynamic recovery pr~cesses,in strain hardening, 3: 363, 366, 369 Dynamic recrystallization, 2: 139, 142 Dy-Nd system, 1: 407 Dyscrasite [mineral name], 1: 10,628,632; 3: 872 Dzhalindite, 3: 12
E9, crystal structure see CFe,VV, structure type E-phases, 3: xxxvii, see also TiKiSi structure type Eagle International Software, 3 872 Easy glide, 3: 362 Economic impact of IMCs meaning of term, 3: 793 values quoted, 3: 794, 796 Economic leverage, meaning of term, 3: 793 Economic value, 3: 791-796 bottom-up measurement of, 3: 793-796 of finished parts, 3 791 of synthesis stage, 3: 791 top-down measure~entof, 3: 792, 793 Edge-centered stacking, 1: 339-340, 341 Effective cluster interactions (ECIs), 1: 28, 30; 3: 194, 196 Effective concentration, Eirective electronic charge Ga m GaAs, 1: 174, 178 III-V compounds, 1: 179, 183 Zn in ZnS, 1: 174 Effective heat of form~tion,definition, 2: 607 Effective-heat-of-formation model, 2: 6 0 ~ 0 9 Eirective-~ediumtheory, 1: 78 EfFective parr interactions, f.c.c.-based Gu-Zn alloys, 1: 44 Effective pair ~n~eractions (EPIs), 1: 29, 30, 36-37 b.c.c.-based alloys, 1: 42 experimental determination, 1: 40 8-N rule, f: 346 Eigenstrain method, for elastic displacement field around dislocation in quasicrystal, 3: 390, 395 Eknian’s rule, 31: 239, 240, 242 EL2 defect [in AsGa], 1: 580 Elastic aniosotropy, dislocation affected by, 3: 364 Elastic compliance, 2: 397 Elastic-compliance-constant matrix, 1: 876, 877 Elastic constants Elastic compression, at high pressuresJ 3: 161 ab mitiaifirst-prinaples ca~cu~a~ions, 196,199-202,203--207,920-921; 148,238 calculated values binary intermetallic compounds, 1: 204-206, 876
978 determina~ionof errors in calculation, 1: 208 deviation from experimental values, 1: 206 ~onatoniicmetals, 1: 203-204 polycrystalline phases, 1: 206 sources of error In calculation, 1: 203-204 effect of magnetic transition, 1: 889 effect of order-disorder transition, 1: 888-8 89 effect of point defects, 1: 580 first-principles calculation, 1: 195-209, 874 of single crystals, 1: 873-874, 875-876 of transition-metal aluminides, 1: 138
see also Bulk modulus; Foisson’s ratio: Shear modulus . . .; Young’s modulus Elasticity theory Eshelby’s method, 3: 393-394 for quasicrystals, 3: 385-389 general equations, 3: 385-387 Stroh theory. 3: 393 Elastic moduli AlFe, 2 207 Al-Fe-Si-V alloys, 2: 184 of AlNi, 2: 56 dynamic modulus, 1: 873 of high-temperature intermetallics, 2: 238, 241-242 magnetic-field dependence, 1: 889, 890 and melting temperatures, 1: 891-892; 2: 252 of polycrystal~ineinte~etallics,1: 882-886 of silicides, 2: 213 static modulus, 1: 873 Elastic (neutron) scattering, 3: 250 Elastic properties, 1: 873-892 anisotropy in single crystals, 1: 874, 876-8 80 beryllides, 3: 42 colored compounds, 3: 240, 241 composition dependence, 1: 886-887 of dislocations, 1: $80482, 883, 884 metallic glasses, 3: 697-698 niobium silicide composites, 3: 555 orientation dependence, 1: 877-880, 881, 882 pressure dependence, 1: 890, 891 temperature dependence, 1: 887-890 Elastic stiffness constants, 1: 874 calculated values listed, 1: 876 measured values listed, 1: 875 Elastic-strain interactions, 1: 867 Electrical appli~ations,2: 469-470, 651-653 Electrical behavior, 1: 941-961 Electrical conduction in intermetallic compounds, 1: 945-960 effect of chemical composition, 1: 953-954 effect of dislocations, 1: 958-960 effect of grain boundaries, 1: 954-958 effect of impurities and vacancies, 1: 954 effect of ordering, 1: 945-948 in metals, 1: 942-945 Electrical conductivity
Subject I d e x in Ag-Mg system, 1: 8 disilicides, 3: 235 of Condal alloys, 2: 525 factors affecting, 1: 265, 943-945 of liquid alloys, 1: 669-671 temperature dependence, 2: 470 temperature effects, 3 246 and thermal conductivity, 1: 1026 of thermoelectric materials, 2: 463,464, 466,467,469 Electrical connectors, 2 541-543 Electrical equipment, value of IMCs, 3 793
Electrical resistance, changes during martensitic transformation, 1: 830 Electrical resistivity calculation of, 1: 944-945 effect of antiphase boundaries, 1: 946-947 effect of order, 1: 777, 778, 946-947, 948 ef€ect of point defects, 1: 567, 568, 580 of heavy-fermion compounds, 1: 212, 213, 214,219 magnetic disorder component, 1: 944 models, 1: 942, 945 order changes studied by, 1: 777, 778, 78 1 phase transitions studied by, 1: 8, 941 point defects studied by, 1: 562, 567, 568, 577 of quasicrystals, 1: 484 of shape-memory alloys, 2: 538 of silictdes, 2: 21 1, 231 of trialuminides, 2: 157 of zirconium aluminide, 2: 134, 136 Electrical steels, 2: 317 Electrical-to-mechanicalenergy conversion, 2: 318 Electric-field-assisted reaction synthesis, 3: 723, 725, 734-735 voltage~resistance/currentinputs, 3: 735
and wave velocity, 3 735 Electricity generators, 2 381-383, 455, 457, 458-460 Electric power transmission Lines, 2: 380-381, 382 Electroanalytical techniques, 3: 30 Electrocataiysis, 2 504 precious metal compounds used in, 3: 78 transition-metal silicides, 2 505 Electrocatalysts, 2 503 E~ectrochemicalapplications, 2 501-51 1, 646 Electrochemical factor, 1: 232, 270 in AF-AN plots, 1: 230, 233 atomic properties grouped under, 1: 23 1 and solid-solution tendency, 1: 244,248 and structure mapping, 1: 241, 242 Electrochemical oxidation, 1: 965, 966 Electrochemical parameter (structure maps), 3: 91 Electrochemical properties, dental amalgam, 2 586-587 Electrochemical reduction, 1: 965, 966 amalgams prepared by, 3: 25 Electrode materials, 2 504, 646, 653 Electrodeposited coatings, 2: 521
Electrodeposition, 1: 652-653, 733; 2: 506-507, 51 I nanostructures synthesized by, 3: 752, 754 thin films synthesized by, 3 670-67 1, 673
Electrodischarge machining, 2: 65, 68 Electrolu~ninescence,2: 418, 424 Electroma~neticapplications, 2 301-471 see also Electrical . . .; Magnetic . .; Magneto-optical . . .; Magnetostriction . .; Optical . .; Semiconductor. . .; Superconductor . . .; Thermoelectric applications ElectrometallL~r~ical recovery, 2 518 Electromigration, 2: 654 Electromotive force (EMF) methods, thermodynamic properties determined using, 1: 94-95, 97, 120, 663 Electron-beam evaporation, synthesis by, 3 666, 781, 784 ~lectron-beamphysical vapor deposition (EB-PVD) process, 2 492, 493 Electron beam welding, 1: 655 Electron compoundsr 1: 763 Electron density, binary alloys, 1: 282 Electron diffraction see Reflection highenergy electron diffraction Electronegativity as atomic property, 1: 231, 232 in close-packed crystal structures, 1: 279, 282, 298 listed for various elements, 1: 102, 233, 279,423 principle, 1: 12 in solid-solubilitymapping, 1: 245-246, 247 in structure mapping, 1: 242, 243, 424 Electronegativity diRerence embrittlement affected by, 1: 907, 908 enthalpy of bonding of Hunie-Rothery phases affected by, 1: 105 enthalpy of formation of HumeRothery phases affected by, 1: 101 interaction between internietallic compounds determined by, 1: 712 liquid alloys, 1: 662 plot vs average of principal quantum number, 1: 357, 420 plot vs electrical conductivity, 1: 265 plot vs ~seudopotentialradii difference, 1: 262 plot vs valence-electron number, 1: 388 structural stability affected by, 1: 420 Electron energy-loss spectroscopy (EELS), 1: 612; 2: 191; 3: 147-148, 150, 785 principle, 3: 148 Electroneutrality rule, 1: 301 Electron hole calculations, 2: 9 Electron factors close-packed structures affected by, 1: 417. MoSi,-type phases affected by, 1: 388 see also ~alence~electron concentration Electronic circuit boards, 3: 512, 676 see also Microelectronic applications Electronic heat capacity, 1: 1018
979
Subject Index Electronic properties, quasicrystals, I: 484-485 Electronics applications, 2: 2 13, 229-233, 327, 330, 332-335, 383-384, 652, 653-654 Electronic structure calculations, 3: 196197 and color, 3: 233-234 of FeSi, 3: 144 investigation of, 3: 135-1 51 Electronic structure, behavior explained by, 1: 21 Electronic theories of phase stability, 1: 21-51, 437 Electron irradiation amorphous phases produced, 1: 694495, 811-812, 813, 814 defects produ~ed,1: 560, 567, 573, 578-579, 793, 794 effects compared with those of ion irradiation, 1: 813 metastable phases produced, 1: 706 order-disorder transformation affected by, 1: 803 recovery after, 1: 794, 795 Electron microscopy weak-beam (WB) technique, 1: 496, 504, 521; 2: 214, 268, 271 see also High-resolution . . .; Scanning .; Transmission electron microscopy Electron mobility, 2 332, 344 Electron-momentum distribution, 3: 288 Electron-per-atom ratio, structural stability affected by, 1: 420, 424 Electron phases, 3: xxxvi Electron-phonon coupling, 1: 153, 157-1.59, 952-353 Electron relaxation time, 3: 270 Electron-photon interactions, applications based on, 2 41 8-432 Electron screening, 1: 128, 157 Electron smoothing of surfaces, 3: 22022 1 Electron spectroscopy for chemical analysis (ESCA), 3: 136 Electron transport, and dynamic atoniic disorder, 3 256-257 Electron vacancy number, and creep resistance, 3: 308 Electro-optical properties, database on, 3: 816,817 Ele~troplating IMGs synthesized by, 3: 670-671 thin films deposited by, 3 670-671 see also Electroch~m€cal deposition Electrorefining, 3: 29 Electro-spar~-deposition(ESD) technique, iron aluminde coatings prepared by, 3: 572 Electrosyntheses, 2: 507 El~inghamdiagrams, 1: 978 Elongation properties see Ductility Elser indexing scheme, 1: 460 Embedded-atom method (EAM), 1: 77-89; 3: 196,765,766-767 a~plicability,1: 88, 507, 536, 569, 570, 611, 618, 874, 876; 2: 21, 58 applications, 3: 768-775 energy expression for alloys, 1: 79-80
for pure elements, 1: 77-78 example calculations, 1: 85-88 free surfaces, 1: 85 grain boundary cohesion in Ni,AI, 1: 86-88 interstitial boron in Ni,AI, 1: 85-86 future developnients, 1: 88-89 irradiation effects modelled using, 1: 797 modified ( ~ E A M ) 3: , 457, 458, 765, 767-768 phonon dispersion curves calculated using, 1: 155, 156, 164 physical interpretation, 1: 78-79 potential fitting for AI-B-Ni system, 1: 82-85 cross potentials, 1: 83-85 for pure elements, 1: 82-83 similar methods, 1: 80-82 and surfaces, 1: 85, 611, 618, 620 see also Glue model; Second moment approximation Embedded-atom potentials, 3: 201, 21 5, 766 Embedded-cluster method (ECM), 1: 24, 25, 26, 30 E~bedded-de~ect (ED) potential, 3: 768 Embrittlement by environment, 1: 927-928, 989, 993, 995; 2: 28-34? 124, 144, 204 by impurities, 1: 904, 907, 908 see aZso Br~ttle-to-duct~~e . .; Environmental . . .; Hydrogen . . ., Oxygen embrittlement Emission spectroscopy, vapor species studied using, 1: 681 Empirical prediction methods, 3: 811-8 12 Empty states distributions, 3: 146 Enantiomorphous structures, 1: 292-293 Enargite [mineral name], 1: 353, 353 ~ ~ c ~ c l o pB a er ~ ~~ ua ~3:~ 873, ~ c a876 , Energy of alloy formation, 1: 850 Energy conversion devices, 3: 698, 702 efficiency of, 2 455-456 electrochemical, 2 507-5 11 photovoltaic, 2: 418, 423 thermoelectric, 2: 453-469 Energy density [for permanent magnets], 2 304 Energy dispersive X-ray (EDX) mapping, 3: 619, 620 Energy gap, 1: 109 correlation with enthalpy of formation for III-V compounds, 1: I10 Energy of mixing see Mixing energy Energy product [for permanent magnets], 2 306, 318 Energy-storage applications, 2: 475, 484-485, 509-510, 637; 3: 107-108, 250 see also Fuel storage; Rechargeable batteries Energy-storage densities, 2 643 Energy threshold criterion, in shockassisted reactions, 3: 742 Engel-Brewer compounds, 3 xxxvii, 68 Engel-Brewer correlation, 1: 63 Engel, Niels N., 3: xliii Engineering, ineta-sites on Internet, 3: 865
Engine thrust, 3: 472 Enthalpy of formation, 1: 92, 234 of alummides, 2: 621; 3: 759 components of, 1: 1 determination, 1: 96, 97 of ~ r i ~ m - ~ o m m e r f ecompounds, ld 1: 110 of Hume-Rothery phases, 1: 100, 101, 102 correlation with entropy of formation, 1: 106-107 of Laves phases, 1: 107, 108 effect of axial ratio, 1: 109 and phase diagrams, 1: 113-114, 258 relationship with atomic fraction, 1: 117 number of stable compounds per system, 1: 258 of solid solutions, 2: 613 of ternary compou~ds,1: 121, 122 of Zintl phases, 1: 111 see aZso Partial enthalpy of formation Entbalpy of phase f o r ~ a ~ i o 1: n ,920 Enthalpy of transfori~ation,1: 104 values listed for various elements, I: 105 Entropy of formation, 1: 92, 119-120 of Nume-Rothery phases, 1: 106-1 07 Entropy of fusion, 1: 120-121 of glass-forming liquids, 1: 736, 737 listed for various compounds, 1: 121; 2: 639, 642 Entropy of mixing, 2: 638 changes on fusion, 2: 641 of liquid alloys, 1: 665, 666 see also Configurational entropy Environment, fracture behavior affected by, 1: 897, 905-906 Environniental effects on creep, 3: 303-304, 320-321 on fatigue, 3: 336337, 337, 339-341, 346, 347 Environmental embrit~lement,1: 989, 993,995 in alum~nides,2 28-29, 124, 144, 204 at elevated temperatures, 2 30-34 graiii-bouiidary brittleness caused by, 2: 28-29 iron aluminides, 3: 339, 340-341, 613 titanium aluminides, 3: 336-337 Environmental resistance, hightemperature intermetallics, 2: 243 Epttaxial films, 3: 779, 784 see also ~olecular-beamepitaxy Epitaxial growth, GaAs on Gap, 1: 179-180; 3: 783. 784, 785 Epitaxial silicides, 2: 231-232 Epitaxy techniques see Atomic-Iay~r. ., Liquid-phase . .; Molecularbeam . . .; Vapor-phase epitaxy Epsilon ( 8 ) phase, 1: 105; 3: xxxvii Equation of state, determination by firstprinciple calculations, 1: 199 Equation of state. universal, 3: 766 Equiaxed investment casting, 3: 609 Equiaxed nanocrystalline metals, 3: 750
980 Equi-channel angular extrusion (ECAE) technique, 3: 628 Equilibrium concept [for formation of IMCS], 3: 28 Equilibrium nanocrystalline IMCs, 3 756-759 ErCd, structure type, 3: 88, 89 Er-Nd system, 1: 407 Erosion resistance, effect of coatings, 2: 489,496 Erosion resistance, ruthenium aluminides, 3: 61-62, 69 Escaig effect, 2: 26 Eshelby's method, for elastic field around dislocation in quasicrystal, 3: 392393 Eshelby twist, 1: 521 Eta (q) carbide compounds, 1: 396, 400; 3: xxxvii Eta (U)phase, 1: 105, 627 Cu,Sb, 1: 627 Hume-Rothery phases, 1: 105 Etch-pitting, dislocation mobility measured using, 1: 523 EuAuGe structure type, 3: 95 EuAuln structure type, 3: 95 EuAuSn structure type, 3: 95 EuMGe structure type, 3: 91, 93, 93 Eutectic alloys features of deep eutectics 1: 689, 691, 737 glass formation in, 1: 692, 737 Eutectic blades, 3: 544 Eutectic crystall~zation,1: 747 Eutectics, directionally solidified, 3: 544 Eutectic reaction, 1: 5 Eutectic transformation, 2: 640 thermodynamic equations for, 2: Excess free energy metastable phases, 4: 688, 693 of mixing, 1: 103, 104, 666 Exc~angecorrelation (XX) function, 1: 132, 138, 197 Exchange correlation (XC) hole, 1: 129 External deformation, 1: 405 Evaporation, thin films deposited by, 3: 666-667, 473, 784 Exo-Melt process, 3: 503, 604-606, 61 1, 612 benefits, 3: 606 Exomold technique, 3: 542 Expert systems, 3: 818-819 Extended X-ray absorption fine structures (EXAFS), 3: 147, 650 Extrinsic toughening, 3: 78 Extruded bars, of Al,5Nb54Ti&r,alloy, 3: 852 Extruded iron aluminides, 3: 505, 655 Extruded products, homogeneity, 3: 628 Extrusion of cast FeAl, 3: 655 comb~~stion synthesis Eollowed by, 3: 736 of powder preforms, 3: 646 of TiAl alloys, 3: 627-630, 636 and yield stress, 3: 630
Extrusion processes, 1: 654-655; 67, 153, I61 Eyeglass frames, 2: 556 F [flat] crystal faces, 1: 170 representation in stereographic projection, 1: 169 FA-I29 alloy, 3: 573 casting of, 3: 613 composition, 3 333 fatigue behavior, 3: 332, 339 FA-186 alloy, 3: 573 Fabrication by multistep forging, 3: 623 of TiAl automotive valves, 3 636 Fabrication methods, 3: 509-51 1 Fabrication techniques, 2: 43, 87 Fabry-Perot ~nterferencefilters, 2: 410 Fabry-Perot interfero~eter,2: 412 Fabry-Perot lasers, 2 338, 339 Face-centered cubic (f.c.c.) compounds deformation twinnin~in, 3: 41 1-415 surface structure, 3: 2 17-2 I9 see aZso B1. .; B3.. .; C1. ...; C15 compounds; Cubic close-packed (c.c.P.) structures; DO, structure Face-centred cubic (f.c.c.) lattice AI-B-Ni system, properties used to fit cross potentials, 1: 83, 84, 85 ground-state diagram, 1: 853 ordering maps, 1: 35, 36 and other cubic structures, 1: 169, 281 stability in transition metals, 1: 58 see m'so B1 . . .; B3 . . ., C1 . . ., C15 compounds; Cubic close-packed (c.c,p.) structures; DO, structure; L1, structure Face-centered cubic (f.c.c.) metals strain hardening in, 3 362-363 surface energies and stresses, 3: 226 Face symmetry, 1: 309 FAL alloys, 3: 573, 613 Famatinite [mineral name], 1: 344,353,359 Fano-like interactions, 3; 142-143 FAP-Y alloy composition, 3: 333 fatigue crack growth in, 3: 339 Faraday effect, 2: 435, 438 Farkas interatomic potentials, 3 772, 773 Farkas-Jones (FJ) interatomic potentials, 3: 774 Farkas-Mutasa-Vailh~Ternes (FMVT) interatomic potential, 3: 769, 770, 77 1 FAS alloys, 3: 573, 613 Fasteners, 2: 541 Fast-ion phases, disorder in, 3 258-252 Fast-neutron irradiation A15 superconductors affected by, 1: 808 properties of AlZr, affected by, 2: 138, 143, 144 Fatigue, 3: 325-350 compared to conventional alloys. 3: 347 and creep, 3 298 of FeAl alloys, 3: 332, 339-341 of Fe,Al alloys, 3: 331-332, 338-339 future research, 3: 348 microstructural effects, 3: 334-335, 337, 347 of Nb3A1alloys, 3: 332-333
of NiAl alloys, 3: 330-331 of Ni,Al alloys, 3: 329-330 of TiAl alloys, 3: 328, 1334337,476, 479480,484 of Ti,Al alloys, 3 328-329, 337-338 variables, 3: 325-326 Fatigue behavior 2: ,586 dental a ~ a ~ g a m iron aluminides, 2 207 nickel a~um~nides, 2: 32, 40, 41, 65 titanium aluminides, 2: 74, 83 Fatigue crack growth, 3: 333-346 effect of coatings, 3: 571 effect of ductile particulate reinforcerilents, 3 336 environmental effects, 3: 336-337, 337, 339-341, 346, 347 frequency effects, 3: 325, 337-338, 339, 347 and hydrogen, 3: 337, 339, 340, 347 iron aluminides, 2: 207; 3: 338-341 MoSi, and composites, 3 342, 344345 nickel alum~nides,3 341-342 niobium aluminides, 3 342 3: 491 ~ ~ o b i u m - s i composi~~s, li~~~ and R ratio, 3: 325, 347 TiAl alloys, 3: 334-338,479-480, 483-
Fatigue properties, s h a ~ ~ m e ~ alloys, ory 2: 538 Fault stabilities, methods of predicting, 2 20-22 Fault~toleranto p t o e l e c t ~ ~ i c components, 2: 414, 415 FauIted dipoles, strain hardening affected by, 3: 367 Faults ab initio ca~culations,3: 457-458 atoinistic calculations, 3: 452 in B2/CsC1 structure, 3: 451-453 in C l l ~ ~ ~structure, o S i ~ 3: 454-458 in DO,/BiF, structure, 3: 453454 in DO,,/Ni,Sn structure, in DO,/Al,Ti structure, displa~mentvectors lis energies of, 3: 4 4 3 ~ ~ 4 , 52, 453,455 difficulties in determining, 3: 462-463 variation with temperature, 3: 46 in L1,JAuCu structure, 3: 442, 44 in Ll,/Cu,Au structure, 3: 440-442 in L2,/Ni2A1Ti structure, 3: 453-454 multi layer^ 3: 464 temperature dependence, 3 461-462 thickness vanation, 3: 462 see also Stacking faults FAX alloys, 3: 573 F.c.c, see Face-centred cubic . . , FeAl alloys see AlFe alloys Fe,Al alloys see AlFe, alloys Feature, meaning of term, 3: 835 FeB structure type, 3: 89, 89 57FeMoss~auerspectra, 3: 167, 168 Fe-Ga-Ni system, 1: 726 Fe-Gd system, 1: 703, 735 Fe-Mn-Ni system, 1: 721, 726 Fe-Mn-Si system, 2: 557, 558
98 I
Subject Index Fe-Ni,AlTi system, two-phase microstructure, 3: 843-844 Fe-Ni-Pd system, 1: 721 Fe-Ni-Pt system, 1: 721, 725, 726 Fe-Ni system, 1: 632,634,683,830; 2: 306 Fe-Ni-Ti alloys, twinning in, 3: 416 Fe-Ni-Ti system, 1 :7 16-7 17, 727 Fe,NW, structure type, 1: 395 Fe-I'd system, 1: 632; 2: 557 Fe-Pt system, 1: 632, 829; 2: 556, 557 Ferchromide [mineral name], 1: 628, 632 Fermi-Dirac statistics, 1: 128, 1018 Fermi edges, 3: 139 Fermi energy, 1: 942, 1018 Fermi energy level in colored compounds, 3: 234, 236 in epitaxial films, 3: 785 Fermi, Enrico, 3: xliii Fermi level, 1: 942 mid quasigap, 1: 57, 63, 73 Fermi liquid, 3: 45-46 F e ~ - l i q u i dmetals, 1: 211 F e ~ i - l i q u i dstate, 1: 214, 21.5 Fernii-liquid theory, 1: 128, 130 Fermiology, 1: 135 Fermion compounds see Heavy-fermion compounds Fermi surfaces, 1: 28 iii alumiiiides, 1: 138, 141, 142, 143 band structure near, 1: 135 distortion near Brillouin zone, 1: 100-101, 132 mapping/measurement, 1: 135, 136, 137; J: 139, 140 Fermi velocity, 1: 942 Ferr~magnetism,1: 439, 935, 938 Ferrites, 2: 317, 390 Ferritic alloys precipitation strengthening in, 3: 64-65 with AlNi precipitates, 2: 277-278 with Fe,SiTi precipitates, 2: 278-279 Ferroalloys, 2: 523 Ferroelastic phases, disorder in, 3: 258259 Ferroelectrics see High-temperature ferroelectrics Ferromagnetic glasses, 3: 691, 698-699 appli~dtions,3: 682, 698, 702 Ferromagnetic matenals, electrical resistivity of, 1: 944 Ferromagnetic moment, 1: 938 Ferromagnetic-paramagnetic transition, 1: 947,949 Ferromagnetic semiconducting compounds, 2: 330-332 Ferromagnetic-to-antiferromagnetic transition, 3: 177, 179 Ferromagnetis~,1: 439, 935, 937 in heavy-fermion compounds, 1: 215 Ferroiiickelplatiiium [mineral name], 1: 628, 631, 632 Ferrous shape-memory alloys, 2 536, 556, 557, 558 Fe-Ru-Zr amorphous system, magnetic phase diagram, 3: 182, 183 Fe-Sb naturally occurring compounds, 1: 632 FeSi structure type, 1: 261, 422, 426 Fe-Si system, 1: 662, 969-970, 999 Fe-Si--Ti alloys, strengthening of, 2: 279
Fe-Sn system, 2 448 FeS, structure type, 1: 388, 389, 428, 429 Fe,R structure type, 2: 3 16 Fe,,R, structure type, 2 316 F e T b system, phase diagram, 2: 396 Fe-Th system, 1: 703, 705 Fe-Zn system, 3: 157 Fe-Zr amorphous system, 3: 181-183 magnetic phase diagram, 3: 182 phase diagram, 3: 182 Fe-Zr system, 1: 696, 735, 740 Fiber-reinforced composites, 290 creep resistance, 3: 318-3 19 fatigue behavior, 3: 337, 338 oxidation behavior, 1: 1007, 1008 Fibonacci, Leonardo, 3: xliii Fibonacci sequence, 1:459, 461, 480 Fibonacci series, 1: 457, 467, 476 Fibonacci superlattices, 3: xxxvii Field-effect traiisistors (FETs), 2: 332-334, 335, 347 Field ion microscopy (FIM), 3: 212, 225 Field ion microscopy (FIM) observations with atom-probe analysis (FIM-AP), 1: 595, 596 grain boundaries, 1: 590-591 Figures of merit magnetic materials, 2 318 magneto-optical materials, 2: 442 superconductors, 2: 366, 367 thermoelect~cmaterials, 2: 456, 461, 462,463, 464, 466,467 Filled-billet extrusion technique, 3: 646 Filled skutterudites, 3: 106 Filled structures, 3: xxxvii Filled zinc blende structures, 3: 238 Filter materials, 3: 272 Filters, 2: 330, 410411 Fine-filament superconductors, 2 357, 366 FINEMET alloys, 3: 702, 754 Fineness [gold jewelry alloys], 2: 559 Finite-difference techniques, oxidative lifetime prediction using, 1: 1009-1010 Finnis-Sinclair (FS) potentials, 3: 20 1, 215, 226, 766 Fire-stfety devices, 2: 545, 546, 550 First-order magnetization processes (FOMPs), 3: 169 First-principles calculations, 1: 195 computational details, 1: 198-199 for construction of phase diagrams, 3: 185, 192 for crystal structures, 1: 230 elastic constants determined, 1: 196, 199-202, 203-207; 2: 148 future developments, 1: 207-208 limitations~1: 196, 208 for surfaces, 1: 611, 617 theoretical background, 1: 19Gi-198 see also uh initio calculations Fission-neutron irradiation, effects, 1: 804-805 V-VI compounds semiconductors, 2: 329 thermoe~e~tric materials, 2 457, 460, 462464,466 Flame spraying, 1: 642 Flinn's model, 1: 546 Floating-zone (FZ) method, 1: 643; 2: 68, 153, 221, 395
3: 545, 546 Float zone proce~sing~ Flory's formula, 1: 675 Flow stress effect of off-stoic~io~etry, 3: 357 and strain hardening, 3: 361 373 Flow-stress anomaly, 1: 914-915; 2: 243; 3: 372 in B2 alloys, 1: 536 in L1, alloys, 1: 532-533 in Ll, alloys, 1: 527, 546-549, 914-915; 2 24-25 locking of screws by elongated KW segments, 1: 548-549 by pinning-point formation, 1: 547-548 Fluorescent lamps, 3 32 Fluxing, heteronucleants removed by, 3 683-684 Foaming agents, 2: 646 Focusons, energy dissipated by, 1: 794 Foiles-Daw (FD) inte~atomicpotential, 3: 769 Foil rolling, 1: 655; 2: 128 Foils ferromagnetic glasses, 3: 698 TiAI, 3: 633 Forbidden twinning, 3: 410, 43 1 Force-constant models, 1: 153-1 54 Forging, 1: 653-654; 2 86, 127, 161; 3: 626-627 canned, 3: 627 and eutectoid temperature, 3 623 isothermal, 3: 626-627 of TiAl alloys, 3: 626-627, 635 and yield stress, 3: 630 Formation of compounds, 1: 623-751 IV-IV compounds, semiconductors, 2: 325, 327-328, 348 IV-VI compounds semiconductors, 2: 325, 328-329 thermoelectric materials, 2: 458,464-465 Formation temperatures aluminides, 3: 676 silicides, 3: 675 Forming, 3: 617-4542 and coniposition, 3: 618-620 effects of canning, 3: 627 of FeAl, 3: 639 of Fe,Al, 3: 639 and microstiucture, 3: 618-620 of Ni&, 3: 639 and phase distribut~on,3: 632 by powder metallurgy, 3: 633 and prism glide, 3: 623 temperature control in, 3: 631-632 of TiAl sheet, 3: 636-638 of Ti,Al, 3: 639 of titanium aluminides, 3: 617-638 Forming processes modeling of, 3 630-631 primary, 3: 626-630 secondary, 3: 631-638 Forsterite packing coefficient, 3 7 str~cturalformula for, 3: 5
982 Foundry revert stock, 3: 607 Fractographs, AlZr,, 2: 141 Fracture, 1: 895-909 aluminides, 2: 27-30, 64-65, 140-143, 202 B2 compounds, 1: 904-906; 2: 64-65 dental amalgam, 2: 586 effect of boron, 1: 896, 898-904,906; 2 27-30, 64, 251 and environment, 1: 897, 905-906 and grain size, 1: 902, 903 L1, conipounds, 1: 896-904, 907; 2: 27-30, 140-143 mechanisms, 2: 33 and processing history, 1: 898 and quantuin-mechanical calculations, 1: 907-908 and stoichiometry, 1: 897, 904-905 trialuminides, 2: 166 see uho Cleavage ., Intergranular fracture Fracture behavior compared to conventional alloys, 3: 474, 476, 495 niobmm-silicide composites, 3: 550, 552 role of surfaces in, 3: 226-227 Fracture maps, 1: 9 18 Fracture strength, 1: 873 Fracture surfhce, of NiAl single crystal, 3 227 Fracture toughness Al-Fe-Si-V alloys, 2 195 aluminides, 2 64-65, 78, 85, 92, 120-122, 143-144 beryllides, 3: 43, 44 disilicides, 2: 221; 3: 242, 485 ininimum for critical components, 3: 495, 550 molybdenum-silicon alloys, 3 489 nickel aluminide, 3: 494 niobium-silicide composites, 3: 490,552 Ti-A1 alloys, 3: 48W81, 495 and twinning, 3: 426-427 see also Toughness . . . Frameworks, 1: 3 17 D(60j Framework, 1: 325, 327, 328, 331, 332 of 1 family, 1: 325-336 1 framework, 1: 317-325 see also 1 framework Framework structure types, 3 18 FrangibolP, 2: 555 Frank criterion, 1: 524 Frank, Frederick Charles, 3: xliii Frank-Kasper phases, 1: 73, 409, 473-475, 479, 480, 740; 3: xxxvii Frank partials, 3: 423, 430 Free energy of formation, 1: 92 determination, 1: 93-97 metastable crystalline phases, 1: 687, 703, 704 metastable phases, 1: 687 of inixing amorphous phases, 1: 696, 698 liquid alloys, 1: 103, 104, 665, 666, 673, 677 and ordering, 1: 758 see also Gibbs . Free energy curves, 3: 188-189, 189-190
Subject Index phase diagrams generated from, 3: 189, 191 Free-energy diagrams, 1: 852, 854 Free surfaces, 1: 609-621 analytical techniques, 3: 2 14-2 15 and atomistic simulations, 3: 21 5 and fracture properties, 3: 2 11 properties, 3: 225-228 property measurement, 3: 215 relaxation at, 1: 85, 616-618 structure, 3 21 1-225 see also Surface., , French gold, 3: 77 Frenkel defect-pairs, 1: 693, 793, 798 irradiation-induced amorphization affected by, 1: 819, 820 Fresnel reflection coefficients, 2: 436 Fresnel transmission coefficients, 2: 436 Friauf, James B., 3: xliii Friauf-Laves phases ,see Laves phases Friction joining, 1: 656 F~edel-Anderson lattices, 1: 214 Friedel-Fleischer statistics, 2 259, 260, 263, 264, 266, 268 Friedel oscillations, 1: 154 Froodite [mineral name], 1: 628, 631, 632 Frozen-phonon calculations, 1: 73, 156 Frustrated f.c.c. alloys, 1: 614 Frustrated Kondo lattice model, 3: 181 Frustration and paramagnetic phases, 3: 181 and phase diagrams, 3: 195, 206 Frustration [magnetic phenomenon), 1: 938-939 Fuel cells, 2: 507-509 Fuel injection valves, 2: 403 Fuller, Richard Buckminster, 3: xliii Fulleranes/fullerenes, 3: xxxvii, 8, 9, 255 Fullerene, bombardment by, 3: 267, 268 Fullerene-like cages, Zintl phases with, 3: 126,127 Fullerides, 3 xxxvii dynamic disorder in, 3: 257-258 Full-potential augmented plane wave (FLAPW) ~~lculations, 3: 202 Full-potential cellular methods, 1: 133 Full-potential linear augmented planewave (FLAPW) method, 1: 31, 60 applications, 1: 47, 69, 70, 501, 505, 506, 507, 874, 879; 2: 148 Full-potential linear muffin-tin orbital (FLMTO) calculations, 3: 199 Full-potential linear augmented Slatertype orbital (FLASTQ) method, 1: 60 applications, 1: 64 Full-potential linearized plane-wave (FLAPW) method, applications, 2: 148 Full-potential linear m u ~ n - t i norbital (FLMTO) method, 1: 60 applications, 1: 67, 874, 876 Furnace components, 3 512,514 Furnace fixtures, 2: 45 Fusion properties, 2: 642-644 Fusion research, 2: 366, 371-376 G phases, 2: 61; 3: xxxvii, 308, 316 GaAs-Co system, 1: 696 Ga,,HoNi,, structure type, 1: 412, 415 Gadolinium, inagnetocaloric properties, 3: 519,520, 527, 530,530
Galfan-coated steel, 2: 520 Galfan coatings, 2: 519 Gallains, 3: 33 Gallides, 1: 364; 2: 621 data sources, 3: 808 prediction of formation, 3: 829-830 Gailim. as metallic solvent, 3: 33 Gallium arsenide, as substrate, 3: 783, 784, 787, 796 Gallium arsenide devices, fabrication of, 3 782-783, 787 Galiiuni-containing Zintl phases, 3: 119, 3 23 Gallium-mercury alloys, 3: 32, 33 Galvanized steel, 2: 519, 520; 3 791, 794-795 Galvanomagnetic coefficients, 1: 943 Galvanostatic i n t e ~ i t t e n titration t technique, 2: 502 Gamma brass phases, 3: xxxvii Gamma-prime phase, elemental partitioning to, 3 66 Gamma-prime (y’j phase, 2 3, 17, 257 elemental partitioning to, 2: 8-9 morphology, 2: 9-13 Gamma-prime precipitates. and creep properties of nickel-based superalloys, 3: 314-316 Gamma-prime rafting, 3: 3 14-3 15 Gamma-surfaces, 3 438, 439 in B2 compounds, 3: 452 in DO, compounds, 3: 454, 454 in DO,, compounds, 3: 447 iii L1, compounds, 3: 445, 446 Gamma (gj surface, 2: 20 Gamma TiAl and alloys, 2 73-88 see also AlTi-based alloys Gamma titanium aluniinide alloys, 3: 477-485 casting of, 3 591-602 see also y-Titanium aluminide Ga--Nb system, 2: 353 Ga-Ni-Si system, 1: 726 Ga-Ni-Ti system, 1: 718, 727 Garnet ion tracks in, 3: 267 three-dimensional packing in, 3: 7 Gas atomization, 3: 644 of aluminides, 3 504, 648, 649, 650, 654, 655 cornpared with PREP, 3: 649, 650 Gas atoniizations [processing technique], 1: 640, 641, 912; 2: 87, 17.5 Gas-cycle refrigeration, compared with magnetic refrig~ration,3: 520, 536 Gas gun, shock-consolidation by, 3: 739 Gas-phase charging [with hydrogen], 2 477, 486 Gas tungsten arc welding, 1: 655 Gas tungsten arc (GTAj technique, iron aluminde coatings prepared by, 3: 572 Ga-V system, 2: 354 Gd-Fe system, 1: 738 Gd-Ge system, 1: 120 Gd-La system, 1: 407 Gd-Nd system, 1: 407 Gd-Pr system, 1: 407 Gd-Si system, 1: 120 GE 48-2-2 alloys, 3 595, 597 GE 1541 alloy, 3: 715
983
Subject Index G c H g system, phase diagram, 3: 24 Ge-In-Ni system, 1: 726 Ce-La system, 1: 735 Ge-Mg system, 3: 157 Ge-Nb system, 2: 353; 3: 161 Generalized Darken Model, 3: 563-564, 568 Generalized 8-N rule, 1: 346 Generalized gradient approximations, 1: 130, 132 Generalized perturbation method (GPM), 1: 24, 25, 26, 28-30 applications, 1: 41, 44 Generalized stacking faults, 3: 438 Genera~"potent~a1 linear a u ~ e n t e d plane-wave method, 1: 197-1 99 see also Linear a u ~ e n t e plane-wave d ., General-use structural applications, 3: 50 1-5 18 GeNiPt, structure type, 1: 408 Genkinite [mineral name], 1: 628, 632 Geoffrey's table of affinities, 1: 4-5 Geometrical stability plots, 1: 249 binary compounds, 1: 250 ternary coiiipounds, 1: 251 see also Structure maps Ge,Os structure type, 1: 388, 389, 429 Geothermometer, 1: 633 Germanldes, 2: 621, 622 data sources, 3: 808 prediction of formation, 3: 830 G e ~ a n i u m%atom , cluster, in Zintl phases, 3: 115-1 16 Gerinanium-mercury phase diagram, 3: 24 Germanium octahedra, 1: 180 Ge-Sb system, 1: 703, 704 Ge-Sc system, 1: 120 G c S i system, 2: 465, 466 GeS structure type, 1: 261, 426 Getters, 2: 646 Ge-U system, 3: 1-77' Geversite [mineral iianie], 1: 628, 632 Ge-W system. 1: 707 Ghost bands, 1: 134 Giamei dislocation lock, 1: 545 Giant cells, 1: 411 Giant magnetocaloric effect, 3: 528, 529, 536, 691 Giant-magnetor~slstance(GNR) materials, 3 104, 215, 691 Gibbs-Duliem equation, 1: 93 Gibbs free energy 3: 521 amorphous phases, 1: 746 in calculation of phase diagrams, 3: 188 of forniation, 1: 92 determination, 1: 93-97 as function of atomic fraction, 1: 97-98 phase stability affected by, 1: 100 metastable phases, 1: 687 Gibbs-~elmholtz equation, 1: 92 Gibbs phase rule, 3: 186, 221 1: 866 G i b b s - ~ h o ~ s oequation, n Gibbs triangle, 3: 186 Giua-Giua-Lollini plot, 1: I2 Glasses devitrification, 1: 748, 749 icosaliedral short-range order in, 1: 479 structural similarities with quasicrystals, 1: 479-480
synthesis, 1: 689-703 vlscoslty, 1: 744 see also Aiiiorphous alloysjphases; Metallic glasses Glass formation and phase diagrams, 1: 691,737,738,750 role of concentration fluctuations, 1: 680-68 1 Glass-forrning alloys, examples listed, 1: 690, 735 Glass-foming tendency, 1: 689, 734 Glass-to-metal solder, 2: 523 Glass transition, 1: 679, 688, 731 temperature, 1: 690. 736 of water, 1: 702 Glass transition temperature reduced, 3: 683 requirements for metallic glasses, 3: 682, 700 typical valucs for metallic glasses, 3: 682 Glassy foils, 3: 698 Glassy rods, in tension, 3: 695 Glassy state, 1: 731-733 specific volume-temperature plots, 1: 732 Glissile-type twinning dislocation, 3: 408 Glossaries, 3: xxxv-xliv IMC families, 3: xxxv-xliv on Internet, 3: 876-877 mini-glossaries, 3: xxx Glow-discharge sputter-deposition techniques, 3: 665 Glue model, 1: 82; 3: 766 Glue term, 3: 215 Gold colored alloys, 3: 76-77, 232 extraction of, 3: 29 reflectivity data, 3 232, 233 Gold aluminides, jewelry applications, 3: 75-76, 77 Gold amalgams, 3: 21 Gold coatings, 2: 519-520 Gold coins, counterfeit, 3: 31 Golden mean, 1: 457 Gold jewelry alloys, 2: 559-573 Ag-Au-Cu alloys, 2: 56 1-569 age hardening of, 2 560, 565-567 aluminum-containing alloys, 2 572-573 classification of, 2 559-560 color variations of, 2: 559, 560, 571 nickel-containing alloys, 2: 570, 571 palladium-containing alloys, 2: 570-572 titanium-containing alloys, 2: 569--570 white gold alloys, 2: 570-572 Gold leaf, bonding of, 2: 519-520, 521 Gold-recovery processes, 1: 4; 2 515-51 6, 5 17-5 18 Goodinan's rule, 1: 247, 249 Google search engine, 3: 857, 858, 859, 860 Gorsky-~ragg-Williams (GBW) method/ model, 3: 191, 201, 202, 206 Gorsky effect, 2: 484 Government, meta-sites on Internet, 3: 857, 865 Graded coatings, 2 498, 499 Grain boundaries A15 compounds, 2: 353
in Al-Ni system, 3: 771-772 amorphizat~onat, 1: 816 atonic configurations, 1: 598, 599 atomic images of, 1: 586, 587 bonding at, 1: 603-604 calculations, 3: 841 corrosion reactions at, 1: 966-967 and creep, 3: 3 19-321 dislocation structures, 1: 588-590 in disordered alloys, 1: 902-903 electrical conductivity affected by, 1 : 954-958 slip transmission across, 1: 902 symmetrical tilt boundaries, 1: 600, 601 vacancies at, 1: 603 Grain-boundary brittleness, 2: 27-28 environmental embrittlement as cause, 2: 28-29 Grain-boundary chemistry AES observations, 1: 591-594, 605 direct observations, 1: 591-595, 605 FIM-AP observations, 1: 591, 595, 605 TEM-EDX observations, 1: 594-595 Gramboundary coheslon, AlNi,, 1: 86-88 Grain-boundary ei~brittlement,1: 904, 907, 908, 927 Grain-boundary enricliinent factor, 1: 594 Grain-boundary flux pinning, 1: 957 Grain-boundary hardening, 1: 595-597, 956 electrical properties affected by, 1: 956-957 elimination, 1: 596 and pest effect, 1: 596-597 Grain boundary hardness, 3: 228 Gram-boundary mobility retardation, 1: 863 Grain-boundary phase transforniations, 1: 955 Grain-boundary pinning, Grain boundary segregati creep properties affected by, 3: 306 in TiAl sheets, 3: 634 Grain-boundary sliding, 1: 917, 925; 3: 622 Grain-boundary structure, 1: 585-605, 863 Al-Zr system, 2: 143 direct observations, 1: 586-591, 604-605 by FIM, 1: 590-591 by TEM, 1: 586-590, 604-605 effects of boron doping, 1: 602 niodeling of, 1: 598-603, 605 computed atomic models, 1: 600-603, 605 geoinetric~cystal~ograp~ic models, 1: 598-600 properties interpreted, 1: 595-596 stoichiometric effects, 1: 600-601 Grain orie~tation,in ~agnetostrictive materials, 2: 395, 399 Grain-refining agents, 2 85 Grain size coarsening rates for Al-Fe-Si-X alloys, 2: 185 and hardness, 3 739, 761-762
984 oxygen embrittleinent aRected by, 2: 33-34 shape-memory alloys, 2: 538 tensile properties affected by, 1: 925; 2: 79 Grain texture, creep resistance affected by, 3: 306 Graphite intercalation compounds, disorder in, 3: 257 Gravity metal rnold (GMM) process, 3: 593 Gray cast iron, 2: 535 Gray tin, 1: 180, 277 see also P-Tin structure type Grazing-incidence X-ray scattering (GIXS), 3: 212 Green’s function method, for elastic field around dislocation in quasicrystal, 3: 391-392, 395 Griffith energy, 1: 86 Griffith’s relation, 1: 873 Grimrn, Hans August Georg, 3 xliii Gri~m-Sommerfeldcompounds, 1: 109-1 10, 118; 3: xxxviii Grimiii-Sommerfeld rule, 3: 114 Grinding, 3: 509 Gross domestic product, 3: 792 Grotthus mechanism, 3 257 Ground state calculations, 3: 197-199 Ground-state-derived properties, prediction by DF theory with LDA, 1: 130 Ground-state diagrams b.c.c. systems, 1: 851 f.c.c. systems, 1: 853 Ground states b.c.c.-based alloys, 1: 42 heavy-ferrnion compounds, 1: 214-2 18, 22 1 insulating-ground-states systems, 1: 2 17-2 18 Ground-state structures, 1: 850-851 Croup 13, Zintl phases, 3 117-121, 123125 Group 14, Zintl phases, 3 115-117, 120, 122, 130-131 Croup 15, Zintl phases, 3: 129-130 Group structure types, 3: I7 Growth kinetics, 1: 759-760 antiphase domains, 1: 773, 865, 866 bainite, 1: 839-840 oxide protective layers, 1: 992, 994, 997 see also Kinetics Growth twins, 3: 403, 410, 431 Griineisen functions, 1: 1017, 1018, 1022 Guided-wave optoelectronics, 2 416 Guinier-Preston zones, 1: 100, 759 ‘Guitar string’ eflect, 1: 1025 Gunn diodes, 2: 327 CW approximation, 1: 130 H phase, 1: 718; 3: xxxviii Habit planes, and martensitic Hahn, Harry, 3: xliii Hahn phases, 3: xxxviii Half-crystal position, 1: 170 Half-Heusler phases, 3: xxxviii Halides, 1: 299, 300, 301 Halite, structural formula for, 3: 4, 5
Subject Index Hall devices, 2: 344 Hall effect, 1: 943 Hall mobility [semicotiductors], 2: 327, 333 Hall-Petch relationship, 1: 902,925; 2 61, 79, 137, 138, 204; 3: 630, 652, 739, 761-762 inverseJnegative relationship, 3: 762 Hardening dental amalgam, 2 582-583, 584 gold-based alloys, 2: 565-567 see also Precipitation hardening; Solidsolution hardening; Work hardening Hardening rate, and shear-stress shearstrain curve, 3: 362 Hard magnets, 1: 939-940; 2 303, 305; 3: 97, 101-103 Hardness beryllides, 3: 43 of Condal alloys, 2: 525 grain-boundary, 1: 8, 595-597, 956 and grain size, 3: 739, 761-762 of high”temperature intermetallics, 2: 248, 249 and melting temperature, 2: 252, 253 nanocrystalline IMCs, 3: 762 Qf nickel a~uminide,2 598 of semiconductors, 2: 327, 328 of trialuminide compounds, 2: 163-1 64 Hard-sphere models, 1: 600 Hartree method, 1: 129 Hastelloy, 3: 304, 418 Hastelloy-X [superalloy], 2: 41, 598 Haucke, Werner, 3: xliii Haucke phases, 3 xxxviii Heat capacity, 1: 1017, 1019-1022 of Al-Cu system, 1: 1020 of aluininides, 1: 1020-1021 of Cu-Zn system, 1: 1019-1020 and magnetic refrigeration, 3: 49, 522, 524 and order-disorder phenomena, 1: 1 19, 1020 point defects affecting, 1: 562 and polymorphic transformations, 1: 118-1 19 of quasicrystals, 1: 480 temperature dependence, Zintl phase, 1: 110 Heaters, materials used, 2 469, 470 Heat of formation, 1: 15, 32; 2: 638 ab initio calculations, 1: 55, 61-65, 66-71, 73 listed for various compounds, 2 642 silicides, 3: 733 Heat of fusion, 2 638 listed for various compounds, 2: 643 Heat of mixing amorphous alloys, 1: 739, 740 liquid alloys, 1: 665, 666, 673 Heat pumps, 2: 455, 457-458, 460, 484, 485, 486 Heat of reaction, 1: 6 Heat of reaction, powder mixtures, 3 743 Heat of solution, 2: 638 Heats of vaporization, listed, 3: 214 Heat sources, 2: 645 Heat-storage applications, 2: 637-644 see also Fuel storage Heavy-fermion behavior, 3: 46, 176
Neavy-fermion compounds, 1: 21 1-222; 3: xxxviii, 46 alloying effects, 1: 221 band structures, 1: 130 electrical resistivity behavior, 1: 212, 213, 214, 219 ground states, 1: 214-218 insulating-ground-s~atessystems, 1: 2 17-21 8 magnetically ordered compounds, 1: 215-216,219, 221, 935 magnetic properties, 1: 217-21 8, 220-221,937 nuclear fuel applica~ions,2: 648 pressure effects, 1: 221 single-crystal growth, 1: 643 specific-heat behavior, 1: 212-213, 216, 217, 219,220 superconducting materials, 1: 216-217; 2: 229 thermal behavior, 1: 1018, 1028 Heavy-ion ~rradiation,effects, 1: 821 Helical magnetic structure, 3: 170, 171 He~ann-Maugin nomenclature, 1: 309 He~aiin-§kilInian calculations, 1: 23 1, 241,423 Heterocoordination, 1: 667 Heterodyne receiver, 2: 343 Heterogeneous disordering, 1: 780 Heterogeneous nucleation, 1: 748 Heterogeneous ordenng, 1: 774 Heterogeneous SRO, 1: 782 Heterojunction bipolar transistors, 2: 332, 335 Heterojunction phot~transistors,2: 430, 43 1 Heterojunctions, 2: 324, 334 see also Double hetero . . Heteronucleants, 3: 683 removal of effect on glass formation, 3: 682, 683, 700 by fluxing, 3: 683-684 by self-fluxing, 3: 684-685 Heteropolarity, 1: 101-102, 319 at grain boundaries, 1: 604 Heusler, Friedrich, 3: xliii Heusler alloys, 1: 15; 3: xxxviii ab initio calculations, 1: 69, 71 creep behavior, 3: 316 electrical properties, 1: 950 magnetic properties, 1: 445,935; 2: 303, 308 prediction of, 3: 831-832 splittings of I lattice complex, 1: 319,320 Hexaferrites, 2: 318 Hexagonal close-packed (h.c.p,) interstitial alloys, 1: 302-303 Hexagonal close-packed (h.c.p.) structures, 1: 278, 287, 288, 291 crystal equilibrium forms, 1: 173 twinning in, 3: 417-418 SCE also A3 . . ., Cl4 . . .; DO,, compounds Hexagonal dipyramid, 1: 404 Hexagonal Hunie-Rothery phases, 1: 105-1 06 Hexagonality, close-packed structures, 1: 406,407 Hexagonal layers, 1: 277, 285-287, 288, 295
S ~ ~Index ~ ~ c t HfCuSi, structure type, 3: 92 Hf-Ir alloys, 3: 70, 71 Hf-Nb-Si system, phase stability in, 3: 548-549,550 H f - N b 4 alloys, twinning in, 3: 415 Hf-Pt system, 3 70, 72 Hf,Sii,Cu structure type, 3: 89 Hg-In alloys, 3: 31 Hg-I system, 3: 25 Hg-K system, 1: 662, 668, 673 Hg-Mg system, 3: 158 Hg-Na system, 1: 662, 664, 673 Hg-Ni system, 3: 258 phase diagram^ 3: 25 Hg-Pb naturally occurring compounds, 1: 632 Hg-Pd naturally occurring compounds, 1: 632 Hg-Pd system, 3: 158 Hg--Pt system, 2: 518 Hg2Pt structure type, 1: 389, 409, 429 Hg-Sn system, 2: 408, 520-521, 576 HgS structure type, 1: 261 see also Cinnabar-structure compounds Hg-Ti system, 3: 158 Hg-Tl system, 3 24 High-coor~~nation-n~imber structures, 1: 414415 Hig~-dampingalloys, 2: 649-650 High-electron-mobility transistors (HEMTs), 2: 332, 334-335 High-energy ball milling, 3: 61 High-energy ion scattering (HEIS) technique, 3: 214-215 High-field magnets, 2: 351, 367-370 High-force deuces, shape-memory alloys used, 2: 555 h-melt~n~-po~nt metals, amalgams with, 3 23-24 High-perform~nc~ systems, introducing new materials, 3: 473477 h pressure, effect on equilibrium, 3: I54 Hi&h-pressur~ phases, prediction of, 3 161 High-resolution electron microscopy (HREM) observations GI-Li-Zr aystem, 2: 191 dislocations, 1: 521; 2: 152 quasicrystals, 1: 462, 467, 468, 475 Hi~h~reso~ution transmission electron microscopy ( H R T ~ observations ~) grain boundaries, 1: 586-591, 604-605, 903 see also Transmission electron
607, ~ 0409~ , High-resolution lattice fringe method, ~ u r ~ evectors rs of dislocations in quasicrystals determined by, 3: 383384
Hi~h-resolutiontransmission micro~copy(HRTEM) Burgers vectors of dislocations in quasicrystals determined by, 3: 384 See ulso Transmission electron microscopy.. .
High Speed Civil Transport program, 3 473 High-temperature aluminum alloys, 2: 178, 194 High"temperatur~ferroelectrics, and struct~iremaps, 1: 265, 266, 435 Highrtemperature intermetallics, 2: 237-255 cost considerations, 2: 250, 252, 254 mechanical properties, 2 240-241 physical properties, 2: 243-246, 249, 253-254 screening of compounds, 2: 241-252 extensive/less-intens~vesurvey, 2: 247-249 lii~~ted-choi~/data-intens~ve survey, 2: 245-247 selection criteria for, 2: 241-245 special tough alloys, 2 250-252 structure types, 2: 240, 244, 245, 249 ternary alloys, 2: 251-252 High-tempera~ureshape-~emoryalloys, 2: 556 High-temperature solution calorimetry, enthalpies of formation determined using, 1: 96, 97 H~gh-temperaturestrength, 2: 36, 60-61, 240, 241 High-temperature structural materials, 2: 650-651 affordability, 3: 476-477, 482-483 beryllides, 3: 42-43, 48 design issues, 3: 475-476 iridium-base superalloys, 3: 70 iron aluminides, 3: 496 nickel aluminides, 3: 492-494 niobium silicide composites, 3 552-553 platinuin-base superalloys, 3: 72 processing of, 3: 474-475 and property balances, 3: 474 refractory silicides, 3: 485-487 rhodium-base superalloys, 3: 70 silican-bearing alloys, 3: 487-492 ~itaniumaluminides, 3: 477-485, 494496 1: 16, ~ i ~ h ~ t e m p e r a superconductors, ~ure 216; 2: 211, 223, 351; 3: 205 band structures, 1: 130, 136 and electron-phonon interaction, 1: 153, 157, 952-953 gram-boundary eEects, 1: 958 heavy-fermion compounds, 1: 2 16-2 17 phase transition, 1: 162 and structure maps, 1: 265, 266, 435 High-tem~eraturethermoelectrics, 2; 465 Hill's appro~iii~atio~, elastic constants estimated using, 1: 883, 884, 885 Hip joints, 3: 514 Historical background, 1: 3-16 precious-metal campoundsp 3: 54-55 Zintl phases, 3: 114 Hitchiner in~estmentcasting processes, 3: 592-593 HoA1, structure type, 3: 88, 8Y Hodgkinsonite, 3: 9-10 structural formula for, 3: 9-10 Ho-La system, 1: 407 Homeotectics, 1: 405 Home page (on Internet), 3: 860, 876 Homer Mammalok@,2: 553 eneous disordering, 1: 781
985 Homogeneous flow, of metallic glasses, 3: 695-696 Homogeneous nucleation, 1: 747, 748; 3: 683 Homogeneous ordering, 1: 774 Homogeneous SRO, 1: 782 Homologous series close-packed structures, 1: 29 I, 293-294, 305,405, 416, 417 hetero~eneoustype, 1: 416, 417 homogeneous type, 1: 417 quasihomogene~ustype, 1: 416,417 H O M ~ L separation, ~ M ~ Zintl phases, 3: 119, 113 Homometric structures, 1: 289, 292-293, 302 Ho-Nd system, 1; 407 Honeycomb structures, 3: 633 Hongshiite [mineral name], 1: 628, 632 HoNi,B, structure type, 3: Y2 Hooke's law, 1: 151, 874; 3: 386, 393 Hot bulging, 3: 635, 636 Hot-corrosion resista~ce,1: 989, 990; 57,489 Hot-die [hammer] forging, 1: 653 Hot dip coatings, 2 518-519 Hot-dippi~gprocess, 1: 646 Hot-electron transistors, Hot extrusion of cast FeAl, 3: 655 of powder preforms, 3: 646 of TiAl alloys, 3: 627430, 636 Hot isosVatic pressnig (HIP), 1: 645, 912: 2 359; 3: 643, 645446 Al-Ti alloys, 3: 595, 619 beryllides, 3: 44 densification mechaixsm map, 3: 637 density--press~remap for AlTi, 1: 644 reactive HIP, 1: 646 Hot pack rolling, 1: 655 of TiAl alloys, 3: 632 Hot-spot initiation mechanrsin, in shockassisted reactions, 3: 742 Hot working and i n t ~ r n astresse, ~ 3: 628 and microstructure, 3: 624 and recrystallization, 3: 623-626 Huang scattering, 1: 562, 573 Hue, meaning of term, 3: 242 Hugoniot curves, of copper, 3: 738 Human perception of colar, 3: 232 Hurne-Rothery, William, 3: xliii Hume-Rothery criterion, 3: 812 Hume-Rothery mechanism, quasicrystals stabilized by, 1: 485 Hume-Rothery phases, I: 13 as catalysts, 2: 647 colored compounds, 3: 74 compared with Zintl phases, 3: 114 electrical resistivity, 1: 953 enthalpy of formation, 1: 100, 101, 102 correlat~onwith entropy of formation, 1: 106-107 forniation tendeiicy, 1: 103, 104, 703 heteropolar bonds, 1: 101-102 meaning of term, 3: xxxviii metallic bonds, 1: 100-101 stability, 1: 100-107 effect of atomic-radius differences, 102-104
986 effect of structure of higlier-valent component, 1: 104-105 valence-electron concentration values, 1: 100, 101, 105, 106, 319, 325, 704 Hui~e-Rotheryrules, 1: 13, 63, 100, 101, 102, 105, 244, 246, 568 Goodman’s extension, 1: 247, 249 Hybrid close-packed structures, 1: 414,417 Hybrid deposition (HD) techniques, 3: 671-672, 673 Hydrides, disordered, 3: 252-253 Hybridization antiferromagnetic exchange interaction arising from, 1: 212 in band structure, 1: 134, 138, 143, 213 Hybridized bonding orbitals, 1: 346; 2: 155 Hybrid magnets, 2: 371 Hydrides, 2: 475-487 amorphous hydrides, 1: 695 applications, 2: 484-486 crystallinity, 2: 48 crystal structures, 1: 297, 298, 299-302, 393; 2: 475, 478 decrepitation, 2: 478, 482 electronic properties, 2 479-48 1 in emb~ttlementof Al-Ti alloys, 1: 974 surface properties, 2: 48 1-483 thermodynamic properties, 2: 476-478 Hydrogen absorption behavior, 2: 476 impurities affecting, 2: 484, 485 diffusion of, 2: 483-484 storage as fuel, 2: 475484485,509-5 10 Hydrogenated rare-earth compounds, 3: 100 Hydrogenation~ecompositioii-desorption-recombination (HDDR) process, rare-earth metal magnets, 3: 98, 102 Hydrogen decrepitation (HD) technique, 3: 101-102 Hydrogen e~brittlenient,1: 927, 967, 970-974; 2: 28-29, 39, 46 compared with stress corrosion cracking, 1: 967 compounds susceptible, 1: 970 plasticity affected by, 1: 927-928 Hydrogen e i n b ~ t t l i neffect, ~ in fatigue crack growth, 3: 337, 339, 340, 347 Hydrogen evolution reaction, 2: 504-506 Hydrogen-generators, 2: 645-646 Hydrogen-induced a~orphization,1: 695, 734 Hydrogen oxidation reaction, 2: 504-506 Hydrogen-storage materials, 3: 48, 78, 253 Hydrostatic extrusion, 1: 655 Hydrostatic pressure, elastic constants affected by, 1: 890, 891 Hydroturbine parts, 2: 45-46 Hyper~ne-enhancednuclear magnetic cooling, 2: 654, 655 Hysteresis loops for ferroinagnetic materials, 2: 304 magnetization, 1: 939-940 rectangular hysteresis loop, 2: 304, 307 for s~ape-memoryalloys, 2 530 square hystercsis loops, 2: 304, 448 1 framework, 1: 317-325
Subject Index see also ..cP,(12i) 1: 335, 336; D(60) 1: 325, 327, 328, 331, 332; ..dI,(60) 1: 328, 332, 333; ..nI(12i) 1: 329, 332. 334, 336 I(4t) framework, 1: 325. 328, 330 I(12i) framework, 1: 329, 332, 334 I(60) framework, 1: 325, 328, 330 I lattice complex, 1: 317 8th order, 1: 319, 320 64th order, 1: 320, 322 splittings, 1: 319-325 27th order, 1: 320, 322, 327 216th order, 1: 320, 325 i phase, 1: 453, 454-465 IC6, casting of, 3: 608-609, 611 IC-50, 3: 61 1 IC-218 composition, 3: 329 fatigue properties, 3: 329, 330 superplastically formed, 3: 505-506 IC-22 1/22 1M casting of, 3: 609, 610, 611 composition, 3: 329, 492, 515 fatigue behavior, 3: 329, 341, 493 grain size, 3: 609 properties, 3: 493, 515 IC-396LZr, 3: 611 IC-438, 3: 611 lcosahedral alloys, searching on Internet, 3: 859-860 Icosahedral compounds, 1: 161-162, 419 quantum structure diagram, 1: 267 Icosahedral glass model, 1: 463, 482 calcuiated patteriis, 1: 464 Icosahedral-pliase quasicrystals, 1: 453, 454-465 and amorphous phase, 1: 740 atomic structures, 1: 483 rational crystal a p p r o x i ~ a ~ t1: s, 476-477 Icosahedral quasicrystals (IQCs), 3: xxxviii, 379 dislocations in, 3: 379, 380, 382-383 elastic constants, 3: 390 elastic equations, 3: 388-389 mechanical properties, 3: 397-398 nanostructures, 3: 755 spectroscopy, 3: 147 Icosahedron, 1: 238, 239, 404, 409, 410, 453 frequency plot, 1: 240 Ideal atomic-radii ratio, Laves phases, 1: 107 Ideal axial ratio Hume-Rothery phases, 1: 106 Laves phases, 1: 109 Ideal glass temperature, 1: 736 Image sensors, 2: 232-233 Images, on Internet, 3: 874 IMI 834 superalloy, temperature dependence of yield stress, 3: 630 Immiscibility gap, 1: 724 Impact collision ion scattering spectroscopy with neutral particle detection (NICISS), 3: 214, 218 Impact resistance aluminides, 2: 66 see also Toughness Impurities diffusion of, 1: 767-768 ductility affected by, 2: 59, 82
electrical conductivity affected by, 1: 954 eiyects on partial size and shape stability, 1: 864, 868 from foundry practice, 3 607 removal by temperature gradient, 1: 189, 191 strength affected by, 2: 82 IN-100 [superalloy], 2: 6, 7, 8, 493 creep behavior, 3: 298, 300 IN-713C [superalloy], 2: 41 fatigue properties, 3: 329 temperature dependence of yield stress, 3: 630 IN7 18 superalloy conipared with TiAl alloys, 3: 481, 482 temperature dependence of yield stress, 3: 6-70 Incoherent scattering, 3: 248, 251, 256 Incoherent twins, 3: 410 Incommensurate crystals, 1: 453 lnconel superal~oys,2: 7, 42, 282; 3: 42 Incubation period, in creep, 3: 299, 302303 Independent-pa~t~cle approximation, 1: 137 Indides data sources, 3: 808 prediction of formation, 3: 829-830, 831 Indium antimonide semiconductor, 3: 795-796 Indiuiii-containing fulleranes, 3: 8,9, 126, 127 1ndium”contain~ngZintl phases, 3: 117119, 118, 123-125 Induction-skull melting (ISM), 3: 592, 619 Inelastic neutron scattering, 3: 247, 248249, 254 Inert gas condensation, synthesis by, 3: 751, 752-753 Information, meaning of term, 3 836 Info~ation-predictionsystems, 1: 270-272; 3: 819-821, 836 Kiselyova’s system, 1: 270, 271 Savitskii’s system, 1: 270, 271 Zhou’s system, 1: 270, 272 Infrared detectors, 2: 232-233, 326, 419, 460, 654 Infrared detector applications, 3: 31, 672, 794,795-796 Ingot break-down, 3: 626-430 Ingot melt techno~ogy,3: 592, 619, 636, 643 Ingot production, 1: 639-640; 2 161, 175, 176 Inhomogeneities, after extrusion, 3 629 Inhomogeneous flow, of metallic glasses, 3: 694-695 Initial flow stress, effect of order, 1: 913 InNi, structure type, 1: 364, 373, 374, 375 separation in structure maps, 1: 389, 429 therniodynamic properties of compounds, 1: 112 Inoculants, 2: 523, 524 In-Pt n a t ~ r a ~occurring ~y compounds, 1: 632 111-Sb system, 3: 158, 161 Insizwaite [mineral name], 1: 628, 632
Subject Index Inspection limit (for flaw detection), 3: 484 Insulating-grouiid-states systems, heavyfermion compounds, 1 : 2 17-2 18 Integrated-circuit applications, 230-232, 332, 333, 346,40 653-654 Integrated-circuit substrates, 3: 663, 676, 783, 796 Integrated density of states (IDOS), 1: 26 Integrated optical devices, 2: 327, 343, 347,432, 438 Integrated photoemission, 3: 137 Intensity modulators, 2: 416 Interaction of intermetallic compounds, factors determining, 1: 712-713 Interatomic distances, and atomic radii, 1: 253, 378-381 Int~ratomicinteractions Laves phases, 1: 109 and phonon dispersion curves, 1: 151-1 56 Interatomic potentials, 1: 77; 3: 769 deter~nation,1: 81, 154 Intercalation compounds, in graphite, 3: 257 Iiiter~hangeshuffling, 3: 417, 424 Interdiffusion, 1: 861-862 amorphous phases formed by, 1: 696-699; 2: 619420 synthesis of intemetallics by, 1: 645-646, 757-760 Interfaces and twinning dislocations, 3: 407-410 and twin nucleation, 3: 4 19-421 Interfacial defects, Pond’s topological theory, 3: 410 Interfacial energy crystal-crystal, 1: 40 crystal-liquid, 1: 480 i up er alloy precipitate coarsening affected by, 2 10 Interfacial-reaction- induce^ amorphization, 1: 695-699, 733-734 Interfacial reactions, kinetics, 3: 674 Interference filters, 2: 410, 41 I as optical switches, 2: 412 lnt~rgranularattack, and pest degradation, 1: 1005-1006 Intergranular corrosion, 1: 966 Intergranular fracture, 1: 895-909 aluminicles, 2: 27-30, 64, 207 examples, 1: 895 L1, compounds, 2: 27-30 meaning of term, 1: 895 trialuminides, 2 166 Intergrowth concept, 1: 374 Inte~ediate-temperatureaccelerated oxidation, 1: 1000-1006 see afso Pest degradation Interniediate-valence cornpounds, 2: 213-214 phanon dispersion effects, 1: 154, 159 Intermetallic compounds definitions, 1: 272-273; 3: 231,246, 663, 876 manufac~ure by casting (q.v.) by powder metallmgy (q.v.1 by wrought processing, 3: number of Internet hits for,
I n t e ~ e t a l l i cfilters, 3: 272 Internal combustion engine applications, 2: 46, 596-597 Internal deformation, 1: 405 Internal surfaces, definition, 3: 437 Internal-tin processes [for fabrication of superconductors], 2: 361-363, 364, 365 International Commission on Illummation, color scheme, 3: 232 International Thermonuclear Experimental Reactor (ITER) program, 2: 361, 374, 375-376 I n t e r n a t i o ~ Union ~l of Crystallography , on nomenclature, 3: 8, 11 In ternet address (URL), 3: 857, 877 directories, 3: 864-865 fee-charging services, 3: 868 glossaries, 3: 876-877 guides and tutorials, 3: 867 meaning of term, 3: 876-877 meta-sites, 3: 865 resources, 3: 857-877 sites with significant IMC coverage, 3: 874-875 useful sites, 3: 871-875 Internet search process, 3: 857-864 hyperlinks, 3: 860 and keywords, 3: 860 and links, 3: 859 meta-search engines for, 3: 857-858 search engines for, 3 855-860 Internet service provider (ISP), 3: 857,877 Interparticle spacing, and creep, 3: 315 Intersection group, 1: 857 Interstitial atoms, in close-packed alloys, 1: 296-299 Interstitial clusters, 1: 573 Interstitial compounds complex close-packed alloys, 1: 303-304 cubic close-packed alloys, 1: 292, 299-302 as diffusion barriers, 2: 623, 624 hexagonal close-packed alloys, 1: 302-303 Interstitialcy mechanism, 1: 760 Interstitial defects, 1: 352, 559 migration properties, 1: 560, 577-578 production of, 1: 560, 566-567 structure, 1: 560, 567-568, 578 Iiiterstitial~electronmodel (IEM), 1: 155-1 56 phonon dispersion curves calculated using, 1: 156, 164 Interstitial energies, 1: 799 Interstitial hydrogen in disordered hydrides, 3: 253 in rare-earth compounds, 3: 98 Interstitial impurities, effects on IMC synthesis, 1: 638 Interstitial migration, 1: 760 Interstitial modification, magnetic properties o f rare-earth compounds affected by, 3: 98, 178-180 Interstitial-vacancy recombination, 2: 134 In-Ti system, 1: 834; 2: 538; 3: 258 Intrinsic coercivity, 2: 304 Intrinsic resistivity, 1: 944 Intrinsic stacking faults, 2: 19-20
987 Invariant lattice complexes, 1: 313, 315 symbols used, 1: 313, 314 Invariant line strain, 1: 829 Invariant plane strain, 1: 828 Inverse creep, 1: 917 Inverse melting transition, 1: 751 Inverse Monte Carlo (IMC) method, 1: 40 Inverse-photoemission spectroscopy (IPES), 1: 135, 136; 3: 149 see aZso Anglo-resolved inversephotoeinission . . . Investment casting processes, 3: 592-593, 608409 Iodin~mercuryphase diagram, 3: 25 Ion beam deposition techniques, 3: 665 Ion-beam mixing, 1: 650, 7005, 733, 792; 2: 614, 625 Ion-beam spectroscopic techniques, 3: 214 Ion explosion spike mechanism, 3: 264265, 266, 270-271 Ionic bonding close-packed alloys, 1: 296 and diffusion, 1: 764 valence-electron transfer in, 1: 346 Ioiiic compounds, 3: xxxviii interstitial atoms in, 1: 301 lattice dynamics, 1: 151-1 52 Ionicity, crystal habit affected by, 1: 176-179, 183, 184 Ion implaiitation, 1: 649-650, 733, 792 Ion irradiation effects on amorphi~ation,1: 809-8 12 compared with electron-irradiation effects, 1: 813 on disordering, 1: 806-807 Ion plating, 2 493 Ion scattering spectroscopy (ISS), 3: 214 Ion tracks, 1: 693; 3: 263-272 conditions for production, 3: 264, 266, 269 energy-loss thresholds, 3: 266 primary ionization energy, 3: 264, 26.5, 266, 269, 270 formation mechanisms, 3: 264-265, 269-27 1 ion explosion spike mechanism, 3: 264265, 266, 270-271 thermal spike mechanism, 3: 269 in IMCs, 3: 265-267 in metals, 3: 267-268 in oxide conductors, 3: 268-269 particles that produce, 3: 263-264 and regis~ration/non-registration,3: 266-267 in superconductors, 3: 269 Ion wakes, 3: 270-271 Iridium demand and price flL~ctuations,3: 79 limitations, 3: 69 melting temperature, 3: 67 world depositslreserves, 3: 80 Iridium-aluminide coatings, 3: 568 Iridium aluminides, 3 58-59 effect of boron, 3: 59 ternary alloys, 3: 59-60 Iridium-base superalloys, 3: 67-69, 71 compositianal substitution, 3: 69 Ir-Nb alloys, 3: 67-68, 70, 71 Iron ion tracks in, 3: 267, 268
988 single crystal, strain hardening in, 3: 362 Iron alloys, magnetic structures, 1: 444 Iron aluminides, 2: 199-208; 3: 496 casting of, 3: 611-614 and barriers to commercial development, 3: 611 COatlngs Of, 3: 571-574 composites, 2: 208 corrosion resistance, 2 207-208 effects of boron, 2: 204, 206 enthalpy of formation, 3: 605 extruded, 3 505 fatigue crack growth in, 3: 338-341 environmental effects, 3: 347 fatigue properties, 3: 331-332 forming of, 3: 639 hi~h-temperaturemechanical properties, 3: 291 hydrogen-embrittlement susceptibility, 1: 971-973 mechanical properties, 2 202-207 melting of, 3: 612 oxidation resistance, 2: 207; 3: 572, 573 polycrystalline behavior, 2: 202-207 elevated-temperature properties, 2: 204-207 room-temperature properties, 2: 202-204 powder metallurgy applications, 3: 654-655 processing of, 2: 201-202 single-crystal, 2 202 strain hardening in, 3: 368, 369, 370, 372, 373 sulfidation of, 3: 712-716 synthesis by mechanlcal alloying, 3 654, 731, 758, 759 thermal vacancy formation in, 3: 282, 289-290, 291 twinning in, 13: 416-417 two-phase, 2: 206 vacancy migration in, 3: 282 see also AlFe alloys; AlFe, alloys; Al-Fe system Iron amalgam, 3: 29 Iron-based binary compounds, magnetooptical applications, 2: 446-447 Iron-~anthanide[ m a ~ e t i ccompounds, ] 2: 314 Iron meteorites, 3: 271-272 Iron-rare-earth compounds crystal structure of Laves phase, 2: 395 magnetic materials, 2: 316-3 17 magneto-optical measurements, 2: 446, 447 m$tallurgy, 2: 395 Iron silicides corrosion studies, 1: 969-970 see also Fe-& system Iron-terbium compounds, niagnetostriction in, 2: 392, 393 Iron tracer diffusivities, 3: 290 Ir-Pt-Ti system, 1: 724 Irradiation alloy stability during, 1: 821 amorphization induced by, 1: 692-695, 733, 791, 808-822 dose required, 1: 811, 812, 815
Subject Index experimental observations, 1: 808-8 17 mechanisms, 1: 8 17-822 temperature effects, 1: 8 13-8 14, 815 atomic-mobility enhanced by, 1: 785 by electrons, 1: 694 by fission fragments, 1: 694 collision sequence resulting, 1: 579, 692 disordering induced by, 1: 692495, 733, 791, 803-808 effects of, 1: 649, 791-822 theoretical modelling, 1: 797-799 order changes induced by, 1: 784-786 ordering enhanced by, 1: 799-803 synthesis using, 1: 649-650, 705-706 see also Electron . . .; Neutron irradiation Irregular atomic e n v i r o n ~ e n(IAE), ~ 1: 367 Ir-Rh-Ti system, 1: 724 Ir-Ta alloys, 3: 70, 71 Ir-Ti alloys, 3: 70, 71 Ir-V alloys. 3: 70, 71 Ir-Zr alloys, 3: 68-69, 70, 71 Ir-Zr system, 1: 740 Ising energy, 3: 199 Ising model, 1: 24,294-296, 500, 504, 507, 850; 3: 192 lsochronal annealing, 3: 283 Isochronal annealing treatments, 1: 78 1, 793 Isocon~~urational structures, 3: 11 Isodesmic compounds, 1: 113 Isoelectrical conductivity plot, 1: 265 Isoelectronic close-packed structures, 1: 417 Isoferroplatinum [mineral name], 1: 627, 628, 631, 632 Isolation procedures, 1: 5 Isomertieite [mineral name], 1: 628, 629, 632 Isomorphism, 1: 415 Isoperm [niagiietic material], 2: 306, 307 Xsopiestic technique, vapor pressure determined using, 1: 94 Isopointal structures, 3 11 Isostructural close-packed strttctures, 1: 417 Isothermal annealing, 1: 781 Isothermal compressibility, 1: 1017, 1022 Isothermal die forging, 3: 635, 636 Isothernial elastic constants, 1: 873 compared with adiabatic elastic constants, 1: 876 Isothermal forging, 1: 653; 3: 626-627 Isothermally forged and annealed (IF-A) Al-Ti alloys, toughness, 3: 426, 427 Isothermal rolling, 3: 632-632 Isothermal time-di~erentiallengthchange measurements, 3 28285 Isotope separation, amalgam used, 3: 30 Isotropic moduli, 1: 202 accuracy, 1: 208 calculated values, 1: 206 see ulso Bulk modulus; Poisson’s ratio; Shear modulus . . .; Young’s modulus Isotypic structures, 3: 11 Itinerant ferromajpetisni, 1: 135, 141, 1020 Itinerant magne~ism,1: 132
in heavy~ferinioncompounds, 1: 215 3: 8 17 I V T A N T H E ~software, ~~ Jagodzinsk~notation, 1: 279 Jagodzins~~-Wycko~ notation, I t 345, 358-3 59 ZnS polytypes, 1: 345 Jarmoljuk-Kr~pyakevich phenomenon, 1: 41 1 Jensen notation, 1: 427, 432, 433 Jewelry applications gold alloys, 3 75-77 platinum alloys, 3: 75, 76 usage of gold, 3: 77 Joining of FeAl and Fe,Al alloys, 3: 510 mechanical, 3: 5 10-5 11 of Ni,Al alloys, 3: 509-510 of Ni,Si alloys, 3: 510 of TiAl alloys, 3: 510, 640 Joining techniques, 1: 655-656; 2: 44-45, 128, 521-523 Jones vector f o ~ a l i s m 2: , 436 Josephmite [mineral name], 1: 633 Jo~e~hson-junct~on technology, 2: 383 Journals on Internet, 3: 868-870 K [kinked] crystal faces, 1: 170 representation in stereographic projection, 1: 169 K5 alloys, 3 481-482, 481, 482 K-cell, 3: 667 K-effect, 3: 307 K-phases, 3 xxxviii Kappa phases, 3 xxxviii Karatage [gold jewelry alloys], 2: 559 Kasper, John S., 3: xliii see" ulso Frank-Kasper phases Kasper polyhedra, 1: 473, 474, 479 Kear-Wilsdorf (KW) configuration, 1: 512, 529, 543-544 role in deformation, 1: 544 Kear-Wilsdorf locking, 3: 73 Kear-Wilsdorf (KW) ~ocks/mechanism, 1: 52$, 529,544,914; 2 13, 14,24,25, 137 Kear-Wilsdorf mechanism, 3: 372 Kerr effects, 2: 435, 437 Kerr rotation, 2: 442, 444, 445, 647, 449 see also Polar Kerr rotation Kesterite [mineral name], 1: 355, 359 Khantha-Cserti-Vitek (KCV) dislocation”1ocking model, 1: 547 Khatyrkite [mineral name], 1: 627, 628, 632,633 Kikuchi bands, 1: 470 Kikuchi-Barker coefficients, 3: 195 Kikuchi’s cluster variation method, 1: 30, 610, 775 Kimberlite [mineral name], 1: 628 Kinchin-Pease function, 1: 999 Kinetic processes and phase transformat~ons,1: 755-868 amorphization, 1: 812 bainite growth, 1: 839-840 Grystallization, 1: 747 decomp~sitionpathways, 1: 853-857 of dental amatgam dissolution, 2: 580-581 domain growth, 1: 865
Subject Index fuel-cell reactions, 2: 508-509 long-range-order changes, 1: 775-779 inarteiisitic transformations, 1: 831 ordering/disordering, 1: 948 oxide-layer growth, 1: 992, 994, 997 pest degradation, 1: 1004 see also Growth kinetics Kinetics, data sources, 3: 805 Kirkendall porosity, 2 356, 357, 572 Kirkendall voids, 3: 580, 581 Kiselyova’s inforination-prediction system, 1: 270, 271 Kitagawa diagram, 3: 483, 484 Knee joints, 3: 514 Knight shift, liquid alloys, 1: 671, 682 Knowledge, meaning of term, 3: 818, 836 Knowledge base, 3: 818, 819, 836 Knudsen cells, 3: 667, 781 Knudsen effusion method, 1: 93, 681 Kohn anomalies, 1: 157 Kohn-Sham equations, 1: 129, 131 Kohii-Sham potentials, 1: 197 Kolymite [mineral name], 1: 629, 632, 633; 3: 21n Komura, Yukitomo, 3: xliii Koniura phases, 3: xxxix Kondo, J., 3: xliv Kondo impurities, 1: 212 Kondo insulators, 1: 217-218 Kondo lattices, 1: 212-213; 3: xxxix frustrated model, 3: 181 Kondo resonance, 1: 218; 3: 46 Kondo temperature, 1: 212 Korringa-Kohn-Rostoker coherentpotential approximation (KKRCPA), 1: 25 applications, 1: 41, 42, 44, 46, 47 K-Pb system, 1: 662 Kripyakevich’s classification (of structure types), 1: 403, 405, 409 Krivoglaz-Clap~Mossapproximation, 1: 27-28,40 Kroll-Betterton process, 2 5 17 ‘Kryptonite’, 1: 633 K-Sb system, 3: 158 Kumakov compounds, 1: 712, 714, 720, 725, 727, 924 Kurnakov, Nikokdi Semyonovich, 3: xliv Kurnakov compounds, 3: xxxix L1, type compounds, 3: 275 ab znitio calculations, 1: 64 APB energies, 1: 506 crack-tip twinning in, 3: 419 crystallographlc elements and parameters of twin modes, 3: 408 crystal structure, 1: 61, 520, 532; 2: 74 dislocations, 1: 532-534, 545 debris, 1: 534, 551-552 displacement vectors of stacking-faultlike defects, 3: 463 elastic constants, 1: 201, 205 flow-stress anomaly, 1: 532-533 fracture of, 1: 907 ion tracks in, 3: 272 Pearson symbol for, 3 xxxii, xxxiv, 15, 439 quasibinarylquasiterna~systems, 1: 720-724 space group for, 3: xxxii, xxxiv, 15, 439 stacking faults in, 3: 442, 444446
strain hardening in, 3: 365 surface-induced ordering, 1: 614-61 5 surface structure, 3: 217 twinning in, 3: 41 1-414 vacancy migration, 1: 575 see also AuCu structure type; Titanium aluminide L1, structure. 1: 61 L1, type compounds, 3: 275 ab initio calculations, 1: 64 A N , and alloys, 2: 1 7 4 7 A1,Ti-based compounds, 2: 155-1 67, 168-170 antiphase boundaries in, 1: 497, 498, 499; 2 19, 20 APB-coupled dipoles, 1: 553, 552 APB energies, 1: 496, 505 variation with temperature, 1: 503 crystallographic elements and parameters of twin modes, 3: 406, 407 crystal structure, 1: 61, 65, 497, 520, 896; 2 17, 18, 148, 156, 212 relation to DO,, and DO,, structures, 1: 499, 520, 530; 2: 148, 156 diffusion in, 1: 766 dislocation debris, 1: 551 dislocations in, 1: 527-529, 542, 545 disordering by milling, 3: 759 displacement vectors of stacking-faultlike defects, 3: 463 elastic properties, 1: 879, 880 energy differences vs DO,, structure, 1: 65, 66 fault types, 2: 21 flow Of, 2 23-27 anomalous behavior, 1: 527, 546-549, 914-915; 2: 24-25 normal behavior, 2: 23-24 flow stress anomaly, 3: 372 fracture of, 1: 896-904, 907; 2: 27-30 fracture modes, 1: 897 grain boundaries, 1: 586 bonding at, 1: 603-604 grain-boundary structure models, 1: 598-603 hydrogen-embrittlement susceptibility, 1: 970-971 ion tracks in, 3: 272 metastable phases, 1: 706, 707 Pearson symbol for, 3: xxxi, xxxiv, 15, 439 quasibinary/quasiternary systems, 1: 72.4-727 solid solution hardening of, 3 353-354 space group for, 3: xxxi, XXXIV, 15, 439 stacking faults in, 3: 440-442, 443-444 strain hardening in, 3: 365 surface reconstruction, 1: 619-620 surface structure, 3: 217 surface terminations, 1: 616 trialuminides, 1: 924 twinning in, 3: 404, 411 vacancy migration, 1: 574-575 see also AlNi, alloys; AuCu, structure type L1,/DO2,/DO,, competition, 1: 65-66, 706, 707 L2, type compounds APB energies, 1: 507 colored compounds, 3: 238
989 crystallogra~hicelements and parameters of twin modes, 3: 407 crystal structure, 1: 62, 520; 2 62 dislocations in, 1: 536-537 displacement vectors of stacking-faultlike defects, 3: 463 Pearson symbol for, 3: xxxi, xxxiv, I S , 439 quasibinary systems, 1: 718 space group for. 3: xxxi, xxxiv, 15, 439 stacking faults in, 3: 453-454 see also AlNi,Ti structure type; Heusler alloys LaFe,Sb,, structure type, 3: 106 LaIrSi structure type, 3: 91, 91, 93 Lamellar eutectics, 3: 6 1, 67 Lamellar microstructure, 1: 864, 926; 2: 75-76, 398 anisotropic deformation properties, 79, 80 Lamellar microstructure of TiAl, 3: 328, 478,619,626, 628 anisotropy, 3: 621, 625 and powder metallurgy, 3: 652 Lamellar strengthening, creep behavior Lamp filaments, 3: 31 La,Ni,B,N, structure type, 3: 104, Z0.5 Landau theory, 1: 110 Lanthanide compounds amalgams, 3: 23 beryllides, 3: 46 magnetocaloric properties, 3: 529-535 see also Rare-earth compounds Lanthanides, 3 85 data sources, 3 807 see a1,s.o Rare-earth elements Lap, structure type, 3: 89, 89 LaPt,Ge, structure type, 3: 92 LaPtSi structure type, 3: 91, 93 LaRe,Si, structure twe. 3: 92 Large- Coil Task (LCT) program, 2 361. 372-373 Large Hadron Collider (LHC), 2: 377 Larson-Miller parameter, 3 300 Larson-Miller plots, 2 62, 244, 293; 3: 480, 553 Laser ablation, 3: 666467, 673 Laser diodes (LDs), 2: 325, 326, 328, 336, 338-339, 346, 347, 418,427-431; 3: 796 Laser-heated surface-film technique, 1: 640-64 1 Laser-induced etching, 2: 413-414 Laser welding, 1: 655 Lattice c/a ratio, 1: 250, 251, 281 A1,Cu-type compo~nds,1: 392, 393 atomic environment affected by, 1: 370 MoSi,-type compounds, 1: 386, 387 see also Axial ratios Lattice complexes examples, 1: 324 meaning of term, 1: 310 nomenckdture, 1: 309 symbols used, 1: 313, 314 types, 3 : 313 see also Invariant lattice complexes Lattice constant, 3: 768 predicted, 3: 775
Subject Index
990 Lattice expansion, by hydrogen, 2: 478 Lattice melting anisotropy, 3: 245 Lattice mismatch creep resistance affected by, 3: 313, 314 in ~I3E-growncompounds, 3: 783, 784 in superalloys, 3: 842 Lattice parameters, 1: 62 ab irzitio calculations for, 1: 62, 68, 69, 70, 73 Al-Fe/Al-Fe-Si systems, 2: 177, I84 Al,Ti, 2: 148 AIZr,, 2 134 Ni and A N , solutions, 2: 10 Se~icond~Ictors, 2: 328, 333 transition-metal a~uminides,1: 138 Lattice parameters, of b.c,c. metals, 3: 843 Lattice resistivity, 1: 944 Lattice softening, 1: 890 Lattice structures, at high pressure, 3: 161 Lattice vibrations, 1: 149 Laue classes, quasicrystals, elastic constants, 3: 387-388, 38SI Laves, Fritz Henning, 3: xliv Laves phases, 1: 14, 107-109, 338, 409411, 417; 3 xxxix in AlNi-based alloys, 3: 316 atomic-size ratio, 1: 107, 409, 763 beryllides, 3: 40, 45 diffusion in, 1: 763 dislocations in, 1: 537 effect of irradiation, 1: 706 as magnetic materials, 2: 310, 317 magneto-optical effects, 2: 446,447 niultilayer Laves phases, 1: 410, 413 phonon dispersion curves, 1: 154 solid solution hardening of, 3: 356 in structure maps, 1: 241, 428 as superconductors, 1: 16 temperature dependence of elastic properties, 1: 889 ternary Laves phases, 1: 107, 108, 413 twinning in, 3: 415, 430 velocity-stress relationships, 2: 238 in wear-resistant alloys, 2: 596, 597 Laves rule, 3: 812 Layer descriptions, 3: 13 Layers, heterogeneous and homogeneous,
3:5 La-Zr system, 1: 740 Leaching, 1: 970 Lead, 9-atom cluster, in Zintl phases, 3: 116 Lead amalgams, 3: 21n Lead-based bearing alloys, 2: 592, 593, 594, 596 Lead refining, 2: 516 Learning set, 3: 836 Length changes, defect-induced, 3: 277278 Levi-Cast process, 3: 593 components produced by, 3 599 Lewis acrd-base interactions, bonding in transition-metal aluminides explained by, 1: 116 Lexan ~olycarbonate),ion tracks in, 3: 264 Liberite [mineral name], 1: 359 Libraries, on Internet, 3: 867-868
Life-cycle costs, 3: 472, 482 Lifetime predictions, 1: 1008-1010 metrics used, 1: 1009 Lift-and-project technique, 1: 483 LiGaGe structure type, 3: 91, 93, 93 Light-emitting diodes (LEDs), 2: 323, 325,326,327,328,330,336,337-338, 346, 347, 418, 426-427; 3: 783, 796 Lighter flints, 2: 647 Lightning protection device, 2 545-546, 548 Li-Mg-Zii system, niultilayer Laves phases, 1: 410, 413 Lindemann [melting] criterion, 1: 8 19, 1023 Linear augmented plane-wave (LAPW) method, 1: 60 applications, 1: 41, 64, 67, 68, 69, 133, 138, 198-199, 874, 876 see also Full-potential linear augmented plaiie-wave (FLAPW) method Linear a ~ g ~ Slater-type ~ ~ ~ orbital e d (LASTO) method, 1: 60 applications, 1: 68, 133, 908 see aLw Full-potential linear augmented Slater-type orbital (FLASTO) method Linear-combin~tion-o f-atomic-orbitals (LCAO) methods, 1: 60-61 applications, 1: 68, 70, 133, 955 Linear elasticity theory, for quasic~ysta~s, 3: 385-389 Linearized muffin-tin orbital (LMTO) method, 1: 31, 60 applications, 1: 41, 65, 66, 69, 133, 505, 506, 507 Al-B-Ni system, 1: 83, 84, 85 Al-Ni compounds, 1: 69,70,114,115 see also Full-potential linear muffin-tin orbital (FLMTO) method Linearized muffin tin orbital-atomic sphere approximation (LMTOASA) method, 3: 202 Line compounds, 1: 350, 355, 356, 378, 897; 2: 134, 170,224,523; 3: xxxix, 23 Linguistic processor, 3: 838 Li-Pb system, 1: 662, 664, 666, 670, 671, 673, 674 Liquid alkali metals, as solvents, 3 34 Liquid lithium, 3: 34 Liquid magnets, 3: 32 Liquid metal cooling (LMC) technique, 3: 543 Liquid metals, surface structure, 3 216 Liquid-phase epitaxy (LPE), 2: 326, 328 Liquid quenching, 1: 689-692, 733, 749, 772 Liquids, 1: 661-686 association in, 1: 675-676 chemical complexes, 1: 672-675 complex formation in, 1: 671-675, 750 energy of formation, 1: 672-4375 quasilattice model, 1: 673-675 strong interaction approximation, 1: 672 demixing of, 1: 677-678 direct observations, 1: 662-4571 concentration ~uctuations,1: 665-668 diffraction experiments, 1: 662-663 diffusion measurements, 1: 668-669
electncal properties, 1: 669-671 phase diagrams, 1: 663, 664, 666 thermodynamic properties, 1: 663, 665, 666 viscosity measur~ments,1: 668 models complex-formation, 1: 671-675, 750 quasilattice model, 1: 673-675 regular-associate~-solution model, 1: 675-676, 750 ordering in, 1: 479, 661, 750 Liquid-solid reactions, beryllides prepared by, 3: 47 Li-Sn system, 1: 95, 97, 662, 667, 668, 669, 673,674 Lithmted carbons, 3: 250, 258 Lithium alwninide, Fdst-ion behavior, 3: 25 1 Lithium [rechargeable] batteries, 2: 510-51 1 Liu-Liii-Chen (LLC) interatomic potential, 3 770, 771 Local-density approximation (LDA), 1: 24, 26, 59, 129-130, 197, 631; 3 196 equation-of-state parameters calculated inte~etalliccompounds, 1: 204 monatomic metals, 1: 203 errors in calculations, 1: 207 Localized point. obstacles, strengthening by, 2: 258-260 Lo~a~"spin-density approxi~ation (LSDA), 1: 130 calcu1atioiis for transition-metal alumiiiides, 1:, 138 Lock-CrispWest (LCW) folding, 1: 137 Loellingites, 3: xxxix Lomer-Cottrell dislocation locks, 1: 540, 545 Lone-electron-pair concept, 1: 181 [footnote] ~ o n ~ i t u d i nacoustic al (LA) mode [of phonon-~ibration~, 1: 159, 1018 Longitudinal-optical-transverse-optical (LO-TO) split tin^, 1: 152 Long-period superlattices, 3: xxxix Long-period superstructures, 1: 46, 49; 2: 148 Long-range degree, for various irradiation temp~ratures,1: 800, 801 Long-range interactions, 3: 201 Long-range ordering, 1: 771, 773-775 effect on electrical resistivity, 1: 945 Long-range order para~eter,1: 772, 851; 3 191 I32 alloys, 3: 849 determination of, 3: 212 and milling, 3: 731, 760 at surface, 3: 222 temperature dependence, 3: 222 Lonsdaleite [mineral name], 1: 343 Loop shearing, magnetic materials, 2: 305 Lorentz forces, 2: 318, 439 Lorentz ratio, 1: 1026 Low-contact-resistance electrodes, 2 653 Low-energy electron diffraction (LEED), 1: 61 I, 616,618,619; 3: 212,218,219, 785 Low-energy ion scattering spectroscopy, 1: 612, 618 Low-energy ion scattering (LEIS) technique, 3: 214, 218, 219
Subject Index Low-melting alloys, as amalgain analogs, 3: 33-34 Low-inelting-point metals, amalgams, 3: 23 Low-pressure plasma spraying, 1: 642-643 Low-temperature t hermoelectncs, 2: 462-464 Lubrication, eflect of tribaloys, 2: 596-597, 601 Liiders bands, 2: 137 Ludwig-Gumbsch (LGF) interatoniic potential, 3: 769, 769, 770 LuFe,B, structure type, 3: 104, 105 Luminance, 3: 232 LuMnGe, structure type, 3: 92 LuNiBC structure type, 3 103, 104, 105 LuNi,B,C structure type, 3: 103, 104, 105 LuNiGe structure type, 3: 91, 93, 93 LuNiSn, structure type, 3: 92 Luster, meaning of term, 3: 242 Lydd~ne-Sachs-Teller relation, 1: 152 M-phases, 3: xxxix MA6000 [superalloy], 2: 240 MacDonald je1ly~ro~l (MJR) process [for superconductors), 2 363, 364, 365 Machatschki symbols, 3: 4 Machinable permanent magnets, 2: 317 Mackay icosahedra, 1: 471,472,473,478, 48 5 Mackay-type clusters, 3: 398 Macrostructures, tltanium aluminides, 2 84 Macrotwins, 3: 411, 431 Madeliiiig factors, interstitial compounds, 1: 302 Magic strain barrier, 1: 207 Magnesides, data sources, 3: 808 Magnesiiim-based high-temperature alloys, 2: 650651 Magnesium, single crystal, shear-stress shear-strain curve, 3: 362 Magnet charactenstics, 2: 304-306 Magnetically anisotropic materials, 3: 99, 100 Magnetically levitated transport, 379-380, 381 Magnetic annealing, 2: 306. 400-401 Magnetic applications, 2: 217, 303-319 Magnetic cooling, 3: 519-524 Magnetic disorder, as resistivity component, 1: 944 Magnetic entropy change, 3: 524, 525 Magnetic exchange coupling, 3: 170 Magnetic exchange interactions, 1: 439-441, 936-937 direct exchange, 1: 440 double exchange, 1: 441 indirect exchange, 1: 936 superexchange, 1: 441 Magnetic force inicroscopy (MFM), 3: 215 Magnetic form factor, 1: 441 Magnetic fusion research, 2: 366, 371-376 Magnetic interactions, 3: 170-173 crystal field interaction, 3: 170-173 exchange interactions, 3 170
Magnetic-martensitic transformation, 3: 533, 534 Magnetic inateriats, 1: 748, 1029; 3: 78 applications, 2 317-3 19 composition notation, 2: 305-306 prediction of formation, 3: 829, 830 rare-earth compounds, 3: 97-101 BFe,,R, compounds, 3: 97-98, 165 prediction of formation, 3: 829, 830 R(Fe,M),, phases, 3: 99 R,(Fe,M),, phases, 3: 98 R,(Fe,M),, phases, 3: 99 R,Fe,,,M, phases, 3: 100-101 RT,,,M, phases, 3: 100 see also Hard . ;Permanent . ; Soft magnets Magnetic ordering elastic moduli affected by, 1: 889 in heavy-fermion compounds, 1: 21 1, 213, 215-216,219 Magnetic ordering temperature, listed for various materials, 3: 528, 529 Magnetic oxides, 2: 317 Magnetic phase diagrams, 1: 447; 3: 165184 amorphous Fe-Zr system, 3: 181-183 metallic glasses, 3: 693 rare-earth compounds, 3: 173-181 CeSb, 3: 180-181 interstitially modified compounds, 3: 178-1 80 La(T,M),, compounds, 3: 174 R,Fe,,B compounds, 3: 173-174, 17.5 R(Fe,T),, compounds, 3: 174, 176, I77 R,(Fe,T),, compounds, 3: 174 Magnetic properties, 1: 15, 935-940 ab 7sifiocalculations, 1: 71 data sources, 3: 803-804 intrinsic properties, 1: 937-939 liquid alloys, 1: 682-683 and structure maps, 1: 431 units used, 2: 304 Magnetic refrigeration, 2: 455, 457, 458, 460,484,486, 654; 3: 519-539 and bulk metallic glasses, 3 692 and (Dy, rErx)A12compounds, 3: 530531, 535-536 and (Gd, .Er,)NiAl compounds, 3: 535-5 36 and Gd,(Si,Ge,,) compounds, 3: 531535 and lanthanide compounds, 3: 529-535 measurement of magnetocaloric effect, 3: 524-525 principles, 3: 519-524 and transition metal compounds, 3: 528-529 Magnetic regenerator, 3: 522-524 Magnetic resonance imaging, 2: 377-379, 380, 381 Magnetic scattering [of electrons], 1: 1028 Magnetic spin-density distribut~on,1: 441, 442 see cclso Spin-density waves Magnetic structures, 1: 439-450, 939 and exchange interactions, 1: 439-441 d e t e ~ i n a t i o nof, 3: 165-170 by Mossbauer spectroscopy, 3: 166169 by neutron diffraction, 3: 165-166
99 1 by single-crysta~magnetizati~nmeasurements, 3: I69 by X-ray powder diffraction,3: 169170 experimental deter~~nation, 1: 440 interactions affecting, 3: 170-173 models, 1: 442, 443 rare-earth compounds, 1: 447-449 rare-earth metals/alloys, 1: 446-447 small-scale magnetic order, 1: 449-450 transition-metal compounds, 1: 444-445 transition metals, 1: 441-444 Magnetic superlattices, 1: 450 Magnetic susceptibility beryllides, 3: 45 heavy-fermion compounds, 1: 2 17-2 18, 220-22 1 liquid alloys and metals, 1: 671, 682, 683 metallic glasses, 3: 692 Magnetite, ion tracks in, 3: 267 ~agnetizationcurves metallic glasses. 3: 699, 700 single crystals, 3: 169, 176 Magnetizat~onenergy, 1: 71 Magnetization intensi~y,2: 304 Magnetization measurements, 3: 524 Magnetocaloric egect (MCE), 3: 519 'caret'-like behavior, 3 519, 520, 524, 525 @ant, 3: 528, 529, 536, 691 magnetic field dependence, 3 530 measurement of, 3: 524-525 'skyscraper'-like behavior, 3: 519, 520, 525, 528, 529 'table'-like behavior, 3: 519, 520, 525 temperature dependence, 3: 520, 525 Magnetocaloric properties, listed for various materials, 3: 526-527, 528, 529 Magnetoc~stallineanisotropy, 1: 937, 939; 2: 303, 304, 305, 306, 390, 392, 478; 3: 170-171 Magnetoelastic coefficient, 2: 398 Magnetoe~asticcoupling, 2 389 Magnetoelectronic semiconductors, 2: 324 Magnetogranulometry, 2: 483 Magnetomechanical coupling, 2 389, 396-399 Magnetomechan~calresonance, 2: 398 Magneto-optical applications, 2 435-450 materials requirements, 2: 441-442 Magneto"optica1 effects phenomenological description, 2: 435-438 physics, 2: 438-441 see also Faraday . , .; Kerr effect Magneto-optical properties, 2: 442-450 amorphous alloys, 2: 449-450 cobalt-based binary compounds, 2: 444-446 iron-based binary compounds, 2: 446-447 manganese-based binary C O ~ ~ O L I ~ 2: 443-444 ternary compounds, 2 448-449 Magnetoresistance, 3 104 Magnetoresistant rare-earth coinpounds, 3 104 Magnetoresistive coefficients, 1: 943
~ S ,
992 Magnetoresist~vedevices, 1: 950; 2: 344 Magnetostriction, 1: 1023; 2: 306, 389-403 atomic theory, 2: 392 frequency response, 2: 400 magnetic"fie1d effects, 2: 393, 394, 402 pressure response, 2 399400,402 temperature dependence, 2: 393-394 Magnetostriction constants, 2: 390 Magnetostrictive actuators, 2 401-403 applications, 2: 402-403 design and control, 2: 402 operating principles, 2: 402 Magnetostrictive devices, 3: 794 Mag~etostrictiveeffects, 3: 170 Magnetostrictive materials applications, 2: 402403 historical background, 2: 390-39 1 rare-earth compounds, 2: 392-393, 394-395 Magnetostxictive response. 2: 394 Magiietovibrational coupling, 1: 159 Magnetron sputtering, 2: 290,493; 3: 570, 571, 572, 581, 665 Magnons, 1: 945 Maldonite [mineral name], 1: 629, 632 Manganese alloys, magnetic structures, 1: 445 Manganese~basedbinary c~mpounds, magneto-optical applications, 2: 443-444 manganese structure, 1: 322, 325, 326, 327, 341 manganese structure, 1: 319, 683 Manganites, perovskite-like, 3: 104 Many-body interactions, 1: 31, 34, 603; 3 765 Marcasite [mineral name], 1: 328, 428 Marcasite structure type, 3: xxxix, 12 Marine applications, 2: 645-646 Market size, 3: 477 Mar-M superalloys, 2: 6, 7, 259 coatings on, 2: 492, 498 Mar MO02 coatings on, 3: 564, S65, 567, 568 density profiles, 3: 569 diffusivities, 3 569 oxidation of, 3: 568 hot corrosion kinetics, 3: 568 Mar M200, creep behavior 306 Mar M247, hig~-temperat~re strength, 3 70 Martensites crystallography, 1: 828-829, 830 and ductility, 1: 845-846 ineaniiig of temi, 1: 827 nucleation of, 1: 831 thermoelastic martensites, 2: 530-532 Martensitic transformations, 1: 143, 163, 715 in ceramics, 1: 835 crystallography, 1: 827-83 1 ~ e ~ n i t i o n1: s ,829-831 diagrams, 1: 715, 716, 717 electrical resistance changes durnig, 1: 830 at high pressures, 3: 160 on Internet, 3: 874, 875 kinetics, 1: 831 magnetic, 3 533, 534 in nickel aluminides, 2: 54-55
Subject Index in plastically deformed quasicrysta~s,3 400 relationship with massive transformations, 1: 843-844 and s~dpe-memoryefFects, 1: 163-164, 832; 2: 54-55, 529, 530 stress- aiid stram-induced, 1: 831, 832 theories, 1: 828 in titanium aluminide~,2: 96 see also Shape-memory effects Martynov-Batsanov electronegativity, 1: 232, 233, 390, 423, 424 see also ~lectronegativity MASC see Metal and Silicide Composite Mass spectrometry, tim~-of-flight(TOF) technique, 3: 214 Massive-like transform~tions,1: 844 Massive transformation$, 1: 843-844 definition, 1: 843 in intermetallics, 1: 843 and optical-memory applications, 1: 846 relationship with martensitic transformations. 1: 843-844 Mass spectroscopy, vapol- species studied using, 1: 681 Master alloys, 2 523 Materials, meta-sites on Internet, 3: 865 Matermls design accuracy of prediction, 3: 835 computer-aided, 3: 8 1 1-839 cybernetical-statistica~approach, 3: 832-834 Materials life-cycle, 3: 792 Mathias criterion, 3: 812 Matthias profiles, 1: 266, 951 Matthiessen's rule, 1: 944 MatWeb, 3: 872 Maximu~-convex-volumerule, 1: 365 Maximum-gap rule, 1: 237, 365 examples, 1: 238, 365, 368-373 Mean-field approximation in calculation of phase diagrams, 3: 1 87- 1 9 1 compared with CVM approach, 3: 195 as single-site model, 3: 195 Mean-field theory, 1: 38-39 Mechanical alloying, 3: 727-732 advantages, 1: 648 aiuminides, 3 61, 652-653, 654 aluminurn alloys prepared by, 2: 175 amalgams, 3: 26 amorphous phases formed, 1: 699-701, 734 beryllides, 3: 47 disadvantages, 1: 648-649 disordering from, 3: 73 1-732 magnetic materials prepared by, 2: 314, 315 mechanisms, 3: 728-730 metastable crystalline phases formed, 1: 706 metastable phases, 3: 756, 757 milling practice and theory, 3: 727-728 modeling of, 3: 730 nanostructures, 3: 7-51, 753-754, 755759 Nd-Fe-B magnets, 3: 102 powder synthesis by, 3: 645, 652-653, 654 process control agent used, 3: 756
quasicrystalline phases, 3 755, 7.56 reaction mechanisms during, 3: 723, 728-730 synthesis using, 1: 647-649, 912; 2: 599400 timescales, 3: 724 types of mill used, 1: 647 Mechani~alanisotropy, TiAl, 3: 630, 634 Mechanical attrition, 1: 700-701,733,734 types of mill used, 1: 701 Mecha~icaldisordering, 1: 701, 786-788 Mechanical energy, equivalence to heating, 3; 761 Mechanically alloyed compounds, 3: 730732, 756 Mecliclnical milling, 3 753, 759, 76U see also Mil~ing Mechanica~properties, 1: 14-1 5, 873-934; 3: 295-467 amorphous alloys, 3: 693-698 beryllides, 3: 4 2 4 dental amalgam, 2: 584-586 disilicides, 2: 220-223 high-temperature intermetallics, 2 240-241, 242 iron aluminides, 2; 202-207 metallic glasses, 3: 693-698 nanostructures, 3 762-763 iimr surfaces, 3: 227-228 Ni-A1 alloys, 3 737 nickel aluminides, 2: 35-42, 58-65 niobium silicide composites, 3: 490491, 550-555 processed TiAL 3: 630 quasicrystals, 3: 397-398 shape-m~moryalloys, 2 538 and structure maps, 1: 431 superalloys, 2: 8, 13-14 superconductors, 2: 358-359 TiAl sheets, 3 633-635 titanium aluminid~s,2: 77-79, 80-8 I, 83,92, 106-113, 120-124 trialuminide compounds, 2: 158-160 zirconium aluminide, 2: 136-144 see also Ductility; Elastic properties: Fracture; Plastic deformation; Ultimate tensile strength; Yield strength Mechanical testing, 1: 912-913 Mechanical-to-electricalenergy conversion, 2: 318 Mechanical twinning, 3: 403, 43 1 in plastically deformed quasicrystals, 3 400 see also Deformation twinning Mechanical yield stress anomaly, and thermal vacancies, IS: 290-29 1 ~echanochemicalsynthesis, 3: 723, 724, 743-745 reaction t h ~ ~ o d y n a m i c3: s , 744 Medical applications, 2: 548-554 Medium-temperature thermoele~trics,2: 464-465 Melting cofd-wall crucible, 3: 592 and disordering, 3: 245, 246 iron aluminides, 3: 612 nickel aluminides, 3: 602, 604-606 titanium aluminide, 3: 592 Melting inethods [for synthesis], 1: 638-644; 2: 43
s ~ b j Index e ~ ~ Melting temperatures ahmmrdes, 1: 114, 121; 2: 56, 68, 134, 199, 293, 525 as atomic property, 1: 231, 232 beryllides, 3: 38, 39, 40 temperatures, 2: 246 and cohesive energy, 1: 232 and density, 1: 266, 268 and elastic constants, 1: 195, 206-207, 891-892 element~lem~nt contour plot, 1: 268, 269 high-temperature ~ ~ t e r m e t ~ l 2: l i ~243, s, 244, 245, 246, 249, 251, 252, 253 listed various compounds, 1: 121,207,268, 269; 2: 642 various elements, 1: 207, 233 NaT1, 1: 110, 121 platinu~-groupmetals, 3: 67, 70 121 as selection criteria, 2: 241 s~~iconductors/semimetals, 2: 326, 327, 328, 465 shape-memory alloys, 2: 518 and shear moduli, 2 252 silicides, 2: 213, 294; 3: 733 and specific gravity, 2: 243 sulfides, 3: 710, 711 thermoelectric materials, 2 463, 465 ~ria~uminide compounds~2: 170 MeIt spinning, p~rmanentmagnets prepared by, 3: 97, 102 Melt spinning [processing technique], 1: 640, 733, 748, 773, 912: 2: 175, 180, 314, 316 Memory effects, 1: 163, 744, 832-834 see also Optical-memory . . .; Shapememory M e ~ r y t e cantiscaId ~ valve, 2: 546, 549 Mendeleev number in structure mapping, 1: 241, 243-244, 390 in weldability plots, 1: 266, 267 Mercury allotropes, 3: 21 applications, 3: 31 corrosion by, 3: 31 environmental aspects, 3: 33 extraction of, 3 21 lustoric importance, 3: 21-22 inte~etallics,3 800 in metal extraction/refin~~~ processes, 3: 29 as nuclear reactor coolant, 3: 31 occurrence, 3: 21 physical properties, 3: 21, 31 recycling of, 3: 33
Mercury amalgams coatings using, 2: 519-521 recovery processes using, 1: 4; 2: 5 15-51 8 see also A m a l g a ~ s
Mercury compounds see Amalgams Mercury-sharing clusters, in Zintl phases, 3: 122 Mertieite [mineral name], 1: 628, 629, 632 Mesotaxial implantat~on,2: 330 Metadata, 3: 836 Metal-base transistors, 2 213, 231; 3: 787 Metal hydrides, disordered, 3: 252-253 Metal-insulator semiconductor field-effect transistors (MISFETs), 2: 332, 335 Metallic bonding crystal structure affected by, 1: 183, 184 d~ffusionmechanisms afYected by, 1: 763 Mume-Rothery phases, 1: 100-101 Metallic compouiids, lattice dynamics, 1: 153-1 56 Metallic glasses bulk, 3: 681-4582 chance discovery, 3: 703 compared with oxide glasses, 3: 681 composition ranges, 3: 686, 688, 691 conditions for formation, 3: 699-701 creep behavior, 3: 696-697 developmeiit of, 3: 68 1-682 elastic properties, 3: 697-6953 ferroma~etic,3 698-699 f o ~ a b i l ~ t3:y ,697 formation conditions for, 3: 699-701 future prospects, 3: 701-703 magnetic hysteresis, 3: 700 magnetic susceptibility, 3: 692 mechanical properties, 3: 693-698, 702 Pd-rich, properties, 3 685493 plastic ffow in, 3: 694-696 precipitation in, 3: 703 stabilization of by solutes, 3: 700-701 synthesis of, 3: 682-4585 Metallic glasses, phonon dispersion effects, 1: 161 Metallic solvents, 3: 33 see also Amalgams Metallic Zintl phases, 3: 118, 122, 125 Metalliding, 1: 653 Metallization [in integrated circuits], 2: 23 1 diffusion barriers in, 2: 622428 Metallograpliy, 1: 7 Metallurgica~c o ~ p ~ e s s1:~ ~ 116 n, Metallurgy applications, 2: 489-499, 515-526 Metal nitrides, growth by MBE, 3: 786 Metal organic c h e ~ ~ vapor c a ~ deposition (MO-CVD), 2: 422; 3: 670 Metal-organic molecular-beam epitaxy (MQ-MBE), 3: 781 Metal organic vapor-phase epitaxy, 2: 324, 326 Metals, ion tracks in, 3: 267-268 Metal separation processes, 3: 29 Metal Schottky field-effect transistors ( M E S ~ ~ T2:S 332-334, ~, 346 Metal and Silicide Composite (MASC), 3: 490-49 1 creep properties, 3: 489, 553 fatigue crack growth data, 3: 491 h i g h - t e ~ p e r a t ~strength, e 3: 552 mechanical properties, 3: 490-491, 552 microstructure, 3: 547 oxidation behavior, 3: 556, 557 thermal expansion coefficient, 3: 555
993 toughness, 3: 490, 552 Metamagnetism, 3: 183 Meta-search engiiies, 3: 857, 877 number of hits for “inte~etallic”,3: 858 Metasites, 3: 865, 877 Metastable crystalline phases, 1: 703-707 synthesis by irradiation, 1: 705-706 by mechanical alloying/attrition, 1: 706 by pressure quenching, 1: 706-707 by rapid s o l i d i ~ c ~ t ~1: o n703-705; , 2: 180, 185 Metastable equ~libria,1: 99, 100,697498, 731 Metastable eutectic diagram, 1: 697, 698 Metastable phases, 1: 687-708; 3; 664,756 definition, 1: 687 free energy, 1: 687 metallurgical examples, 1: 687-688 production, 1: 688 quenched, 3 162 see also Amorphous alloys/phases; Metastable crystalline phases eta stable states, energy, 3: 198 Metastable transition states, 1: 850 Meteorites, as sources of inte~etallics,1: 3, 626, 633 Meteoritic materials, ion tracks in, 3 265, 27 1-272 Metglas foil, 3: 698 MgAgAs structure type, 3: 91, 93, 93 prediction of compounds, 3: 832 MgCu, structure type, 3: 88, 89, 91 twinnin~in, 3: 415 MgCuAl, structure type, MgN, structure type, 1:
Mg-Sn system, 1: 662 Mg-Zn system, 1: 662, 735 MgZn, structure type, 1: 389, 409-41 1, 423, 428, 429; 3: 88, 89, 91, 91, 93 Microalloying of AINi, 2: 51, of Al,Ti, 2: 147, 154 Microc~in~, s t ~ c t u fr o~ ~ u for, ~ a3; 5 Micro-electronics applications, 3: 512, 663,666,676,783,786-787,795-796 Microhardness testing, 1: 8, 595-597, 956; 2: 202, 248 M~cro-indentation,metallic glasses, 3: 697 Micropyretic synthesis, 3: 723, 724 see also Reaction synthesis Microscopy, first used, 1: 7 Microstruct~ralcoarsening, 1: 8 6 ~ 8 6 8 Microst~ucturalstability, 1: 849-868 Microstructure A2 and B2 phases, 3: A2 and E2, phases, 3: from canned extrusion, 3: 628 effect of casting on, 3: 594-596, 609610, 613 and forming, 3: 618-620 gamma-gamma-pr~~e phase, 3: 843 and hot working, 3: 624 niobium silicde composites, 3: 547-548 shock-induced, 3: 741 thin films, 3: 664
994 titanium aluminide, 3: 318, 478 Van Gogh’s sky-like, 3: 851-854 ~icrostructur~-proper~y relationships, titanium aluiiiinides, 2: 77-79 ~icrostructures bearing alloys, 2: 592, 593 creep affected by, 2: 115, 116 dental amalgain hardened amalgam, 2 581-582, 583 silver-tin alloy powder, 2: 580 Dy-Fe-Tb system, 2 397 effect o f processing, 2: 126, 127 N b S i system, 2: 296 properties affected by, 1: 924-925 effect of second phases, 1: 925-927 titanium aluminides, 2: 75-77, 84, 97, 98-101, 126, 127 t~aluminides,2: 161 Microtwins, 3: 411, 431 Microwelding technique, 3: 572 Mictomagnetis, 1: 440; 3 174, 177 Miedema analysis, 3: 66, 761 3: 91 Miedema e~ectroneg~tivity, rare-earth compounds, 3: 93 Miedema’s map, I: 234, 264 Miedema’s model, 1: 99, 116, 232, 234, 258, 381, 390, 391, 569; 2 609 Miedema’s surface segregation relationship, 1: 612 Miedema’s vaca~cy-formation rel~tionship,1: 568-569 Migration enthalpy, study of, 3 277 Milling disordering by, 3: 759-761 energy stored by, 3: 761 and long-range order, 3: 731 powder synthesis by, 3: 444-645, 652, 727-730, 753-754 Minerals on Internet, 3: 871, 872 see aiso specific ~ i ~names, e ~e.g.~ l Altmarkite; Poterite Mineral species see Naturally occurring compounds; specific mineral names, e.g. auricupride, cabriite, etc. Minii~umcreep rate effect of element additions, 3: 311 modeling of, 3: 299-302 Minimum structure type, 3: 19 Mirror Fusion Test Facility, 2: 373-374 Mirrors, 1: 3; 2 408, 409-410; 3 31, 231 Miscellaneous applications~2 527-655 Miscellaneous mnterrnetallic compounds, 2: 237-255 2: 313, 647 ~ i s c h m e t aalloys, l Miscibility gap (MG) phase diagrams, 3: 188, 189 Miscibility gap, in Pd-Rh alloys, 1: 50 Misfit strain, 1: 867 and primary creep, 3: 303 Mis~in- arka as (MF) interatomic potential, 3: 769, 770 Mismatch [in semicoiiductors], 2: 326 Mixed conductors, 3: 250 Mixed-gas environments (O/S),reaction to, 3 708-709 Mixing energy, 1: 28, 43, 850 b.c.c. alloys, I: 852 f.c.c. alloys, 1: 44, 854
Subject Index Mixing entropy, 1: 665, 666; 2: 638, 641 MnP structure type, 1: 261, 422, 424, 426, 431 Mn-Sb system, 2: 444 Mn-Si system, 2: 468, 469; 3: 158 Mn,Si, structure type, 1: 120; 3: 89,89, 90 Mn,,Th structure type, 1: 398,433; 2 31 1, 315-3 16 Mn,,Th, stixcture type, 1: 385, 396-397, 447-448; 2 315 Mn13Y6hydrides, 1: 448 Mii-Zr system, 1: 740 Mobile species, effects, 3: 245-261 Modeling casting processes, 3: 595, 598 diffusion-limited grain growth, 3: 674 forming processes, 3: 630-631, 635-436 mechanical alloying, 3: 730 minimum creep rate, 3: 299-302 powder consolidation, 3 647-648, 658 of structure, 3: 841 thermodyna~ic,of phase diagrams, 3 191 track formation, 3: 269-271 Models, 3 821 amorphization, 1: 820-821 amorphous alloys, 1: 739-742 bainite formation, 1: 841-842 brittle fracture, 1: 907-909 complex formation in liquids, 1: 671475,750 for construction of phase diagrams, 3: 187-191, 195 Deal-Grove model, 3: 674 diffusion, 1: 762 dislocation~locking,1: 547 dislocations, 2: 23, 238 electrical resistivity, 1: 942, 945, 959, 960 flow-stress anomaly, 1: 546-547 glass-formation, 1: 679-680 Gorsky-~ragg-Williams model, 3: 191, 201, 202, 206 grain boundaries, 1: 598403, 605 for interdiffusion, 3: 563-564, 568 Ising model, 3: 192 magnetic structures, 1: 442, 443, 450 Miedema’s model, 3: 66 Nabarro creep, 3: 315 Orowan’s twin-nucleation model, 3 418 short-range order, 3: 73-74 strengthening, 2: 260 surfaces, 1: 610-61 1, 620 triple-defect, I: 559, 566, 572-573, 575 Modified aluminide coatings, 2: 491 commercially available, 2: 492 Modified embedded atom method (MEAM), 3: 457,458, 765, 767-768 applications, 3: 774-775 Modulated phases, 3 xxxix Modulation-doped heterojunctions~ structures, 2: 334, 425 Modulators optical, 2 342, 416 photoelastic, 2 416, 417 Mohr-Coulomb yield criterion, 3 694 Mold preheat, 3: 595 Molecular-beam epitaxy (MBE), 1: 651; 2: 324, 421; 3: 667, 779-788 advantages, 3 782
device ap~lications,3: 786-787 growth chamber, 3 781 Cs growii by, 3: 780, 783-786 applications, 3: 786-787 CsC~-structurecompounds, 3 784-786 NaCl-structure compounds, 3: 786 materials grown by, 3: 780 metal-base transistors fabricated by, 3: 787 quantum wells grown by, 3: 787 technology, 3: 780-783 vacuum requirements, 3: 779, 780 Molecular beams, generation of, 3: 781 Molecular dynamics (MD) calculatio~s, 1: 77 defect structures formation, 1: 799 grain boundaries, 1: 87-88 irradiation-induced amorphization, 1: 817, 820 Molecular dynamics (MD) simulations Internet image, 3: 875 for phase diagrams, 3: 197, 201 for undercooled liquids, 3 694 Molecular-field theory, 1: 949 Molecular-orbital calculations, Zintl phases, 3: 119 Molybdenum, properties, 3: 545 Molybdenum disilicide, 3: 485-486 fatigue properties, 3 342, 344-345, 346 synthesis of, 3: 657, 734, 735, 758 M o l y ~ d e n ~disilicide m based composites creep behavior, 3: 317-318, 485, 486 fatigue crack growth in, 3: 344345,346 Molybdenum disilicide coatings, 3: 574 Molybdenum silicides high-temperature properties, 3: 485487 nanocrystalline, 3: 731 powder metallurgy applications, 3 657-658 sulfidation of, 3: 717-719 Momentum theory, for quasicrystals, 3: 386 Mo-Ni-Si system, 2: 226 Monitor, in AI system, 3: 818, 819, 820 Mo-Ni-W system, 1: 728 Monoatomic metals, elastic properties calculated, 1: 203-204 Monolithic integration, 2: 342 Monte Carlo methods, 1: 39, 128, 132, 762, 776 in phase diagram calculations, 3: 195 surfac~-~nduced ordering studied by, 3: 223 see also Inverse Monte Carlo method Mooij correlation, 1: 484 Mooser-Pearson plots, 1: 241, 242, 357, 419, 420 see also Structure maps Morphology, of scale on Ni,Al, 3: 712, 713, 714 Morphotropy, 1: 415 Morse potentials, 1: 82, 83, 154, 155; 3: 766, 768 Mo~cliellandsbergite[nlineral name], 1: 629, 632; 2: 576; 3: 21 Mo-Si alloys, 3: 487-490 Mo-Si system, 1: 998-999 MoSi2structure type, 1: 340, 385, 386-391, 392, 396, 721; 3 88,89 in structure maps, 1: 389, 428, 429
Subje~tIndex Mo,Si,C structure type, 3: 89 Mo-Si system, defect calculations, 3: 774775 Mo-Si-W system, 2 616 Mo-Si-Zr system, 2: 225, 226 Mossbauer spectroscopy, 1: 440,444,450, 479 magnetic structure determined by, 3: 166-1 69 Matt insulators, 1: 130, 197 Mo-Zr system, 1: 740 Muffin-tin (MT) approximation, 1: 133 see also Linearized muffin-t~norbital (LMTO) method Mullite [mineral tiame], 1: 999 Multicomponent refractory silbdes, 2 225-227 Multidimensional classification rules, prediction by, 3: 812-8 13 M~iltifilamentarysuperconductors, 2: 358, 360, 361, 370 Multilayer coatings, 2: 499 Mult~layerfaults, 3: 460, 464 Multiphase systems microstructures, 1: 859 stability of, 1: 861-862 Multiple-compound [diffusion] barriers, 2: 627 Multiple defects, superimposing effects, 3: 359 Multiple diffusion, 3: 841 Multjple quantum-well (MQW) inodulators, 2: 341, 342 Multiple quantuin-well (MQW) structures [in semiconductors], 2: 328-329,425,426 ~ u l t i s t e pforging, 3: 623, 639 Munta metal, 1: 3 Muon diffusion, 2: 484 Mu phases, 3: xl Muscovite, structural formula for, 3: 5
difYusivity, 3: 750 disordering of, 3: 759-761 ductility, 3: 763 hardness, 3: 762 mechanical properties, 3 762-763 solute efYects, 3: 758 synthesis of, 3: 751-755 by devitrification, 3: 740, 751, 754755 by electrodeposition, 3: 751, 754 by inert gas condensation, 3: 751, 752-7 53 by mechanica~alloying, 3: 751, 753754 by rapid solidification, 3: 751, 755 by spray conversion, 3: 754 toughness, 3: 763 see also Nanocrystalline IMCs Nanotwins, 3: 411, 419, 431 NaO structure type, 1: 262, 426 Na-Pb system, 1: 662, 667, 680 Narrow f-band compounds, 1: 214 NASAIR 100 [superalloy], 2 63 Na--Sb system, 3: 158 Na-Sn system, melting behavior, 3: 255, 256 NaTl structure type, 3: 251 NaT1-type Zintl phases, 1: 110, 319 Naturally occurring compounds, 1: 10, 625-634 characterization, 1: 625-626 listed, 1: 627-631 selection procedure, 1: 626, 632 sources, 1: 626 NaZn,, structure type, 1: 385, 397-398, 433; 3: 174 NbBe, structure type, 3: 88, 89 Nb-Ni system, 1: 735 N-body potential, 3: 766, 772 Nb-Rh alloys, high-temperature strength,
Nabarro creep model, 3: 315 Nabarro-Herring creep, 1: 9 17, 9 19 NaCl structure type, 3: 89, 89, 180 growth of compounds by MBE, 3: 786 see also B1 type compounds; ClNa-structure Nanocrystalline IMCs, 3: 755-761 equilibr~~im phases, 3: 756-759 metastable phases, 3 756, 757 quasicrystalline phases, 3: 755, 756 shock consolidation of, 3: 739 synthesis of by mechanical alloying, 3: 652, 654, 724,727,728, 729.731,753-754, 755 by shock compression, 3: 724 Nanocrystalline materials, 1: 748, 749; 3: 750 Nanoglasses, 3: 750 Nanoindentation with atomic force microscopy (NIAFM), 3: 215, 227, 228 surface properties determined using, 3: 227-228 ~ A N O P E R Malloys, 3: 702 Nanoquasicrystals, 3: 750 Nanostruct ures, 3: 749-764 characteristics, 3: 750-75 1 classification of, 3: 750
Nb-Si alloys crystallization under pressure, 3: 161 fatigue crack growth in, 3: 345-346 Nb-Si system, 1: 999 Nb-Si-Ti system, phase stability in, 3: 548,549 Nb-Sn system, 2 353, 355 Nb-Zr system, 1: 740 NdAs, structure type, 3: 89, 89 NdNiGa, structure type, 3: 92 NdRuSi, structure type, 3: 92 Nearest-neighbor bond models, and vacancy-fo~ationenthalpies, 3: 280 Nearest-neighbor interactions close-packed structures, 1: 280 see also Next neighbor . .; Secondnearest . . Nearly-free-electron theory, 1: 56, 945 Near-net-shape casting, 3: 492 Near-net-shape fabrication, 3: 643, 654, 682 Near-net-shape forgmg, 3: 635, 636 Near-net-shape production, 2: 42, 67, 68 Necking [in tensile testing], 2: I42 ‘Necklace, ~icrostructure,3: 60, 67, 478 Niel temperature, 1: 938; 2 331, 391 Negative crystal impurities/inclusions, removal of, 1: 189, 191 Negative crystals, 1: 189, 190
3: 70
99s Neighborhood diagrams, Al-Ti system, 1: 432, 433 Nets body-centred cubic structures, 1: 336-341 close-packed structures, 1: 410, 412 ‘Netspeak’ vocabulary, 3: 876 Network compounds, 3: 122-127 Neumann’s principle, 2 438; 3: 439 Neural networks, 3: 824 Neutron absorption capture cross section, AlZr,, 2: 134, 144 Neutron diffraction/scattering, 1: 10-1 1; 3: 247 heavy-fermion compounds, 1: 2 18 Laws phases, 1: 109 limitations, 3: 166 liquid alloys, 1: 662-663 magnetic structures d e t e ~ ~ i n using, ed 1: 439.440,442,444,447,450,936: 3: 165-166 phonon dispersion curves dete~mined using, 1: 149, 150 vacancies detected by, 3: 277 Neutron irradiation, effects, 1: 803, 804-808, 813 Neutron multiplier, 2: 648 Neutron scattering, 3: 248-250 inelastic, 3: 247, 248-249 quasielastic, 3: 249-250 news group^, on Internet, Next-neighbor histograms, 1: 238, 365. 368-373 NG45F [resistive] alloy, ..nI(12i) framework, 1: 329, 332, 334, 336 NiAl alloys see AlNi alloys Ni,A1 alloys see AlNi, alloys NiAI- based alloys see AlNi-based alloys Ni,Al-based alloys see AN,-based alloys Niccolite, 1: 10 Nickel, reflectivity, 3: 232, 233 Nickel aluminide and alloys, 2: 17-69 as bearing materials, 2: 598-599, 600 see also A N - . . .; AlNi,-based alloys Nic~el-aluminidecoatings, 3: 565-567 growth mechanisms, 3: 566 plati~¶uni-mo~i~ed, 3: 64, 65, 72, 74, 564, 565, 567, 568 diffusional transport in, 3: 568-570 Nickel aluminides, 3: 55, 492-494 atomistic calculations, 3: 768-772 bery~lium-substituted,3: 45 casting of, 3 602-61 1 creep behavior, 3: 297, 309-312, 493 economic factors, 3: 795 enthalpy of formation, 3: 605 fatigue crack growth in, 3: 341--342,347 fatigue properties, 3: 329-331 forming of, 3: 639 joining of, 3: 509 -5 10 length”change measurements, 3: 283286 and platinum group metals, 3: 64 and positron lifetime, 3: 279 powder metallurgy applications, 3: 655-657 properties, 3: 493 solid solution hardening in, 3: 353-354 strain hardening in polyc~stallineinte~etallics,3: 368373
Subject Index
996 single crystals, 3: 365-367 sulfidation of, 3: 710-712 synthesis of, 3: 655-657 by mechanical alloying, 3: 656-657, 728-730, 757-759 by reaction synthesis, 3: 656, 736 ternaiy alloys, 3: 60, 302 thermal conductivitv, 3: 603 thermal vacancy foimation in, 3: 280, 281, 282 see also AlNi-. . .: AlNi,-based alloys Nickel-based alloys, corrosion behavior, 1: 966-967 Nickel-based bearing alloys, 2: 596-597 Nickel-base eutectic alloys, 3: 3 18 Nickel-based superalloys, 2 3-14 aluminide coatiiigs on, 3: 494,563, 566, 567 belyllides compared with, 3: 42 cooling of, in turbines, 3: 541 creep behavior, 3: 297, 300, 302, 309312,404 design of, 3: 842-843 liquidus temperat~~res, 3: 544 notch rupture behavior, 3: 298 platinum group metals as alloying additions, 3: 65-66 precipitation hardening in, 3: 3 13-316 precipitation in, 1: 859, 867 precipitation strengthening in, 3: 64 silicide coatings used, 2: 228 so~~d-solution strengthenlng in, 3: 307308 temperature limits, 3: 542 Nickel-based white gold alloys, 2: 560,570 Nickel beryllides, 3: 43 Nickel-metal~hydridebatteries, 3: 107108 Nido clusters, Zintl phases, 3: 114, 115, 120, 125 Niggliite [mineral name], 1: 629, 630 Ni-Hf system, 1: 696 Ni-Mg system, 1: 696 270, 272, 273-276 oxidation behavior affected by, 1: 984-986, 993; 2: 81 plasticity of AlTi, alloys affected by, 2: 91 properties, 3 545, 800 see also Nb . . Niobium aluminides fatigue crack growth in, 3: 342, 343 fatigue properties, 3: 332-333 twinning in, 3 414 Niobium beryllides dislocations in, 3: 44 fracture toughness, 3: 44 Niobium silicide composites, 3: 490-492, 541-560 creep behavior, 3: 489, 553-555 elastic properties, 3: 555 fatigue crack growth in, 3: 491 fracture behavior, 3: 550, 552 high-temperature strength, 3: 552-553
oxidation of, 3: 556-558, 658 and phase stability, 3: 548-550 powder metallurgy app~~~dtions, 3: 658 processing of, 3: 545-547 and quaternary/higher-order systems, 3: 549-550 stress rupture data, 3: 553 and ternary systems, 3: 548-549 thermal properties, 3: 555-556 toughness, 3: 490, 552 Niobium silicides fatigue crack growth in, 3 345-346 nanocrystalline, 3: 73 1 Ni-Pd-Ti system, 1: 716 Ni-Pt naturally occurring compounds, 1: 632 Ni-Pt system phase diagram calculations, 3: 205 surface-induced ordering, 3 223 surface structure, 3: 218 Nisbite [mineral name], 1: 629, 632 Ni-Sb system, 1: 632, 703, 704 Ni-Si alloys, 3: 503 joining of, 3: 510 shock-induced synthesis of, 3: 741 Ni-Si-Pd system, 2: 616 Ni-Si-Pt system, 2: 615-616 Ni-Si system, 1: 649, 999; 2: 643 Ni-Si-Ti system, 1: 258, 261, 898-899, 900; 2: 215-216, 616 Ni-Si-Zr system, silicidation reactions, 2: 615 Ni-Sn system, 1: 767,843,846; 2: 506-507 Ni,Sn structure type, 1: 430, 431 Ni-Ta system, 1: 735 Ni-Ti alloys, twinning in, 3: 416 Nitinal alloy, 1: 832, 835; 2 649, 650 Ni--Ti system, 1: 696, 742, 832, 835; 2 536, 537, 538, 556, 649 NiTi, structure type, 1: 385, 392, 393-394, 396 in structure maps, 1: 389, 429 Ni,Ti structure type, 1: 430, 431 Nitrides, 1: 299-302 Nitrogen, oxidation affected by, 1: 991-992, 993 Nivco-10 alloy, 2: 650 Ni-Zn system, 3 IS8 Ni-Zr system, 1: 696-699, 735, 738, 740, 742, 750 Noble metal alloys, phase diagram calculations, 3: 205 Noble metal coatings, 3: 584 Non-bonding orbitals, 1: 346 Non-ferrous shape-memory alloys, 2: 538 Non-metals, mercury compounds, 3: 25 Non-octet compounds crystal structure, 1: 242 structure maps, 1: 421, 422, 424 see also Octet cornpounds Non"stoichiometric compounds, 1: 299 Normal adamantane"structure types, 1: 347, 349, 359 Normal tetrahedral structure, 1: 346 Normal-valence compounds, 1: 346 meaning of term, 3: XI Notch rupture behavior, 3: 298 Notch sensitivity, 2: 143 Notch strength ratio (NSR), 2: 143 for AlZr,, 2: 143, 144
Nowick-Berry relaxation model, 1: 778, 782 Nowotny, Hans, 3: xliv Nowotny compounds, 3: xl Nuclear fission gas bubble formation, factors affecting, 1: 815, 816 Nuclear fuel applications, 1: 791; 2: 133, 647-648 Nuclear magnetic resonance (NMR) spectroscopy, 1: 440, 479; 2: 369, 470-37 1 Nuclear moderators, 2: 475, 648 Nuclear radiation detectors, Nuclear reactor coolant, mercury as, 3: 31 Nuclear reactor structural materials, 1: 791; 2 133-145, 648-649 requirements, 2 133 Nucleation bainite, 1: 838-839 crystals, 1: 690 disordered regions, 1: 780 ~drtensite,1: 831 new phases, 1: 759, 859 Nucleation barriers ainorphous-phase formation, 1: 696 metastable crystalline phase formation, 1: 704 0 ~ o r t h o r h o phase ~c~ in Al-Fe-Si-V system, 2 180, 181 in Al-Nb-Ti system, 1: 858; 2: 94, 95, 104-105, 106, 292, 293 structure, 1: 520 Object, meaning of term, 3: 836 Occupied electronic states, investi~ation of, 3: 135-145 ~ctagonal-phasequasicrystals, 1: 453 Octahedral intersticies atoms, 1: 296 -299 Octahedral slip, 3: 304 Octahedron, 1: 238, 239, 281, 404 frequency plot, 1: 240 structure map, 1: 245 Octet compounds, 1: 420; 3: xl crystal structure, 1: 242 structure maps, 1: 420, 422, 424 see also Non-octet compounds Off-stoichiometry effects in aluminides, 2 30, 32, 60-61 on flow stress, 3: 357 in trialu~inidecompounds, O h c contacts, 2: 213, 231 Olivine [mineral name], 1: 328 Omega phases, 3: XI Omega (0) type structures, 1: 857, 858 One-electron approximation, 1: 57 One-phonon scattering, 3: 249 One-way shape-memory effect, 1: 832, 833; 2: 534, 535 applications, 2: 536-643 Onofrite [mineral], 3: 21 Optical applications, 2: 336-343, 407-432 Optical bandpass filters, 2 410-41 1 Optical communications systems applications, 2: 336, 342, 346, 413, 414, 428
Optical conductivity tensor, 2: 439 Optical coupling, 2: 414 Optical data storage applications, 2: 435, 438,441
997
Subject Index Optical detectors, 2: 339-342 Optical filters, 2: 330, 410-41 1 Optical glass fibers, 2: 346 Optical imaging float zone (OIFZ) process, 3: 545, 546 Optical-m~~ory alloys, 1: 846 Optical modulators, 2 342, 416 Optical NOR gate, 2: 412 Optical phonons, 1: 149 Optical properties, 3: 231-244 databases on, 3: 816, 817 Optical recording media, 2 431 Optical reflectivity, 3: 232, 233 disilicides, 3: 233, 235 Optical signal repeaters, 2: 414, 415 Optical signal storage/retrieval, 2: 336, 428 Optical switches, 2: 342, 343 photoconductivity-dependent, 2 421 and photorefractive effect, 2: 41 1-413 Optical waveguides, 2: 413-414 Optoelectronic integrated circuits, 2: 327, 343, 347,432 Optoelectronlc se~iconduc~ors, 2 324, 336-343, 426 Orbital quenching, 1: 937 Order, 1: 771-788 and ball milling, 1: 700, 788 and creep resistance, 3: 308-312, 321 and electrical resistivity, 1: 777, 778, 946947,948 nucleation, 1: 780 radiation-enhanced relaxation, 1: 784 and scattering measurements, I: 781 and sputtering, 1: 773 strain liardening affected by, 3: 364-365 see also Long-range . .; §hort-range order Order changes ir~adia~ion-induced, 1: 784-786 mechanically induced, 1: 700, 786-788 thermally induced, 1: 771, 772-784 see also Aiiiorphization Order-disorder phase diagram, 1: 854 Order-disorder transition, 1: 119, 712 and creep resistance, 3: 308-309 elastic moduli affected by, 1: 888-889 electron states influenced by, 1: 1018 at grain boundaries, 1: 590 Hamiltonian describing, 1: 24 and heat capacity, 1: 119, 1020 near surfaces, 3: 222-223, 228 physical properties affected by, 1: 21 second-order, 3: 846 and surfaces, 1: 615, 618 see also Disorder-order . . Ordered adaman~ane-structuretypes, 1: 352-3 56 binary defect types, 1: 353-354 binary normal types, 1: 352-353 quaternary defect types, 1: 356 quaternary normal types, 1: 356 ternary defect types, 1: 350. 355-356 ternary normal types, 1: 354-355 Ordered Cu,Au rule, 2 606 Ordered substitution in close-packed structures, 1: 405, 416 Ordered twinning, 2: 150 Ordering congruent, 1: 853
continuous, 1: 851 cooperative, 1: 1020 and diffusivity, 3: 364 by distribution of relaxation times, 1: 782 hardening of gold alloys by, 2: 562-569 heterogeneous, 1: 774 homogeiieous, 1: 774 irradiation-enhanced, 1: 799-803 and magnetic permeability, 2: 303 non-equilibrium, 1: 857 relaxation trajectories, 1: 783 sponodal, 1: 773, 803, 859 at surfaces, 3: 223-224 Ordering energy, 3: 28, 31 graiii-boundary structure studied using, 1: 600 listed for various b.c.c.-based alloys, 1: 42 and surface structure, 3: 219 Ordering instability, 1: 852, 855 Ordering kinetics, 1: 771 experimental determination, 1: 776-779 theory, 1: 775-776 Ordering maps, 1: 34-35, 36-37 Order parameters, 1: 775, 854 amorphous phases, 1: 700, 701 Order relaxation, 1: 775-776 temperature dependence, 1: 776 Order shufiing, 3: 41 7 Order strengthening i~echanism,2 266-269 Order-twinning, 3: 406, 43 1 Ore deposits, as sources of intermetallics, 1: 626 0,Re structure type, 1: 428, 430 Organometallic precursors, synthesis using, 1: 650-651 Orientational disordering, 3: 245, 254 and ferroelastic transitions, 3: 258 Orientation imaging microscopy (OIM), 3: 306, 322 Orowan equation, 1: 547; 2: 260, 268-269 Orowan relationship, 3: 315 Orowan's twin-nucleation model, 3: 41 8 Orpiinent structure type, 3: 19 Orr-Sherby-Dorn parameter, 3: 300 see also Larson-Miller parameter Orthoclase [mineral], ion tracks in, 3: 264 Orthodontic application [for shapememory alloys], 2: 553 Orthohexagonal arrange~ents nets in, 1: 336-341 notation used, 1: 337 Orthopedic applications [for shapememory alloys], 2: 548-553 Ort~orhombicalloys, 3: 494-495, 849 see also AINbTi, alloys Ortliorhombic phase see 0 phase Orthorhombic structures, 1: 339-341 Orthorhombic symmetry in Al-Nb-Ti system, 2 94,95, 104-105, 106, 243 see aZso 0 phase Orthorhombic system, point groups, 1: 311 Oscillatory [surface] relaxation, 1: 6 17, 618 Osmium, world deposits/reserves, 3: 80
Osprey [thermal deposition] process, 1: 642 OsSi, structure type, 1: 388 Os-Zr system, 1: 740 0,Ti structure type, 1: 389, 428, 429 Overlay coatings, 2: 489, 490, 491-492 deposition processes used, 2: 492 Oxidation failure mechanisms during, 3: 564-565 fundamentals, 3: 575-577 and sulfidation, 3: 707 Oxidation-affected zone (OAZ), 1: 1001, 1002 Oxidation behavior, 1: 977-1017 accelerated, 1: 1000--1006 active, 1: 997 Al-Ir system, 3: 59 of alurninides, 1: 981-997; of beryllides, 1: 999-1000; 3: 44-45 coatings on Nb-based allays, 2: 229 fundamentals, 1: 977-981 high-te~pe~ature intermetallics, 2: 244 niobium silicide composites, 3: 556558, 658 of silicides, 1: 997-999; 2: 227, 228 y-titanium aluminide, 3 481, 575-577 see also Past degradation Oxidation resistance iron aluminides, 2: 207 molybdenum silicides, 3: 486, 657 nickel aluminides, 2 57 oxide layers providing, 1: 912,977-978, 981, 998, 1000 silicides, 2 213, 227 titanium aluminides, 2 81-82, 124 trialuminides, 2 167-169 zirconium aluminide, 2 144 Oxidation-resistant coatings, 3: 48, 65, 488,494, 574, 577-580 Oxidative lifetime predictions, 1: 1008-1 0 10 metrics used, 1: 1009 Oxide conductors, ion tracks in, 3: 268Oxide-dispersion-stren~thened 269 NiAi alloys, 3: 656-657 Oxide dispersio~~~trengtlien~d (ODS) alloys, 1: 925, 1006 Oxide phase maps, 1: 982, 992 Oxide protective scales adherence, 1: 979-980 growth rates, 1: 978--979 ino~hologles,1: 98 1, 982, 990 reformability, 1: 980 requirements, 1: 978-980 stability, 1: 978 transport mechanisms in, 1: 990 Oxides close-packed structures, 1: 299, 300, 301, 302 diffusion coefficients, 1: 979 removal from substrate, 3: 782 self-diffusion coefficients, 3: 711 stability, 3: 708 thcrinodynainic stability, P: 978 vapor pressures, 1: 978, 997 Oxide super conductors^ 2 352 Oxygen dissolution reaction, 2: 508 Oxygen embrittlement, 2: 30, 46, 124 effect of grain size, 2: 33-34
998 Oxygen-scavenging effect, twinning affected by, 3: 424 P phases, 1: 411, 412; 3: xl Pack aluminizing/chromizing, 2: 49049 1, 495 Packing density close-packed structures, 1: 278 cubic structures, 1: 281 Pack processes, aluminide coatings produced by, 3: 566, 570, 572, 577-579 Paddle-wheel migration, 3: 255-257, 258 Paidar-Pope-Vitek (PPV) [dislocationlocking] model, 1: 547, 548; 2: 24 modifications, 2: 25-27 Pair correlation functions, 1: 38 Pair distribution functions, 1: 739 Pair interactions, 1: 28, 29, 786 Pair potentials, 3: 194 Pairwise interaction model, 1: 850 Palarstanide [Inineral name], 1: 629, 632 Palladium demand and price ~uctuations,3: 79 ductility of AINi, affected by, 2 39 gold alloys with, 2: 560, 570-571 world deposits/reserves, 3: 80 see also Pd- . . . Pa11adiL~m”containing metallic glasses, 3: 685-693 Paolovite [mineral name], 1: 629, 630, 632 Paradocrasite [mineral name], 1: 629, 632 Paramagnetic moment, 1: 938 Paraschachnente [mineral name], 1: 629, 632; 3 21 Paris-Erdogan equation, 3: 325, 333 Parkes process, 1: 5; 2 516 Partial atomic volume, partial enthalpy of formation affected by, 1: 114-1 17 Partial enthalpy of formation, correlation with partial atomic volume, 1: 114-117 Partial pair distribution functions, 1: 739, 741 Particle accelerators, 2: 376-377 Particle interactions, calculations, 3: 841 Particulate-reinforced alloys, 1: 925 oxidative behavior, 1: 1006, 1007 Parting [in copper-gold alloys], 1: 970 Partition functions, complex formation described using, 1: 676-677 Passivation, 1: 967, 969; 2: 482, 502, 503 Passive corrosion, 1: 966 Passive oxidation, 1: 997 Patents, 1: 3 [footnote), 5; 3: 487, 488 Internet resources, 3: 865, 866 Patho-probability method, 1: 775, 779 Patio process, 2 515 Pattern recognition, 3: 822, 824 Pattern-recognition applications, 2 41 2 Pauling’s electrovalence rule, 1: 300, 301, 305, 354, 356 Pauling’s principles, 1: 260 Pauling’s rule of parsimony, 1: 280, 294, 302, 305 Pauling’s triacontahedron, 1: 471, 475 Pauli paramagnets, 2 479 PbFCl structure type, prediction of compounds, 3: 831, 832 Pb-Pd naturally occurring compounds, 1: 632
Subject Index Pb-Sb system, 3: 161 Pb-Te system, 2: 465 P b Z n system, 1: 662 PCT curves, 2: 476 Pd-Cu-Ni-P glasses, 3: 490-691 composition range, 3 691 Pd-Cu-P glasses, 3: 688-690 composition range, 3 688 Pd-Fe-Ni-P glasses, 3: 691-693 composition range, 3: 691 Pd-Ni-P glasses, 3: 684, 685-4588 composition range, 3: 686 Pd-Rh system, 1: 50-51 Pd-Sb system. 1: 632, 703 Pd-Si system, 1: 691, 735, 738 Pd-Sn naturally occurring compounds, 1: 632 Pd-U system, 2 469 Pd-Zn system, 3 158 Pemson’cyHandbook, 3: 797 Pd-Zr system, 1: 740 Peak strength correlation with antiphase”b0undary energies, 2: 283, 284 listed for various alloys, 2: 284 model prediction, 2: 281 Peak strengthening, 2: 268, 283 Peak stress and strain rate in TiAl, 3: 637 Pearson classification [of structme types], 1: 403 Pearson structure symbols beryllides, 3: 38, 39, 40 rare-earth Compounds, 3: 165 relationship to other nomenclature, 3: XXXi-XXXiV, 13-15, 238, 439 Pearson symbols relationship to other nomenclature, xxiii-xxvi for various crystal structure types, 1: 364, 373 Peierls instability, 1: 57 Peierls stress, 1: 873; 2 19, 23 Peltier eEect, 2: 454, 457 Peltier refrigerators, 3 49, 106 Pencil glide, 3 362 Pen nibs, 2 556, 651 Penrose lattice, 1: 480, 482 Penrose tiling, 1: 460, 480-482, 483 Pentlandite [mineral name], 1: 630 Percolation model for amorphization, 1: 820-821 Periodic bond chain vector concept, 1: 167, 170-171, 182 Periodic parameter (structure maps), 3: 91 Penodic-system representation atomic properties of elements, 1: 233,423 binary AB, compounds, 1: 376, 377, 380-381 Pettifor’s relative order number (for structure mapping) defined, 1: 425 Periodic table, on Internet, 3: 871 Peritectoid transfo~ations,2: 134-136 Permalloy [magnetic material], 2 306, 390 Permanent-magnet materials, 3: 97, 101, 165, 171 economic factors, 3: 794 Permanent magnets, 1: 748; 2 305, 307-3 17 applications of, 2: 3 18 machinable, 2: 317 in magnetic refrigeration, 3 524
manu~actureof, 3: 101-103 Permeability [of magnetic materials], 2: 304, 305 at constant length, 2 398 at constant stress, 2: 396 Permeable-base transistors (PBTs), 2 213, 231, 330, 332, 335 Permendur [magnetic material], 2: 307, 308, 390 Permittivity, semiconductors, 2: 327, 328, 345 Perovskite defect derivative, 3: 12 ~ a3 5 structura~f o ~ u for, Perovskite-like manganites, 3: 104 Perovskite structures, 3: XI Perovskite-type ceramics, 1: 419 Per~stentslip bands, 3 326, 348 Pest degradation e~ect/phenomenon,1: 16, 596. 597, 898, 927, 997, 1000-1006; 2: 2007, 288, 507 abatement~e~mination of, 1: 1002-1003, 1005 activation energy for, 1: 1002 Al--Nb system, 1: 1000-1002 aluminides susceptible, 1: 1000-1003, 1005 beryllides susceptible, 1: 1001, 1005 explanations, 1: 1001-1002, 1003, 1004-1005 features of phenomenon, 1: 1000 and intergranular attack, 1: 1005-1006 Mo-Si system, 1: 1003-1005 silicides susceptible, 1: 1001, 1003-1005 Pest phenomenon beryllides, 3: 44 molybdenum disilicide, 3: 657 silicide coatings, 3: 574 Pettifor [structure] inaps, 1: 390, 439420, 424-434 Ab compounds, 1: 425-427 AB, compounds, 1: 389, 390, 393, 396, 428, 429 AB, compounds, 1: 428,430,431 AB,, compounds, 1: 433,434 AB,, compounds, 1: 433, 434 AB,, compounds, 1: 433, 434 A,B,, compounds, 1: 433, 434 A,B, compounds, 1: 400 colored IMCs, 3: 237, 242 extension to pseudobinary phases, 1: 243. 431 extension to ternary phases, 1: 243-244, 435-437 relative ordering number used, 1: 425, 43 1 ternary chalcogenides, 1: 435-437 see also Structure maps Pewter, 1: 3 PHACOMP [phase computation] analysis, 2: 9 Phase boundaries, and pressure, 1: 688 ~hase-changematerials, 2: 637, 638 Phase compatibility, multiphase alloys, 3: 843-847 Phase decomposition, 2: 608 Phase diagrams, 1: 21 A15 compounds, 2: 353-354, 355 nb irzitio calculations used, 1: 73,114, 1 15 Ag-Au-Cu system, 2: 560, 562, 563, 567, 568
Subject Index Ag-Au system, 2: 561 Ag-Bi system, 1: 102 Ag-Cu system, 2 561 Ag-Ge system, 1: 103 Ag-Hg-Sn system, 2: 579 Ag-Hg system, 2: 517, 578 Ag-In system, 2: 522 Ag-Sn system, 2: 522, 577 Al-Ca system, 1: 738 Al-Ce-Fe system, 2: 179 AI-Cr-Ti, 3: 581 Al-Cu system, 2: 537, 595 AI-Fe-Si-V system, 2: 183, 193 AI-Fe system, 1: 114; 2: 200 Al-Fe-V system, 2: 184 AI-Ir, 3: 57 Al-La system, 1: 680,738 AI-Li system, 1: 47, 48 AI-M-Ni system, 2: 9, 35 Al-Nb-Ni system, 2: 651 Al-Nb system, 1: 994; 2 353 Al-Nb-Ti system, 1:863; 2: 93 AI-Ni system, 1: 114, 982; 2 5, 55, 557, 607 AI-Ru,3: 57 AI-Ti, 3: 618 AI-Ti system, 1: 113, 638-639, 992; 2 75 Al-Zr system, 2 135 As-Pt system, 2 524 Au-Cu-Ni system, 2: 571 Au-Cu system, 2: 561, 563 Au-Hg system, 2: 516 Au-Si system, 1: 691 Au-Ti system, 2: 569 Ba-Hg, 3: 23 Bi-Te system, 2: 462 B-MO-Si, 3: 487 calculation of, 3: 185-208 basic principles, 3: 186-202 classical approach, 3: 186-187 cluster approach, 3: 191-194 cluster-expansionfree energy, 3: 1 9 4 196 electronic structure calculations. 3 196-197 example, 3 202-205 ground state analysis, 3 197-199 mean-field approach, 3: 187-191 static displacive interactions, 3: 199201 thermal effects, 3 201-202 Cd-Hg system, 2: 652 Cd-Ni system. 2 520 Cd-Fe system, 2: 648 Cd-Sb, 3: 155,160 classification of. 3 185, 844 CO-MO-Si system, 2: 596 computation with empmcal interactions, 1: 39, 47 from thermodynamic properties, 1: 97-100 CO-U system, 1: 738 Cr-Fe system, 1: 98 Cu-Hg system. 1: 633 Cu-Sb-Sn system, 2: 592 Cu-Sn system, 2 594 Cu-Zn system, 1: 44, 45, 101; 2: 536 data sources, 3 800-803 effectsof pressure, 3: 154-155, 159-160 eutectic-type topology, 3: 187, 843.844
Fe-Gd system, 1: 738 F e N i system, 1: 634 Fe-Tb system, 2: 396 F e Z r amorphous system, 3: 182 first-principles calculations, 3: 185, 192 example, 3: 202-205 first used, 1: 5 fitted diagrams, 3: 185 construction of, 3: 191 Ga-Nb system, 2: 353 Ga-V system, 2 354 Ge-Hg, 3 24 Ge-Nb system, 2: 353 glass-forming systems, 1: 680.691, 737, 738, 750 graphical representations, 3: 154-1 55, 155. 191
Hg-I, 3 ’ 25 Hg-Ni, 3: 25 HE-TI, 3 24 history of, 1: 5, 6 on Internet, 3 871 Laves phases, 1: 107 Li-Pb system, 1: 666 liquids studied, 1: 663, 664, 666 magnetic, 3: 165-183 magnetic structures, 1: 447 meaning of term, 3: 185 Mn-Si system, 2: 468 Nb( + Cr)-Ti,AIMo, 3: 845 Nb-Si-Ti. 3: 549 Nb-Sn system, 2: 355 Ni-Ti system, 2 537 Ni-Zr system, 1: 738, 750 oxide phase maps superimposed, 1: 982, 992 Pb-Te system, 2 465 Pd-Si system, 1: 691, 738 peritectic-type topology, 3: 187 precipitation reactions. 1: 859, 860 prototype diagrams, 3: 185 construction of, 3: 191 pseudo binary systems, 1: 711-730 AIFe,-Fe,Si, 2: 218 MoSi&,Ti, 2: 219 pseudo ternary systems. 1: 71 1-730 see also Quasibinary/quasiternary systems P-Tphase diagrams, 3: 155, 159, I62 P-T-Xphase diagrams, 3: 155, 159 isobanc cross-sections, 3: 155, 160 rare-earth compounds, 3: 85,86 Sb-Sn system, 2: 592 schematic diagrams, 3: 187 schematic example, 1: 6 Si-Ti. 3: 733 Si-T(system, 1: 738 Si-Zr, 3: 737 software for calculations, 3: 187, 191, 206, 803 Ta-Ti,AIMo, 3: 846 Ta-(Ti,Zr),Al(Mo,Nb), 3 846 tie-lines in, 3: 187 water, 1: 702 Phase distribution, in Ti-AI alloys, 3: 632 Phase enthalpy, 1:920 and Young’s modulus, 1: 920 Phase equilibria AI-Fe-Si system, 2 217, 218 AI-Fe systems, 2 183-184, 190&200 high-temperature intermetallics, 2: 243
999 titanium aluminides, 2: 75, 95,96 Phase eauilibrium calculations. aluminum-transition metal systems, 2: 192-193 Phase formation mechatnsms, in combustion synthesis, 3: 732-735 Phase formation sequence, prediction of, 2: 606, 608 Phase instability, and creep resistance, 3: 308 Phases classification of, 1:11 decomposition of, 2: 608 pressure effects on boundaries, 1: 688 see alro under names of individual phase types (Laves, Heusler. Zintl, etc.) Phase separation, in bulk metallic glasses, 3 687 Phase stability, 1: 11 of aluminides, 2: 54-55, 74, 199 electronic theories, 1: 21-51 in niobium-silicide composites, 3: 548550 predicting, 3: 775 of superalloys, 2: 9.74 thermodynamic considerations, 1: 100 of tnaluminides, 2 155-157 Phase transformations quasicrystals, 1: 465 at surfaces, 1: 618-619 experimental evidence, 1: 618 theoretical studies, 1:618-619 btanium aluminides, 2: 96-101 and twinmng, 3: 410 see also Martensitic . .;Quenching .; Transformahons Phase transitions kinetics, 3: 159-160 at surfaces, 3: 221-223. 228 see also Martensitic transformation Phason-based transformations (of quasicrystals), 1: 465 Phason stram, 1: 461 in quasicrystals, 1: 463, 477 in tilings. 1: 482 Phason stram fields, in plastically deformed quasicrystals. 3: 400 Phenomenologcal relative ordering number, 1: 425,431 see also Mendeleev number; Pettifor . . . maps Phenomenological theory of martensite crystallography, 1: 828-829 Phillips--Van Vechten plots. 1: 357-358, 419,422,424 see also Structure maps Phonon anomalies, 1: 153, 157 Phonon behavior, 1: 1018-1019 Phonon density-of-states, 1: 161 Phonon dispersion, 1: 135, 149 relation measurements, 1: 150-151 Phonon dispersion curves, 1: 149-164, 1018-1019 ob initio calculations, 1:156157 disorder effects, 1: 159-162 expenmental determination, 1: 149, 15&151 and interatomc forces, 1: 151-156 covalent compounds, 1: 152-153
1000 fundamentals of model analysis, 1: 151 ionic compounds, 1: 15I -1 52 metallic compounds, 1: 153-156 Phonon dispersion relation, 3: 258, 260 Phonon-electron coupling, 1: 153, 157-1 59, 952-953 Phonon-phason coupling icosahedral quasicrystals, 3: 391 22) quasicrystals, 3: 390 Phonon-phonon interactions, 1: 1027-1 028 Phonon resistivity, 1: 944 Phonon softening mode, 1: 163, 164, 465 Phonon stress tensor, 3: 386 Photoabsorption coefficient, 3: 145 Photoabsorption cross-section, 3: 145 switches, 2: 421 Photodetectors, 2: 328, 340, 342, 418-421 Photodiodes, 2: 339-342, 418, 419 Photoelastic modulators, 2: 416, 417 Photoelectrocheniical cells, 2: 51 1 Photoelectrons, 3: 137 Photoelectron spectroscopy, of hydrides, 2: 479,481 Photoemission angle-resolved, 3: 137 integrated, 3: 137 inverse, 3: 149 Photoemission spectroscopy (PES), 1: 135, 136; 3: 136-140, 150 core-levef spectra, 3: 138, 239 and XES, 3: 144-145 see also Angl~~resolved p~otoe~nission spectroscopy Photolunnesceiice, 2: 424, 425 Photoluminescence spectra, 2: 425, 426 3: Photon a b s ~ ~ t ~and o n pseudogap, , 234 Photon-electron mteractions,
Phototransistors, 2: 430, 431 Photovoltaic cells, 2: 418, 423 see also Solar cells Physical properties Al-Fe-Si-V alloys, 2: 184-185 effect of crystal structures, 1: 264-268 higli-temperature interme~allics,2: shape-memory alloys, 2: 538 silicides, 2 213 titanium aluminide, 2: 73, 74 zirconium aluminide, 2: 134, 136 see also named property (resistivity, density, elastic moduli, etc.) Photovoltaic detector, 3: 32 Photoyield current, 3: 145 Physical-chemical system, 3: 836 classification of, 3: 819 Physical properties, data sources, 3: 803804 Physical vapor deposition (PVD), 1: 651-652; 2: 492-493, 628; 3 664-667 advantages/disadvantages, 1: 652 see also ~vaporation;Sputtering
Subject Index Physicist’s view of color, 3: 232-233 Physicochemical analysis, 1: 7, 8, 71 1 Physics, meta-sites on Interiiet, 3: 865 Piezoelectrically determined crystal polarity, 1: 178 Piezoelectric coefficient, experimental determination, 1: €78 [footnote] Piezornagnetic strain coefficients, 2: 397 Piezoresistive devices, 2 345 Pile-up mechanisms (for twinning), 3: 422 Pilling-Bedworth ratio (PBR), 3: 710, 721 Pinning, locking of screw dislocations by, 1: 547-548 Pinning centers, 2: 353 Pipe couplings, 2: 538-540 Pi ( E ) pbase alloys, 1: 704, 705 Pi phases, 3: xl Pitting, 1: 966, 969 Planar-defect energies, ab irzitio calculations, 1: 71-72 Planar faults in AINi,, 2: 19-22 in Al,Ti, 2: 149 in AlZr,, 2: 140 Planar flow casting, 1: 733 Planar opto-electronic integrated devices, 2: 347 Plasma arc melting (PAM), 3: 592, 619 Plasma-assisted doping, 2 326 Plasma-enhanced chemical vapor deposition (PEGVD), 3: 667, 668 Plasma-melted induction-guiding gas atomization (PIGA), 3: 649 Plasma melting technique, 1: 640 Plasma rotating electrode process (PREP), 3: 644, 648 compared with gas atomization, 3: 649, 650 Plasma spraying, 1: 642; 2: 492, 493-494 beryllium and beryllides, 3: 47 structural alloys, 3: 507 see also Low-pressure plasma spraying Plasinon peaks, 3: 148 Plastic deformation, 1: 91 1-928 controlling factors, 1: 919-928 bond characteristics, 1: 919-921 crystal symmetry, 1: 921-922 environinental factors, 1: 927-928 microstructural features, 1: 924-927 point defects, 1: 580 stoichiometric composition, 1: 922-924 and deformation twinning, 3: 400, 427 and dislocations in quasicrystals, 3 397-401 high-temperature, in quasicrystals, 3: 400-40 1 mechanisms, 1: 913-9 I8 diffusional creep, 1: 9 17-9 18 dislocation creep, 1: 916-917 slip, 1: 913-915 twinning, 1: 9 15-91 6 of metallic glasses, 3: 694 homogeneous flow, 3 695-696 in~omogeneoLisflow, 3: 694-695 of trialuminides, 2: 166 Plasticity, and twinning, 3: 425-426 Plastic spikes, 1: 803 PlatigemTM,3 75-76 Platinum demand and price fluctuations, 3: 79
melting temperature, 3: 70 recovery of, 2 518 separation from gold, 2: 517-518 world deposits/reserves, 3: 80 Platinum~aluminidecoatings, 3: 567 Platinum amalgams, 3: 21n Platinum-based magnets, 2: 308-309 Platinum-base superalloys, 3: 70, 72-73 short-range order in, 3: 73-74 Platinum group aluminides, 3: 55-59 in colored jewelry, 3: 75 composite microstructures, 3: 61 effect of boron, 3: 59 high-temperature performa~ce,3: 6162 productioii of, 3: 60-61 ternary alloying, 3: 59-60 Platinum group, 3: metals as alloying additions to aluminides, 3: 62, 64 to ferritic alloys, 3: 64-65 to nickel-base superalloys, 3: 65-66 as coatings, 3: 584 demand fluctuations, 3: 79 factors affecting price/supply, 3: 53-54, 77, 79-80 in Ni-MH batteries, 3: 108 price fluctuations, 3: 79 in sigma phase formation, 3 62 world deposits/reserves, 3: 80 see aZso Iridium; Osmium; Palladium; P~at~num; ~henium~ ; utheniu~n P~at~num-modified aluminide coatings, 2 491, 495, 496, 497; 3: 64, 65, 72, 74, 564, 565, 567, 568 diffusion~ltransport in, 3: 568-570 Plumbopalladinite [mineral name], 1: 630, 632 Plutonium IMGs, 3: 800 Pnictides, 1: 154; 3: xl Pockels effect, 2: 416 Point charge model (PCM) calculations, 3: 171, 179 Point defect energies, 3: 769 in Mo-Si system, 3: 776 in TiAI, 3: 773, 774 Point defects, 1: 559-581 in A15 compounds, 1: 578-579 accumu~ationduring irradiation, 1: 817 in B2 compounds, 3: 354 in C15 compounds, 3: 354 clustering of, 1: 564 concentration required for amorphization, 1: 818-819 and dislocations, 1: 580 and electrical resistivity, 1: 577 experimental investigations, 1: 560-564, 793-797 formation propert~es computation, 1: 569-570 experimental determination, 1: 560-562 theory, 1: 568-570 identification of, 3: 358-359 in iron aluminides, 2 199-200 and lattice parameter, 1: 561, 565 low-temperature retention, 1: 573 migration properties, 1: 574-578 experimental d e t e ~ i n ~ t i o1: n, 562-563 and ordering energies, 1: 564
Szabject hdex physical properties affected by, 1: 580 production of, 1: 560 recovery of, 1: 564 after irradiation or quench, 1: 573-5 74 in semiconductors, 1: 579-580; 2: 327 theoretical modelling, 1: 569-570, 797-799 thermodyna~~i~s, 1: 570-573 in transition-~etalcarbides and nitrides, 1: 579 triple-defect model, 1: 559, 566, 572-573, 575 and twinning, 3: 422-424 typical examples, 1: 559 see also Antisite . , .: Interstitial . . .; Vacancy defects Point groups (cr~stallographic), 310 cubic system, 1: 312 meaning of term, 1: 30911 orthorhombic system, 1: 311 ~ e t ~ a g o nsystem, a~ 1: 311 trigonal system, 1: 312 Point-group symmetry, 1: 294 Point sets, 1: 255-256, 262, 373 Poisson’s ratio, 1: 202, 876; of alloys, 1: 887 calculated values, 1: 206 coniposition dependence, 1: 887 of nickel aluminide, 2; 56 orientation dependence, 1: 877, 878, 880 of polycrystals, 1: 885 estimation from s~ngle-crystal constants, 1: 884, 885 of silicides, 2: 213
Polarization curves, 1: 967, 968-969, 972 P o ~ ~ r i z a tmodulators, io~ 2: 416, 438 Polarization rotation, 2: 416-417 Polar Kerr ellipticity, 2: 437,443,444,445 Polar Kerr rotation, d: 437,443,444,445, 447, 448 Polarography, 3: 22 Pole mechanisms (for twinning), 3: 422 Polk model, 1: 741-742 Il”olyanionic-valence compounds, 1: 346 ~olycarbonate,ion tracks in, 3: 264 Poly~ationic-valencecompounds, 1: 346 Polycrystalline AlNi, 2: 58, 60, 61 Polycrystalline AlNi,, 2: 40-41 Polycrystalline inteimetallics, strain hardening in, 3: 368-373 Polyionic structure compounds, meaning of term, 3: XI Polynierically bonded magnets, 2: 317 Polymorphic crystallization~1: 746-747 P o l ~ o r p h i cmodi~cations~ crystal structure affected by, 1: 260-261 Polymorphi~transfos~ations,and heat capacity, 1: 110, 118-119 Polymor~hism,1: 41 5 Polysynthetically twinned (PST) crystals creep anisotropy in, 3: 305, 317 fatigue damage in, 3: 326, 327 formation by twinning, 3: 404, 431
strain hardening in, 3: 365, 366 twinning in, 3: 407, 408, 425, 431 Polytope [3.3,5], 1: 473 Polytypes stacking notations used, 1: 345, 358-359 ZnS, 1: 344-345 space groups, 1: 359-360 Pond’s topological theory, 3: 410 Pore-collapse curves, 3: 738 Pore volume fraction, in trialuminide compounds, 2 161 Porosity, and combustion synthesis, 3: 733-734 Porous metallic filters, 3: 643 Positive crystals, 1: 189, 190 removal of impurities, 1: 189, 191 Positron annihilation experiments?3: 277 Positron annihilation spectroscopy (PAS), 1: 41, 135, 136 see also Angular-carrelation-ofannihi~ation-radiation. . Positron annihi~ationtechniques, vacancies studied using, 1: 561-562, 564 Positron-electron annihilation photon line 280, 288 Positron-lifetime spectroscopy, 3: 276277 in high-pressure cell, 3: 286 results, 3 279-283 Positron satu~at~on trapping, 3: 288 Positron trapping, 3: 287-288 Positron trapping rate, 3: 276-277 temperature dependence, 3: 280, 286, 287 time dependence, 3: 282 values, 3: 281, 282 Potarite [mineral name], 1: 644-647, 912; 2 523 for alumin~dealloys, 2: 43-44, 87, 202 techniques consolidation, 1: 644-645 hot isostatic pressing, 1: 645, 912; 2: 359 reactive sintering, 1: 646-647, 912; 2: 44 sintering, 1: 645 Potentials, semi-empirical, 3 765, 766768 Poterite [mineral], 3: 21n Powder metallurgy, 3: 643-662 advantages, 3; 643 applications amalgams, 3: 29-30, 643 beryllides, 3: 46-47 iron aluminides, 3: 654-655 nickel aluminides, 3: 655-657 p~atinumgroup aluminides, 3: 61 rare-earth metal magnets, 3: 97, 101102 refractory metal silicides, 3: 657-658 structural materials, 3: 475, 504-506, 643 y-TiAl, 3: 648-653 powder billets, deformation processing of, 3: 646-648 powder compact density, 3: 647, 648 powder coiiipressibility, 3: 726 powder consolidation, 3: 645-646 powder extrusion, 3: 646 powder forging, 3: 646
1001 forming by, 3: 633 limitations, 3: 643 powder pressing, 3: 645-646 powder processing of n i o b i ~ silicide composites, 3: 658 reactive, 3: 653 powder rolling, 3: 646 Powders gas-atomized, and oxygen, 3: 649 heats of reaction, 3: 7#3 hot pressing of, 3: 646 isothermal compression of, 3 650 reactivity in shock ~ompression,3 743 shock-consolidation of, 3: 725-727, 738-740 shock response of, 3: 737-738 synthesis of by atomi~ation,k 644 by inechanical alloying, 3: 645, 652653, 753-754 by milling, 3: 644-645, 652, 727-730 by reaction synthesis, 3: 645, 652653, 732-737 by shock compression, 3: 740-743 work-hardened, 3: 645 Powder synthesis, 3: 644-645 Power-generation equipment, life-cycle cost, 3: 472 Power-law creep, 1: 916, 923; 2: 117 Power-trans~issionlines, 2: 380-381, 382 Precious-meta~alloys, 2: 65 I recovery processes, 1: 4; 2: 515-516 see also Gold; Palladium; Platinum; Silver Precious-metal compounds, 3: 53-84 applications, 3: 61-77 €32 conipounds, 3: 55-62 colored compounds, 3: 75 composite structures~3: 61 hig~-temperaturepe~formance,3: 6162 production of, 3: 60-61 shape deformation in, 3: 56 ternary alloying, 3: 59-60 Precious metal modified coatings, 3: 567568 Precious metals as alloying additions, 3: 62-66 availability and supply, 3: 77, 79-80 see also Gold; Platinum group metals; Silver ~recipitateshape stability, 1: 867 Precipitation, 1: 859-861 Precipitation hardening, 2: 162, 165, 569, 594-595 of AN-based alloys, 3: 3 16-3 17 in iridium-base superalloys, 3: 68 in nickel-base superalloys, 3 3 13-3 16 Precipitation stsengthenin~,2: 257 in Al-Ni alloys, 2: 271-273 alu~inum-lithium alloys, 3: 202 in ferritk alloys, 3: 64-65 mechanisi~s,2: 266-269 for overaged alloys, 2: 268-269 for underaged alloys, 2: 267-268 nickel-base superalloys, 3: 64 platinum alumini~es,3: 73 Precursor effects, 1: 149 Prediction of AB& compounds, 3: 825,826
1002 of AB&, compounds with ThCr,Si, structure, 3: 828-829 Of AB.&, COliipOUiidS, 3: 825--827, 828 accuracy of, 3: 835 classification rules, 3: 8 12-8 13 compared to new data, 3: 833 empirical criterialrules, 3: 8 1I , 812 of Heusler alloys, 3: 831-832 of Neiisler-like compounds with cornposition ABCo,, 3: 832 of Heusler-like compounds with composition ABCu,, 3: 832 of IMCs, 3 823-832 meaning of tern, 3: 836 of phases with Al, Ga and In, 3: 829830, 831 quantum-mechanical approach, 3: 81 1, 812 Preoxidation, in pack mixtures, 3: 578 Pressure ~xperiiiien~L1 studies, 3: 154 four-phase equilibriu~iiaffected by, 3: 156-159 phase diagrams affected by, 3: 154-1 55, 159 -160 Pressure-assisted investment casting processes, 3: 592-593 Pressur~ompositionisotherms, 2 476 Pressure effects, 3: 153-16 elastic constants, 1: 890, 891 heavy-femion compounds, 1: 221 Pressure-induced amorphization, 1: 701-703 Pressure quenching, I: 688, 706-707 Pretransitional phenomena, 1: 716 Pressur~tempcrature~omposition phase diagrams, 3: 155, 159 database on, 3: 816 isobaric cross-sections, 3: 155, 160 Pr~ssure-te~perature phase diagrams, 3: 155, 159, 161 Price of IMCs, 3: 794 meaning of term, 3: 793 of precious metals, 3: 65, 79 of zinc, 3: 794 Prunary creep, 1: 917; 2: 119; 3: 299, 302303, 321 in nickel-base superalloys, 3: 302 in titanium aluminides, 3: 302-303 Primary ionization, 3 264 as measure of track damage, 3: 264265, 266 Primary twin system, 3 405 Principal quantum number as atomic property, 1: 231, 241 as indication of metallic bonding, 1: 183, 184 as measure of bond directionality, 1: 238 in structure mapping, 1: 241, 242, 357, 420 Printed-circuit board manufacture, 2 507 Prismatic loop debris, 3: 365 Process control agent @CA), m niechanical alloying, 3: 756 Process efficiency maps, AlTi, alloys, 2: 125 Processing, 1: 637-657, 912 of amalgams, 3: 29
Subject Index of dental amalgam silver-tin alloy powder, 2: 579-580 of diffusion barriers, 2: 628-629 ductility affected by, 1: 898 forming techniques, 1: 656-657; 2 43-44, 53, 67-69, 136 of iron aluminides, 2 201-202 joining techniques, 1: 655-656; 2: 44-45, 128 melting methods ingot production. 1: 639-640; 2: 161, 175, 176 single-crystal growth, 2: 395 i~icrostructuresarising, 2 126 of nickel aluminides, 2: 43-45,53,67-69 of niobium silicide composites, 3: 545547 of semiconductors, 2: 326, 327, 328, 329, 347 of silicides, 2: 217 of superconductors, 2: 355, 360-363, 364 of structural materials, 3: 474-475, 503-509 thermomechaiiical operations, 1: 653-655 casting, 2: 83-85 extrusion, 1: 654-4555; 2: 43-44, 67, 153, 161, 202 forging, 1: 653-654; 2 86, 127, 161 rolling, 2: 655 wrought processing, 2: 85-86 of titaniun~aluminides, 2 83-87, 125-1 28 of trialuminides, 2: 161 see also Synthesis techniques Process metallurgy applications, 2: 515-526 Processor, conversational, 3: 818, 819, 820 Production costs, 3 477 Program schedmler, 3: 818 Properties of materials on Internet, 3 872 searching for, 3: 81 1 see also Elastic.. .; Mechanical.. .; Optical properties Property-coniposition curves, first used, 1: 7, 8 Property map, 3: 823 Proteus@Safety Link, 2: 545, 548 Proton difhsion, in hydrogen-bonded crystals, 3: 257 Proton irradiation, amorphization caused by, 1: 817 Prototype structures, 3 11 structural notations for, 3: xxxi-xxxiv, 13-15 PrPdSi structure type, 3: 91, 91, 93 Pr-Tb system, 1: 407 Pseudo-atom concept, 3: 114 Pseudobinary compounds, 1: 305; 2: 392, 394 structure mapping of, 1: 243, 431 Pseudobinary phase diagrams AlFe,-Fe&, 2: 218 MoSi,-Si,Ti, 2 219 see also Quasibinary/qu~sit~rnary systems Pseudo-elastic behavior, 3: 404 see also Shape memory effect
Pseudoelasticity, 1: 834 Pseudogap, 1: 220, 671, 948; 3: 139, 234, 236 see also Quasigap Pseudopotential methods, 1: 134, 319, 945; 3 196 Pseudopotential radii, 1: 23 1 232 listed for various elements, 1: 233, 423 r various plot vs atomic n u ~ b e for elements, 1: 229, 230 in structure mapping, 1: 242-243, 262, 422 Pseudopotentials, 1: 59, 134 Pseudostab~eordering states, 1: 856 Pseudosymmetrical [atomic] configurations, 1: 598, 599 Pseudoternary systems, 1: 71 1-730 see also Quasibinarylquasiterna~y systems Pseudo-twinning, 3: 404, 405, 406, 410, 431 P,Th, structure type, 1: 385, 399-400 Pt-Rh-Ti system, 1: 724 Pt-Sb system, 1: 632, 735 Pt-Sn naturally occurring compounds, 1: 632 Pt-Ti system, 3: 70 Pt-Zr systems, 1: 740; 3: 70, 72 Puckered hexagonal structures, 3: 6 PuGa, structure type, 3: 88, 89 Pugh [brittleness] critenon, 2: 166167; 3: 240 typical values, 3: 241 Pulsed-electrodischarge,3: 723 Purchasing of IMCs, 3 51 1 ‘Purple glory’, 3: 75, 76 ‘Purple plague’, 2: 410, 572, 654; 3: 74-75 PWA superalloys, 2 6 PWA-1484 superalloy, creep behavior, 3: 489 Pyramidal network, 3: 819, 822 Pyrite, structural forniula for, 3: 4, 5 Pyrites [mineral name], 1: 428 Pyroelectric detectors, 2 419 Pyrometal~~rgical recovery processes, 2: 5 15-5 18 Pyro-optical sensors, 2: 41 7 Pyrophoric alloys, 3: 794 Pyrophoric compounds, 2: 647 QE22A alloy, 2 651 Qualitative property, 3: 836 prediction of, 3: 834 QLiantitative property, 3: 836 prediction of, 3: 834 Quantum boxes, 2 425 Quantum-con~nedStark effect, 2 342 Quantum dots, 2 425 Quantuin mechanical calculations, 1: 21. 127; 3: 812 boron bonding, 1: 907-908 electrical conductivity, 1: 943 magneto-op~icaleffects, 2: 439-441 phase-formation enthalpies, 1: 920 Quantum mechanical interactions, 3: I53 ~ u a n t u mmechanics and color theory, 3: 233-234 prediction using, 3 811, 812, 841 Quantum number as indication of inetallic bonding, 1: 183, 184
Subject Index as measure o f bond directionality, 1: 238 see also Principal quantum number Quantum structure diagrams. 1: 195, 265, 266, 267, 382 Quantum wells, 3: 787 Quaiituin wells [in semiconductors], 2: 327, 328,424-426 Quantum wires, 2 425 Quasiatom method, 1: 78 Quasibinary/quasiternary systems, 1: 107, 392, 711-728 I32 type compounds, 1: 713-719 DI, type compounds, 1: 727-728 definitions, 1: 711 L1, type compounds, 1: 720-724 L1, type compounds, 1: 724-727 Quasibinary/quasiternary systems, A N BeNi system, 3: 49 Quasicrystals, 1: 71, 453-486 atoinic structures, 1: 483-484 commercial app~ications,1: 485-486 crystal approximants, 1: 466, 471-478 decagonal calculation of elastic displacement field around dislocation, 3: 395397 characteristics, 3: 389 elastic constants, 3: 395 plastic deformation of, 3: 399, 400 decagonal phase, 1: 453, 4654.71; 2: 179, I85 atomic structures, 1: 483-484 classification based on periodicity, 1: 467 crystal approximants, 1: 477478,483 diffraction-pattern indexing, 1: 470-47 1 diffraction patterns, 1: 467-470 general metallurgy, 1: 465467 deformed, stacking faults in, 3: 400 dislocation density in, 3: 399-400 dislocations in, 1: 463, 537, 538 dislocation mechantsm of deformation, 3: 400-401 dislocation motion in, 3: 399 dislocation reactions in, 3: 398-399 dislocations in, 3: 379-402 Burgers vectors for, 3: 381-385, 385 and dislocation arrangements, 3: 399 and displacement field, 3: 381 elastic fields around, 3: 389-397 and plastic deformation, 3: 397-401 Volterra process for, 3: 379-381 disorder softening of, 3: 401 elastic constants, 3: 387-388, 390, 391 elasticity theory, 3: 385-389 general equations, 3: 385-387 for IQCs, 3: 388-389 electronic properties, 1: 484-485 growth morphologies, 1: 457, 467, 468 high-temperature plastic deformation of, dislocation mechanism for, 3: 400-401 icosahedral dislocations in, 3: 379, 380, 382-383 elastic constants, 3: 390 elastic equations, 3: 388-389 inechariical properties, 3: 397-398 plastic deformation of, 3: 399-400
synthesis by mechanical alloying, 3: 755 icosahedral phase, 1: 453, 454-465; 2: 180, 185 atomic structures, 1: 483 crystalline phases related, 1: 472-475 diffraction-pattern indexing, 1: 460-46 1 diffraction patterns, 1: 456, 457 general metallurgy, 1: 454, 455, 457 phase transformations, 1: 465 projection from higher-d~mensional space, 1: 457, 459 rational crystal approximants, 1: 476-477 structural defects, 1: 461-465 on Internet, 3: 873 meaning of term, 3: xli mechanical properties, 3: 397-398 naiiostructured IMCs, 3: 755 octagonal, elastic constants, 3: 387388,389 octagonal phase, 1: 453 phonon dispersion effects, 1: 161-162 plastically deformed, straiii fields in, 3: 400 plastic deformation of, 3 397-401 precious metal compounds, 3: 78 prediction of formation, 1: 485 quasilattice models, I: 480-482 real-space structures, 1: 480-484 strain fields in, 3: 400 structural similarities to liquids and glasses, 1: 479-480 and structure maps, 1: 265, 267, 435 surface structure, 3: 216 t-phase, 1: 465 see also Quasicrystals, decagonal phase Quasielastic Mossbauer spectroscopy, 3: 290 Quasielastic neutron scattering (QENS), 3: 249-250 applications, 3: 252, 252, 253, 254, 255, 257. 258,259 Quasigap, I: 57 in Fermi level, 1: 58, 63 see also Pseudogap Quasigap effect, 1: 57, 65, 66, 67 Quasilattice, 1: 453, 480 Qudsilattice models complex formation in liquids, 1: 673-675 concentration ~ u c t u ~ t i o n s calculated, 1: 675 limiting cases, 1: 674-675 quasicrystals, 1: 480-482 Quasiyarticles, 1: 128; 3: 46 Quasiperiodic lattice, 3: 379 Quasiternary systems, 1: 71 1-728 see nlso Quasibinary/quasiternary systems Quasitriangular diagram, covaI~i~cy-ionicity-average~quan t umnumber plot for tetrahedral compounds, 1: 184 Quasi-twinning, 3: 410, 431 Quasizones. 1: 485 Quaternary alloying, silicides affected by, 2: 215-216
1003 Quaternary compounds, close-packed structures, 1: 305- 306 Quaternary defect adaniantane-structur~ types, 1: 356 Quaternary normal adamantane~ structure types, 1: 356 Quenched-in vacancies, 3: 275 Quenching amorphous phases produced by, 1: 689-692, 733 disordered states, 1: 772-773 see aZso Liquid . .; Pressure quenching QLienching transfo~ations,in titanium aluminides, 2: 96-97 R phases, 1: 475; 3: xli R ratio, in fatigue behavior, 3: 325, 336, 347 Radiant burner tubes, 3: 510, 512 Radiation damage, 1: 649. 791-822 Radiation detectors, 2: 232-233, 326,419, 457 Radiation-enhanced order relaxation, 1: 784 Radiation-iiiduced ductility (RID), 3: 423, 430 Radical distribution f~nctions,1: 739 Radio-frequency cavities, 2 376-377 Radiofrequeiicy (RF) sputtering, 3: 665 Rndiopolarography, 3: 30 Raman spectroscopy, 3: 116, 572 Ramsdell notation, 1: 176, 345, 358, 359, 360; 2: 54, 532 Raiiey nickel, 1: 15; 2: 647 Raoult’s law, deviations from, 3: 23, 24 Rao-Woodward-~arthasarathy (RWP) interatomic potential, 3: 769, 770, 772 Rao-~Oodwdrd-Simm~ns-~imiduk (RWSD) interatomic potential, 3: 772, 773 Rapidly quenched glasses, 3: 681 Rapid-solidification methods Al-Fe-based alloys, 2 178, 202 AlTi-based alloys, 2: 87 amorphous phases, 1: 640-642, 689-692, 733 advantages, 1: 641 disad~antages,1: 641-642 magnetic materials, 1: 748-749; 2: 314 Rapid solidification processing (RSP), 3: 751, 755 Rare-earth compounds, 1: 215; 3; 85-1 I1 applications, 3: 97 arsenides, 3: 784, 786 beiyllides, 3: 46 binary compounds, 3: 87-90 RX compounds, 3: 89,89 RX2 c o ~ p o u n ~3:s ,88-89, 89 RX, compounds, 3: 88, 89 R5X1compounds, 3: 89-90,89 crystal structures, 3: 87-96 data sources, 3: 806-807 energy-storage applications, 3: 107-108 ferromagnetic compounds, 3: 99 frequent compositions, 3: 88, 90 hexagonal structures, 3: 94 hydrogen uptake by, 3: 100 interstitial compounds, 3: 98, 99, 178-180 magnetic phase diagrams, 3: 179-180
1004 magnetic phase diagrams, 3: 173-183 CeSb, 3: 180-181 interstitial compounds, 3: 179-1 80 La(T,Mj,, compounds, 3: 174 R,Fe,,B compounds, 3: 173-174, 175 R(Fe,T),, compounds, 3: 174, 176, 177 R,(Fe,T),, compounds, 3: 174 WT,X, compounds, 3: 174, 176178 magnetic properties, 3: 97-101 R,Fe,,B compounds, 3: 97-98 R(Fe,N),, phases, 3: 99 R2(Fe,Mj,7phases, 3 98 R3(Fe,M)29phases, 3 99 R,Fe,,.M, phases, 3: 100-101 RT, .M, phases, 3: 100 magnetic structures, 1: 447449, 935, 1028 ~ a ~ e t o c a l oproperties, r~c 3: 529-535 magnetocrystalline anisotropy, 3: 170 magneto-optical applications, 2: 446, 447, 450 magnetoresistance, 3: 104 ~agnetostrictionin, 2: 392-393 metamagnetic transitions, 3: 183 permanent-magnet applications, 3: 101-103, 165 in p e ~ a n e n m t a ~ e t s 3: , 97, 101 phase diagrams, 3: 85, 86 physical properties, 3: 96108 preparation of, 3: 98 single-crystal growth, 3: 103 stability domains, 3: 88-89 structural types, 3: 88-90 superconducting borocarbides, 3: 105 superconduct~vity,3: 103-104 technological applications, 3: 97 ternary compounds, 3: 90- 96 RTX compounds, 3: 90, 91, 93 RTX2 compounds, 3: 90, 92 RT&, compounds, 3 90, 92 thermal stability, 3: 85-86 thermoelectric properties, 3: 106-107 volume effects, 3: 86-87 Rare-earth compounds, 3 temperature dependence of elastic properties, 1: 889 thermal conductivity, 1: 1028 Rare-earth elements magnetocaloric properties, 3: 526-527, 528,529 miienionic for, 3: 108 sequence, 3: 108 see also Lanthanides Rare-earth metals atomic properties listed, 1: 233, 423 magnetic structures, 1: 446-447 see also Lanthanides Rare-earth permanent magnets, 2: 303-304, 316-317, 478 Rarc-earth-traiisition-metal borocarbides, 3: 103 Rb-Sb system, 3: 158 Rational approxima~ts[in quasicrystals], 1: 459,465, 466,476477 Reaction layer formation, 1: 696, 861 Reaction mechanisms in mechanical a ~ ~ o y i3:~ g723, , 728-730 in reaction synthesis, 3: 723, 726 Reaction synthesis, 3: 645 advantages, 3: 645
Subject Index electric-field-assisted, 3: 723 of iron aluminides, 3: 654 of nickel aluminides, 3: 656, 657, 737 shock-induced, 3: 723, 737-743 solid-state, 3: 645, 724-727 of titanium aluminide, 3: 653 Reaction temperatures amorphous alloys with aluminum, 2 625 with silicon, 2: 618-620 transition metals with silicon, 2: 616-618 Reactive consolidation/sintering, 1: 646-647, 912; 2: 44 Reactive diffusion, 3: 673-676 advantages, 3: 673 disadvantages, 3: 674 Reactive diffusion, growth rates in, 1: 768 Reactive hot isostak pressing (RHIP), 3: 61, 63, 653 Reactive sputtering, 3 665 Rearrangement of atoms, ~ e c h a n i s ~ s for, 1: 803 Rechargeable batteries, 2: 475, 486, 510-51 1 Recoil (neutron) scattenng, 3: 249 Re~ombinationmechanisms, 1: 803 Recombination structures, 3: 6-7 Reconstruction of surfaces, 1: 183, 6 19-620 Recovery, dunng high-temperature deformation, 3: 617, 622 Recovery processes? 1: 4; 2: 515-518 Recrystallization, 1: 654, 863 Al-Ti alloys effect of A1 content, 3: 624-625 efiect of B content, 3: 625 in hot working, 3: 623-626 Recrystallization twins, 3: 410, 431, 622 Recursion method, 1: 81 Recycle program, 3: 5 15-5 16 Recycling, of mercury, 3: 33 Red mercury, 3: 32-33 Red shift, 2: 332 Reduction in area AITi,-based alloys, 2: 113 AlZr,, 2: 140, 142, 144 see also Ductility Reduction reactions, ama~gamsused, 3: 30-3 1 Reduction techniques, synthesis using, 2: 650-651 Reed contacts, 2: 653 Re-entrant spin glass, 3: 182 Reflectance, as function of incidence angle, 2: 409 Reflection high-energy electron difiraction (RHEED), 3: 782 applications, 3 782, 783, 785 Reflectivity, 3: 234 and color, 3: 232, 233 of disilicides, 3 233, 235 of gold, 3 233 Reflectors, 2: 408, 409-410 Refractive index, 2: 437 Refractory materials, beryllides, 3 48 Refractory metal interme~alliccomposites (RMICs), 3: 545 B2 phases, 3: 847-851 mechanical properties, 3: 550-555
microstructure and phase stability, 3: 547-550 oxidation behavior, 3: 556-558 physical properties, 3: 555 processing of, 3: 545-547 thermal properties, 3: 555-556 Refractory-metal silicides, 2: 2 19-227, 329 Refractory silicides fatigue properties, 3: 342, 344-346 high-temperature properties, 3: 485487 oxidation resistance, 3: 486, 657 powder metallurgy applications, 3: 657-658 Refractory superalloys, 3: 67-74 i ~ d i u ~ - b a s e 3: d , 67--69, 71 platinum~based~ 3: 70, 72-73 short-range order in, 3: 73-74 rhodium-based, 3: 69-70 Refrigeration and active ~ a g n e t i cregenerator, 3: 522-524 by adiabatic demagnetization, 3: 521522 Refrigerator applications, 3: 48-49, 519539 Refrigerators, 2: 455, 457-458, 460, 484, 486, 654 Regular-associated-solution model (for liquids), 1: 675-676, 750 Reg~larities(with crystal structures), 1: 236-264 Regularities, o f uiikiiowii compounds, 3: 826 Reinforcement phases [in intermetallic composites], oxidation behavior affected by 1: 1006, 1007 Reinforcements continuo~s,2: 287, 290 discontinuous, 2 289-290 Reinforcing phases, beryllides as, 3: 49 Reinsch test, 2: 647 Relative cooling power, 3: 525 based on magnetic entropy change, 3: 527, 528 based on temperature change, 3: 527528 Relaxation behavior amorphous phases, 1: 742-745 ordered structures, 1: 775-776 surfaces, 1: 612, 616-618 Relaxation methods hydrogen diflusion studied by, 2: 484 stepwise relaxation method, 1: 86 Relaxatioii rates, in study of LRO changes, 1: 778 Relaxation trajectories, in SRO, 1: 783 Relaxation volumes, interstitial defects, 1: 567 Reliability considerations, 3: 483485 ~emanence[for p e ~ a n e n magnets], t 2 306, 319 Remnant vacancies, 3: 275 clustering of, 3: 286 Remote ~lasma-enhancedchemical vapor deposition (RPECVD), 3: 667, 668 Renk 80 superalloy, fatigue behavior, 3: 331, 332 R6ne 95 superalloy, temperature dependence of yield stress, 3: 630
Subject Index RenC N6 superalloy, creep behavior, 3: 489 Rene superalloys, 2: 42, 60, 61, 62, 65 Renormalized interactions, 1: 30 Replacement collisions, 1: 803 Replace~entcollision sequences, 1: 784, 795, 796 Replacive energy, 3: 192, 199 Repulsive interactions, close-packed alloys, 1: 282, 285, 295 Research techniques, 3: 72 1-788 Research topics, on Internet. 3: 863 Residual resistance ratio, 1: 1017 Residual resistivity, 1: 944 Resistivity anomaly, 1: 947-948, 949 Resistors, electronic, 2: 469, 470, 652, 653 ReSi, structure type, 1: 388 Re,Si,U,, structure type, 1: 408 Reuss’s approximation, 1: 883, 884 R ~ ( F e , . ~ ~ ~ ) ~phases, - t y p e crystal structure. 3: 93-96 ~henium price (~996),3: 65 production data, 3: 65 Rhenium clusters, 3: 307 Rhodium demand and price fluctuations, 3: 79 world deposits/reserves, 3: 80 Rhodium-base superalloys, 3: 69-70 Rhombic dodecahedron, 1: 238, 239, 404 frequency plot, 1: 240 structure map, 1: 245 Rh-Si system. 1: 696 Rh-Ta alloys, 3 70 Rh-Ti alloys, 3: 70 Rh-Zr system, 1: 740 Rice-Thomson [brittleness] criterion, 2: 167; 3: 241 typical values, 3 241 Rietfeld analysis, 1: 440 Ri~id-bandbehavior, 1: 141 Rigid-ion model, 1: 151-1 52 Ring diffusion mechanism, 1: 742 Rippled surfaces, 3: 220, 228 R(o) structure, 1: 716, 717 Robot arms, 2 556 Roll anisotropy, 2: 306 Roll compaction, 3: 504, 506-507, 508, 655 Rolling order changes induced by, 1: 786-787 synthesis by, 1: 655 of TiAl sheets, 3: 631-633, 651-652 Rotating-electrode process, 1: 640, 641 Rotating machinery, 2: 382-383 Rotor phases, disorder in, 3: 253-255 Rubber-like behavior, 1: 834 Rub~ni-Ballone (RB) interatomic potential, 3 769, 770 Ruderman-Kittel-Kasuy a-Y oshida (RKKKY) interactions, 1: 215, 440441,447,450, 945; 3: 692 Rup~ure~ e n1: ~91 1~ , RustenbLirgite [mineral name], 1: 629, 630, 632 Ruthenium demand fluct~ations,3: 79 price, 3: 65, 79 production data, 3: 65 world deposits/reserves, 3: 80 Ruthenium aluminides, 3: 55-58
ductility, 3: 55-56, 58 effect of boron on mechanical properties, 3: 59 point defects in, 3: 354 production of, 3 60-61 ternary alloys, 3: 59-60 Rutherford backscdtteriiig spectrometry, 2: 625 amorphous alloys, 2: 619, 620 and crystal habits, 1: 183 Rutile [mineral name], 1: 328, 428 Ru-Zr system, 1: 740 RV 8413 alloy, 3: 715
S phase, in AI-Li alloys. 2 188 S (stepped) crystal faces, I: 170 representation in stereographic projection, 1: 169 Safety devices, 2: 545-546 Samson, Sten O., 3 xliv Samson’s giant-cell compounds, 3: xli Sand casting, 3: 607-608, 612-613 Saturation (color), meaning of tern, 3: 242 Saturation magnetic moments, 1: 7 1 Saturation magnetization, 3 702 effects of fission neutrons, 1: 805 values listed, 2: 444, 445, 447, 449 Savitskii’s information-prediction system, 1: 270, 271 Sawtooth [diffraction] gratings, 2: 415-416 Sb-Sn system, 1: 632, 704 Sb,S,, structural foriiiula for, 3 7 Sb,Tl, crystal structure, 1: 322, 325, 326, 327, 341 Sb-Zn system, 3: 158-159, 161 Scald-prevention safety device, 2: 546, 549 Scales adhesion of, 3: 573 failure mechanisms, 3: 564-565 growth mechanisms, 3: 566 on Ni,Al, 3: 712, 713, 714 Scalloped dislocation images, 2: 268 Scandium, recovery of, 2: 518 Scanning probe microscopy (SPM) techmques, 3: 215 Scanning tunneling microscopy (STM), 1: 467,468,612; 3: 211, 212-214,215 examples of app~ications,3: 218, 219, 223 Scattering measurements, order changes studied using, 1: 781 ScAuSi structure type, 3: 91, 91, 93 Schachnerite [mineral name], 1: 629, 630, 632; 3 21 Schottky barrier, 3: 785 Schottky barrier heinhts, silicides, 2 230 Schottky-type devices, 2: 213, 230-23 1, 232, 347 Schwarz-Labusch statistics, 2: 263, 266, 273 SC IN-6-7, creep behavior, 3: 305 Screening approaches, high-temperature intennetallics, 2: 241-252 Screw dislocations, 2: 23, 105 locking of, 1: 547-548 pinning of, 3: 365-366 SCS-6 fiber reinforcement, 3: 337, 338, 582 Sc-Zr system, 1: 740
100.5 Seals, 2: 541 Search engines, 3: 857, 877 directories covering, 3: 865-866 number of hits for ~‘intermetallic’’,3: 858 Searching for properties, 3: 81 1 for ternary and higher-order conipounds, 3: 81 1 Secondary creep rate, high-teinperature materials, 3: 489, 554 Secondary-ion mass spectrometry (STMS) observations, 1: 990; 3: 214, 780 Second moment approximation, I: 80-82 Second-nearest-neighbor interactions, 1: 850, 856 Second-order phase transitions, 3: 222 Seebeck coefficient, rare-earth compounds, 3: 106, 107 Seebeck effect, 2: 454, 457 Segregation in AI-Nb-Ti alloys, 3: 620 in close-packed structures, 1: 282, 284, 287, 289 at grain boundaries of boron, 1: 593, 594, 638, 899, 901, 904 electrical conductivity affected by, 1: 955, 956 liquid alloys, 1: 667 during solidification, 3: 474, 619, 620 surface, 1: 612-613 to antiphase boundaries, 1: 503-504 Seinajokite [mineral naineJ, 1: 630, 632 Selected-area diffraction (SAD) patterns, 3: 417 Selected area electron di~raction(SAED) observations, 2: 150, 151, 565 Selenides, 1: 626 prediction of formation, 3: 826 as thernioelectrics, 2: 469 Self-association, in demixing liquid alloys, 1: 677-678 Self-consistent band~structure calculations, 1: I3 1 Self-coord~nationnumbers close-packed structures, 1: 281, 282, 283, 288 ho~ologousstructures, 1: 293 homomet~cstructures, 1: 289, 292 interstitial atoms, 1: 297, 301 quaternarylternary alloys, 1: 305 Self-destructing corroding links, Self-diffusion activation energy, 1: 575; 3: 300 in metal oxides and sulfides, 3: 721 and thermal vacancies, 3: 289-290 Self-di~usioncoe~cients,1: 574, 575, 979 Self-diffusivities,tem~eraturevariation, 3: 289, 290 Self-fluxing, heteronucleants removed by, 3 684-685 Self-interaction correction, 1: 130 Self-interstitials niigration of, 1: 797 structures determined, 1: 567-568 synthesis (SHS), 1: 646; 2: 44; 3: 645, 724, 732-737 mechanically activated powders, 3: 744
1006 and milling, 3: 757-758 see also Combustion synthesis Semantic networks, 3: 824 Semantic searching algorithm, 3: 864 Semiconductivity, heavy-fermion compounds, 1: 218 Semiconductor applications, 2: 323-348 Semiconductor diode lasers, 2: 418, 427-43 1 Semiconductor injection lasers, 2 428 Semiconductors, 1: 16, 109, 424, 953; 3: 31, 246 broken bonds at surface, 3: 216 crystal structure, 1: 424 dabbase on phase diagrams, 3: 814 data sources, 3: 805-806 economic factors, 3: 794 economic value, 3: 795-796 energy gap, 1: 109, 110 grain-boundary effects, 1: 958 interactions with binary alloys, 2: 6 14-622 amorphous alloys on Si, 2: 618-620 bilayers on Si, 2: 615-616 on GaAs, 2: 621 single-element films on Si, 2: 614 solid solutions on Si, 2: 616-618 nanocrystalline, 3: 732 phase diagram calculations, 3: 205 quantum phenomena in, 3: 779 therrnal properties, 1: 1021-1022, 1024-1025, 1027, 1029 thermoelectric applications, 2: 463, 464-465 transformation to metallic compounds, 3: 161 vacancies in, 3: 288 Semi-empiricalpotentials, 3: 765, 766-768 Semi-iiisulating crystals, 2 332 Sernirnetals, 2: 329 Sendust [magnetic material], 2 217-21 8, 307 Sequestering agents, 2: 524 Service costs, 3: 477 SESF see Superlattice extrinsic stackingfault * . . Sessile dislocations, 1: 543-546 Set for prediction, meaning of term, 3: 836 Shade (color), meaning of term, 3: 242 Shape approximations, 1: 133 Shape-memory actuators, 2: 545 Shape-memory alloys, 1: 163-1 64, 832 A1-Ni based, 2: 538, 556 applications of, 2: 529-558 chemical composition, 2: 537, 538 coil-spring behavior, 1: 833; 2 535 copper-based, 2 536, 538, 543-544 corrosion resistance of, 2: 538 as couplings, 2: 536, 538-540 database on, 3: 809 density of, 2 538 economic factors, 3: 794 in eyeglass frames, 2: 556 fatigue properties, 2: 538 gold-based alloys, 3 55, 76-77 grain size effects, 2: 538
Subject Index high-temperature shape-mernory alloys, 2: 556 hysteresis loops for, 2: 530 limitations of, 2: 556 mechanical properties of, 2 538 medical applications, 2: 548-554 cardiovascular applications, 2: 554 orthodontic applications, 2 553 orthopedic applications, 2: 548-553 superelastic devices used, 2 553-554 nielting temperatures, 2 538 NiTi, 3266 non-ferrous alloys, 2: 538 phonon studies for, 1: 164 physical properties of, 2: 538 quasibinary systems, 1: 717, 722, 724 specific damping capacity, 2: 535, 538, 555 stress-strain diagrams for, 1: 834, 835; 2: 532, 533, 534 superelasticity, 1: 834, 835 thermal conductivity, 2: 238 thermal expansion, 2: 238 Shape-memory effects, 1: 832-834; 2: 529 and rnartensitic transformations, 1: 163-164, 832; 2: 54-55, 529, 530 one-way memory, 1: 832, 833; 2: 534, 535 applications, 2: 536-543 processes involved, 1: 833; 2: 533 and pseudo-twinning, 3 404 rubber-like behavior, 1: 834 superelasticity, 1: 834 two-way memory, 1: 832-834; 2: 534, 535 applications, 2: 543-548 Shape-memory films, 2: 432 SHAPE software [crystal habit representation], 3: 188 Shear defects, in metallic glasses, 3: 693694 Shear displacement, and magnetic-martensitic transformation, 3: 533 Shear modulus, 1: 202, 876; 3: 2#1 calculated values, 1: 206 changes during amorphization, 1: 814-8 15 composition dependence, 1: 887 of high-temperature intermetallics, 2: 252 and melting temperature, 1: 891 of nickel alumiiiide, 2: 56 of polycrystals, 1: 202, 206, 209, 885 estimation from single-crystal constants, 1: 884, 885 of silhdes, 2: 213 temperature dependence of, 1: 888 of trialuminides, 2 149 Shear stress Al,Ti, 2: 154, 1S5 see also Critical resolved shear stress Shear-stress shear-strain curves, 3: 241, 362 Sheet processing, 3: 506-507 Sheet rolling, 3: 631-633, 651-652 Sheet structure types, 3: 18 Shock compression, for mechanochemical activation, 3: 744 Shock-compression synthesis, 3: 723, 724, 725-727, 740-743 timescales, 3: 724
Shock"conso~idationtechniques, 3: 723, 738-740 liardness and grain size, 3: 739 Shock densification processes, 1: 649, 912 Shock-induced reactions, 3: 723, 740-741 Shockley partial dis~ocations,3: 409, 423, 430,437 Shock-modified materials, 3: 743 Shock pressure, and hardness, 3 738 Shock wave, ion wake compared with, 3: 270-27 1 Shoemaker's exclusion rule, 2: 478 Short-range order, 1: 781-784 in Cu-Pd alloys, 1: 49-50 in Cu-Zn alloys, 1: 46 in liquid alloys, 1: 479, 661, 671 in Pd-Rh alloys, I: 50-51 Short-range order parameter 1: 304, close-packed al~oys~compounds, 305 liquid alloys, 1: 677 tetragonal single-layer stru~tures,1: 284 Short-range order (SRO) disperse order model, 3: 74 lattice defect model, 3: 74 in metallic glasses, 3: 689, 690, 700 microdomain model, 3: 74 in Ni-Pt allays, 3: 223 in palladium-base alloys, 3: 74,689,690 in platinurn~basealloys, 3: 73-74, 223224 statistical model, 3 73 at surfaces, 3: 223-224 SHS diagrams, 3: 732 see also Self-propag~tinghigh-temperature synthesis Shuffle parameter, 3: 405 Siberia, platinum group metal reserves, 3: 80 Sigma ((T)phases, 1: 119, 249, 250, 253, 409, 412; 3: xli Silicate glass, ion tracks in, 3: 264 Silicidation reactions, 2: 614, 615 Silicide coatings, 3: 574-575 Silhdes, 1: 388, 647, 652, 969-970; 2 211-233; 3: 485-487 A15 silicides, 2: 223-224 coatings based on, 2: 211, 227-229 crystal structure, 2: 212 data sources, 3: 808 electrical resistivity, 2: 21 I, 231 ~lectrochemicalbehavior, 2: 505-506 epitaxial silicides, 2: 231-232 fatigue crack growth in, 3: 342, 344346, 346 5:3 silicides, 2 224-225 3: 675 formation temperat~~res, heats of formation, 3: 733 mechanically alloyed, 3: 757 multicomponent silicides, 2: 225-227 oxidation behavior, 1: 997-999, 1001, 1003-1005 oxidation resistance, 2: 213, 227 oxide protective scales, 1: 998 physical properties, 2: 2 13 powder metallurgy applications, 3: 657-658 prediction of formation, 3: 829 refractory-metal silicides, 2: 2 19-227, 329
Subject Itidex semiconductors, 2: 329-330, 614 superconductors, 2: 21 1, 223, 229-230 ternary silicides, 2: 225-227, 615, 616, 619, 622 as thermoelectrics, 2: 469, 470 thin films, 3: 674-675 Ti-based alloys strengthened by, 2: 282 transition-metal silicides, 2: 213-219,469 see also Disilicides Silicon 4-atom cluster, in Zintl phases, 3: 117, 118 9-atom cluster, in Zintl phases, 3: 117, 118 Silicon, oxidation behavior affected by, 1: 984-986, 989, 993 Silicon carbide fibers, 2 292 whiskers, 2 295 Silicon octahedra, 1: 180 Silicon(1IIj surface phase transition, 3: 221-222 reconstruction of, 3: 216 Silmanal [magnetic material], 2: 308-309 Silver amalgam, 3: 21, 795 Silver amalgam coatings, 2: 520 Silver clialcoge~ides,mobile species in, 3: 250-251 Silver, extractioii of, 3: 29 Silver-recovery processes, 2: 515 Silver-tin alloy powder, 2: 578-580 microstructures, 2 580 preparation and processing of, 2: 579-5 80 Silver-tin-mercury alloys, 2: 575, 576-577 see also Dental amalgam Simple cubic (s.c.j lattice combmation with f.c.c. structure, 1: 169 crystal model demonstrating bonding, 1: 170 Simplex coordinate mapping, 1: 186, 187 Simplicity principle, 1: 255-256, 264 Simulated annealing, 1: 259 Simulation methods dislocation fine structure, 1: 523; 2: 23 supercooled liquids, 1: 679-680 Simulation techniques, 3: 765-778 in calculation of phase diagrams, 3: 201 see also Atomistic simulations Single-crystal aluminides, 2: 41-42, 67, 202 strain hardening in, 3: 365-367 magnetization curves, 3: 169, 176 m~gnetizationmeasurements, 3: 169, 183 processing techniques, 3: 542, 543, 602 superalloys, 3: 543 creep behavior, 3: 553 thermal fatigue, 3: 556 Single crystals, 1: 11 ailisotropy of elastic properties, 1: 874, 876-880 elastic constants, 1: 873-874 growth methods, 1: 643-644 Single-crystal trialumiaides, 2 153 Single-environment structure types, 1: 273, 280, 294, 435 Sing~e-particleHamiltonian, 1: 127 Single-site mean-field theory, 1: 25 Sintering, 1: 645 reactive sintering, 1: 646-647, 912; 2 44
Sintering aids, 2: 523 Sintering, Nd-Fe-B magnets, 3: 102 Sintering mechanism maps, 3: 647 Sinusoidal model (of magnetic structures), 1: 442, 443 Si-Pd-Pt system, 2: 616 SISF see Superlattice intrinsic stackmgfault . . . Si-Ta system, 1: 999 Site-set symbols, 1: 316-3 17 Si,Th structure type, 1: 364, 372, 373, 375, 389,429 Si-Ti powders shock consolidation of, 3: 740 shock-densified, reaction of, 3: 745 shock-induced synthesis of, 3: 741 Si-Ti system, 1: 696, 735, 738, 999 SizTistructure type, 3: 388, 389, 390,428, 429 Si-Ti system, phase diagram, 3: 733 Si-U system, 3: 159 Si-V system, 1: 696 Si-W system, 1: 999 Si,W, structure type, 1: 262 Size factor, 1: 232, 270 in AP-AN plots, 1: 230, 233 atomic properties grouped under, I: 23 1 composition of ternary c.p. compounds affected by, 1: 407 compounds, 3: xli interaction between intermetallic compounds determined by, 1: 712 and solid-solution tendency, 1: 244,248 in space filling, 1: 254 and structure mapping, 1: 241, 242-243, 420 see &o Atoinic-size differences/ratio Size mismatch, structural stability affected by, 1: 102-104, 420, 422 Size parameter (structure maps), 3: 91 rare-earth compounds, 3: 93 Si-Zr system, phase diagram, 3: 737 Si,Zr structure type, 1: 388 Skull melting technique, 1: 640 Skutterudite structure types, 1: 332; 3: 12, 106 filled skutterudites, 3: xxxvii, 106 Slab waveguides, 2 413-414 Slater-Nkel-type diagram, 1: 440 Slip bands, persistent, 3: 326, 348 Slip [deformation] mechanisms, 1: 9 13-9 15 AlTi, alloys, 2: 102-106 Slip geometry, creep resistance affected by, 3: 304-305 Slip systems, 1: 541-543 nickel aluminide, 2: 17 and strain hardening, 3: 364, 370-371 trialuminide compounds, 2: 154, 155, 162 zirconium aluminide, 2: 143 Slip-twin conjugate relationships, 3: 41 1, 414 smart^ windows, 2: 41 1 Sm,Ge, structure type, 1: 120 Smirnova’s classification (of structure types), 1: 403 SmSb, structure type, 3: 89, 89 SnNi, structure type, 3: 88, 89 Sn-Te system, 1: 704
1007 Sn-U system, 3: 159 Sn-Zn alloys, 2: 521 Sobolevskite [mineral name], 1: 630, 63 1, 632 Sodani-Vitek (SV) dislocation-lock in^^ model, 1: 547 Sodizun-zinc amalgam, 3: 33 Soft ferromagnets, 1: 939-940 Soft magnets, 1: 749; 2; 303, 305, 306-307 Soft-sphere model, 1: 679-680 Software, phase diagram calculations, 3: 187, 191,206, 803 Soft X-ray emission spectroscopy, 3: 136 Solar cell applications, 3: 663, 672, 673 Solar cells, 2: 330,418, 421-424 characteristics, 2: 424 construction, 2: 422, 423 Solar dynamic power systems, 2 637, 643 Soldering processes, 2: 522-523, 654 Solidification directional, 3 541-544 segregation during, 3: 474, 619 Solid-solubility mapping, 1: 244-248, 264 binary systems, 1: 244-248 ternary systems, 1: 248, 394 see also Continuous solid solutions Solid-solution hardening, 1: 923; 2: 35-37, 42, 46, 163, 200, 206, 238-240 Solid-state amorpbization, 1: 696-699, 733-734, 739 Solid-state diffusion, 1: 645-646 Solid-state diffusion coatings, 2: 521 Solid-state reaction synthesis, 3 645, 724727 Solubility extension, I: 703, 705, 706 Solubility limits transition metals in aluminurn, 2: 176 Solution calorimetry, enthalpies of formation determined using, 1: 96,97 Solution hardening, 3 351-360 B2 compounds, 3: 354, 355, 356 C15 compounds, 3: 356 conventiona~~echanisms,3: 352-353 L1, compounds, 3: 353-354 and modulus difference, 3: 352 recent literature, 3 351-352 and size misfit, 3 352 test procedures, 3: 356, 358 Solution strengthening, creep resistance affected by, 3 307-308 Sommerfeld, Arnold J. W., 3: xliv see also Grimm-Sommerfeld compounds Sommerfeld theory, 1: 942 Sonar transducers, 2: 402-403 South Africa, platinum group metal reserves, 3: 80 Space-based applications, 2: 229, 287, 289, 409, 419, 458, 459, 546-548, 551-552 Space filling, 1: 248-254 Space-filling ratio, 1: 249 Space-filling ratio plots, 1: 249, 250 see also Geometrical stability plots Space-group numbers, distribution of compounds, 1: 254,255,259, 264 Space-group operators, 1: 31 1 Space groups, 1: 309, 31 1 assignment of lattice complexes, E: 317, 318
1008 r a r e - ~ ~ r compounds, th 3: 165 relationship to crystal structure nomenclatures, 1: xxv-xxviii; 2: xxiii-xxvi; 3: xxxi-xxxiv, 13-15, 439 ZnS and Sic polytypes. 1: 359-360 Space Shuttle experiments, 3: 32 Space principle, 1: 248 quantitative indicators, 1: 249-254 Spaepen-Turnbull nucleation theory, 1: 690, 692 SpangoldTMconcept, 3: 76-77 Spar~-plugelectrodes, 3: 61-62, 70, 736 erosion resistance, 3 62, 69 Special-points method (for Brillouin-zone sampling), 1: 132 Specific damping capacity [for shapememory alloys], 2 535, 538, 555 Specific electrical resistivity, listed for various metals, 1: 956. 960 Specific gravity beryllides, 3: 38, 39, 40 higli-temperature intermetallics, 2: 24 1 , 242, 243, 244, 245, 249 and lattice parameter, 1: 561 and melting temperature, 2: 243 see also Density Specific heat amorphous alloys, 1: 736, 737 heavy-feimion compounds, 1: 212-21 3, 216, 217, 219, 220 liquid alloys, 1: 665, 666, 674 quasicrystals, 1: 484 seniiconductors/semi~etals,2: 328 shape-memory alloys, 2 538 Specific strength, 2: 165, 24'1 Spectroscopy, 3: 135-1 5 I of icosahedral Al-Pd-Mn quasicrystal, 3: 147 of occupied states, 3: 135-145 of unoccupied states, 3: 145-150 see also Bremsstrahlung isochromat. . .; Electron energy loss.. .; Photoemission. ..; X-ray em~ss~on,. ., Xray photoabsorption spectroscopy Speculum, 2: 507 SPEX mill, 3: 727, 728, 756, 757, 758 Sphalerite-type compounds, 1: 320, 343-358 crystal equilibrium forms, 1: 168, 172 crystal etch habits, 1: 181, 182 crystal growth habits, 1: 180, 181, 182 energy gap, 1: 109, 110 enthalpy of formation, 1: 109, 110 heat capacity, 1: 118 perspective view of structure, 1: 175 see also ~ i n c - b l e n ~structures e Spin-density-wa~emodel (of magnetic structures), 1: 442 Spin-density waves, 1: 440, 442, 445 see also Magnetic spin-d~nsity distribut~on Spin-disorder scattering effect, 1: 945, 946, 949 Spinel ion tracks in, 3: 267 mean~ngof term, 3: xli structural formula for, 3: 5 Spinel structure, 2: 330 Spin-glass behavior, 1: 484, 935
Subject Index Spin glasses, 1: 439-440, 450; 3 174, 181, 182, 692 Spinodal decomposition, 1: 852,855; 2: 565 Spinodal ordering, 1: 773, 803, 859 Spinodal trans~tion,1: 160 Spin-orbit coupling, magnetic behavior affected by, 3: 170 Spin-orbit coupling, magneto-optical effects affected by, 2: 441 Spin-reorientation (of magnetization direction), 3: 98, 167-169 Spiral magnetic structure, 3: 170, 171 Spiral model (of magnetic structures), 1: 442, 443,445,450 Splat-cooled Al-Fe alloys, 2: 178 Splat cooling, Hum-Rothery phases formed, 1: 104 Split-Hopkinson-bar testing, 3 371, 373 Split interstitials, 1: 559, 578, 785, 796 Splitting diagram, adamantane-structure Compounds, 1: 348, 349 Splitting of (crystallographic~I-complex, 1: 319-325 Spot welding, 2: 523 Sputter-deposited films, 2: 316, 431-432 Sputtered films, 2: 316 Sputtering, 1: 773 beryllides prepared by, 3: 47 coating by, 3: 570, 571 limitations, 3: 672 thin films deposited by, 3: 665-666,672 see also Physical vapor deposition Square (magnetic) hysteresis loops, 2: 304, 448 SrZn, structure type, 1: 413 Stability plots see Structure maps Stacking fault energy (SFE), and creep resistance, 3: 308 Stacking fault interfaces, 3: 437-467 Stacking faults in AINi,, 2: 19-20 in Al,Ti, 2 149 in AIZr,, 2: 139, 140 in B2 structure, 3: 451-453 in C11, structure, 3: 454-458 in C40 structure, 3: 458-460 m C54 structure, 3: 458-460 complex stacking faults (CSFs), 2: 20, 149 in complex structures, 3: 460-461 and crystal structure, 3: 439-461 in DO, structure, 3: 453-454 in DO,, structure, 3: 448-451 in DO,, structure, 3: 444-448 in deformed quasicrystals, 3: 400 generalized, 3: 438 in L1, alloys, 1: 534 in LI, structure, 3 442, 444-446 in Ll, structure, 3: 440442 in L2, structure, 3: 453-454 in martensitic transformation, 2: 530 in plastically deformed quasicrystals, 3: 400 in quasicrystals, 1: 467 temperature dependence, 3: 461-462 and twin nucleation, 3: 419, 420 see a2so Complex . . .; Extrinsic . . ., Intrinsic . . ., Superlattice stacking faults Stacking notations, 1: 279-280, 358-359, 406
Stacking sequences, 2 156, 532 close-packed structures, 1: 277-282, 305 hexagonal layers, 1: 277-278, 287 homome~ricstructures, 1: 293 NiTi,-type compou~ds,1: 394 Stackiiig symbols, 3: 5 , 6 Stacking variants disordered alloys, 1: 305 interstitial alloys, 1: 303 ZnS polytypes, 1: 344-345 Stage 2 deformation, 3: 362, 368 Stainless steels iron aluminide overlays on, 3 573,7 15716 sigma phase in, 3: 62 Stair-rod partial dislocations, 3: 414, 421 Stannite [mineral name], 1: 344, 355, 359 Stannopalladinite [mineral name], 1: 630, 632 Static inductance transistors, 2: 332, 335 Statistical approach to design of materials, 3: 833 Statistical i~echanism,1: 21, 24 Statistical substitution i n close-packed structures, 1: 406, 416 Steady-state oxidation, transition from transient oxidation, 1: 908, 998 Steady-state unzipping process for dislocations, 2 259 see also F ~ e d e l / ~ l ~ i s cstatistics her Stepwise relaxation method, 1: 86 Stevens constant, 3: 172, 176 Stevens factor, 2: 316 Stibnite crystal structure, 3: 6-7 Stibiopalladinite [mineral name], 1: 629, 630 Stiffness see Young's modulus Stistaite (mineral name], 1: 630, 632 ~toichiome~ric restraint approach, compound-for~ationpredicted using, 1: 257-260, 264 Stoichionietry of I32 phases, 3: 849 binary compounds, 1: 257-258 and compositional control, 1: 637-638 and creep resistance, 3: 309-312, 31 5 ductility affected by, 1: 897, 922-924 effects on properties, 1: 8, 9 electrxal conductiv~tyaffected by, 1: 953-954 term first used, 1: 9 [footnote] ternary compounds, 1: 258-260 atomic environment approach, 1: 259-260 space-group approach, 1: 258-259 Stoner parameter, 1: 141 listed for transition~metalaluminides, 1: 138 Strain-energy interactions, 1: 867 Strain hardening, 3: 361-377 in b.c.c. metals, 3: 363 and compression testing, 3: 371 in disordered materials, 3: 362-364 and dynamic recovery processes, 3: 363, 366 effects of APBs, 3: 367 in f.c.c. metals, 3: 363 in high-temperature plastic deformation of quasicrystals, 3: 401 and order, 3: 364-365
Subject Index in polycrystalline inte~etallics,3 368373 rates, 3: 370 in single crystals, 3: 365-368 stram-rate dependence, 3: 363 strain-rate sensitivity, 3: 370, 371 temperature effects, 3: 363, 365 titanium aluiiiinides, 3: 365-366, 368, 369, 370, 371, 372, 621-622 Strain-hardening behavior, 1: 913, 925 Strain-indLiced niartensitic transformation, 1: 831, 832 Strain~inducedresistimty change, 1: 958-959 Strain-rate sensitivity aluininides, 2: 65, 66 silicides, 2 218 Strains, tetragonal lattice, 1: 201 Strength analyaing and modelling of experimental data, 2: 269-270 of dental ama~gam,2 585 effect of APBs, 1: 925 effect of second phase, 1: 925-926 and grain size, 1: 925; 2: 79 or iron aluminides, 2 203, 205 of link (in economic terms), 3: 793 of iiickel aluminides, 2: 36, 37, 38, 42, 60-62 of superalloys, 2: 8, 12, 41, 74 temperature d~pendence,2: 36, 38, 41, 60-61, 86, 113, 137-138, 153-154, 164-166 of titanium aluminides, 2: 74, 83, 86, 92, 107, 108, 113, 114 of trialum~nides,2: 164-1 66 of ~irconiumaluminide, 2: 136-1 38 Streiigthening of Al-Ni alloys, 2 271-273 by bimodal dispersions, 2: 279-280, 281-282 by extended obstacles, 2: 260-261 by finitc obstacles, 2: 262-263 by mixtures of obstacles, 2: 261-262 of Co-Cr-Ni alloys, 2: 276-277 computer siinulatron of, 2: 260, 262 of ferritic alloys, 2: 27'7-279 generalized addition rule for, 2: 261, 282 implications for alloy design, 2: 282-284 mechanisms, 2: 266-269 in overaged alloys, 2: 268-269 in peak-aged alloys, 2: 268 precipitation strengthening, 2: 257 of superalloys, 2: 273-276 in underaged alloys, 2: 267-268 Stress-assisted grain-boun~aryoxidation, 2: 33 Stress corrosion cracking, 1: 927, 967, 969, 970 compared with hydrogen e~brittlement,1: 967 Stress dipoles, hardening by, 3: 358 Stress exponents [in creep], 2: 63, 115, 585 Stress-induced ~ ~ r t e n s i t i c t r a n s f o ~ a t i o n1: , 831, 832, 834; 2: 529 and ductility improvement, 1; 845-846 Stress-intensification factor, 3: 647, 648
Stress relaxation behavior, 3: 298 Stress-rupture behavior, titanium aluminide alloys, 2: 118 St~ess-rupturecurves AlNi alloys, 2: 62 superalloys, 2: 7, 62 Stress-rupture life 2: 14 super~lloys~ titanium aluminide alloys, 2: 92 Stress-strain diagrams A N , , 1: 88 AI-Ru alloys, 2: 251 Al,Ti, 2: 154 AlZr,, 2: 138 calculated/predicted curves, 2: 238, 239 CuTi-NiTi alloys, 1: 724 Cu-Zn system, 1: 834 effect of order, 1: 914 quasibina~/quasiterna~ systems, 1: 724 serratsons in, 1: 915; 2: 154 shape-memory alloys, 2: 532, 533, 534 superelasticity loop, 1: 834 Stress-strain-temperature diagrams, shape-memory alloys, 1: 835; 2: 535 Stress-teinperature diagrams, martensitic transformation, 1: 832 S~rip-projectionmethod, quasicrystals studied using, 1: 457, 459 Strohs method extension to 12D formalism, 3 394 generalization of, 3: 393-395 Structural alloys basic properties, 3: 502 design data, 3: 513 specific strengths, 3: 472 Structural applications, 2: 1-300 affordability, 3: 476-477, 482-483 Al-rich internietallics in altrmrnuni alloys, 2 194, 195 AI,Ti alloys, 3 495-496 attributes of IMCs, 3: 501-503 composites, 2 28'7 FeAl alloys, 3: 496, 503, 514 Fe& alloys, 3: 496, 502-503, 513 general-use, 3: 501-5 18 high-temperature, 3: 47 1-499 miscellaneous iiovel IMCs, 2: 237-255 molybdenum silicides, 3: 485-487 molybdenum-silicon alloys, 3: 487-490 PJiAl alloys, 3: 494. 502 Ni,AI alloys, 3: 492.493, 501-502, 512 Ni,Si alloys, 3: 503 nickel alu~inides,2 45-46, 53 niobium-silicon composites, 3 490-492 si~con"bearingalloys, 3: 487-492 TiAl alloys, 3: 477-485, 503, 515 Ti,AlNb alloys, 3 494-495 Ti,Al alloys, 3: 494-495 titanium aluminides, 2: 73, 87-88 trialuminide compounds, 2: 169-1 70 Structural classification, 3: 3-4 Structural derivatives, 3: 11-12, I9 Structural energy, 3: 197 Structural energy differences, L1, vs DO,, crystal structures, 1: 65 Structural formulas, 3: 4-9 Structure, modeling of, 3: 841 Structural notation, 3: 4-9 application to structure type prototypes, 3: xxxi-xxxiv, 12, 13-15 Structural relation restraints, 1: 260-264
1009 Structural stability, 1: 850-861 and antipliase-boL~nda~ energies, 1: 499-500 factors affecting, 1: 102-104, 420, 422, 424 Structural transfor~at~ons, physical properties aEwted by, 1: 21 Structural units, 3 4 Structural vacancies detection/determination techniques, 1: 560-561 empirical prediction, 1: 568 formation of, 1: 565-566 Structure factor, liquids, 1: 661, 663 Structure factors, 3: 248, 249 Structure iiivenron methods (StM), 3: 194, 201 Structure mapping, 1: 419-437 Structure maps, 1: 241-244,264,420-437; 3 842 adamaiitane-structure compounds, 1: 356-358 binary compounds, 1: 241, 242-243, 264, 420-434 ciosc-packed intcrstitid alloys, 1: 300, 303 close-packed structures, 1: 286, 287, 289, 295, 296 colored IMCs, 3: 237, 242 disordered close-packed alloys, 1: 304 hexagonal layer structures, 1: 285, 291 Mooser-Pearson maps, 1: 241, 242, 357, 419,420 ordering number defined, 1: 425 Pettifor maps, 1: 243-244, 390, 393, 396, 400,419-420, 424-434, 435-437 rare-earth compounds, 3: 88 ternary compounds, 1: 241, 243-244, 264, 434-437 tetragonal layer structures, 1: 282-285, 29 1 W a r s (three-dimensional) maps, 1: 244, 245, 262, 357, 358, 381, 390, 393, 419, 424 Structure separation plots see Structure maps Structure types classification of. 1: 403 Structures chains, 3: 3, 4 close-packed blocks, 3: 6 clusters, 3: 3 frameworks, 3 3, 4 puckered hexagonal, 3: 6 sheets, 3: 3, 4 Structure types, 3: 10-1 I ordering of, 3: 12, 16-18, 19 rare-earth compounds binary compounds, 3: 88-90 ternary compounds, 3: 90-96 for structure prototypes, 3: 13-15 symbols proposed for, 3: 11,439 Strukturberrcht designations, relationship to other aornenclatures, 1: relationship to other nomenclature, 3: xxxi-xxxw, 13-15, 238, 439 Stuffed structures, 3: xxxvii Sub-grain boundaries,
1010 Subgrain formation, 1: 916, 925 Sublattice mechanism of diffusion, 1: 761 Sublattice occupancy, 3: 774 Sublattice occupation probabilities, 1: 775 Sublattices, 3: 189 Substitution derivatives, 3: 12 Substitutional disorder, phonon dispersion affected by, 1: 160-161 Subtransus processing, AlTi, alloys, 2: 98, 109-112 Subtransus structures, 2: 77 Sudburyite [mineral name], 1: 630, 632 Sulfidation, 3: 707-71 9 barrier layers for, 3: 710 with complex scales, 3: 708 doping to slow, 3: 709 as liigh-temperature corrosion, 3: 707 of iron aluminides, 3: 7 12-7 16 as irreversible process, 3: 719 kinetics, titanium aluminides, 3 7 16717 of molybdenum silicides, 3: 7 17-7 19 of Mo,Si, susfaces, 3: 715 of nickel aluminides, 3: 7 10-7 12 and oxidation, 3: 707 and oxide scale formation, 3: 708 scale formation, 3: 708 and surface appearance, 3: 71.5 thesmogravinietnc data, 3: 712, 714718 of titanium al~iminides,3: 7 16-7 17 and weight change of iron aluminides, 3: 714, 71.5 of Mo,Si,, 3: 718 of nickel aluminides, 3: 712 of titanium aluminides, 3: 716, 717 Sulfidation resistance, 1: 999 Sulfidation-resistaiit materials, 3 7 10-7 19 development of, 3: 709-7 10 Sulfides, 1: 626 melting temp~rature,3: 710, 711 physicochemical properties, 3 711 prediction of forniation, 3: 826, 827, 827 s ~ ~ f ~ d ~ f f ucoe~cients, sion 3: 711 stability, 3: 708 thermodynamic data, 3: 711 Sulfur, in fossil fuels, 3: 707 Sulfur, segregation to grain boundaries, 1: 904, 967 Superalloys alloying of, 2: 8-9 AlNi, misfit, 2: 10 compared with intermet~~lics, 2: 41,42, 43, 56, 57, 62, 63, 65, 74, 240 compositions listed, 2: 6, 270 crack propagation rates, 2: 42 development advances, 2: 4 economic factors, 3: 794 historical perspective, 2: 3-4 microstructura1 control in, 1: 859 microstructures, 2: 6, 11 oxidation resistance, 2: 7 phase stability considerations, 2: 9, 74 physical metallurgy, 2 4-8 precipitate morphology, 2: 9-13 strengthen~ngof, 2: 273-276, 283 stress coarsening in, 2: 11-13 stress-ruptuse curves, 2: 7 stress-rupture life, 2: 14
Subject Index temperature limits, 3: 542 tensile strength, 2: 7, 13 yield strength, 2: 8, 12, 41 see also specific alloys: A286, Astroloy, B- 1900, CSMX-2, Hastelloy X, IN-100, IN-713C, IN-738, MA6000, MarM200,NASAIR- 100, Nimoiiic, PWA-1480, Rent, TDNi, TRW-NASA IVA, Waspaloy, Iridium-base.. .; Nickel-base.. .; Platinum-base superalloys Supercells, 1: 59, 66 Superconducting c~tical/transitio~ temperature, 1: 951 effect of point defects, 1: 578-579, 580 effect of stoichiometry, 2: 356 as function of valence electrons per atom, 1: 266, 951, 952 listed for various compounds, 2 352, 482 long-range-order changes studied, 1: 777 metastable crystalline phases, 1: 704, 705 superconducting integrated circuits, 2 653-654 Superconducting quantum i~terference devices (SQUIDS), 2: 383, 384 Superconducting Super Collider (SSC), 2: 377 Superconductivity, 1: 16, 135, 211, 950-953; 2: 351 effect of grain boundaries, 1: 957 effect of strain, 2: 354 Superconductors, 2: 351-384 appli~ations,2: 364, 366, 370-384 beryllides, 3: 45 current-carrying capacity, 2: 366, 367 data sources, 3 806 effect of crysta~lization~ 1: 748 fabrication techniques, 2: 355, 360-363, 364, 365 high-temperature, 3: 205 hydrides as, 2 481 ion tracks in, 3: 269 nanostructured, 3: 759 oxides as, 2: 352 phonon dispersion effects, 1: 153, 157-1 59 prediction of formation, 3: 828-829 quantum structure diagram, 1: 265,266 rare-earth compounds, 3: 103-104, 105, 759 silicides as, 2: 21 1, 223, 229 specific-heat anomaly, 1: 216, 217, 220 see also Higli-temperature superconductors Superconductor wires, 3: 794 Supercooling, 1: 479 Supercorroding alloys, 2: 645-646 Superdislocations, 1: 15, 521, 913 in aluminides, 2: 18, 74 in B2 phases, 3: 851 climb, 3: 621 fine structure in relation to crystal structure, 1: 523-541 and strain hardening, 3: 369-370, 372 Superelastic behavior, 2: 529 Superelastieity, 1: 834 Superelastic medical devices, 2: 553-554 Superionic conductors, 3: 250
1: 521, 913; 2: ~ u p e r ~ a tdislocations, ti~ 18-19, 204 fatigue properties affected by, 3: 332 Internet image, 3: 874 Superlattice extrinsic stacking-fault (SESF) dipoles, 1: 534, 551-552 Superlattice extrinsic stackiiig-fault (SESF) energies ab z~itiocalculations, 1: 71, 72 DO,, conipoun~s.3: 447-448, 449 determination of, 3: 407 in Li, structure, 3: 442 ‘Superlattice’ [in semiconductors], 2: 324, 327, 334,421,425 Superlattice intrinsic stacking-fault (STSF) energies, 2: 20, 153 ab initio calculations, 1: 71, 72, 73 and antiphase boundaries, 1: 512 and APB energies in dissociation of dislocations, 1: 525, 526 C40 compounds, 3: 459 DO,, compound, 3: 450,451 DO,, compounds, 3: 447448,449 determination of. 3: 407 use in fitting of Ni-A1 cross potential, 1: 83, 85 Ll, compounds, 3: 445,447 L1, compounds, 3: 442.443444 and twinning, 3: 424, 428 Superlattice intrinsic stacking faults (SISFS),2 19-20, 139, 140, 149 in C1 1, structure, 3: 455, 457 in C40 structure, 3: 4.59 in C54 structure, 3: 460 in DO,, structure, 3: 448-449, 449 in DOz2structure, 3: 447, 448 in L1, structure, 3: 444, 445, 445 in L1, structure, 3: 440, 441, 441 in twinning, 3: 41 1 in twin nucleation, 3: 419, 420 Superlattice lines, 1: 10 Superlattices, rare-earth magnetic materials, 1: 450 Superlattices, t w ~ n n i nin, ~ 3: 41 1-418 Superlattic~stacking fault (SSF) dipoles, 1: 551 ‘Superlattice’ structures, 3: 779, 783 meaning of term, 3: 779n Supermagnets, 2: 3 19 Superparamagnetism, 3: 69 1, 693 Superplastic forming (SPF), 1: 656-4257, 925; 3: 298-299, 636-638 in industrial facilities, 3: 640 with diffusion bonding, 1: 657 Superplasticity, 1: 728, 749 dental amalgam, 2 586 nanocrystalline IMCs, 3 762 precious metal compounds, 3: 78 silicides, 2 213-214 TiAl sheets, 3: 637 titanium alwinide alloys, 2: 86,127-128 Supersolvus structure, 2 77 Superstruc~ures,1: 71 1-712 body-centred cubic structures, 1: 338, 339 close-packed structures, 1: 405, 406-408 MgCu,-type, 1: 109, 410 MgZn,-type, 1: 410, 413 MO%,-type, 1: 338, 386 Super~superlattice,1: 7 12
1011
Subject Index Supe~-twinning,1: 845; 3: 404, 405, 431 Suppliers of IMCs, 3: 511-513 Surface composition, 3: 222-223, 224 Surface crystallization, 1: 748 Surface diffraction techniqu~s,3 2 12 Surface diffusio~,3 225 Surface enegies3: 225-226 ab initio calculations, I: 71, 72, 73 1: 496, 504 antiphase bou~ida~es, of ordered Cu,Au, 3: 226 of relaxed surfaces, 3: 226 in TiAl alloys, 3: 774 see also Antiphase b o u n d a ~energies Surface energy, trialuminides, 2: 167 Surfkce free-energy, effect on crystal habits, 1: 171, 182 Surface-induced disordering, 3: 614, 618 Surface~inducedordering, 1: 614-615, 618 Sur~ace-inducedorderin 223-224 Surface layers composition of, 3: 222-223, 224 rec~nstructionand termination of, 3: 216-217, 228 B2 compounds, 3: 225 relaxation of, 3: 220-221, 228 Surface melting, 1: 733, 773 Surface microscopy, 3 212-214 Surface order-disorder transition, 3: 222223 Surface phase transitions, 3: 221-223, 228 Surface preme~t~ng, 3 221 Surface properties, hydrides, Surface reconsti.uction, 1: 183, 619 -620; 3: 216-217, 228 B2 compounds, 3: 225 Surface relaxation, 1: 85, 616-618; 3: 220221, 228 effects on surface segregation^ 1: 612-613 Sur~~ces calculations, 3: 841 composition effects, 1: 612-613 degree of order near, 1: 614-615 effects on alloy properties, 1: 6 electron smoothing of, 3 220-221 experimeatal observations, 1: 611-612 expei~mentaltechniques, 3: 2 12-2 1 5 in fracture processes, 3: 22&227 mechanical properties near, 3: 227-228 Monte Carlo simulation of, 3: 223 order at, 3 223-224 order near, 3: 221-224 phase transitions at, 1: 618-619 preparation of, 3 212 properties, 3: 225-228 rippling of, 3: 220, 228 r i p ~ l ~ aofg surface layers, 1 segregation at, 1: 6 12-613; 3: 224-225 models for ordered intermetallics, 1: 613 models for substitutional dilute alloys, 1: 612413 terminations of c r y s ~ l ~ o g r a p ~ i ~ c structures, 1: 615, 616 theoretical calcu~ations atomistic models, 1: 610-61 1 first-principles calculations, 1: 61 1 thermodynamic models, 1: 610 Surface se~regation,3 224-225
~ Smface-site ~ o r p h o l o1: ~174-183 Surfbce stress, 3: 226 Surface structure and properties, 3: 21522 1 atomistic ssmulation t~chnsques,3: 2 15 B2 compounds, 3: 219-220 expermental methods, 3: 212-215 analytical techniques, 3: 214-21 5 diffract~ontechni~ues,3: 212, 782 ~easurementsby local techniques, 3: 215 microscopy techniques, 3: 212 -214 fcc compounds, 3: 217-219 notation for, 3: 216 theoretical prediction, 3: 215 Surface termination, 3: 216217, 225, 228 Surfing on Internet, 3 860, 877 Sutherland-Einstein [diffusion] theory, 3 23,24, 26-27 Suzuki effect, 3: 352, 461 Surface structures, 1: 285, 287, 609 Swelling behavior, 1: 79 1, 815-8 16 Switches laser-based, 2: 431 optical, 2: 342, 343, 4114.13, 421 Symmetrical (atomic) con~gurations,1: 598, 599 Symmetrical tilt boundaries, 1: 600, 601 Symmetry principle, 1: 254-255, 256, 264 Syiichronous generators, 2: 381-383 Synchroshear mechanism, 3: 41 5,417,459 Synonyms, when searc~ingon Internet, 3 859-860 Synthesis of TMCs by centrifugal atomszation, 3: 644 by chemical reactions, 3: 740-741 by combustion/self-propagatsng hightemperature method, 3: 732-737 by devitrification, 3: 740, 7-51, 754-755 by electrodeposition, 3: 670, 671, 751, 754 by electron-beam evaporation, 3: 666, 784 by gas atomization, 3: 644 by inert gas condensat~on,3: 75I , 752753 by meclianical alloying, 3: 644-645, 652-653, 654, 727-732, 751, 753754, 755 mechanochemical, 3: 743-745 by milling, 3: 644-645, 652, 727-730, 753 non-equilibrium phases, 3: 732 by s~ock-compression,3 740-743 of shock-densified Si-Ti powder mixture, 3: 745 by spray conversion, 3: 754 thermodyn~m~c ~imita~ions, 3: 734 Synthesis techniques, 1: 637-657, 912; 3: 591-706, 723-748 for A1,Ti-based alloys, 2: 158-161 composition control, 1: 637-638 iinpurity effects, 1: 438 stoichiometry effects, 1: 637-638 electrodeposition methods, I: 652-653 melting methods, 1: 638-644 directional solidification, 1: 643-644 homogenization, 1: 644 ingot production, I: 639-640 rapid solidification, 1: 640-642 sing~e-cy~ta~ growth, 1: 643-644
thermal spraying, 1: 642-643 p o w ~ consolidation ~r step, 1: 644-645 solid-state reactions, 1: 645-652 compound r e d u c t ~ otechnique^, ~ 1: 650-651 interdiffusion, 1: 645-646, 757-760 ion-beam-induced mixing, 1: 650 ion implantation, 1: 659-650 mechanical alloying technique, 1: 647-649 reactive sintering, 1: 646-647 shock compactioii/synthesis, 1: 649 vapor-phase produ~tionmethods, 1: 651-652 see also Processing SZii structure type, 1: 261 t phase 1: 465 T phase, 1: 721 in AI-Li alloys, 2: 188 Taenite [mineral name], 1: 631 T a k e ~ ~ c h i - l ~ u r ~(TK) m ~ t dislocationo locking model, 1: 547 Tantalum TMCs, 3: 800 Tantalum silbdes, nanocrystallin~, Tape casting, 3: 507, 655 Ta-(Ti,Zr),Al(rvro,Nb) system, niultiphase alloys, 3 845-847 Tau phases, 3 xli TbFeSi, structure type, 3: 92 TbGe, structure type, 3 Tammann ~ e ~ p e r a t u r e , Tandem [diEusionJ barn Tangent co~dition,1: 97-98 Tap s, 2: 360 Tau Technology demonstration of MiAl alloys, 3: 604 of TiAl alloys, 3: 598-601, 638 Tekezawa-Sat0 [bainite transfor~ation~ model, 1: 841-842 Tellurides, I: 626 Tellurides, prediction of formation, 3: 826, 827,828 Temperature control in Exo-Melt process, 3: 606 in forming processes, 3: 631-632 Te~peratureeffects, colored co~pounds, 3: 240 Temperature-history indicator, 2: 544, 546 Tcnsile aniso~ropy,of TiAl, 3: 630, 634 Tensile creep behavior, 3: 299 Tensile ductility, at room temperature, 3: a 53-8 54 Tensile properties, of TiAl sheets, 3 633, 634,652 Tensile strength Al-Fe alloys, 1: 905 AI-Ni alloys, 1: 904 AlZr,, 2: 144 brittle-to-ductile transition, 1: 819 superalloys, 2: 13 see also Ultimate tensile strength
. . .; Yield
1012 Tensile testing, at room temperature, 3: 853 Tension twin, 3: 417, 43 1 Terfenol-D, 2: 392-393 applications in ma~netostrictive actuators, 2: 402-403 magnetic annealing of, properties, 2: 395-400 Terminolo~y,when searching Internet, 3: 859-860 Ternary c o ~ p o u n d s AB, compounds with random atoinic mix in^, 1: 377 beryllides, 3: 49 chalco~enides,1: 435-437 close-pac~edstructures, 1: 305-306 compound-formation tendency of, 1: 234-236, 713 database on, 3: 815-816 data sources, 3 801-802 enthalpy of formation of, 1: 121, 122 high-temperature i n t e ~ e ~ a l l i c2:s , 251-252, 254 hydrides, 1: 393 ~agneto-opticalproperties of, 2: 448-449 NiTi2-typephases, 1: 393-394 P,Th,-type phases, 1: 399-400 precious-metal compounds, 3: 59-60 rare-earth compounds, 3: 90-96
stoichiometric ratio restraint of, 1: 258-260 structure maps for, 1: 241, 243-244, 264,434-437 structure types possible, 1: 434 thermal vacancies in, 3: 288-289 thermodynamic properties of, 1: 121-122 total number possible, 1: 228, 241 trial~~minides, 2: 158-160, 164, 165 Ternary defect adamantane-structure types, 1: 350, 355-356 Ternary Laves phases, 1: 107, 108, 109 Ternary normal adamantane-structure types, 1: 354-355 Ternary phase diagrams Ag-Au-Gu system, 2: 560, 562, 563, 567, 568 Ag-Hg-Sn system, 2: 579 Al-Ce-Fe system, 2: 179 Al-Co-Fe system, 1: 718 Al-Co-Ru system, 1: 718 Al-Cu-Pd system, 1: 714, 721 Al-Fe-Ni system, 1: 719, 726 Al-Fe-V system, 2: 184
Au-Cu-Ni system, 2: 571 Co-Fe-Ni system, 1: 715 Cu-Ni-Ti system, 1: 722 Cu-Sb-Sn system, 2: 592 Ni-Fe-Ti system, 1: 717 Ni-No-W system, 1: 728 Ni-Pd-Ti system, 1: 716 Ni-Ti-Si system, 1: 261 see also Phase diagrams
~ ~ ~Index ~ e c t Ternary solutes, effect on crystal structure, 1: 66 Terne plate, 2: 519 Tetraauricupride [mineral name], 1: 631, 632 Tetradymite, 1: 10 Tetrataenite, 3: 272 Tetraferroplatinum [mineral name], 1: 628, 631, 632 Tetragonal alloys, 1: 419, 431 Tetragonal lattice, 1: 281 strains, 1: 201 Tetragonal layers, 1: 282-285, 288, 291, 295 Tetragonal pyramid, 1: 404 Tetragonal structures, 1: 339-340; 2: 74 see also C l l , . .; C16 .; L1, compounds Tetragonal system, point groups for, 1: 311 Tetrahedral intersticies, 1: 296-299 Tetrahedral-structure compounds, 1: 109 valence-electron rules, 1: 345-346 Tetrahedral-structure equation, 1: 346 Tetrahedral structures, 1: 175-176, 238, 240 Tetrahedral structures, meaning of term, 3 xli Tetrahedron, 1: 238, 239, 404, 405 frequency plot, 1: 240 Tetrahedron cluster variation method approximation, 1: 114 Tetrahedron method [for ~~illouin- one sampling], 1: 132 Tetrataenite [mineral name], 1: 631, 632, 633 Tevatron [particle accelerator], 2: 377 Textron, 3: 582x1 Texture, of TiAl sheets, 3: 631, 634 Thallium-containi~gZintl phases, 3 119, 120,121 Thallium-mercury phase diagram. 3: 24, 27 ThCr,Si, s t ~ c t u r etype, 3: 90,92, 104, 174 prediction of compounds, 3: 828-829, 830, 832 Theory of intermetallic phases, 1: 11-14 Thermal analysis, 1: 5 Thermal barrier coatings, 1: 646; 2: 490, 494, 499 Thermal behavior, 1: 1017-1029 Thermal conductivity, 1: 1017, 1025-1028 alu~inides,2: 56-57, 74, 293 effect of disorder, 1: 1025 hydrides, 2: 486 of Ni41 alloys, 3: 603 semiconductors/semimetals, shape-memory alloys, 2: 538 silicides, 2: 294 thermoelectric materials, 2: 463, 464, 466, 467, 469 Thcrmal disordcring, 1: 771, 779-7-781 Thermal expansion, 1: 1017, 1022-1025 Al-Fe-Si-V alloys, 2: 185 aluminides, 1: 1024; 2: 57-58, 134, 638 Au-Cu system, 1: 1023 shape-memory alloys, 2: 538 silicides, 2: 224 III--V/II-Vi compounds, 1: 1024-1025 vacancy-affected, 1: 561 Thermal expansion coefficient niobium silicide composites, 3: 555-556
t i t a n ~ ualuminide ~ coatings, 3: 570 titanium aluminide composites, 3: 584 Thermally ind~cedorder changes, 1: 771, 772-784 Thermal mechanical fatigue, 3: 298 Thermal parameters, 3: 814 Thermal properties, practical signi~cance,1: 1028-1029 Thermal q u e ~ c h i n1: ~ ,688 Thermal spike model, 1: 803, 808, 821 Thermal spike (track formation) mechanism, 3: 269 Thermal spray coating, 3: 509 Thermal spray processing~1: 642-643; 492 advantages/disadvantages, 1: 643 Thermal spraying, beryllides ~ r e p ~ r by, ed 3: 47 Thermal sta~ility,1: 862-866
Thermal vacancies, 3: 275 characteristics positron-i~feti~e sp~troscopy results, 3: 279-283 time-di~erentialdilatometry results, 3: 283-286 concentrations in TiAl, 3 422 experimental techniques for study, 3: 27~278 di~erentialthermal expansion technique, 3: 277 neutron diffract~ontechniques, 3 277 positron-lifetime spectroscopy, 3: 276-277 time-differential dilatometry, 3: 277-
in ternary compounds, 3 288-289 variation of co~icentrationwith ternperature, 3 276 on which sublattice formed, 3: 287-288 and yield stress, 3: 290-291 T h e ~ o C a l csoftware, 2: 192; 3: 187, 191, T h e ~ o d y n a m i cinstability? 1: 855 Thermodynamic modeling, of phase diagrams, 3 191 Thermodynamic models, surfaces, 1: 610 T h e ~ o d y n a m i cproperties, 2: 502 of alloys, 1: 15 of amorphous alloys, 1: 736-737 bond charact~ri%ation from, 1: 91-122 data sources, 3: 799-800 determination, 1: 93-97, 120, 663 future requ~rements,1: 122 electroche~cal,2: 501-502 of hydndes, 2: 476478 of liquid alloys, 1: 663, 665 phase equilibria calculated from, 1: 97-100 see also: individual properties (enthalpy, entropy, etc.) Thermo~ynamics,laws, 3 521 Thei~odynamicstability, 1: 861 Thermoelectric materials precious metal compounds, 3: 78 rare-earth co~pounds,3: 106-107
~ T h e ~ o e l a s t i cmartensitic transformation, 1: 831; T h e ~ o e l a s t i cproperties, 1: The~oelectricapplications, practice, 2: 456-460
9 2 454 Thermoelectric conversion materials, 2: 329 Thermoelectric coo~ers/conditioners, 455, 457-458, 460 Ther~oelectricdevices, 2: 329 Thermoelectric figure of merit, 2: 456,461 factors ~ffecting,2: 461-462, 468 temperature dependence, 2: 462, 463, 464, 466, 467 Thermoelectric generators, 2: 455, 457, 458-460 Ther~oelectricheat pumps, 457458,460 Thermoelectricity conferences, 2: 453-454 physics, 2: 454-455 Thermoelectric materials ications, 2: 467-469 rature then~oelectrics,2:
low-temperature the~oelectrics,2: 462-464 m-tem~~rature ther~oelectrics, 464-465 r e ~ u i r e ~ e n t2:s ,460-461 theory, 2: 461-462 T h e ~ o e l e c t power, ~c 1: 943 liquid alIoys, 1: 669-671 T h e ~ o e ~ e c t rradiation ic detectors, 2: 457 The~oelectromotiveforce (EMF), temperature dependence, 1: 949, 950
T h e ~ o n u c l e a rfusion research, 2: 366, 371-376 Thermo-physical pro~erties,data sources, 3: 807-808 Thermopower, 2: 454,463, 469 T h ~ n - ~ ld~position, m amorphous phases produced by, 1: 696, 73 T h i n - ~ ~photovoltaic m cells, Thin films 3: 663, 676 appli~at~ons, co~positioncontro~,3: 781-782 deposition of, 3: 663-680, 779 chemical vapor, 3: 667-670 electrochemical, 3: 670-671 electroless, 3: 671 by ev~poration,3: 666-667, 784 hybrid method, 3: 671-4572, 673 by molecular-beam epitaxy, 3: 779788 by reactive di~usion,3: 673-676 by sputter tech~iques,3: 665-666 examples of C ~ ~ - d e p o s i t eIMCs, d 3: 669 and grain size, 3: 664 stoichio~etry,3: 664 Thiogal~atemineral name], I: 355
~t Index ~
j
e
ThMn,, structure type, 3: 96, 96, 99, 174 Th,Ni,, structure type, 3: 96, 174 Thomas-Fermi theory, 1: 58 Thoinson effect, 2: 454 T h o ~ u mIMCs, 3: 800 Thorium, recovery of, 2: 518 Three-dimeiisional maps compound-formation maps, 1: 236, 237, 264 structure maps, 1: 244, 245, 262, 357, 358, 381, 390, 393,419, 424 see also Villars , maps IIX-V compounds, 1: 16, 109, 1 18, 343 anti-ionic bonding in, 1: 176-179 catalysts, 2 647 crystal growth habits, 1: 182 effective electronic charge, 1: 179, 183 enthalpy of formation, 1: 110 heat capacity, 1: 1021-1022 phonon dispersion effects, 1: 151, 152, 153, 157 point defects in, 1: 579-580 semiconductors, 2: 324, 325, 327, 347 thermal properties, 1: 1024-1025, 1027 Threshold displacement energies, 1: 567, 570, 794, 797, 798 a-ThSi, structure type, 3: 89, 89 Th,Zn,, structure type, 3: 96, 96, 98, 174 Th-Zn system, 3: 159 TiCu, s t ~ c t u r etype, 3: 88, 89 ~iemannite[mineral], 3: 21 TiFeSi structure type, 3 91, 93, 93 Tight-binding (TB) approach, 1: 25, 33-34, 51, 56 ap~lications,1: 41, 63, 71, 611 second moment approximation, 1: 80-82 Tight-binding (TB) method, 3: 196 second-moment approximation, 3: 21 5 Tilt boundaries, atomic structures, 1: 600, 601 Time- differentia^ dilato~etry defect concentrations determined by, 3: 277-278, 281, 283-286 high-temperature data, 3: 288-289 Time-of-~ight(TOE;) mass spectrometry, 3: 214 TIMET 1100, oxidation of, 3: 571 Tim~temperature-transformation (TTT) diagrams, 1: 747, 835, 836; 2 135, 136 TiMnSi, structure type, 3: 92 Tin amalgam coatings, 2: 408, 520-52 Tin-based bearing alloys, 2: 592, 593, 594 P-Tin structure type, 1: 183, 184, 277 see also Gray tin Tinidur [superalloy], 2: 6 TiNi, structure type, 3: 88, 89 TiNiSi structure type, 3: 91, 93, 93 prediction of incidence, 3: 829-830, 831,831, 832 Tin plate, 2: 507, 518-519 Tint, meaning of term, 3: 242 Titanium, ion tracks in, 3: 267, 268 Titanium alloys, b~ryllium-contain~ng, 3 42 titanium aluminide alloying additions to, 3: 62, 597 available alloys, 3: 481-482 bor~de-reinforced creep behavior, J: 318-319, 320
1013
~ casting of, 3 591-602 coatings for, 3: 575, 577-580 composition, 3: 477
duplex creep behavior, 3: 318 fatigue crack growth in, 3: 335 microstructure, 3: 318, 478
forging of, 3: 635
lamellar creep behavior, 3: 317, 318 fatigue crack growth in, 3: 334, 335, 347 e properties, 3: 328, 347 structur~,3: 318, 478, 619,626,
microst~cture,3: 318, 478 oxidation of, 3 481, 575-577 oxidation-resIst~ntcoatings, 3: 577-580 powder ~ e t a l l u ~ gapplications, y 3: 648-653 p r o p ~ ~ofi ~various s alloys, 3: 482 reliabil~tyconsiderations, 3: 483-485 sheet material mechanical properties, 3: 633-635 rolling of, 3: 631-633, 651-652 stiffness, 3: 478-479 strength, 3: 479 synthesis, 3: of 648-653 twinning in, 3: 411, 413-414, 424 ~orkability,3: 651 see also AlTi alloys
atomistic calculations, 3: 772-774 casting of, 3: 591402 creep behavior, 3: 317, 318 d~sloca~~ons in, 1: 530-531 fatigue crack growth in, 3: 334-338 environmental effects, 3: 347 fatigue properties, 3: 328-329 forming of, 3: 617-638, 639 hot-working limits, 3: 626 hydrogen-embrittlement susceptibility, 1: 973-974 joining of, 3: 510 mechanically alloyed, 3: 754, 756-757 microstructural control in, 1: 859, 865
369, 370, 371, 372, 621-622 sulfidation of, 3: 716-717 superplastic f o ~ i of, n ~1: 657 supe~lasticity,3: 637, 762 synthesis of, 3: 756-757 twinning in, 3: 411, 413-414, 422, 424425, 621 see also AlTi- . . .; AlTi~-basedalloys
1014 Titanium-clad steel plate, 2: 523 Titanlum disilicide film, 3: 675 Titaniuin silicide coatin Titan~um-transition~me~dl quasicrystals, 1: 454, 457, 463 Titanium trialuminlde, 2: 147-170; 3:
167-168. 188 LI,-structure conipounds, 2: 155-167, 168-169 twinning in, 3: 414 sec also Al,Ti alloys; A1,Ti-based compounds Ti-Zr system, 1: 740
Tool-steel replaceinent, 1: 748 Topological disorder, in quasicrystals, 1: 463 Topological short-range ordering, 1: 744, 745 Topologically close-packed (TCP) structures, 3: xli, 415 see also Laves.. .; Mu.. .; Sigma phases Toughening, and twinning, 3: 426-427 beryllides, 3: 43, 44 high-temperat~reintermetallics, 2: 240, 24 1 minimum for critical components, 3: 495, 550 room-temperature toughness required, 2: 249 titanium aluminides, 2: 78, 83, 85, 92, 120-1 22 zirconium alu~inide,2 243-144 see also Fracture toughness Toughness improverneat, 1: 860, 862, 927 Toxicity beryllium, 3: 49 mercury and amalgams, 3: 21 Trace-element alloying additions, 1: 898, 899 Track formation, 3: 266, 269-271 modeling, 3: 269-271 Track registration, 3: 266-267 Tracks in conducting materials, 3: 267-269 in IMCs, 3: 265-267 USCS, 3: 271 -272 in Lexan polycarbonate, 3: 264 in orthoclase, 3: 264 phenomenology, 3: 263-265 in silicate glass, 3: 264 see ulso Ion tracks 1: 857 T ~ d n s f o ~ a t i ohierarchy, n Transformation~~nduced plasticity (TRIP), 3: 78 Transfoi~a~ion-induc~d twinning, 3: 403, 410,431 Transfo~ations applications to materials, 1: 845-846 ductility improvement, 1: 845-846
~ ~ ~ jIndex e c t optical-memory applications, 1: 846 bainite, 1: 835-842 crystallograp~ic,1: 827-946 martensitic, 1: 143, lG3, 715, 829-831; 2: 54-55, 96 massive, 1: 843-844 massive-like, 1: 844 in titanium aluminides, 2 96-101 Transient liquid phase, and densification, 3: 725 Transient nucleation models, 1: 698 Transient oxidation, 1: 980 transition to steady-state oxidation, 1: 980, 998 Transistor applications, 2: 213, 231, 327, 330, 332-335 Transition fatigue life, 3: 329 Transition metal alloys, 3: 799 Transition metal aluminides, formation temperatures, 3: 676 Transition-metal compounds atomic volume as function of atomic fraction, 1: 115, 116 bonding characteristics, 1: 116, 920 e1ectronic”structurecalculations, 1: 137-144 liquid alloys, 1: 682--683 magnetic compounds, 2: 3 10-3 17 magnetic structures, 1: 444-445 m a ~ e t o ~ ~ l oproperties, ric 3: 528-529 magneto-optical applications, 2: 450 point defects In, 1: 579 Transition metals atomic properties listed, 1: 233, 423 diffusivity in aluminum, 2: 176 magnetic structures, 1: 441-444 solubility in alum in^, Transition-metal silicides, 2: 2 13-2 19 electrochemical behavior, 2: 505-506 formation temperatures, 3: 675 semiconductors, 2: 329-330 Translational disorder in^, 3: 245, 254 Translation-rotation coupling, 3 254, 255 Translations, on Internet, 3 870 Transmission electron microscopy (TEM) observations Ag-Au-Cu alloys, 2: 565 Al-Ni alloys, 2: 262 Al-rich dispersoids in aluminum alloys, 2: 178, 189, 190, 192 Al-Ti-based alloys, 2: 101 , 102-1 06, 163, 166 amorphous-phase production, 1: 694 antiphase boundaries/domains, 1: 496, 497, 501, 502-503, 504, 508-509 B2 aluminides, 3: 848-850 deformed quasicrystals, 3: 399-400 with energ~dispersiveX-ray analysis (TEM-EDX), 1: 594-595 Fe,Al films, 3: 786 grain boundaries, 1: 586-591 irradiation effects, 1: 806, 807-808, 816 mechanical alloying processes, 2: 599-600 pest degradation, 1: 1001-1003, see also pest degradation quasicrystals, 1: 456, 457, 458, 464, 466-467,474,478 twinning, 3: 408, 409, 413
see aZso High-resolution transmlsslon electron microsco~y Transportation equip~ent,value of IMCs, 3: 793 Transverse acoustic (TA) mode [of phono~-vibration],1: 163, 164, 1018 ~rialuminides,2 147-170 as dispersoids in a l u ~ i alloys, n ~ 2: 187-188 energy differences between LI, vs DO,, crystal s t ~ c t u r e s 1: , 65 oxidation, 1: 984, 985, 986, 994-995 stabilization of cubic structure, 1: 433 synthesis by mechanical alloying, 1: 647, 706 see also A1,Ni- . . .; A l ~ T i ~ b a s e ~ compounds; Titanium trialuininide Triangular mapping [crystal habits], 1: 185, 186 Triangulat~~n method [multicomponent phase equilibria], 1: 71 1 Tribaloy [wear-resistant] alloys, 2: 596-597 advanta~es,2: 600 compositions, 2 597 corrosive wear data, 2: 597 Tribotogy applic&.ions, 2: $91 -601 Triboplasma, 3: 743 Trigonal d ~ p y r a ~ i1: d ,404 Trigonal prism, 1: 404 Trigonal prismatic phases, 3: xli Trigonal system, polnt groups in, 1: 312 Trinickel al~iminidealloys see AIM, alloys Triple defect disorder, 3: 759, 760 Triple-defect model, 1: 559, 566, 572-573, 575, 775; 2: 54 Triple points, 3: 154 Tritides, 2: 477 T r i t i m fusion reactors, 2: 648 Tritium getters, 3: 30 True~twinn~ng, 3: 405, 410, 431 T R ~ - ~ A IV ~A A ~s~peralloy], 2: 7 TTT diagrams, 1: 747, 835, 836; 2: 135, 13G Tubular disordering, 3: 291 Tubular tin source (TTS) superconductor, 2: 361-362, 363 Tulameenite [mineral name], 1: 628, 631, 632 Tumbaga, 1: 3 Tunable lasers, 2: 332 Tungsten alloy, 3: 69 Tungsten carbide electrodes, 2: 508, 509 Tunneling microscopy, 3: 21 1, 212-214, 215 Turbine applications, 2: 44, 55, 65-67, 68, 73, 88,287, 289, 294 coatings, 2 489-499 Turbine blades, 3: 476,493,494,541, 598, GO 1 cooling of, 3 541, 542 fabrication of, 3: 602, 603 Turbine-en~neefficiency, 3 472 Turbine vanes, 3 602, 609 Turbocharger parts, 2: 45, 88 Turbochargers, 3: 503, 599-600 Tweed microstructure, 1: 774; 2: 97 Twin boundaries, 1: 587, 589 Twin energies, ab initio calculations, 1: 71, 72, 73
Subject Index Twinned cubooctahedron, 1: 239 Twinning, 1: 827, 844-845, 915-916: 3: 403-436 in Al-Ti alloys, 2: 103, 151-153 in Au-Cu alloys, 2: 564 in B2 compounds, 3: 416 in C15 compounds, 3: 415 in C40 compounds, 3: 458-460 in C54 compounds, 3: 458-460 at crack tips, 3: 419 in creep, 3: 428, 430 crystallography of, 3: 405 in DO, compounds, 3: 416-417 in DO,, compounds, 3: 4 17-4 18 in DO, compounds, 3: 414-415 in DO, compounds, 3: 415 in DI, compounds, 3: 418 defined, 3 403 dislocation models for, 3: 407-408, 42 1-422 and ducti~itylplasticity,3: 425-426, 427 in duplex mic~ostructures~ 3: 424, 428 elements defined, 3: 404, 405-407 fracture toughening affected by, 3: 426427 general literature references, 3: 436 glossary, 3: 430-431 and impLirities, 3: 424 in L1, compounds, 3: 41 1-414 in Ll, compounds, 3: 41 1 modes, 3: 407, 408 in ~ulticomponentalloy systems, 3: 427-428 iii other noii-cubic superlattices, 3: 418 and oxygen-scavengingeffect, 3: 424 and phase transfo~ation,3: 410 in plastically deformed quasicrystals, 3: 400 plasticity affected by, 3: 425-426 and point defects, 3: 422-424 and processing, 3: 405 and stacking fault initiation, 3: 419,420 strengthening effects, 3: 424-425 in s~~erlattices, 3: 411-418 and surfaces and interfaces, 3: 419-421 and tension/coiiipression asymmetry, 3: 404 in titanium aluminides, 3: 41 I, 413-414, 422, 424-425, 621 and toughen~ng,3: 426427 traiisforiiiatioii-induced, 3: 403, 410 types, 3: 403 see also D e f o ~ a t i o ntwinning; Growth twins; Transforniation twms Twinning dislocations, and interfaces, 3: 407-4 10 Twinning shear, 3: 411, 412 and axial ratio, 3: 416 Twins in C40/CrSi2structure, 3: 458-460 in C54/TiSi2structure, 3: 458-460 coherent, 3: 410 in complex structures, 3: 460-461 crystallographic elements, 3: 415 formation of, 3: 421 images, 3: 408, 409 incoherent, 3: 410 nucleation of, 3: 418-424 thickness, 3: 413 Twin toughen in^, 3 426-427
Two-dimensional angular correlation of positron annihilation radiation (2D-ACPAR) spectroscopy, I: 41, 136 II-VI compounds catalysts, 2: 647 semiconductors, 2: 325, 326-327, 348 thermal properties of, 1: 1021-1022, 1024-1025, 1027 see also III-V compounds Two-phase materials, 3: 842-847 Two-way shape-memory effect, 1: 832-834; 2: 534, 535 applications, 2: 543-548 biased actuator devices, 2: 544-545 iionkiased actuators, 2: 545-546 UB,, structure type, 1: 398 Udimet superalloys, 2: 6, 7, 8, 42 a-UFeGe structure type, 3: 95 Ultimate tensile strength Al-Fe-Si-V alloys, 2: 194 AlTi,-based alloys, 2: 113, 114 Condal alloys, 2 525 liigh-temperature mteriiietallics, 2: 244 shape-memoiy alloys, 2: 538 see ulso: ~ e c h a n i c aproperties l Ultra-fine-grained materials, 1: 748 Ultrasonic wave generation applications, 2 346 Ultrasonic wave velocity, 1: 873, 1018 Ultraviolet photoemission spectroscopy (UPS), 3: 136, 780 Uniaxial anisotropy [of magnetic materials], 2: 304, 306, 307 Unit cell axes, 3: 9-10 Unit-cell dimension ratio c/a plots, 1: 250, 25 1 Unit-ceil-parameter-radii relationships, 1: 250, 252, 253, 254 Universal equation of state (UES), 3: 766 Universities, Iiiternet listings, 3: 873 Unoccupied clectronic statcs, mvestigation of, 3: 145-150 Upper critical field, effect of uniaxial strain, 2: 357 Uraiiium pnictides, 1: 154 Urvantsevite [mineral name], 1: 628, 631, 632 USA gross domestic product, 3: 792 platinum group metal reserves, 3: 80 USAF Wright Laboratory, 3: 824 Usenet, 3: 870 V-phases, 3: xlii Vacancies, 1: 352, 559 clusters, 3: 286 constitutiona~,3: 275 dislocations as sources/sinks, I: 918 and EAM poteiitials, 3: 771 electrical conductivity affected by, 1: 954, 955 formation of, 1: 560, 564-566 energies predicted, 1: 568-569 factors affecting, 2: 61 1 volume change, 1: 561 at grain boundaries, I: 603, 918 and Hume-Rothery rules, 1: 568 iiiigration properties, 1: 560, 573-577 A15 compounds. 1: 579
1015 €32 compounds, 1: 575, 576 B11 compounds, 1: 575, 577 L1, compounds, 1: 575 L1, compounds, 1: 574-575 mobility, 1: 574, 577 and off~stoichiometry,1: 572 a i d quasibiiiary systems, 1: 719 remnant, 3: 275 structural vacancies, 1: 560-561, 565-566, 568 in ternary compounds, 3: 288-289 thermal vacaiicies, 1: 561; 3: 275 types, 3: 275 and yield stress, 3: 290-291 see also R e ~ n a nvacancies: t Therrnal vacancies Vacancy activation volumes, 3: 286-287 Vacancy clusters, 1: 574 Vacancy concentration, 1: 57 1-572, 761 Vacancy defect compounds, 3: xlii Vacancy-energy model, 1: 571, 572 Vacancy f o ~ a t i o nenthaipy determination of, 3: 276. 280 as function of composition, 3 282 values listed, 3: 281, 285 variation with temperature, 3: 282 Vacancy formation volume, 3: 276, 287 Vacancy-induced bardening, 3: 372, 373 Vacancy mechanism of diffxision, 1: 761 Vacancv-mediated tracer diffusivity, 3: 285, 290 Vacancy migration enthalpy, 3: 282 determination of, 3: 276 values listed, 3: 281 Vacancy migration process, 3: 282, 283 Vacancy migration volume, 3: 287 Vacancy-ordered phases, 1: 299, 300, 566 Vacancy supersaturation, 3: 423 Vacuum arc remelting (VAR), 1: 640; 3: 592, 619 Vacuum induction melting, 1: 639 Vacuum seals, 2: 523 Valence compounds, 3: xlii Vaience-e~ectroiic o n ~ ~ n t ~ a t (VEC), i o n 1: 346 adamantane compounds, 1: 347-348 A1,Cu-type compounds, 1: 392 close-packed structures, 1: 410, 413, 417 correlation with structure. 3: 114 Grimm-Sonimerfeld compounds, 1: 109 ~ u m e - ~ o t h ephases, ry 1: 100, 101, 105, 106, 319, 325 rare-earth compounds, 3: 91, 93 tetmhedral-structure compounds, I: 346 Valence-electron factor, 1: 232, 270 in AP-AN plots, 1: 230, 233 atomic properties grouped under, I: 23 I and solid-solution tendency, 1: 244,248 and structure mapping, 1: 241, 242 see also Angular valence-orbital factor Valence-electron number as atomic property, 1: 231, 232, 233 listed for various elements, 1: 233, 423 plot vs electronegativity difference, 1: 388 Valence-electron rules, adamantanestructure compounds, 1: 345-348
1016 Valence electrons per atom, superconduc~ingcritical/transition temperature plotted against, 1: 266, 951, 952 Valence orbitals, angular c~aracteristics, 1: 390, 420 Valency difference, grain-boundiry structure affected by, 1: 604 Value bottom-up, 3: 793-796 of TMCs, 3: 192, 793 meaning of term, 3: 793 top-down, 3: 792 Value of sector, 3: 793 Valves, exhaust, 3: 503, 515, 600-601, 635, 636 Vanadium aluminides, compressive ductility, 3: 414 V a n a d i u ~h y d r ~ d neutron ~, scattering in, 3: 253 Van Gogh's Sky (VCS) microstructure, 3: 851-854 contrast, 3: 852 and dendritic segregation, 3: 852-853 foimatioii of, 3: 851-852 Vaporization heat, listed, 3: 214 Vapor-phase deposition techniques, 1: 651-652, 773; 2: 410; 3: 6 6 ~ 6 7 0 see also G h e ~ ~ c .a.l .; P h y s i ~ vapor ~~ deposition Vapor-phase epitaxy, 2: 326 Vapor-phase siliconizing, 1: 652 Vapor-pressure ~ e t h o d s Gibbs , free energy of formation d ~ t e ~ i n e d using, 1: 93-94 Vapor species, 1: 681-682 Varley ~ e c ~ a n ~ 3s m 265 , Vdrma-Weber theory, 1: 158, 159, 164 'Vegetables' (heavy-fermion compounds), 1: 211, 214-215, 219, 221 Vertex correction, 1: 28 Vertical-cavity surface-emitting lasers, 430,431 Vicalloy [magnetic material], Vidoz-Brown dislocation mechanism, 1: 512-513, 514 Villars [compound-formation] model, 1: 234 Villars three-dimensional compoundformation map, 1: 236, 237, 264 Villars three-dimensionai structure maps, 1: 244, 245, 262, 357, 358, 381, 390, 393,419, 424 application to ternary systems, 1: 265, 266, 434-435 Virtual ordered state, 1: 44 Virtual two-way shape-memory devices, 2: 543 Viscosity, liquid alloys, 1: 668 Vitek-Ack~and--Gserti (VAC) interatoinic potential, 3: 769 Vitek-Girshick-Siegl-Inui-Yd~aguchi (VGSIY) i~teratomicpotential, 3: 772, 773 Vogel-Fulcher equation, 1: 690 Voigt approximation, 1: 883, 884 Voigt notation, 1: 874
Subject index Voigt parameter [for ma~neto-optical effects], 2: 437, 439 VOlterra process, for dislocations in qudsicrystals, 3: 379-381 Volume change on formation, 1: 92 Volume contraction, in rare-earth compounds, 3: 86-87 Volume of mixing, liquid alloys, 1: 665. 666, 674 Volumetric energy storage densities, 2: 638 listed for various compounds, 2: 643 Von Mises criterion, 1: 921; 3: 425, 426, 621, 694 Von Neumann's rule, 1: 184 Voronoi polyhedron construction, 1: 365, 479 Voter-Chen (VC) interatomic potential, 3: 769, 771 V-Zr system, 1: 740 Wade's rules, 3: 113, 114, 120, 127 Wairauite [mineral name], 1: 631, 632 Walser-Bent [phase-fo~iation]rule, 2: 604, 608, 609 Warped MT approximation, 1: 133 Warren-Cowley SRO parameter, 1: 27, 40,47, 783, 803 Waspaloy [superal~o~], 2: 6 Watch parts, 2: 651 Water, amorphization of, 1: 702 Water atomiz~tion,of aluminides, 3: 504, 655 Waveguide photodetectors, 2: 340, 342 Waveguides, optical, 2: 413-414, 425 Wave speed, in powder mixtures, 3: 742 Wear resistance Tribaloys, 2: 597 see also Bearing materials; Tribology applications Web address (URL), 3: 857, 877 Weight-reduction egects, 2 55, 73, 289, 409 Weishanite [mineral name], 1: 631, 632; 3 21 Weldability aluminide alloys, 2: 45, 202 factors affecting, 1: 266, 655 plots, 1: 266, 267 Weld-filler wire, 3: 502 Welding, 2 44-45, 128 Welding methods, 1: 655 Welding of IMGs, 3 509-510 Weld overlay process, 3: 507 Weston cell, 3: 22 Weston standard cell, 2: 651-652 White gold alloys, 2: 560, 570-572, 649 W~dmanstattenferrite, I: 835 Widemann-Franz relation, 1: 1026, 1029; 2: 136,462 Wigner-Eckart theorem, 3: 172 Wigner-Seitz cell, 1: 365 Wilson ratio, 1: 220 Window coatings, 2 41 1 Wire resistivity method, vapor pressure determined using, 1: 94 Wirkungsbereich [of atom], 1: 365 Wood surface structure notation, 3: 216 Woodford correlation for tensile ductility, 3: 637 Workability~titanium aluminide, 3: 65 1
Workability, titanium aluminide alloys, 2: 86 Work liardening see Strain hardening Work hardening, AlZr,, 2: 138-139, 142 Wrong-pair bonds. 1: 999 Wrought alloys aluinmides, 2: 77, 83, 85-86 aluminum alloys, 2: 185-1 86 superalloys, 2: 7, 8 3: 617, 626-627 W r o ~ i ~ lprocessing, it advantages, 3: 630 manufacture by, 3: 635-636 model in^ of, 3: 630-631 W,Si, structure type, 3: 89, 89 WulfFs [crystal habit] plots, 1: 172, 177, 178 Wulff's law, 1: 167 WulF s sp~cific-surfacefree-energy diagram, 1: 173 Wurtzite structure, 1: 175, 176; 2: 324 Wurtzite"s~~cture compounds, 1: 343-358 crystal etch habits, 1: 181, 182 derived crystal habits compared with observed habits, 1: 180 , separation in s t ~ c t u r e~ a ~I:s420, 422, 424,426 see aZsa Adamantane-st~uct~~re compounds Wurtzstannite [mineral name], 1: 355, 359 Wyckoff notation, 1: 385; 3: 238 Wyckoffpositions, 1: 311, 314, 317, 318, 319 Wyckoff sequence, 1: 360 Wyckoff sets, 1: 31 1, 318 W-Zr system, 1: 740 XI3 alloys, 3: 328, 336,481,482, 595-596, 597 XD~M-processed alloys, 2: 86 XDTM-processedcomposites, 1: 1008 X-ray absorption near-edge structure (XANES), 3: 147 X-ray diffraction, 1: 10, 11 amorphous phases, 1: 699, 700 liquid alloys, 1: 662-663 long-~diige-orderchanges studied, 1: 776, 777 X-ray diffraction studies dynamic disorder, 3: 255 pressure effects, 3: 162 surfaces, 3: 212 X-ray emission spectroscopy (XES), 3: 140-143, 150 characteristi~~, 3: 142 combined with XAS, 3: 146, 147 and PE§, 3: 144-145 principle, 3: 142 and XAS, 3: 145, 146 X-ray masks, 2: 41 I X-ray photoabsorption spectros~opy (XAS), 3: 149-147, 150; 3: 145-147, I50 combined with XES, 3: 146, 147 principle, 3: 145 and XES, 3: 145, 146 yield technique, 3: 145 X-ray photoelectron diffraction, 3: 785 X-ray photoelectron spectroscopy (XPS), 1: 611; 2: 481; 3: 136
Subject Index applications, 3: 780, 785 of metallic glasses, 3: 689-690 monochroinatized, 3: 576 X-ray powder di~raction,magnetic structure determined by, 3: 169-170 X-ray radiography, first used, 1: 7 Yahoo, 3: 858, 864 YAlGe structure type, 3: 91 Yan-Vitek-Chen (YVC) interatomic potential, 3: 769, 770 a-YbAuGe structure type, 3: 91, 93, 93, 95 y-YbAuGe structure type, 3: 95 Yield anomaly AlFe,, 2: 205 AlNi,, 2 24, 36 and APB energy anisotropy, 1: 498 see also Flow-stress anomaly Yield criterion, metallic glasses, 3: 694 Yield strength AI-Fe alloys, 1: 905 Al-Fe-Si-V alloys, 2: 194 AI-Ni alloys, 1: 904 Condal alloys, 2: 525 efFect of grain size, 2 79, 137, 138, 204 see also Hall-Petch relationship effect of irradiation, 2 138 high-temperature ~~t~rnietallics, 2: 240, 24 1 iron aluminides, 2: 203, 205 nickel almunides, 2: 36, 37, 38, 42, 60, 61, 598
60-41, 86, 113, 137-138, 194, 205,
trialuminides, 2: 164-166 zirconium alumiiiide, 2: 136-1 38 Yield stress anomaly platinum aluminides, 3: 73 and thermal ~ ~ c a n c i e3:s ,290-291 see also Flow-stress anomaly Yield stress, effect of orientation, Yixunite [mineral name], 1: 631, YIrGe, structure type, 3: 92 YLiSn structure type, 3: 91 YNiAl, structure type, 3: 90, 92 Young’s modulus, 1: 202, 876; 3: 241 of Al-Fe-Si-V alloys, 2: 194 of aluminides, 2 56, 74, 134, 598 calculated values, 1: 206 composition dependence, 1: 886, 887 of high-temp~ratur~ ii~termetallics,d: 245, 249 of MoSi,, 2: 294
orientation dependence, 1: 877, 878, 880 and phase enthaipy, 1: 920 of polycrystals, 1: 885 esti~ationfrom single-crystal constants, 1: 884, 885 sliape-memory alloys. 2 538 temperature dependence, 1: 888, 889; 2: 74 of trialuminides, 2: 149 YPdSi structure type, 3: 91, 93, 93, 95 YPtAs structure type, 3: 91, 93 Y-Zn system, 3: 159 Y-Zr system, 1: 740 Zeeman interaction, 3: 168, 170 Zee-Wilkes [irradiation-induced ordering/disordering] model, 1: 80~802 Zener-Holloinon parameter, 3: 623, 635, 637 Zeolite ZK-5, 1: 329 Zero-insertion-force (ZIP) connector, 2 541, 542 phase, 1: 105, 106; 3: xlii Zeta Zhang Bangwei [solubility] plots, 1: 247, 248 Zhanghengite [mineral name], 1: 631, 632 Zhdanov notation, 1: 172, 176, 345, 359, 360: 2: 532 Zhou’s information~predictionsystem, 1: 270, 272 Zimbabwe, platinum group metal reserves, 3: 80 Zinc-blende structures, 1: 343-358 crystal structure, 2: 324 filled, 3: 238 separation in structure maps, 1: 262, 420, 422,424 space groups listed, 1: 359, 360 thermal expansion data, 1: 1025 see also Adamantane . . .; Sphaleritetype compounds Zinc-centered clusters, in Zintl phases, 3: 120-1 2 I Zinc chalcogenides, as semiconductors, 2 326 Zinc-coated steel, 2: 519 Zinc, consumption and price, 3: 794 Zinc-iron compounds, 3: 794 Zinc refining, 2: 517 Zinc-sharing clusters, in Zintl phases, 3: 121, 122 Zintl compounds/phases, 1: 13-14, 110-1 12, 686 Zintl, Edward, 3: xliv, 113 Zintl ions, 3: 254 Zintl phases, 3: 113-131 clathrates, 3: 130-131 clusters, 3: 114-122
(c)
Index ~ o ~ ~ iby l ePaul d Nush
1017 heteroatomic, 3: 121-122 colored compounds, 3: 74, 238, 239, 240, 242 compared with Hume-Rothery phases, 3: 114 with delocalized bonding group 13 compounds, 3: 117-121, 123-125 group 14 compounds, 3: 115-1 17 dynamical disorder in, 3: 25 1,252,254255 fast-ion behavior in, 3: 251, 252 with fullerene-like cages, 3: 126, 127 historical overview, 3: I14 with localized bonding, 3: 128- I3 I group 14 compounds, 3: 130-131 group 15 compounds, 3: 129-130 meaning of term, 3t xlii, 113 mole~ular-orbitalcal~ulations,3: 119 networks of clusters, 3: 122-127 heteroatomic, 3: 125-127 structures, 3: 114-131 synthesis, 3: of 114 ternary, 3: 126 Zircaloy alloys, 1: 791, 813, 819; 2: 133, 144-145, 648-649 Zirconium ion tracks in, 3: 267, 268 IMCs, 3: 800 Zirconium aluminide, 2: 133-145, 649 compared with Zircaloy alloys, 2: 144-145 mechanical properties, 2: 136-144 physical properties, 2: 134, 136 processing of, 2: 136 Zirconium beryllide coating, 3: 47 Zirconium silicides, synthesis of, 3: 736 Zirconiuin trialuminide, 2: 187, 188 ZnS-type compounds crystal equilibrium forins, 1: 168, 172 enthalpy of formation, 1: 110 heat capacity, 1: 118 stacking variants, 1: 344-345 see also Sphalerite . .; Wurtzite .; Ziiic-blende Zonal twin dislocations, 3: 405 Zone refining method, 1: 643 see also Float zone processing . . ZrBeSi structure type, 3: 91, 93 ZrNiAl structure type, 3: 91, 93, 93 prediction of compounds, 3: 831, 832 ZrSi, structure type, 3: 89, 89 Zunger’s pseudopotential radius, 1: 232, 233, 390, 422, 423 Zunger’s structure maps, 1: 357, 358,419, 420, 421, 422, 424 Zvyagintsevite [mineral name], 1: 627, 631, 632
he labour and the patience, the judgment and the netration, which are required to make a good index, is only known to those who have gone through this most ut least praised part of a publication. as it is, I think it indispensably necessary manifest the treasures of any inultifarious collection, facilitate the knowledge to those who seek it, and invite then1 to make application thereof.
Formulas in Figures and Tables are indicated by italic page ~ ~ ~ b e r s , Compounds are listed in alphabetical order of their elements (except in fulleranes) and then in ascending order of subscript. Elements in parentheses are in a~phabeticalorder within the parentheses. The following letter symbols are used: D =deuterium, Ln = lanthanon element, M or TM = transition metal, MM == rnisch metal, R or RE = rare-earth element, x = another atomic species. Note that some representations, e.g. A12,Fe7, do not imply a specific conipound of that atomic ratio, as with AIFe and A1Fe3 for example, but indicate the atomic percent composition of an alloy which would contain one or more intermetallics, in this case A1Fe3.
AgAl 1: 107 Ag2Al2: 522 Ag3AI 1: 9-10, 104, 107, 319 AgAlLi2 3: 239 Ag2AITi 3: 849 AgAsMg 1: 321 AgAu 1: 41 AgAuCd 3: 55 (Ag,Au),Hgz 1: 631 AgBe,, 3: 40 AgCd 1: 105-7, 319, 535, 653; 2: 538; 3: 238, 240 AgCd3 1: 105-7 AgSCd, 1: 107, 763 AgCd21nTe4 1: 356 Ag2CeSb23 104 AgCu 1: 41 AgErGa 3: 526 AgGaLi2 3: 239 AgGaS2 2: 330 AgGa(S,Sel -J2 2; 41 1 AgGaSe, 2: 330, 410-1 1 AgGe 3: 161 AgHgl.25 3: 27 Ag1.1r-Xg0.9 1: 630 Ag1.26Hg0.74 629 Ag2Hg 3: 794 Ag2HgI4 1: 349-50 Ag2Hg3 1: 3, 321, 629; 2: 576-7, 581-6, 588-9 Ag3Hg2 1: 629 AgjHg4 1: 4 Ag9Hgll 2 576-7, 588-9 AgI 1: 172, 179-81, 183-4, 261; 3: 250 A g h 2 2: 522 Ag$n 2: 521-2 Ag3h 1: 105-6; 2: 521 AgInLi2 3: 239 AgInSe4Sn 1: 356 AgIn5Se, 1: 355. 360 Ag$S 3 250 Ag3Ll0 1: 9 AgLizPb 3: 239, 240 AgLizSn 3: 239, 240 Ag2LiSn 3: 239
AgMg 1: 8-9, 536, 565, 568, 575, 595-7, 763, 765, 875, 891, 923, 956; 2 252; 3 227, 311,312, 354, 358 AgMg3 1: 8 AgMgl2Nd2 2: 651 AgMg16Zn311: 257 Ag,Mo,Sr; 3: 833 AgZNd 1: 369-70, 375, 379 AgNdSi 1: 377 AgPS4Zn 1: 356, 360 Ag2Pd3 1: 943 Ag2Pr 1: 379 Ag2RE3Si4 1: 377 Ag2S 1: 320; 3: 250, 250, 251, 251, 757 Ag3Sb 1: 10, 286, 288, 628; 3: 872 Ag$b 1: 105-6 Ag2Se 3: 250, 251, 757 AgSi 2: 505-6 Ag2S11 2: 582 Ag3Sn 2: 507, 522, 576-81, 583-6, 589 Ag5Sn 1: 105-6; 2 521-2 AgTb 2: 935, 939 AgTb,-,Y, 1: 939 Ag2Te 1: 670; 3: 250, 251 Ag25Te751: 705 AgTeTI 1: 375 AgTh2 1: 392 AgTi I: 261, 720; 3: 412 AgTi, 1: 388 Ag2Ti 1: 387 AgVb 1: 261 AgZn 1: 107, 319, 577, 763, 765, 836,846 AgZn3 I: 107 Ag2Zn3 2: 516 Ag5Zn8 1: 107, 322, 763 AlAs 1: 110, 179, 182, 184; 2: 413, 430, 621; 3: 732, 780, 784, 785, 786, 787 A10,3AsGao.72; 342; 3 783 AlAsGa 2: 325,327,334-5,337,343,346, 413-14, 423-4, 430-1; 3: 782, 783, 784, 785, 787 A10.36ASGa0.64 2: 422 A10.4AsGao.6 2; 412 All -,AsGa, 2: 421 Al,AsGa,-, 2: 336, 425-6; 3: 783 AlAsIn 2: 335, 341-2, 421, 425 AlAu 3: 238
AIAu~1: 320, 338-40; 2: 410 AS2Au 1: 6, 117, 673; 2: 410, 572, 654; 3: 74, 75, 75, 76, 76, 231, 238, 240 A14Au 2: 253 AlllAt16 2: 559, 572-3; 3: 75 ASAuCu 3: 56, 75, 76 Al2AuLi,.3 3: 239 AIB 1: 85 A1B2 1: 251-1, 261, 273, 364, 367-8, 370, 372-6, 405, 417, 705 AI~B~C 2: O 307~ ~ A1B2Cr2 3: 7 AlBCrNi, 3: 330 AlB4Mg 1: 378 AIBGNb2 1: 378 AIBN,, 1: 973 AlBgTa3 1: 378 Al2Ba 2: 646 AI4Ba 1: 241, 252, 320, 339, 364-5, 373, 375, 404; 3: IS6 AlgBa3 3: 156 A15Ba4 3: 156 AIl3Ba7 3: 97, 156 A1Be4Fe 3: 49 (A1,Be)Ni3 3 42 AlBiMn 2: 443 Alel ICr0.06Mn,,84 3: 526 AICl:ICrO15Mni75 3: 526 A l C ~ , l C r 0 , ~ ~ M n3:2 526 :~4 A1CEr3 3: 526 AlCFe, 1: 302, 305-6, 921-2 A1C3Fe3 1: 305 AlCo.92Mn3 3: 527 AlCMn, 3: 527 A1Cl,lMn2,93: 526 AICTi2 1: 638; 3: 317 A1CTi3 3: 584 A12Ca 1: 121, 154,875,891,920; 2: 252-3, 642 A13Ca 2: 524, 653 A14Ca 2: 253, 651 AlloCaCr2 1: 339 A12Ce 1: 159-60, 219, 875, 1028 A13Ce 1: 213-15, 221, 407; 3 416, 417 AI4Ce 2: 179 A12&eCr2 3: 90 A110.5Ce2CUg.5 1: 417
1020 AlloCeFe2 2: 178-80 A113CeFe3 2: 179 AlzoCeFe52: 179-80 A152Ce4Fe122: 179 A18C~Mn41: 405,413; 2 779 A~2(Ce,Th)2: 646 AICI, 3: 580 AlCo 1: 116, 137-44, 2067,467, 506, 535, 560-1, 565, 567, 580, 694, 714, 718-19, 763, 765-6, 784-5, 874, 876, 885-7,891,908,920,922-3,925,985, 988, 995-6, 1010, 1024; 2: 237-9, 245-6, 252-3,490-1; 3: 43, 56, 219, 220,225,225,231,232,238,313,351, 354,355,357,358,359,758, 780,783, 785, 785, 787 A15C02 1: 718, 1001 A19G02 1: 718, 1001; 2: 253; 3: 676 A113C04 1: 477-8, 485, 718; 3: 676 AlCoCrY 3: 577, 580-582 Al3Coo 6cu1,4 1: 112 A165COl5CU20 1: 467; 3 756 A165C02oCLll5 1: 435 3: 701 A125C05Cu5NiloLa55 AICoo sFe0.2 1: 921-2 AlCo2Fe 2: 449 3: 107 A10 3C00 7 5 ~ m M n4Ni3.55 0 Al~~co4~M 5 1: n887 A150CO48.5Mn1.5 1: 887 Al(Co,Ni) 3: 310 AlCo, 8Nio2 1: 922 A1Co,"i3~, 1: 122 A170C01~N1151: 435,484.; 3: 384, 399,400 Al2CoO4 3: 609 A150C045Re51: 887 1: 887 A150C048.5Re1.5 A12Co 1 5 S ~ 23: 98 A1C02Ti 1: 504, 921 A12CoTi 3: 849 A145C050Ti51: 887 A148.5C050Ti1.51: 887 (AlxC0~,)2Y 3: 150 AICr2 1: 387-8, 390,404, 985, 988, 996 AICr3 1: 1001 A14Cr 1: 94
AlCrFe2 2: 449 Al(Cr,Fe), 2: 650 A128Cr4Fe68 3: 461, 461 A128Cr6Fe66 3 4-73 AlCrFeY 3: 577, 580-582 Al,,Cr,Mg, 2: 175 Ali3Cr4~o~oNb44Re6Ti23 3: 848 AlCrMX 3: 566 AlCrNb 1: 994 A14Cr2Ni61: 72 Al6cr2Ni4 1: 72 AlCrNiY 3: 577, 580-582 AlCrTi2 3: 845, 847 A165CrgTi27 2 166 A167CrsTi252: 168-70 A120Cr2Y 2: 187 AlCu 1: 886, 1020; 2: 253 AICLI, 1: 320 AlCuj 1: 9,319,957; 2: 536,538,544, 5561, 594, 650; 3: 671
C o ~ ~ o Index u n ~ A12Cu 1: 266, 340, 391, 404, 627-8, 759, 655, 694, 714, 717-19, 763, 765, 767, 764,966,1019-20, 1024,1028; 2: 253, 784-5, 815, 864, 866, 885, 905-6, 503, 595, 642, 651; 3: 5 908, 913, 916, 920, 922-3, 925, 928, A-I~CU 1:~1020 969-73, 985, 988, 995-6, 1001, 1010, A23CuZ 1: 113 E 27,144, 199-209,237,646; 3: A14C~32: 253 55, 64, 212, 219, 220, 224, 224, 225, 651, 846, 1020 A ~ ~ C U1:S321-2, ) 225,226,227,228,238,275,277,280. A17C~g1; 763 280,281,282,283,284,287, 288,289, AlCuFe 1: 161,454-5,457-8,462-3,484 290,291,307,313,332,339-342,358, 368, 373,479,496,502,503, 504,505, A17Cu~Fe3: 142, 143, 146, 146 A162oC1.1255Fe12 5 3: 384, 400 505,506,507,508,509, 510,511-513, A165Cu20Fk151:'435, 537; 3: 756 514, 516, 572, 611-614, 639, 645, A140CuloGe25Mn253: 756 654-655,713,716, 757, 780,783, 784, 10 0-5 5La 1: 417 785 ~ ~ ~ ~ ~ ~ ~457~ 6 1 , 4 5 5 ,AlFe2 1: 767 2: 188, 282 AI~CULI AlFe, I: 71, 113-14, 161, 306,444, 502, ALjCuL13 1: 475, 485, 537-8 504, 506-7, 521, 537, 564, 589, A16CULlj 1: 435, 538 766-7, 794, 806, 875, 886-7, 891, A12CuMg 2 188, 282, 651 907, 915-16,928, 946,965, 967, AlCu,Mg, 1: 410 969-71, 973,985, 988,995; 2: 62, 199-209, 217-19, 252, 306-7, 447, A14Cu8Mgs;z 3: 756 530-2,646,650; 3 225,275,279,284, AlCu2Mn 1: 15, 319-20, 935; 2: 303, 308 290, 300, 309, 320, 325, 331-332, 2: ~175M ~ ~ A ~ ~ o C U 338--339,340,346,368,369,370,372, A1Cu,Ni3 _ n 1: 122 373, 416, 451, 454, 455, 474, 496, A123C~12N1753 462 502-503,502,504,505,505,506,507, A)7,5CU17*5Ni1oZr,yj3 682 511, 513, 564, 571, 572, 583-584, A~~oCu30Ni5Zr55 3: 682 611-614,639,645,654,713-715,739, 435 A ~ ~ ~ C U ~1: O OS~~ 740, 757, 786 A1CuPd2 1: 712-14, 719 Alo.3FeO.7 1: 946-7 AlCuRu 1: 457, 463, 484 A12Fe 1: 114, 767; 3: 220, 224, 320 A161.5CUpjRU13.5 3 144, 144 A13Fe 1: 746-7, 1001; 2: 176-8, 180, 183, A I ~ ~ C U ~ O1: R 435; L ~ , 3: ~ 78, 144, 144, 148, 193-4, 253, 519, 646; 3: 320, 757 756 A15Fe2 1: 114, 766, 1001; 2: 253, 519, 646; AI~OCUI~RUI~ 3 756 3 320, 676, 7.57, 758 AlCuS? 2 325, 330 &Fe 2: 176-80, 195; 3 676, 757 AlCuTa 1: 162 A18Fel2 1: 72 A ~ ~ ~ C U ~1: O467 TM A19Fe2 2: 177 A165CLt20TM15 1: 467, 480 Al13Fe41: 114, 485; 2: 177, 179 A175C~pjViO1: 417, 479; 3: 756 A12,Fe79 3: 280 Al(Cu,Zn) 1: 627 A123,7Fe763 3: 281 A12CuZn 1: 628 AlZ4Fe7~ 3: 279, 280 AI2Dy 3 524, 525, 526, 530, 531,535 A12,,,Fe7, 3: 455 AI,Dy 1: 407 A126,9Fe73 1 3: 455 A12(DYo.~OEro90) 3: 526 Al27.@e72,2 3: 455 A12(Dy0.z5Er0:75)3: 526 Al,,Fe72 3: 455 A12(DY0,40Er0.60)3: 526 A128.3Fe71,73: 455 520, A ~ ~ ( D Y o . ~ o3: E~ o . ~526 ~) A130Fe703: 454, 455 3; 526, 535, 536, 536 A12(~yo,55Ero.45) A13~Fe6~ 3: 279, 280 A12(DY0.70Er0.30) 3 526 A139f;e613: 279, 280, 281, 282, 282, 283, A1,(~Y0.85~r0.15) 3: 526 285, 287, 287, 288, 288 Al2(Dyl-$rs) 3: 530-531. 535-536 A140Fe60 1: 573; 3: 290 Al3Dy1-,Cdx 1: 408 A145Fe55 3: 278, 283, 284, 284, 285 A12(Dyo 50Ho0 50) 3: 526 Al1.,Fe, 3: 785 A12Er 3' 524, 525, 526, 530 (Al1Fe)3Hf 3: 758 A13Er 1: 407 (Al,S"e),,La 3: 174. 177 Al2(Eroap6Gdo,l4)3: 526 AlFe2Nn 2: 449 A13Er1-,Gd, 1: 408 A15FeNb2 1: 66 Al(Ero~3Gdo,7)Ni 3: 528, 529 A~ioFe~oNb60Ti2o 3: 848 Al(Ero.4Gdo6)Ni 3: 528, 529 AlloFe2qNd703: 682 Al(Ero,46Gdo,54)Ni3: 5 19, 520, 528, 529, Al(Fe,Nl) 3: 311-312, 313, 757 535, 536, 536 3: 313 AlFeo.~N~o.8 3: 528, 529 A1(~ro*5Gd~.5)Ni Al(Ero.55~do.45)Ni 3: 528, 529 Alo.s(Fe0.2~i1.0)1: 923 A1Feo$?ig,2 3: 313 A1(Ero~6Gdo,4)Ni 3: 528, 529 Al(Feo,8Nio,2)1: 923 A1(Ero~8Gdo,2)Ni 3: 528, 529 AI(Fe,Mi) 1: 916-17, 922, 925 Al(ErxGd1-JNi 3: 535-536 Al(Feo,2Nio.8)1: 918, 919, 923 AlErNi 3: 526, 535 AlFezNi 2: 308,449 AI~Eu1: 407 A150Fe10Ni403: 288-289, 289 AlFe 1: 71-2, 113-14, 13741, 143-5, A 1 7 5 F e l ~ ~3:i ~7.56 o 204-7, 504, 506, 508, 513, 535-6, AlFePd 1: 454 563, 565, 569,572, 574,577,596,653,
C Al,Fe,2-x,3R 3: 178 Al(Fe,Ru) 3: 65 AlFeSi 2: 177-8, 180, 183 3: 784 A10.~~Fe3Si0.~7 A13FeSi 2: 177 A14FeSi22: 177 AI9Fe2Si22: 183, 193 A113(Fe,Si)42: 519 AlFeTa 2 240 AlFeTi2 3: 845, 848 A15FeTi2 1: 66; 2: 139; 3: 461 A122Fe12Ti663: 848 A123.5Fe6Ti70.53: 848 A164~e~Ti25 2 165 AlB7Fe8Tia52 157 A17(Fe,V) 2: 183, 194 A1,2(Fe,V)2Si2 2: 184
Al,3(Fe,V)3Si22: 182 (A1,Fe)3Zr 3: 758 (A1,Ga)As 3: 780, 782, 783 (Al,Ga,In)As 3: 780 (Al,Ga,In)N 3: 780 AlGaInP 2: 336 (Al,Ga,In)P 3: 780 (A1,GaJn)Sb 3: 780 A1,Ga2-,La 1: 378 AlGaLnP 2: 325 AlGaMgZn 1: 485 (A1,Ga)N 3: 780 ALg8CU4Gd 1: 417 A12Gd 1: 875, 889, 891; 2: 252; 3: 526.530 Al2Gd3 3: 528, 529, 530 A1,Gd 1: 407 A1,GdXHo1-, 1: 408 A13Gd,L~1_-~ 1: 408 AlGdNi 3 528,529 A13GdxTbl-, 1: 408 A13GdXTml- 1: 408 AlGeMn 2: 309,448-9 A150Ge30Mn203: 756 A 1 5 ~ G ~ 1 0 M ~ 23:0 ~756 i~o (A1,Ge)Nb3 2: 360 A10.75Geo.25Nb32: 353 AlHf 1: 375 A13Hf 1: 65,433, 647, 706-7, 727; 2 148, 157, 187-8, 606, 654; 3: 676, 758 AlHf2Nb 3: 848 Al~0~f20Nb60W~0 3 848 A1HfNi2 1: 69; 3: 316 Al2Ho 3: 526 Al3Ho 1: 291,407 (A1,In)P 3 780 AlIr 1: 116, 138-40, 144, 714, 985, 988, 996; 3: 55, 58-59, 78, 784 A131r 1: 996 Al2IrLi 3: 239 A1lrl-,Ni, 3: 58 AlIrRu 3: 78 AIIrl,Rti, 3: 58 A1241r37R~39 3: 67 A l 5 0 1 ~ s ~ , R3: ~ ,59 Al2La 1: 216, 378,673, 875, 887,889, 891, 920; 2: 252 A13La 1: 65, 407 A15La2 1: 378
O Index ~
A1,,La3 1: 328-9 AI3LaNi2 1: 4 13 A1,yLaNi5-, 2: 509 AlLi 1: 41-2, 62, 67-8, 70, 566; 2 502, 510, 648; 3: 251 Ai0.51L10.49 3: 252 A12Li3 1: 68, 70; 3: 204 A13L1 1: 47, 70, 875-6; 2: 188, 190-1, 195, 282; 3: 202 A14Li9 1: 70; 3: 204 A1Li2Pd 3: 239 A12LiPd 3: 239, 240 AlL12Pt 3: 239 AI,LiPt 3 239 A1,LiRh 3 239 A12LiRu 3: 239 Al+u 1: 407 Al2Mg 2: 253 Al3Mg2 2: 253, 642; 3: 757 Alg8Mg5 2: 253 A112Mg17 1; 322; 2: 253; 3: 757 A12Mg04 1: 266; 3 7, 267 AlMgZn 1: 471 AlMg4Znll 2: 282 A16MgllZnl 1: 265, 267 A1,Mg3Zns-,, 3: 756 A1Mn 1: 15,4568,460, 462-3,465,467, 469-71, 483,714, 719; 2: 303, 308-9, 3 17; 3: 757, 784 AI3Mn 1: 477-8, 1001 AI4Mn 1: 465 A16Mn 1: 404, 465; 2 253; 3: 757 Al12Mn 1: 71; 2 186; 3: 757 A I ~ ~ M1:P480 ~I~ A10.4Mn0,6Ni31: 970 A16Mn2Ni41: 72 A I M ~ , N I ~ -1: , ~122 AlMnPd 1: 454, 457, 467, 470 A170Mn10Pd203: 756 A170,4Mn8.4Fd21,2 3: 382, 382, 383, 384, 400 3: 397, 398, 399 A170,5Mn8.5Pd21,0 AlMnSi 1: 472,475 A19Mn2Sil.*1: 475; 2: 186 A19Mn2Si2 1: 473, 475, 477, 483-5 Al15Mn3Si22: 182, 186 Al61MnloT129 2: 165 Al,7Mn*T!z5 2: 157, 165 A168Mn9Ti232 165 AlMo, 1: 638, 996; 2 352; 3: 845 AI~MO 2: 525 A13Mo 1: 1001; 2 253 A14Mo 3: 757, 845 Al5Mo 2 627; 3 757 AlgMo3 1: 288-91, 294, 996; 3: 757, 845 A112M0 1: 473; 2: 627; 3: 676, 757 A122*5M05Nb~5Ta5Ti52.5 3: 848 Al10Mo~0Nb60Ti20 3: 850 3: 848 A~15Mo15Nb40Ti30 Al,7Mo17Nb32Tj343 850, 850 A118.5X/lo5Nb24.5Ti523: 848 3: 848 A124,5Mo10~b16Ti49.5 A125M04Nb4Ti59V4 3: 848 Al(Mo,Nb)(Ti,~r)23 845-847, 854 A117M017Td32Ti343: 848 A120M0~0Ta1~Ti5~ 3: 848 A120M020Ta~0Ti40 3: 848 AIMoTi2 1: 860; 3: 845,846,847,848,850 A123.5Mo~Ti70.53: 848 Al25M020Ti~~ 3: 848 A125M025Ti503: 848, 849
~
~
~
~
1021 ~
A12(M0.26T~0.74)1: 924 A166M9Ti251: 924 AIN 1: 110, 179, 184, 352; 2 327, 345; 3: 656, 657, 663, 669, 780, 786 AlNTi, 1: 638; 3: 653, 843-844 AlNb2 1: 409,411-12,414, 856,984,988, 994-5; 2: 95,244-6, 249, 253, 650-1; 3: 757, 845 AlNb, 1: 136,638,640,788,803,969,984, 987, 994; 2: 244, 246, 253, 290-1, 352-5, 357, 360, 366, 372, 376, 384, 650-1; 3: 264,332-333,342,346,461, 583, 676, 73 1, 759, 845 A13Nb 1: 65-6,433-4, 530,768,916,922, 969, 985, 988, 992, 994-5, 1000-6, 1010; 2: 147, 157, 169, €87-8, 240, 245-6, 248-9, 253, 291, 651; 3: 414, 645, 676, 731, 757, 758, 845 AlNbNi 1: 917, 927,995; 2 240,294, 650-1; 3: 757 AlNbNi, 1: 69, 927; 2: 296, 650-1 A137Nb13Ni37Ti~3 3: 737 Al45Nb5N145T?53: 737 A149.25Nb0 5N150Ti0.25 3: 737 A149.5Nb0.75Ni49.25Ti025 3: 737 Al,(Nb,Ta) 1: 1001 Al(Nb,Ti) 3 583 (A1,Nb)Ti 3: 739 AlNbTi2 1: 538, 844, 858, 860; 2: 91, 94, 95, 290, 292; 3: 337, 370, 474, 494495, 584, 845 AlNb2Ti 3: 849 Al(Nb,Ti), 3: 583 Al(PGb,Ti), 3: 739 A12NbTi 3: 849, 850 Alz(Nb,Ti) 1: 992 A115Nb35Ti50 3: 849 A115Nb45Ti40 3: 453 Al15NbyjTf303: 848 A115Nb75T~19 3: 453, 848 A119,5Nb25Ti53.53 853, 854 A120 5Nb18.0T161.5 3: 84x A12,NbzlTi58 3: 853, 854 A125Nb25TiSO 3: 847, 848, 849 A125Nb50Ti253 849 A15,Nb2,Ti25 3: 849 A115Nb51Ti15Zn3: 853 A115Nb54Ti30Zr 3: 852, 853, 854 A125Nb55Vz0 3: 461 AlNbZr2 3: 848 A13Nd 1: 407 AlNi 1: 41-2, 62,69, 71-2, 82-3, 85, 114, 121, 137-44, 163, 2067,467, 501, 504, 506, 508, 512, 523, 535-6, 541, 544, 565-6, 568-70, 574, 577, 580, 595, 610, 615-17, 620, 640, 649, 651, 655, 694,714,718-19, 763, 765-6, 784-5, 794, 797-8, 811, 815-16, 8454,860,875,880,882,884-5,891, 895. 904-6, 908, 91 1-13, 916-18, 920-8, 956,969, 9814,987, 989, 991-2,994-6,1001, 1005-7,1009-10, 1018, 1024, 1027; 2: 5, 18,27, 53-72, 105, 237, 240, 252, 257, 277-8, 284, 289-90, 293-4, 296, 307-8, 490-1, 496, 538, 556,599,607,609,621,647, 649-51; 3: 43, 44, 55, 56, 60, 64, 212, 217,219,220,224225,225,226,227, 227,228,228,231, 232,237,238,275, 277,280,281,284,286,289, 290,297, 300,304,305,307,308,309,309,313,
I022 316, 317, 319, 320, 325, 326, 327, 330-331,341-342,347,348,351,352, 365, 366,366, 367, 368,370, 371, 372, 372,437,451,452,453,476,479,494, 502,502,513,542,543,544, 563, 564, 566, 583,583, 602-604, 606, 607, 656657,717,728,729, 730,737,739, 757, 757,758,763,768-772, 780,783, 784, 784,785, 785,787, 794,842, 843, 874, 874, 875 AlNi, 1: 65, 68-9, 71-2, 82.-8, 91-2, 114, 121-2, 155-6, 204-7, 222,498, 502--5, 50'7-8,511-13,521-4,526-9,539-41, 544-7, 550-2, 564, 570, 573, 586-8, 590-5, 600, 602-3, 610, 616-18,620, 638,640, 643, 646,649, 651, 653, 655, 725-7, 759, 766-7, 773, 778, 784-8, 794, 798, 803, 808, 845, 859-60, 866, 874-6, 879-82, 885, 887, 891, 896-901,9034,908,914--15,917-18, 920-5,927--8,955,967-72,981-4, 986, 989, 1007, 1019-21, 1024, 1026-8; 2: 3-15, 17-51, 55, 74, 137, 139,142-3,162,164-5,167,214,238, 252, 257, 266-8, 270, 272, 283, 290, 296,496, 498, 598, 607-9, 650-1; 3 64, 66, 69, 201, 217, 218-219, 220, 221,227,228,275,279,281,290,298, 302, 307, 307, 308, 309, 309, 317, 319-330,325,326,328,341,346, 347, 348,352,353,354,35.5,359,365,366, 368,369,370,371,372,372,373,404, 410,411,422,437,440,442,443,462, 462,474,475,492-493,501-502,502, 505-506,506,507,509-5 10,511,512, 513,515, 516, 541, 542, 543,563,583, 604-61 1,639,645,655-656,7 11-7 13, 717,731,736, 757,758, 759, 760,767, 768-772, 785, 795, 843, 874,874 A13Ni 1: 65, 69, 114, 650, 773, 920, 1001; 2: 55, 253, 596, 607-9, 651, 653; 3: 676, 757, 757, 785 A13Ni2 1: 112-14, 566, 649, 816-17, 1001; 2: 55, 253, 491, 607-9, 651; 3 566, 757, 758, 785 A13Ni5 1: 68-9, 114; 2: 607-8 A14Ni3 1: 114 A123 5Ni76.53: 279, 280 A124Ni76 3: 463 A125Mi753: 280, 281 3: 279, 280 A125,9Ni74.1 A148Ni52 3: 283, 285, 286, 286 AlSoNlso 3: 279, 737 A1~$?!47 3 278, 285, 285, 286 A125N~20La55 3: 701 (Al,Ni),Nf 3: 758 ~Al,Ni)7Hf22: 296 AllJ?i13Sin 1: 121 Al(N1,Pt) 3 567-570 A1NiPt2 3: 66 AINi,Rul-x 3: 60 Al2NiRu 3: 60 AINizTa 1: 69 A124.7Ni74.3Ta3: 463 A1Ni2Ti 1: 69, 504, 507, 718, 773, 860, 921-2,925; 2: 62, 294 Alo 4Ni3Tio,63: 307; 3: 843444, 854 Alo.SNi3Tio.53: 307, 31 1 Alo,~Ni~T10.4 3 307
A10 gNi3Tio,23: 307, 31 1
A16iNi8Ti252: 166 A13Ni2U 1: 216 AlNi2V 1: 69 Ai4Ni6V21: 72 A16Ni4V2I: 72 A1Ni2Zr 1: 69 (Al,Ni),Zr 3 758 A1203 1: 639; 2: 29, 61, 81, 124, 168, 207-8,295; 3: 502, 504,509,564, 567, 568,573,576,577,578,580,581, 582, 583,583,584,585,608,610,612,645, 657, 708, 711, 716, 717, 736, 737 A1203.Be0 3: 45 AI3O2Tis 3 575 AlOs 1: 13840, 144; 3: 784 Al,Os2 1: 320, 339 A113oS4 1: 404 AIP 1: 16, 110, 179, 182, 184, 349 AlPd 1: 64, 137-40, 144, 712-14, 719; 2: 622; 3: 237, 676, 757 AlPd, 1: 719 A1,Pd 3: 676 Al3Pd2 1: 112-1 3; 3 584, 676 Al4Pd 2: 622; 3 78, 676 Al,(Pf,Si) 1: 117 A13Pd2IJ 1: 216 Al3Pm 1: 407 A1,Pr 1: 407 AlPt 1: 13740, 144; 2: 621; 3: 55 A1Pt2 1: 996; 3 74 AlPt3 1: 512,549-50,874, 876,996; 2: 23, 162; 3: 72, 72, 73 A12Pt 1: 117, 428, 985, 988, 996; 2: 491-2, 495-7, 621; 3: 75, 75, 76, 238, 567 AI3Pt2 1: 112-13; 3: 676 AI$%j 1: 996 A14Pt 3 676 A1,Pu 1: 291 AlRe 1: 714; 3: 784 AlRh 1: 116, 138-40, 144; 3: 784 AlRu 1: 71-2, 116, 137-41, 1434, 196, 204-7, 506, 535, 595, 714, 718- 19, 788, 885-7, 891, 906, 922, 985, 988, 996; 2: 237-8, 243, 2454, 249-53: 3 55-56, 58, 59, 60-62, 65, 70, 78, 219, 220,272,351, 354, 358,359,370,370, 371, 757, 759, 784 A12Ru 1: 718; 3 65 A13Ru 3: 58 Al&u 1: 718 A113Ru4 1: 718 A147Ru53 3: 62 A1xRul-x 3: 78 AIyjR~SoSc251: 887 Al4oRu5oScio 1: 887 A143R~52S~5 1: 887 A I ~ ~ R u 1: ~ ~887 SC~ AI$, 1: 360; 3 711, 716 Al2S4Zn 1: 355, 360 AlSb I: 16, 110, 118, 121, 152-3, 172, 178-9,1814,763, 1001,1027; 2: 421, 639, 642; 3: 161, 780 AlSc 1: 714 A13Sc 1: 65, 72, 499, 505, 876, 896-7; 2: 139, 167, 282; 3 444 A12Se3 1: 360 AlSiCoCu 1: 454, 465-8, 470, 483 (A1.Si)Feq 3: 780. 784 iAlb,o;Si&)2Mo' 3: 459 Al3Sn7 1: 407
AISn 2: 253 A14Sr 2: 253 AlTa, 1: 986, 988, 997; 3: 676 A13Ta 1: 65, 433, 986, 988, 997, 1001; 2: 253, 624, 627, 654; 3: 583, 676 (A1,Ta)Ni3 3: 302, 310 AlTaTi, 3: 848 A117Ta25Ti583: 848 A12sTa2sTi503: 848 A l ~ o T a ~ o T i 4 ~3: Z r853, 3 ~ 854 A1,Tb 1: 407 AlTc2 1: 390 AIT112 1: 392 Al3Th 2: 253; 3: 416 AiTi 1: 41, 68, 70, 72, 88, 113, 200, 205-7, 407, 506, 510, 527, 532-4, 540, 543, 551, 638-9, 6444, 649-51, 653-5, 720, 862, 885, 907,915-16, 918-19, 925,927-8,969-70, 974, 984, 987, 991-4, 1007,1028; 2: 73-90, 166, 170, 253,290-1; 3: 226,227,275, 279,281, 290,297,298, 300,302,305, 307,308, 312, 317,318,318, 319,320, 321, 325, 326,327,328,336337,346,346,347, 348,351,365,366,368,371,407,408, 408,409,409,410,411,412,413,414, 419,420,421,422,423,424,425,425, 426,427,428,430,437,442,445,446, 446,447,464,474,475,476,477485, 491,502,503, 507, 509, 510, 513,515, 542,570,571,574,575,577-578,581, 581,583,584,585,591-602,617438, 639440,645,647,648-653,716-717, 756, 757, 757, 762, 763, 772-774, 845 AlTi, 1: 65, 70, 113, 119, 433, 527, 534-5. 539,541-6,552,63840,647,649-50, 653-7, 772, 859, 928, 970, 973, 984, 987, 991-3, 1007, 1018; 2: 75, 77, 91-131, 168, 287, 290-1; 3: 227, 275, 279,290,326,328-329,337-338,346, 368,416,417418,419,421,424,425, 425,426,446,450,4.52,463,474,477, 494495,502,571,578,581,581,583, 583, 584,618,623,632,639,648,649, 731, 738, 756, 757, 762, 773, 845 A12Ti 1: 68, 113,639,924,994; 2: 166,477; 3: 570, 577,578 A13Ti 1: 65-6,70, 113, 288-91, 294, 407, 4334,499,503,505,508,528,530-3, 540, 542-3, 546, 439, 647, 650, 707, 816, 845, 874-6, 878-80, 882, 885-6, 907, 916, 924, 969, 984, 987, 991-2, 994, 1010; 2: 23. 124, 147-73, 187-8, 253,290,606,6234,654; 3: 199,370, 414,424,425,425,428,444,446,447, 448,449,495-496, 570,571,577,578, 673, 676, 736, 756, 757 A15TJ2 1: 113, 639 AI5T1? 3: 460 A111T15 1: 113 Al22.pfTi771 3: 279 A133.6T166.4 3: 281 A134.4Ti6s.63: 279, 280 A148Ti521: 540; 3 279, 280 A15k.5FI?4F.5 3: 279, 281 (A1,Ti)Ni 3: 757 (Al,Ti)Ni3 3: 307, 308, 313, 312, 404 A16Tig0V43: 286 A12S.0T161.5V13,S3: 848 AlTM 1: 454 A13TM 1: 407
Compound Index A12U 1: 214, 219; 2: 648 A13U 1: 815-16 A14U 2: 648 A1V 2: 352 Al3V 1: 65,433, 530, 542; 2: 155, 188,253: 3: 414, 676 A13(Fe,V) 2: 184 A16V 2: 184 A17V 2: 183, 184, 193 A18V5 1: 885, 891; 2: 245-6, 252-3, 523 AlioV 1: 1021; 2: 184, 193; 3: 676 (A1,V)Ni3 3: 310 A13(V,Zr) 2: 185 AbW 3: 675, 676 A15W 1: 291-2 A1,2W 1: 329, 332; 2: 622-3; 3: 675, 676 A12Y 1: 875, 883, 885, 891, 920, 1001; 2 245-6, 252-3 A13Y 1: 65-6 A13Yb 1: 407 A1Zn2 1: 627 AlzZn 1: 627 (Al,Zn)4,Mg,2 1: 472; 3: 756 AlZr 1: 997; 3: 757 AIZr2 1: 997; 2: 135-6, 142,253 AlZr, 1: 305, 526, 528, 694-5, 791, 794, 803,806-7,809-10,815,817-18,897, 915, 997; 2 22, 133-46, 253, 649; 3: 265, 271,444,474 Al2Zr 1: 109,883,885,891; 2: 245-6,252-3 A12Z~32: 135, 253 Al3Zr 1: 65-6, 288-92, 294, 306, 647, 706-7, 727, 874-5, 880, 882-3, 885--6, 891; 2: 148, 157. 175, 184, 187-92, 195, 245-6, 252-3, 282; 3: 199, 676, 758 A13Zr2 1: 404; 3: 757 Ai3Zr4 1: 405, 409, 412, 414, Plate 1 Chapter 17 Vol. 1 A1,Zr5 2: 135, 253 A14Zr3 1: 41 1 AsAuCa 1: 375 As2Cd 3: 156 As2Cd3 3 156, 161 AszCdGe 2 330; 3 161 AszCdS4 3: 827 AsCo.2: 517; 3: I56 A s C O ~3: I56 AS~CQ 3: ~I56 As~CO 3: ~I56 A s ~ C O1: 329, 332 AszCr 3: 161 A s ~ C 3: S ~129 AsCU, 2: 517, 521; 3 34 A s C U ~ S1:~174, 240, 353, 359 As4CuGaCe2 2: 323 ASCLIS1: 240 AsCuSe 1: 240 AsCuSiZr 1: 320 AsEr 3: 780, 784, 786 As(Er,Sc) 3: 780, 784 AsEr,Scl_,, 3: 786 As2Fe 1: 630 AsFeMn 3 161 As2FeZn 2: 323 AsGa 1: 110, 151-4, 157, 172, 174-5, 178-82, 184, 191, 343, 350. 352, 578, 580,651,875,953,958,1022,10254, 1029; 2 253, 323-7, 330, 332-7, 339, 342-7,407.412-17, 421-6, 428, 430-2, 614, 621, 623, 626-7; 3:
97, 161, 213, 543, 663, 667, 669, 670, 672,732, 76'0,782,783,784,785, 785, 786, 787, 794, 796 AsGaIn 2: 335. 341-2, 420 As(Ga,Iii) 3: 780 AsGa0.471n0.532: 340, 429; 3: 783 AsGa0.741n0,262: 425 AsGaInN 3: 780 AsGaInP 2: 325, 336-8, 346, 418, 421-1, 425-6 As0.61 Ga0.291n0.73P0.39 2: 427 As4Ga3K3 3 128 AsGaNi? 2 621 Aso.lSGS;P0.852 325 As0,,GaPo.4 2: 325 (As,Ga)Sb 3: 780 As4GeIn2Zn 1: 356 As2GeZn 1: 351, 356 ASH, 3: 781 A s h 1: 110, 172, 178-9, 181-4, 350-1, 356, 707; 2: 324, 327, 332-3, 344-5, 418, 428; 3: 161, 780, 783 As31n2K33 128 ASK 3 156 ASK, 3: I56 A s ~ K3 156 As4K-3 3: 156 As4K-j 3: 129, I56 As,,K, 3: 1.56 AsLi 1: 242 AsLiMn 2: 321 AsLu 3: 780, 784 AsMn 2: 444; 3: 161 Aso.91MliPo.0, 3: 326 Aso,92MnPo,083: 526 (As,N)Ga 3 780 AsNi 1: 10, 112, 242, 250-1, 261-3, 303, 404, 566; 2 517, 627 ASP 3 670 (As,P)Er 3: 780, 784 (As,P)(Ga,In) 3: 780 AsPd, 1: 627 As2Pd 3: 156 As2Pd5 1: 629; 3: 156 As2PdllSb2 1: 628 AsPdSTl 1: 321 As2Pt 1: 628 AsjPt2 2: 523 As4Rb5 3 129 A s ~ 2: S ~412 AsSb, 1: 629 AsSi 3: 156 As2Si 3: I56 AsSn 2 432 AsU 1: 154 A~23V772 352 AsYb 3: 780, 784, 786 AsZn 3: I56 As2Zn 3: 156 A S ~ Z 3: ~ ,156, 161 AuBe, I: 405, 407, 410 AuBe12 3: 40 Au2Bi 1: 629; 2: 432 A u C ~1: 421 AuCd t: 102, 107,242,261,286-9,2934, 407, 5745,763,765, 830,875; 2: 529, 538, 642; 3: 28, 29, 54, 55, 56, 412 AuCd, 1: 107, 763 Au2Cd 1: 105-6 Au,,,Cd 1: 291 Au2,gCd 1: 286-7, 291
1023 Au3Cd 1: 286-8, 291-2, 294; 3: 28 Au3.SCd 1: 286-7 Au42CdI2 1: 287-4 Au72Cd261: 287-8 AuCe I: 261 Au2CeSi2 1: 2 15 Au4Gr 1: 727 AuCs 1: 319, 322, 670, 682, 953 AUCU1: 3, 10, 41, 67-8, 242, 245, 261, 288-91, 293-5, 304-5, 406-7, 504, 510, 560, 567, 572-3, 627, 630-1, 720-1, 763, 774, 784,794-6, 947, 1019, 1023, 1028: 2: 559, 561-9, 572, 649, 651, 653; 3: 28, 54, 77, 218, 412, 413, 422 AUCUJ1: 3, 10, 65, 68, 88, 160, 245, 266, 268, 288-91, 293-4, 304-6,405-9, 498-9, Pate I Chapter 17 Vol. 1, 503-5, 508, 528, 560, 564-5, 567, 572-4, 577-8, 580, 589, 600, 603, 613-16,618-19,627,725-6,763,773, 784, 793-6, 800-3, 806, 820, 844, 874-5,888-9,891,897,908,913,915, 923-4,946-7,1018,1020,1023; 2: 22, 139, 252, 561-3, 568; 3: 54, 140, 141, 217,217,218,221,222-223,226,226, 228,265,280,290,364-365,364,370, 404, 410,411, 761 AuCuj 1: 407 A u ~ C U1: 65, 68, 573: 2: 562-3 AuCuzPd 1: 305 AuzCuPd 1: 305 (Au,Cu)Pt 3: 54 AuCuZn2 2: 538; 3: 56 AuDy 1: 261 Au6Dy 2: 570 AuEr 3: 261 Au4J3 2: 570 AuFe13Nd63: 100 AuFePd2 1: 947 AuGa 2 653 AuGa2 2: 621, 653; 3: 75,238, 780,785 AuGaL12 3: 75, 239 AuCa2Li0 3: 239 AuCd 1: 261 AuGe3: 161 AitNf 1: 63 Au,Hg 2: 518 AuWo 1: 261 AUWGHO2: 570 AuIn 2: 649; 3 28, 29 Au1112 2: 431, 649, 653-4; 3: 75, 238 Au21n3 1: 113 Au$n 3: 28 Au411i 1: 105-6 AuInLi2 3: 75, 239 AUI~~L 3: I239 ~.~ Au3,41n39,8Na233: 126, 127 AL& 1: 63 Au2K,,Nal2Tl48 3: 120 Au$18T1203: 119, 120 AuLiMgSn 3: 240 AuLi2Pb 3: 75,239, 240 AuLiSb 3: 7.5, 239, 240 AuLi2Sb 3: 239 AuLi2Sn 3: 75, 239, 240 AuLu 1: 63 AuMg3: 28 Au3Mg 1: 291 AuMgSb 3: 239
1024 AuMgSn 3: 75,239, 240 AuMn 1: 261, 445,935; 3: 28, 56 Au2M11 1: 445 Au4Mn 1: 727-8 A u ~ j M n1:~288-91 AuzMnSb 1: 950 AuNb-3 2: 352 Au2Nb 3: 759 AuNd 1: 261 AuNiTi 3 56 AuOs 1: 63 AuPb2 1: 121, 393; 2: 432, 605, 642 AuzPb 2: 432 AuPd 2: 649 (Au,Pd),Fe 2: 652 (Au,Pd)3Mo 2: 652 {Au,Pd),V 2: 652 (A~,Pd,Pt)3(Pb,~n) 1: 63 1 (AuPd)Te2 1: 705 AuPm 1: 261 AuPr 1: 261 AuPt 1: 63 AuPt4U 1: 214 AuRb 1: 319, 322,421 AuRe 1: 63 Au25Sb75 1: 705
Au75Siz5 3: 681 AuSm 1: 261 AuSn 1: 112, 121,764; 28, 29 AuSn2 3: 28 AuSq 2: 654 Au5Sn 1: 286, 288, 291-2 Au6Sn 1: 105-6 AuTa 1: 63 AuzoTa- 8o 2 352
Au30Te70 1: 705 AuTi 1: 261, 714-17; 3: 56 AuTm 1: 261 A u ~ U2: 570 AuV, 1: 638 Au4V 1: 727, 935 Au24V-762: 352 AuW 1: 63 AuYb 1: 261 AuZn 1: 4, 121, 319,673, 875; 642; 3: 27, 28, 29, 30, 56 AuZII:, 1: 336 Au3Zn 1: 763; 2: 572
B2BaNi2 1: 415 B6Ba2Ni9 1: 415 B4C 1: 404; 2: 208, 294; 3: 669 B ~ C 5 C r 4 F e 6 7 ~ a 4 ~ 03:4 P702 * BbCa 1: 404 B2CaRh2 1: 415 B C C ~ ~ L u N i ~3:- x103 BCCuxNil-~Y3: 103 B2Ce2C051: 414,416 B2CeCo3 1: 407,412-17 B ~ C Q C O1:~413,416
B4Ce3Coll 1: 413,416 B4CeCo41: 266, 329 BCeCo4 1: 413, 416 3: 268 BCeCo4 3: 7 BCo4Sm 2: 312 B3C02Fe5Pr22: 312 B2Co3Gd 2: 312 B4Co@Id 1: 329 2: 312 BzCO~SIY~ B~COW 1: ~328 B2 - xCoj.+ .xZr 2: 3 I0 B,Co,Zr 2: 309 BCr 1: 242,261-2, 364, 372-3, 375; 3: 6 B4CrMo 1: 377-8 B5,5C$34Fe65,5Ga4M04P123: 698, 699, 699, 700 B8Cr2Nb2 1: 378 B4CrTa 1: 378 B4CrTi 1: 378 B4CrV 1: 378 BzDy3Ni7 1: 414,416 BFe 1: 242,261-2,328,364,372-3,375; 3: 731 BFe2 1: 915; 3: 731 B20Fe801: 742; 3: 681, 731 BFel4 3 794 B F e ~ 4 ( ~ r ~ 3: N d174, ) ~ 175 BFe14Gd2Hx 3: 179, 180 BFe14Nd2 1: 227, 434, 748-9; 2: 31 I , 314-15, 318-19, 478-9; 3: 97-98, 97, 101, 102, 165, 167, 168, 169,169, 171, 172, 173 B6Fe7Y,6Nd13.2Si1,2 1: 748 B6Fe27Nis3P,41: 745 BFe14Pr22: 315, 319 BFe14R2 3 173-174, 794 B20Gd7Nb3 1: 378 BSOGe1: 257 B2Mf 2: 293; 3: 657 B4HfNb 1: 378 B4HfTa 1: 378 B4HfTi 1: 378 B4NfZr 1: 378 B$rMo2 1: 328 BGLa 3: 97 B2La3N3Ni23: 103 B ~ L u R u1: 266 B8Mn2Mo2 1: 378 BMo 3: 6, 657 B M o ~3: 487 BsMozNbz 1: 378 B4MoTi 1: 378 (B)Mo3Si 3: 487,488 B2M05Si3: 227, 486, 487, 488, 657 B ~ M o ~1:Y378 B4MoZr 1: 378 BN 1: 110, 181-2; 2: 327; 3: 669, 672,736 B4NbTi 1: 378 B4NbV 1: 378 B4NbZr 1: 378 B2Nd3Ni131: 413 BNi 1: 83, 85 BNi, 3: 265, 266,267, 268 BNi13 1: 83, 85 BI9Ni8, 1: 741-2 B36Ni641: 742 (BNi),CY 3: 104 B~NPPu1: 378 B203 3 683, 684,685, 698, 700 BP 1: 352
B~oRh~4Sr5 1: 415 B6Sm 1: 213, 217,219 BTa 3: 318 B4Ta3 1: 328 B4TaTi 1: 378 B4TaV 1: 378 B4NbZr 1: 378 BTi 3: 597 B2Ti 2: 61-2, 85, 204, 289-90, 295, 507, 511; 3: 328, 346,495,583, 584, 597, 657, 669, 610, 671 B,TiV 1: 378 B22TiI0W 1: 378 B4TiZr 1: 378 B,,U 1: 398, 404 B,,Y 1: 257; 3: 87 B12Yb 1: 217 B2Yr 2: 206, 295, 507 B2Zr 3: 583, 657 BaCdll 1: 252 Ba2CazCu309T1 3: 269 Ba2Cu30GY1: 266 Ba2Cu307 1: 266; 3 269, 759 Ba2Cu30xY 3: 672 BaFz 3: 780 BaGe2 1: 375 BaHg,, 1: 252, 398 BaI-lg,, 3: 26 BaMg2 1: 109, 667-8, 681; 2: 432, 646 BaMg4 1: 681 BaMgg 1: 681 B a 2 ~ g 2: ~ ,646 Ba~Mgz1: 667-8, 681 Ba3Mg 1: 667-8, 681 Ba6Mg2, 2: 646 BaM~2Sn21: 340 BaMn2Sb2 1: 241 BaNiSn, 1: 339, 374 Ba0,Ti 3 55 BaPb, 1: 291, 405, 407 LPa2Pb 1: 111; 3: 28 Ba(Pbo.,TI, 2)s 1: 291 BaPtSb 1: 374, 376 BaS 1: 111 Base 1: 111 BaSi 1: 375 BaSi2 1: 404; 3: 161 Ba2Sn 1: 111-12 BaTe 1: 9, 110-ll BaZn5 1: 413 BCDyNi2 3: 103, 104 BCErNi2 3: 103 BCHoNi2 3: 103 ~ C ~ u3: ~103i 2 BCNi2R 3: 103, 104 BCNi2Sc 3 103 BCNi2Tm 3: 103 BCNi2Y 3: 103, 104 BCNi3Y 3 104 BCPdzY 3: 103 BC~.,Pd~Y 3: 103 Be2CaGe2 1: 339 Be13Ca3: 39 Bel3& 3: 40 BeCo 3: 39-40 Be12C0 3: 39 Be17C023: 39 BezlC05 3: 39 Be2Cr 3: 37,38,40, 44 Bet2Cr 3: 38, 40
~ BeCu 1: 714; 2: 523; 3: 40 Be2Cu 3 40 Be,,.,Cu12.5Ni1,Ti13.7j~r41.2j 3: 695
Be22.sC~lz,sNiloTi,3.8zr41.2 3: 682, 682, 684, 685,685, 701, 703 Be13Dy 3: 40 Be13Er 3: 40 BeI3Eu 3 40 BeFe2 2: 521 BeFa3 3: 39, 40, 404, 416 Be,Fe 3: 39,40. 49 BeSFe 3: 39, 49 Be7Fe 3: 39 BellFe 3: 39 BelzFe 3 39 BeZOFegO 3: 698 Be13Gd 3 40, 46 Be2Ho,05Ti3: 48 BezH3Ti 3: 48 Be2H15Zr 3: 48 Be2N2,,Zr 3: 48 BeHf 3: 38,40 Be2Hf 3: 38,40,48 BejHf 3: 40 Be13Hf 3: 38, 39 Be17Hfz3 38, 40 Bel3Ho 3: 40 BeIr 3: 39 e2Ir 3: 35, Be& 3: 39 Be17Xr2 3: 39 Be13.La 3: 40 BeLi2 2: 481 BeLi 2 481 Be&u 3: 40 B e & & 3 39 Be$& 3: 38, 40 Be8Mn 3: 38 Be12Mn 3: 38, 40 BeMo3 3: 38, 40 Be$& 3: 38, 40 Be12Mo3: 38, 40 Be2,Mo 3: 38, 38 BeN$i 1: 349, 353, 359 BezNb 3: 38,40 Be2Nb3 3: 38, 40 Be3Nb 2: 316; 3: 38, 40 BeSNb 3: 38, 40 Be,,Nb 1: 883, 885, 891, 979, 1000-1, 1005; 2: 245-6, 249,252-3, 648-9; 3: 38, 40, 42, 43, 43, 44, 46, 47, 757 Be17Nb21: 891, 1000-1, 1005; 2 245-6, 252-3; 3: 38,40,43,43.44,46,47,49, 757 Be,,(Nb,Zr) 3: 42 Be13(Nb,Zr) 3: 42 Be17(Nb,Zr), 3: 42 Be13Nd 3: 40 BeNi 1: 714, 718, 1000; 3: 39, 40, 43, 48 BeNi3 3: 39 Be13Np 3: 40, 46 Be13Np0.68U0.323: 46 Be0 1: 179, 183, 240, 422; 3: 44 BeOs 3 39 BeOs2 3: 39 BezOs 3: 39 BesOs 3: 39 Be170s33: 39 BeP, 1: 240 BePd 3: 39 BePd3 3: 39
o
i
~ Index ~ o
~
~
Be2Pds 3: 39 Be3Pd4 3: 39 Be5Pd 3: 39 Be12Pd3: 39, 40 Bel$% 3: 522 Be13Pr 2: 655; 3: 40 BePt 3: 39 BePtls 3: 39 BesPt 3: 39 BelaPt 3: 39, 40 Be,,Pts 3 39 Be13Pu 3: 39, 40, 46 Be2Re 3: 38 BegRe 3: 38 Be16Re0.923: 38 Bel6Re 3: 38 Be21Re53: 39 Bez2Re 3: 38 BeRh 3: 39 Be2Rh 3: 39 Bea6Rh 3: 39 Be17Rhz 3: 39 Be2Ru 3 39 Be3Ru2 3: 39 Be5Ru 3: 39 BeloRu3 3: 39 BelaRu 3: 39 Be17Ru23: 39 Be17Ru33: 39 BeI3Sb 3: 39 Be5% 3: 40, 40 Bel3% 3: 39, 40 Be13Sm 3: 40, 46 Bel3% 3: 39 BeTa 3 38 BeTa2 3 38 BeITa 3: 38,40 BezTas 3: 38 Be12Ta 1: 1000; 3 37, 38, 43, 43 Bel,Ta2 1: 1000; 3: 37, 38, 40, 42, 49 BeI3Th 3: 40, 42 BeTi 3: 38 Be2Ti 3: 38, 40, 45, 46, 47, 48, 49 Be3Ti 3 38,40 BeIoTi 2: 253 Be12Ti1: 883, 885, 893; 2: 245-6, 252; 3: 38, 40, 42, 43, 43, 44 ReI7Ti23 38, 40 Be13Tm3: 40 Be13U 1: 21 1,21447,221, 1018; 3: 40,42, 45, 46, 47, 49 Be2V 3: 38, 40 Be,,V 1: 1000; 3: 38, 40, 43, 43 Be2W 3 38.40 Be12W 3: 38, 40 Be,,W 3: 38, 38 Be13Y 3: 39, 40 Be13Yb3: 40, 46 BeZr 3: 40 Be2Zr 3: 38,40, 48 Be3Zr 3: 38 Be5Zr 3: 40 Be13Zr 1: 597, 883, 885, 891, 1001, 1005; 2: 245-6, 252-3,648-9; 3: 37,38, 39, 42, 43, 43, 49 Be17Zrz3 37, 38, 40 Be43Zr571: 742 BizCaCuzOgSr23 269, 269 Bi2CaMg2 2: 517 BiCe 3: 238 Bi3Ce4 3: 27
lO25
d
Bi4Ce3Pt31: 217, 219, 221 Bi2Cs3 3: 129, 129 Bi4Csg 3: 129, 129 Bi6CS7In4 3: 128, 128 BiCuMn 2: 448 Bi4Cu4Mn31: 321 BillEu14Mn 3: 104 BiF3 1: 245 Bi3Gd4 3 527 Biin 1: 1022, 1025; 3: 161 Bih2 1: 375 BiR2 3: 28 Bi2K 2: 432 Bi2K3 3: 129 Bi4K5 3: 129 Bi4[K-crypt], 3: 114 BiLi 2: 510 BiLi3 1: 320; 2: 502 Bi2Mg3 1: 14, 669-70, 673, 941; 2: 642; 3: 252 BiMgNi 3: 239 BiMn 1: 15; 2: 309, 435, 4414, 446, 4484,654; 3: 97 BiNb-, 2: 352 BiNi 1: 744 BiO 1: 240 BiPb2 1: 4; 2: 653 BiPbPd, 1: 630 BiPd 1: 630 Bi,Pd 1: 628, 631 BiPr 3: 522 Bi2Pt 1: 628 Bi2Rb3 3: 129 Bi4Rb 1: 332 Bi4Rb5 3: 129 (Bi,Sb)z(Se,Te)3 2: 463 (Bi0.25Sb0.75)3Gd5 3: 527 (Bi0.5Sbo.s)~Gds3: 527 (Bio 75Sb0.,,),Gd5 3: 527 Bi2Se3 2: 329, 462-3 2: 329 Bi2Se0.1~Te2.8~ Bi3Sni4 3: 161 BiTe 2: 432 Bi2Te3 1: 10, 943, 1018, 1026; 2: 329,457, 460,462-4, 642; 3: 97 Bi2Ti 1: 368-9, 375 BisT12 2: 642 Bi2U 1: 375 Bi3Yb4 3 161 BiZr 2: 352 BrCu 1: 172, 180, 1834, 261 C2Ca 1: 404 C C O 1~: 301, 303 CCr 3: 670 C,Cr3 3: 754 C24Cs 3: 257 C,DyFe7 2: 31 1 CxDy2Fe142: 3 11 C0.gEr2FeI73 168, 168 CFe2 1: 328 CFe3 1: 398; 2: 447; 3: 312 CFe4 1: 404 CFe14Pr22 315 CFe14La2 2: 315 Co.4Fe17Nd22 314 CFe14Nd2 2: 311, 315 C0.9Fel7S1112 2: 314 C,Fe17Sm2 2: 314 CFe3W3 1: 328, 332, 395
1026 CFe6WG1: 328, 332 C,.sGd 1: 301 CGd2 1: 299-302 CHTi2 I: 305 CHf 2: 61 C21rU2 1: 320, 340, 374 C6&3 3: 257 C6,LiI2 3: 258, 259 Ci.77MnI7Pr22: 314 CMo 1: 304 CMO, 3 583 C2Mo3 1: 304 (C,N)3Fe17S~2 3: 98 CNTi 3: 663, 665, 734, 736 C,Nl ,Ti 3: 669 CzNa2043: 30 CNb 3: 544 CNb2 1: 301, 303, 579 C0.83Nb 1: 579 C3Nb4 1: 299-300, 302 C,Nb 1: 954-5 CNi3 1: 303 C J P U 1~: 266 C24Rb 3: 257 CS1 1: 172, 175, 179, 181-4, 345, 358-9; 2: 208, 213, 227, 287, 289, 291, 295, 325, 3274. 407, 409--11, 470; 3: 288, 317,319, 337,338,345,479,485, 583, 583, 584, 657, 668, 669, 670 CS&TiO.s 2: 410 C2SiTi3 1: 304 CTd 1: 579; 3: 544 CTaz 1: 301, 303, 579 CzTa2V 1: 304 CTb2 1: 304 C0.77Th 1: 304 CTi 1: 300-1; 2 295,407,409; 3 544,583, 584, 657, 669, 736 CTi2 1: 299-302, 304, 579 CTi, 1: 302 C0.5Ti 1: 301, 579 CSTi8 1: 299-302 C,T1 1: 954-5 CV2 1: 579 C,,,sV 1: 304 cv 3: 754 CSV, 1: 304 CSV, 1: 292, 299-302, 304, 579 CbV6 1: 293 C7V8 1: 299-302 CW 1: 242, 266. 364, 373, 375; 2 409; 3: 754 CW2 1: 303 CaCd2 1: 109 CaCuS 1: 404-5,407,409,411-17, Plate I Chapter 17, Vol. 1; 2 642; 3: 6 CaF2 1: 191, 404-5; 2: 217; 3: 241, 241, 612, 751, 780 CaFe2 3: 794 CaFe4Sb123: 106 CaGaN 1: 320 Ca7Ge 1: 288-9 Ca33Ge 1: 257 CaH4Ni52: 477 CaHg 3: 156 CaHg8 3 27 Ca2Hg 3: 156 Ca3Hg 3: 156 (3t3Hg2 3 156
C
o
~ Index ~ o
~
CasHg2 3: 156 CasWg3 3 156 Ca14Hgcj1 3: 156 CaIn2 1: 364, 367, 370-1, 373, 375 3: 97 Ca0.33La0,G7~n03 CaLi 2 524 Ca7L13 2: 524 CaMg 3: 156 CaMg2 1: 109, 121, 875; 2: 253, 432, 518, 642; 3: 156 Ca20Mg7Zn431: 435 CaO 1: 639; 3 612 Ca03Ti 1: 266; 3 12 CnPb 3: 156 CaPb, 2 518, 594; 3: 156 Ca2Pb 1: 1 11 Ca3Pb 3: 156 Ca5Pb 3 156 CasPb3 3: 156 Ca3Pd2 2 483 CaR2Te4 3: 827 CaS 1: 111 CaSb 2: 432 Case 1: 111 CaSi 2: 524; 3: 157 Ca2Si 1: 111; 2: 524 CaSi2 3: 157 Ca2Si 3: 157 Ca2Sn 1: 111 CaTe 1: 14, 110-1 1 CaZn 2: 646; 3: 157 CaZnz 2: 642; 3: 157 CaZns 3: 157 CaZnll 3: 157 CaZn133: 157 Ca3Zn 2: 646; 3: 157 Ca7Zn43: 157 Ca7Zn203: 157 Cd2Ce 1: 374 CdCI2 3: 10 CdCr204 2: 331 CdCr2S4 2: 331 CdCr2Se4 2: 331-2 Cd2CscjT111 3: 126 CdCu 1: 454; 3: 27 CdCuS 1: 763 Cd3Cu 2: 596 Cd,Cuz 2: 642 Cd&Us 2: 502, 596 CdCu2GeS4 1: 355, 359 Cd24Gas6Na35J: 125 CdGazS4 1: 349-50, 354-6, 360. 566 CdCazSe4 1: 347, 350 CdGSHg35 1: 388 CdHgNa 1: 9 CdHgTe 2: 326,4 17,419 (Cd,Hg)Te 3: 780 Cd0,3Hg0.7Te2 418 Cd,Hgl -,Te 2: 418 CdIz 3 8 CdIn2Se4 1: 360 Cd,In2-,K. 3: 125, 126 CdxIn69-xK373: 126 CdRGNa14T1,83 120 Cd3K16NagT118 3: 119, 120 Cd&jPb, 3: 122 CdgISl4T121 3: 126 CdLi 1: 319; 2 510; 3: 238 CdL13 3: 238 CdMgj 1: 404-5,407, 534, 542, 915, 921; 3 416,417
~
~ Cd,Mg 1: 534; 3: 417 (Cd,Mn)Te 2: 326 Cd2Na 1: 11 Cd20Nai3Pb7 3: 126-127 Cdz0Na1,Sn73: 126-127 C d ~ ~ . ~ N a 4 g S3: n 3127, ~ 128 Cd7N12: 519 Cdz,Ni5 2: 596 CdP2 1: 240 CdPd 3: 412 Cd41Ptg 1: 329 CdS 1: 172, 176, 178-81, 183-4, 261, 875; 2 345,423-4, 511; 3 670, 671, 780 CdSb 1: 242; 2: 418, 642; 3: 28, 155, 157 Cd4,Sbg 1: 702 CdSc I: 178 CdSe 1: 172, 176, 179, 181, 183-4, 261; 2: 345, 511; 3: 780 CdSeZn 2: 336, 339,426 Cd0.2&3eZn,,80 2: 429 Cd0.2SeZn, 8 2: 428 Cd,pjSm,, 1: 321, 326, 329 CdTe 1: 172, 179, 183-4, 191, 350, 676, 953, 1021, 1025-7, 1029; 2 326, 346, 421, 423-4, 511; 3 28, 29, 31, 667, 780 Cd,Hg,-,Te 3: 28 2: 423 Cd,TeZnl CdTi 1: 404, 407 CdTi2 1: 387 CdiiU 1: 219 (Cd,Zn)Te 2 326 CeSCD82: 432 CeCI, 1: 653 CeCo5 1: 416-17; 2: 313 CejCoI9 1: 414 Cez4Coll 1: 405 CeCc12-,Si2+~ 1: 949, 950 Cel,Co,Fe4,Sb12 3: 106 CeCu 1: 375 CeCu6 1: 212,214,219-21,413-414, 1018 CeCuzln 1: 221 3: 138, 149 Ce(Cu,Pd,J3 CeCu2Si, 1: 2 14-1 6,936; 2: 229-30; 3: 45, 46, 141 CeD3 1: 245, 301 CeFe2 1: 889, 891; 2: 252, 524, 647 CeFe5 2: 524 CeFe4P12 1: 217 Ce2FeI7N33: 98 Ce Fe6,Ni,Sb,z 3: 107 Cebe4Sb123 106 Ce{Mn,Ni)], 3 101 CeNi;! 3 143, 143 CeNi3 1: 414 Ce2N$ 1: 414 Ce2Ni15S121: 417 CeNiSn 1: 217,219 CeNisSn 1: 414 CeO, 1: 842 CePd3 1: 159 CeRlizSi2 2: 229 CeRhSb 1: 217 CeRu2 2: 479, 483; 3: 78 CeRu2Si2 1: 214, 219, 221 CeSb 2: 448; 3: 180-181 CeSn2 3 27 CeTe 2: 448 Ce2Znl7 1: 417 Ce32n22 1: 416 ClCr 3: 579
-,
Compound Index c13cr 3 579 ClCs 1: 242, 245, 252-4, 261, 263, 762 ClCs8Gal 3: 119 ClCs81nl 3: 119 ClCu 1: 172, 180-1, 261, 1025 c1,cu 3: 10 CI2Fe 3: 509 CIH3Si 3: 580 C12H,Si 3: 580 C1,HSi 3: 580 ClNa 1: 242, 245, 252-4, 261-3,266,268, 320-1, 404 ClNa02 3: 30 Cl4Si 3: 580 CoCr 1: 417; 3: 582 Co3Cr 1: 683 Co47Cr531: 409 CoCrMn 1: 417 C O C ~2~ 331 S~ (Co0.92-xC"0.06FexZr0.02)7.hS1n 3: 98 ColC~yjPd451: 948 C03C~55Pd451: 948 C O ~ D 1: Y 891; 2: 252; 3: 526 CoxDyFelI-xTi 3: 177 Co2DySi23: 183, 183 Co2Er 1: 891; 2: 252; 3: 526 Co,ErFel l-xTi 3: I77 C O ~ E ~ ~1:. 415 ~ , G ~ ~ C O ~ E ~ , , ~3:T104 I~~,~ CoFe 1: 118, 535-6, 567-8, 573,627, 631, 633, 714, 717-18, 721, 774, 776, 784, 915,921; 2 303,307-8,447,650,653 CoFe, 2: 447 Co3Fe 2: 447 CoFe2Ga 2: 449 Co2FeGa 2 449 Co,Fel l,GdTi 3: 177 CoFeGe 2: 449 Co2FeGe 2: 449 CoFe2Ge 2: 449 Co,FeH(CO)l, 1: 651 CoXFell+HoTi 3: 177 Co2FeIn 2: 449 Co,Fe4xLal-ySb12 3: 106 C0Fe3La~.~Sb12 3: 107 CoFeNi 3: 671 (Co,Fe,Ni)S 3: 671 (Co,Fe,Ni),V 1: 924 Co2FeSi 2 449 CoFeSn 2: 449 CoxFel]-,TbTi 3: 177 Co,FeI1-,TiY 3: 177 CoFeV 1: 970 (C0~.~Fe0,3)3V 1: 913, 914 (c00.78Fe0.22)3V2: 142 ( C O ? ~ F ~ 1: ~ 772, ~ ) ~788 V (Co,Fe)3V 1: 924, 928 CoGa 1: 563,5654,572,574-5,577,763, 765,777,788; 2 621,649; 3: 238,276, 277,280,290,3 11,759, 780,783,784, 784, 785 CoGa5Ho 1: 405, 408-9 CoGagHo2 1: 405, 408-9 1: 561, 566 C048Ga~~ Co2GaMn 2: 449 CozGd 1: 890-1; 2: 252,479 c05Gd 2 312,445 C07Gd2 1: 414 C017Gd2 2: 445 CoGe 3: 161 CoGe2 2: 502
Co3Ge2 1: 112 Co5Ge7 1: 305 CoGeLi2 3: 239 CohGehLi1: 416 COgGegY 1: 416 CoH,Hf2 2: 484 CoH5Mg2 3: 142, 142 CoHf 1: 535, 544; 2 244, 246 COHf2 2 484 Co2Hf 1: 875, 887, 891, 920; 2: 252 Co7Hf 2: 445 Coz3Hf 2: 445 CO,,Hf6 2: 445 Co2HfSn 2: 449 C O ~ H 1O: 875, 891; 2: 252; 3: 526 CosLa 2: 445 Co7La22: 445 Co13La2: 445; 3: 100, 174 Co2.5LaNi2.52: 510 Co,LaP,! 1: 375 CO, LaSi12 2: 449 CoZMg 1: 109; 2: 445; 3: 157 COSMM 2: 313 Co3Mn 3: 417 CoMnSb 1: 321 Co2MnSb 1: 950 CoMnV 1: 417 CoMo 1: 417 Co3Mo 2: 649 CogM02 2: 445 C040M060 1: 409 Co,Mo2Si 2: 596-7 Co3N1.,Fe8NdTi 3: 99 Co17NOzS1112 3: 794 CoZNb 2: 244, 246, 249, 253; 3: 583 CoNb4Si 1: 340 C020Nb60Zr20 3: 845 C02Nd 1: 889-91; 2: 252 Co5Nd 2: 444 C017Nd2 2 444 Co9NdSi4 3: 100 CoNi 2: 307-8 (Co,Ni),(Al,Ti,Zr) 2: 650 (Co,Ni)Si2 2: 216 (Co,Ni),V 1: 924 CohNi73: 307 CoPW 3: 671 Co4PbXSb123:107 CoPd 2: 445 Co2Pr 1: 889, 891; 2: 252, 310 Co3Pr 2: 310 Co7Pr2 2: 310 CoI9Pr5 2: 310 Co9PrSi4 3: 100 Co13-xPrSix 3: 100 Co5(Pr,Sm) 2: 313 CoPt 1: 946-8; 2: 308-9, 445; 3: 54, 223, 413 COPt, 1: 777, 797; 2: 450; 3: 223 C00.3Pt0.7 1: 946-7 CO$%3: 223 C075Ptyj 3: 223 C O ~ ~ R 1:U887 ~ ~ T ~ ~ ~ Co20R~30Ta50 1: 887 C O , ~ R U ~ 1,:T887 ~~~ c o s 1: 112 CO,S~3: 711 CoSb 1: 112 CoSb, 3: 105 Co4Sb12Snx3: 107
1027 CoSi 2: 216,230,240,253, 329,609,61617; 3: 6 75 CO%, 1: 650, 875, 885, 999; 2: 211-13, 216-17, 230-1, 244, 246, 253, 324, 330, 502, 505-6, 616, 619-20, 622, 624-5; 3: 231,232,233,234235,235, 236.237,238,240,241,241,242, 367, 45811, 583, 675, 731, 787 CO2% 1: 1 11, 364, 373, 375; 2: 216, 609, 617; 3: 675 Co3Si 2: 445 C O ~ S1:I ~889-91; 2: 252 Co3Sm 2: 316 C05Sm2 310, 312-14, 317-18,475,478 C O ~ ~ 2: SM ~ 312-14, 318,475, 479 C019Sm5 1: 414 CuSn 1: 242, 766-7; 2: 521 CoSn, 1: 764, 766-7; 2: 521 Co3Sn 1: 764, 767 Co3Sn2 1: 112, 766-7 C O ~ 2: T 253 ~ Co2Td 1: 890 CoTi 1: 512,535,544,712,714-17,722-4, 875; 3 328, 328, 365, 874, 874 Co2Ti 1: 706 Co3Ti 1: 65,72,490,505,508,527-8,545, 549-52, 897-9. 908. 915, 924, 970; 2: 29 c o u 1: 319 2: 445 CO~U CoV 1: 417 CO~V 1: 924 CO~Y 1: 889, 891; 2: 252 1: 701; 2: 445 CO~Y COSY1: 701; 2: 313, 445 C O ~ Y1: Z 701; 2: 445 C017Y2 1: 701; 2: 445 CoZn 2: 445 CoZn2 1: 319 Co5ZnZ11: 322 CoZr 1: 535-6, 544; 3: 759, 760, 761, 845 CoZr2 1: 392, 396 Co2Zr 1: 875, 891, 920; Co5Zr 2: 309-10 Co11Zr22: 445 2: 310, 445 Cr2CuS4 2: 331 Cr2CuSe4 2: 331 Cr2CuTe4 2: 331 CrF6NiRb 1: 328 CrFe 1: 42-3, 98, 119, 242, 252-3, 266, 417; 2: 314, 523 CrFe2Ga 2: 449 CrFez.,Ni 1: 305 CrFe, 1: 683 Cr1-xFe3 1: 627 Cr2Fe3 1: 409 1: 628 Cr46Fe541: 473 Cr,Fey 1: 250 CrFeMn 1: 417 Cr2FeS4 2: 331 (cr0.172Fe0,828)29cysn~3 3: 99 Cr2FeloY 2: 448 Cr4Ge32 621 Crl.8Hl,7Ti 2 477 Cr2H3,8Zr2: 475, 479 Cr2Hf 2: 295 Cr2HgS4 2: 331 Cr2HgSe4 2: 331-2
1028
Compound Index
Cr822r182: 352 Cr,LaNi2-. 2: 509 CrMn 1: 417; 3 55 CrMnj 1: 409 CrMnNi 1: 417 CrzMnS4 2: 331 CrMo 2: 41 (Cro 03M00.97)Si23: 459 CrMoZr 2: 505 CrN 3: 665 CrNaO, 3: 34 Cr2Nb 1: 647, 885, 891, 920; 2: 244-6, 249, 252-3, 290; 3: 43, 370, 415, 419, 461, 583, 658 CrNi 1: 41 CrNi, 1: 305, 970; 3: 307 CrNi, 1: 683 Cr13Ni5Si21: 409 Crz03 3: 578, 579, 611, 708 Cr730~27 2: 352 Cr204Zn 2: 331 Cr,Pd5~xTi503: 56 CrPt, 3: 72, 72 Cr3Pt 2: 521 Cr7gRh,, 2: 352 Cr72RuZ82 352 Cr2S3 1: 301 Cr3S, 3: 711 Cr5S6 1: 301 Cr2S4Zii2: 331 CrSi, 1: 67, 388, 403-4, 875, 885; 2: 212-13, 222-3, 225. 227-8, 230-1, 253,329,618,622,624-5,654; 3: 157, 458, 459, 733, 757 Cr2Si 1: 641 Cr3Si 1: 252, 261, 266, 404-5, 409-10, 412, 414, Plate I Chapter 17 Vol. 1, 473, 638, 652, 885, 891, 999, 1008; 2: 211-13, 224-5, 243-6, 249, 252-3, 245-6253; 3: 157,675, 733 CrTe 2: 448 Cr3Te4 2: 448 CrTi 1: 41-3 CrTi, 1: 388 Cr2-,yTi 2 484 Cr,Ti 1: 706, 885, 891; 2: 240, 245-6, 252-3 Cr3Tiz 2: 253 Cr40Ti601: 751 cI’60Ti40 1: 891; 2 245-6, 252 CrV 1: 42 CrW 1: 41 :238,475,479,481; 3: 356, 356, 359 Cs8GaII3: 119 Cs4Ge, 3: 115, 116 Cs8Ge136Na~6 3: 131 CsgGegZn 3: 121, 122 CsH 1: 297, 299 Cs,HgSns 3: 122 CssXnll 3 119 CsLi 1: 681 CsgNa16Si1363: 131 CsPb 3: 254, 255,255,257 CsqPb4 1: 686; 3: 254
Cs4Pbg 3: 116 CsPrS2 3: 825 CsRb 1: 681 CsSb 3 157 CsSb, 3 157 Cs2Sb 3: 157 Cs3Sb 2: 418; 3: 97, 157 Cs3Sb7 3: 157 C~gSb43: 157 CslzSi173: 117 C S ~ S I1:S ~682 Cs3Sn51: 682 CsgSnq 1: 682 CsfjSng 1: 682 CsTl 3: 120, 255 CsST17 3: 255 CsST111 3 119, 255, 256 Cs T1 3: 119, 123 ~ul;~u’~+s7si2I: 359 CuFe2 1: 359 C~Fel3Hy~Nd6 3: 100 (Cu,~e,Ni,Pd)*oP,o3: 697, 701 CuFe(Pd,Pt)3 I: 628 CuFePt, 1: 431 CuFeS2 1: 172, 181, 240, 353, 359; 2: 330 Cu,FeS4Sn 1: 240, 344, 355, 359 Cu2(Fe,Zn)S4Sn 1: 355 Cu2(Fe2+ ,Zn)S4Sn 1: 359 Cu,Fel -xZr2 1: 392 CuGa 1: 568 CuGa, 2 432 CuGaGe2P4 1: 356 CuGa,Inl _,Se2 2: 421, 423 CUgG821L113 3: 125 Cu3~Ga27gNalo,3 125 CuGaS, 1: 349-50; 2: 325, 330 CuCaSea 2: 423 CuGaTi 1: 371-2, 375 CuGd 1: 949 Cu,GdNil-, 1: 949 C U__~,GdZ, 1: 949 Cu,Ge 2 502, 521 CusGe I: 957 CuGeLiz 3: 239 CuZGeLi 3: 239 CuGePS 1: 356 Cu2GeS4Zn 2: 345, 347 Cu,GeSe3 1: 240, 351, 353-4, 359 CuSHf 2: 653 cu7Hg6 1: 629, 633 CuHg2Ti 1: 245, 319 CuqHg3 2: 449 CuI 1: 179, 184, 261 Cu$n 1: 112 CuIn,,Ga,Se2 3: 672 Cu42nMa 1: 410 CulnS-2 3: 757 CuInSe, 1: 351-2; 2: 330,421,423; 3: 663, 670, 672, 673, 757 CuIn2Se4 1: 354 Cu0.391n1.203Se2 360 Cu141n,b,6Se3, 1: 360 CuInTez 1: 350 Cu,LaNi5-, 2: 509 Cu2LaSi2 2: 229 Cu2LiSi 3 239 CuLi2Sn 3: 239 CuMg 3: 157 CuMgz 1: 404, 953; 2 641-2, 645 Cu2Mg 1: 107-8,250,252,405,407, 410-12, Plate 1 Chapter 17Vol. 1,537,
763,875,891,953: 2: 238,252-3, 524, 641; 3: 6, 157, 351, 356, 359 CuMgSb 3: 239 Cu3MgZSi1: 407, 410 C u ~ ~ M g 6 1: S i322, ~ 397 CuMgSn 3: 239 Cu4MgSn 1: 405, 407, 410 CuMgZn 1: 109 CuzMnSb 1: 950 ~u,MnSn I: 71,935; 2: 303 Cu4Nb5Si41: 328 CuNi 1: 41 C u ~ 0 N i , o P ~ ~ P3:d 4701 ~ Cu20~iz0P,QPd403: 698 C U ~ ~ N 3: ~682,~ 691, ~ 696, P 696, ~ ~ P ~ ~ 697 Cu40Ni20P20Pd,o3: 69 1 (Cu,Ni Pd)goP,O 3: 691, 691 Cu4Ni*+S7SI21: 359 CuNiTi 1: 711,713, 721-2,724; 3: 638 Cu2NiZn 1: 948; 3: 444 CU~O 1: 320 Cu3P 2: 523-4 C U ~ O P , O P3:~ 698 ~~ CuPd 1: 41, 282, 714, 721; 2: 649, 651; 3: 56 CuPdzSn 1: 627, 631 CuPd3 1: 65 C~O.55Pd0.451: 948 CUo.60Pd0.40 1: 948 Cut -,I’dx 1: 948 Cu3Pd 1: 65,604-5,725-6,763,806, 897, 908, 924 (C%.37Spd0.625)81P193 690 3 688, 688, 689 (Cu0.5pd0.5)81p19 3 6,g9 (CuO.5PdO 5)82.5p17.5 3: 689, 690 (CuQ. 5)83.SP 1 6.5 3: 689 (Cuo.5Pdo 5)lwxPx 3: 688,688,689,689 (Cu,Pd,Pt)2Sn 1: 631 (Cu,Pd,Pt)3Sn 1: 631 C U , P ~ , ~ S3:~682 ,~ CU,Pd77.~si,,j53: 695 CuzPr 3: 522 Cu5Pr 3: 522 Cu6Pr 1: 218; 2: 655; 3: 522 CuPt 1: 35,41,68, 288-92,407, 628, 720 CuPt3 1: 65, 288-9 Cu3Pt 1: 65, 498, 725-6; 3 54 Cu3Pt2 1: 631 CuRE2Si2Zn 1: 377 CuRh 1: 41 CU,,S 3: 756, 757 Cu3S4Sb I: 352-3, 359 Cu2S3Si 1: 359 CUS~TI,3 827 CuzSb 1: 42364,373,375,404,627; 2: 521, 647; 3: 34, 236, 242,418 Cul0Sb3 1: 286-8 Cu(Sb,Sn) 1: 630 Cuz(Sb,Tl) 1: 627 CuSc 1: 714, 719 CuSe, 3: 756, 757 Cu2Se 2: 469 CuzSe3Sn 1: 352 Cu2Se4Sn 1: 352 1: 352 Cu2Se3T11: 375 Cu3si 1: 957, 1001; 2: 616, 622 CU-~SI 1: 319, 957; 2: 5 0 5 4
1029
Compound Index CulsSi4 1: 410 Cu3,Si8 1: 322 CuSiZr 2: 653 CuSi2Zr 1: 320 Cu4Si4Zr31: 329 Cu5Sm 2: 313-14 GuSn 1: 3; 3: 28, 671 Cu3Sn 1: 3,2864,761,763; 2: 538,5767, 581, 589; 3: 28, 34,671,671,874,875 Cu5Sn 1: 13 Cu5Sn81:326 Cu6Sns 2: 507, 576-8, 581-2, 588-7, 589, 592-3; 3: 671 CuxSns 2: 506 Cu31Sng 1: 322; 2 594 CU~~SIIII 1: 321, 325-6, 329 Cu7Tb 1: 416; 2: 310, 316 CuTe 1: 670 CuTh2 1: 377 Cu2Th 1: 377 CuTi 1: 88, 242, 261, 320, 559, 575, 577, 580, 699-700,711, 713, 720-4,728, 798-9,811, 817, 820; 3: 671,759, 760 CuTi, 1: 386-1, 570, 694, 699-701, 788, 799, 820; 3: 671 Cu2Ti 1: 404 Cu3Ti 1: 284-9, 291, 294; 3: 671 Cu4Ti3 1: 320, 694, 817, 820 Cu4Ti 2: 282 CU~U 1: 215 Cu,Ni1_.xZr2 1: 392 Cu,iVi50-~Ti~01: 728 CUZII 1: 3-5, 8, 9. 13, 41-2, 44, 101, 107, 109, 319-20,498, 503, 506, 511, 524, 535-6, 546, 562, 564, 571, 575, 631, 714,763,767,875,880,882,886,889, 891,915,970, 1019-20, 1028; 2 252, 5324,536,538,541,544,555-6,591. 595, 646; 3: 27, 28, 29, 290, 309, 416, 761 CuZnz 1: 101, 374-5, 627; 3: 28 CuZn3 1: 44, 107, 325, 763 Cu3Zn 1: 44, 46; 3: 28 Cu5Znx 1: 101, 107,321-2, 325, 327, 329, 341, 404, 627, 763 Cu51Zn493: 280 CuZr 3: 271 Cu3Zr 2 653 1: 742 D3Ho 1: 301 D3Mn2Ti 2: 475 D3Mn2Zr 2: 475 D 3 ~ ~ n 2 ~2:T 479 h6 D23Mn23Y62: 481 DPd 1: 300 DO.7Pd 1: 297, 299 Do'gPd 1: 299-300, 302 DPdz 1: 299 DyFe2 1: 889; 2: 3 9 1 4 , 4 4 6 7 DyFe3 2: 394 DyFeloSi, 2: 448 Dy0.7FeTbo.272: 396 DyO~7Fel,95Tb0,3 2: 400; 3 97 Dy0,7Fe2Tb0.32: 395 DY0.73Fe1.95Tb0.2~2: 401 DY0.73Fel .9STbO.27 2: 398--9 Dy0,73Fe2Tb0,27 2: 392-4 DyxFe2Tb,, 3: 794 DyFeItTi 3: 174 DyNi 1: 375
DyN& 3 526 DyNiSb 3: 104 (DY0.2Fr0.8)Fe10.5N,vl.5
DyPtS 2: 502 Dy2S3 3: 756 DySi 1: 261 DySi, 2: 230
3: 99
Er(As,P) 3: 780, 784 ErFe2 1: 885, 891; 2: 252, 392-3, 446-7, 483; 3 528, 529 ErFe10Si22: 448 ErFellTi 3: 176 ErGa4V2 1: 328 ErNi, 3: 526 Er3Ni 3: 526 ErPt, 2: 502 ErRh3Si2 1: 415 ErSi, 2: 230 Eu14MnSbll3: 104 ELIS3: 526 EuSe2T!TT13: 825 EuSi 1: 261 EuTe2Tl 3: 825 F6GeIS2 1: 653 F7KzNb 1: 653 FLi 1: 523; 2: 239; 3: 268 F3Nb 1: 428 FPb 3 250 F~SS 3: 780 F,S 1: 328 F3Ta 1: 428 1: 428 F~ZS Fe3(A1,Si) 2: 21 1; 3: 780, 784 Fe17CySm23: 98 Fe2Dy 1: 891; 2 252 EeGa 1: 714,717 FeGa3 2: 447 Fe2Ga 2: 447 Fe3Ga 1: 725-6, 897,915,924 Fe3Ga2 2 447 Fe3Ga4 2: 447 Fe13GaHzo.2Nd63: 100 Fe2GaNi 2 449 Fe16.66Ga0.34N?sm2 3: 98 FeGaSc9 3: 90 Fe2Gd 1: 889, 891; 2: 252,446-7,479 Fe5Gd 2 446 FeI7Gd22 446; 3: 85 FeGd2S4 3 827 FeloGdSi22: 448 FellGdTi 3: 176 FeGe2 2: 502 Fe3Ge 1: 501-2, 514, 541, 549-50, 552, 725, 915; 2: 447; 3: 417, 759 Fe3Ge2 1: 112; 3: 759 Fe13Ge31: 319-20 Fe3,Ge2$Ni4, 3: 443 FeH .9T12: 475 FeH3Hf2 1: 393 FeJ2.-,HYM,R 3: 99 Fe2Hf 1: 953; 2: 447; 3: 731 Fel 91(Hf0,83Ta017) 3: 528, 529 F ~ ~ , O ( H E ~3:, 528, ~ ~ 529 T~~,~~) Fe2.09(Hfo.83Tao.17)3: 528, 529 F~~H 1: 889, o 891; 2: 252, 392-3, 446-7 Fe3Ho 3 528,529 Fe2(Ho,Tb) 1: 889 FeHTi 2: 475,477 FellHoTi 3: 176
FeInNi 2 449 Fe4LaP,2 1: 329 Fe4LaSb12 3: 106 Fe9LaSi4 3: 100, 101
Fel ,LuTi 3: 176 (Fe,M),2R 3 99, 172 Fe14.-,M,R6 3: 100-101 ( F e ) M ) l 7 3: ~ ~98 (Fe,M)29R33: 94, 99 Fe16MSq 3 98 (Fel-,Mx)2~Srn33: 99 Fe3Mn 3: 417 (Fe,Mn)29Nd3 3: 94 Fez5MnzlSni121: 397 FeMnSiTi 1: 457 (Fe,Mn)Ti 2: 485 FeMnV 1: 417 FeMo I: 417; 2: 523 Fe,,Mo2Y 2: 448 Fe2N 1: 301, 303,404 Fe3N 1: 303 Fe4N 1: 405 FellN,NdTi 2: 316; 3: 178 FeNNi I: 302, 305 3: 98 Fe14N2~6Si2Sm2 FeI7N2,3Sm22: 314 FeI7NxSm22: 312, 314, 316 Fe17NySm23: 98, 102 Fe2Nb 1: 953; 2: 253, 523 Fe2+,Nbl-, 1: 953-4 Fe3Nb 3: 583 (Fe,Nb)29Gd3 3: 94 Fe17Nd52: 316 FeNd2S4 3: 827 Fe4NdSbi~3 106 Fe13Nd6Si3: 100, 101 Fel0NdSi22: 448 FellNdTi 2: 315; 3: 176 Fe,7-xNd2Ti, 3: 98 FeNi 1: 3, 631, 633,688,720-1; 2: 306-8; 3: 272, 671 FeNi3 1: 3, 498, 503, 505, 508, 528, 540, 567,618,627,631,6334,725-7,794, 874-5, 879-80, 882, 885, 891, 897, 908, 923-4, 970-1; 2: 28, 139, 252, 303, 306, 447; 3: 272, 443, 461, 441 (Fe,Ni)(As,Sb)2 1: 630 (Feo.8Ni0,2)(A~0.3Sb1.7)1: 630 Fe3Ni 1: 3, 683, 1029 (Fe,Ni)3Ge 1: 915 Fe17Ni23P20Pd403: 698 Fe17.,Ni22 sP20Pd403: 692, 692, 693 3: 691 Fe,Ni4~P~OPd4~, F e ~ ~ i 4 ~ ~ P 23:0691, ~ d 692,693 40 FeNiPt, 1: 628 (Fe,Ni)3V 1: 924, 970 F e ~ o N i ~ 8 s i ~3:~ 756 ~i56 Fe203 2 207 Fe304 2: 207; 3: 267 Fe50,2Y3 3 267 Fe204Zn 3: 267, 759 Fe2P 2: 523 FePd 1: 720-1; 3 413, 663, 874 FePd3 1: 498, 725 Fe3Pd 1: 1029; 2: 557; 3: 56 Fe(Pd Ptl-,), 1: 949 FePd'fi 3: 56 Fe13E"r6Si3: 100 FePt I: 631, 720-1; 2: 308, 319,447; 3: 54, 78, 365, 413, 419
C ~ ? ~ pIndex o u ~ ~
1030 FePt3 1: 627-8, 725, 794, 803-4; 2 649 Fe3Pt 1: 725-6, 829-30, 1023, 1029; 2 319, 447, 557; 3: 56 FePtSb 2: 449 Fe17R2 3: 99, 172 Fe17RE22: 314, 391 Fe23RE62: 391 (Fe,Re)&3rn3 3: 94 FeRh 1: 444, 714, 717; 2: 446 Fe0.49Rh0.51 3: 527 FeRh 3: 56, 519,527, 528 FeRu2Sn 2: 449 Fe7.5Ru42.5Ta50 1: 887 Fe15Ru35Ta501: 887 FeRuTi, 3: 78 Fe,exRuxZrlo 3: 182, 183 FeS 1: 112; 3: 711, 757 FeS2 1: 186, 320, 328, 405, 428 FeSb 1: 112 FeSb, 1: 328, 630 Fe$% 2: 654 Fe2& 1: 953 FeSbV 3: 161 Fe4SblzYb 3: 106 FezSe 2: 447 FeSi 1: 217, 242, 261, 403-4; 2: 505; 3: 139, 140, 144, 144, 574,675,731, 757 FeSi, 1: 388, 1001; 2: 225, 231, 329, 502-3, 505-6, 523; 3: 574, 675, 757 Fe2Si 1: 373, 375 Fe,% 1: 404, 507, 537, 806, 876, 888, 999; 2: 21 I-13,217--19,225,233,307,447; 3: 275, 279, 281, 283, 290,454, 731, 759,786 Fe&! 1: 999; 2: 211, 225, 446-7; 3: 731 3: 279, 280, 281, 290 Fe752Si24.8 3: 454, 455 F e 7 ~ S 3: i ~282 ~ Fe7&,, 3: 279 FeglSi19 3: 290 FelqSi2Th 2: 448 FeSiTi 1: 475 Fe2SiTi 2: 278-9, 282, 284 I
Fe12Sm2: 315; 3 174 Fe17Sm22: 313--14; 3: 178 Fe17Srn52 316 FellSrnTi 2 315; 3: 176 FeSn 1: 767; 2 448 FeSn, 1: 764, 767; 2: 506-7, 519, 521 FeSn5 2: 506 Fe13Sn5Th43: 101 Fe7Ta 1: 963; 2: 253, 523 Fe;+,Tal-, 1: 953-4 Fe2Tb 1: 405,411, 885, 891; 2 252, 317, 391-4, 396, 446-7; 3: 527 Fe3Tb 2: 3924, 396 FeI7Tb22 392-3, 396 Fe23Tb62: 392, 394, 396 Fel ,TbTi 3: 176 FeloTb2Y 2: 448 Fe,(Tbo.2Yo.8) 3 528, 529
Fe2(Tb0.33Y0.67) 3: 528, 529 Fe2(Tb0.45t(05 s ) 3: 528, 529 Fe2(Tb0.8Y0,2)3: 527 Fe3Th7 483 FeTi 1: 42,88,282,616,694,712,714-17, 799, 815, 820; 2: 475,478-9, 483-4, 524: 3: 78 (Feo.933Ti0.~67),,N~sn1~ 3: 99 (Fe,Ti)?gNd3 3: 94, 96 (Fe,Ti),9~33: 94 (Fe,Ti)&m 2: 3 15 Fe7Ti 1: 706, 953: 2: 523-4 Fe2+xTil-x 1: 953-4 Fe3Ti 2: 253 FellTiTm 3: 176 FellTiY 3: 176 FezTm 2: 392-4 Feu3 1: 791 Feu6 1: 694, 815-16; 2: 648 Fe2U 1: 953 FeV 1: 41-2,417, 714; 2: 523-4 Fe3W2 2: 523 Fe7W6 1: 414, 417 FeloW,Y 2: 448 Fe2Y 1: 885, 889, 891, 953; 2: 252, 446; 3: 527, 535 Fe3Y 3: 527, 528, 535 Fel7Y2 3: 177 FeZn3 2: 519; 3: 794 FeZn7 1: 9; 2: 519; 3: 794 FeZn13 1: 212, 214; 2: 517, 519 Fe3Zn101: 322 Fe5Znzl 1: 768 FeZr2 1: 392 FeZr3 1: 811-15, 819 Fe2Zr 1: 154-5, 161, 953; 2: 484, 523-4 Ga(As,N) 3: 780 GaGdNi 3: 526 Ga4GdU 1: 378 GaGeMn 2: 448 Ga4GeSeg 1: 347, 350 GaZHf 1: 288-90 Ga2Hg5Te8 1: 347 Ga2.4H~Niz,6 1: 412, 415 GaIn 3: 670 (Ga,In)(As,P) 3 780 (Ga,In)N 3 780 GaInP 3: 669 (Ga,In)P 3 780 GaIr 3: 784 GaIrLi2 3: 239 GazIrLi 3: 239 Ga3K 3: 123 Ga9K3 3: 123 GaI3R33: 123 Ga28.83K3Li93: 123 Ga49,57K4N&13 3: 123 GaLi 2: 510 Ga7Liz 3: 123 GagL15 3: 123 Ga19.57L~3Na5 3 123 GaLi2Pd 3: 239 Ga2LiPd 3: 239 GaL12Pt 3: 239 Ga2LiPt 3: 239 GaLi2Rh 3: 239 Ga2LiRh 3: 239 GazLiRu 3: 239 Ga67Li38Zii343: 125 GaMg 1: 121
GaMg2 1: 121 GaZMg 1: 121 Ga5Mgz 1: 340 GaaMg5 1: 121 GaMgZn I: 454, 457 GaMn2 2: 444 GaMo3 2 352 G d N 1: 110; 2: 325; 3: 663, 669, 780,786 Ga13Na7 3: 123 Ga39Na223: 123 GalONalONi3: 120 Ga243Na10zZn72 3: 125 GaNb, 1: 792; 2: 352-3, 384 Ga13Nb51: 288-91 GaNi 1: 572, 595,714,718,720,784,956; 2: 621,649,653: 3 227,238,216,280, 780, 783, 784, 784, 785 GaNi, 1: 498, 505, 508, 512-13, 527-8, 545-7, 551-2, 593, 725-7, 875, 897, 899, 904, 908, 924; 2: 22. 27-8, 35, 137, 143; 3: 66 Ga2Ni3 1: 112 Ga4Ni3 1: 319, 321 GaNi2Ti 1: 718 Ga2Np 1: 703, 705 Gap 1: 110, 172, 174, 179-82, 184, 191, 352, 875; 2: 324-6, 424, 465, 626; 3: 780 GaPd 3: 236 GaPci2R 1: 399 GasPd 2: 432 GaPt 1: 651; 2: 621; 3: 236 GaPt, 1: 651; 2 621 GaPt3 1: 320, 512, 897, 899, 904; 3: 72, 72 Ga,Pt 1: 651; 2: 621; 3: 238 Ga3Pt5 1: 288-91 Ga3Pu 1: 291 Ga7Rb 3: 123 GaRh 2: 621; 3: 784 GaRu 3: 784 Ga2S3 1: 353-4, 360, 566, 568 GaSb 1: 110, 121, 172, 178-82, 384,763, 875, 1027; 2: 253, 333, 345, 426, 621, 639-40, 642; 3: 161, 780, 785 Ga2Se3 1: 347, 350, 354, 360 GaTi 1: 720 Ga3Ti2 1: 288-91 GaU 1: 340 GaloU2Y3 1: 378 edV3 1: 777, 803, 957, 1017; 2: 352-5, 357, 360-1, 366-8, 384; 3 759 Ga2Zr 1: 288-90, 294, 404 GdGe 3 86 Gd5(Geo.,Si3.5) 3: 527, 534 Gd5(Ge*,*Si,.,) 3 527, 529 Gd5(GeSi3)3: 527, 530 Gd5(Gel 5Si2.5) 3: 527 G d 5 ( G ~ ~ , 9 ~ S i3: z . 527 ~6) GdGe2 3: 86 Gd5Ge3 1: 120; 3: 86 Gd5Ge4 1: 120; 3: 528,529,531, 533,533 GdGe2Pt2 1: 375 Gd5(Ge2Si,) 3 524, 527, 530, 532, 533, 533, 534,534 GdS(Ge2.28Si1.72) 526 Gd5(Ge2 3: 526 Gd5(Ge3Si)3: 526 3 526 Gd5(Ge3,~S~~.9) Gd5(Ge3,2S~0.8) 3: 528, 529 Gd5(Ge367Si0.33)3: 519, 520, 528, 529, 529,535
1031
Compound Index Gd5(Ge3.$i0 15) 3: 528, 529 Gd5(Ge&3ix)3: 97, 531-535, 536 GdInNi 3: 526, 529, 536 GdMg 3: 85 GdMn,, 3: 85 GdMnSi 3: 519, 521 GdNi 1: 261; 3 526 GdNi, 3: 526 GdNilOSiz1: 417 Gd,(O,S)3.8H,O 3: 519, 522 GdPb 3: 86 Gd+jPb, 3: 86 GdPd 3 525,526 Gd2PdSi3 3: 526 GdSz 1: 381 Gd4Sb3 3: 527 GdSi2 2: 230; 3: 86 Gd3Si5 1: 378 Gd5Si3 1: 120 Gd5Si4 3: 527, 531, 532, 533, 533, 534 GdSn 3 86 GdgSn, 3: 86 GdZn 3: 527, 529, 530,530, 534, 536 GdZn121: 417 Ge2Gd,Mn2Sml-, 3: 177, 179 GeHg2Se4 1: 355 Ge3H02 1: 378 Ge$r3 1: 325 GeK 1: 242, 328-9 Ge7Li123: 130 GeLiNi, 3: 239 GeLiPd2 3: 75, 239 GeLi2Pd 3: 239 GeMg2 1: 111, 764, 1027-8; 3: 157 GeMg2S4 1: 328 GeMn, 1: 939; 3: 416, 417 Ge2Mn52: 444 Ge3M115 2: 444 Ge2Mn2Sm 3: 177, 179 Ge2Mn2SmI-_xVx 3: 177, 179 Ge2Mn2U 2: 448 Ge23M0772: 352 Ge136Nax3: 131 Ge136Nai6Rb83: 131 GeNb, 1: 16, 578, 748, 794, 803; 2: 353, 355, 357, 381, 384 Ge2Nb3 1: 653 Ge3Nb, 1: 653 GeNi 2: 502, 622 CeNi, 1: 593-4, 725-6, 875, 879-80, 882, 895, 897, 899, 904, 908, 915, 924-5; 2: 27-8, 35, 137, 143; 3: 66, 290, 443 Ge2Ni3 1: 112 Ge4Ni7 1: 374-5 Ge23pni76., 3: 443 GeNiPt, 1: 408 GeP2Zn 2: 330 GePd2R 1: 399 GePr 1: 261 GePt 2: 621-2 GePt, 2: 622 Ge2Pt 2: 621 Ge3Pt2.2: 621 GePtTi 2: 622 CeRe 3: 161 GeRh 3: 161 GeS 1: 242, 261, 263 GeS4Zn2 1: 349 Ce,oSb70 1: 705 Ge3S~51: 120 Ge4Sc51: 120
GeSe2 1: 347 GeSi 2: 622; 3: 780 GesSi3 2: 253 Ge4Sm5 1: 405 GeTa-3 2 352 GeTe 2: 432, 465 Ge3Ti5 1: 891; 2: 245, 252 Ge3U 3: 157 Ge3U5 3: 157 GeV, 1: 875, 891; 2: 252 GeV-, 2: 352 Ge2W 1: 707 Ge3W2 1: 707 Ge3W5 1: 707 H1.6Hf 1: 297 H3Hf2Mn 1: 393 tlzHf2Rh 1: 393 HK 1: 397, 299 H1,95La1: 297 H2La 1: 301 M2.,La 1: 301 H3La 1: 297, 301 H6LaNi52 477, 479, 484; 3: 253 H6 ,LaNis 2: 478 H6,7LaNii52 475 H7LaNiS 2: 479-80 HLi 1: 297, 299 H1.9Lu 1: 297 H1.1Mg2Rh 1: 393 W4Mg2Ni 2: 475, 477, 479,484 HMNi 3: 794 HGMNi52: 477 HMnPd3 1: 306 H3Mn2Zr 2: 479 H0.92M02Zr3: 253 HNa 1: 297, 299 H2Nb 1: 297 H2.7Nd 1: 297 H,Nd 1: 298 HNiTi 2: 475 H3NiTi2 1: 393 HNiTi2 2: 475 H,O 1: 702 H3P 3: 781 Ho,5Pd 1: 301 H0,gPd 1: 301 HPd2 1: 301 HzPdZr2 1: 374 WRb 1: 297, 299 H3RhZr3 1: 695 H$c 1: 297 HTa 3 253 HI.,TaV2 3: 253 H2Th 1: 321 HTi 2: 95 H2Ti 1: 297; 2 217 H3U 1: 329, 336 HV2 3 253, 253 H2V 1: 297-9 H4,,V2Zr 2: 475 Hg.lV2Zr 2: 479 HI.,Zr 1: 297-8 H2Zr 2 217 HfN 3: 552, 665 (Hf,Nb),Si3 3: 550 (Hf,Nb)Vz 3: 419 HflVi 1: 714 Hf2Ni 1: 396 Hf6Ni16Si, 2: 61, 282; 3: 308, 316 Wf2Pd 1: 396
HfPt 1: 64-5 HfzPt 1: 396 HB2 3: 711 HfSi 2: 230; 3: 675 HfSi2 1: 67; 2: 219, 231; 3: 675 H€,Si 3: 548, 550 Hf5Si3 3: 548, 550 HBiTe 1: 328, 375 (W€,Ti),Si3 3 553 HFV, 1: 916 HgIz 2 346 HgInlI 3: 26 Hg31n2Te61: 347 HgIn&8 3: 119 HgK 1: 9 Hg2K 1: 9, 668, 673, 681 Hg3K 1: 9, 681 Hg9K2 1: 9 HgioKl 1: 9 Hg2KNa 1: 9 HgK6Nal4T118 3: 120 HgLi 1: 66 Hg3L1 3: 27 HgMg 2 432; 3: 158 HgMn 1: 320 HgMg, 3: 158 Hg2Mg 3: 258 Hg2Mg5 3: 158 Hg3Mg.j 3: 158 Hg2,5Mn 3: 27 Hg5Mn2 1: 404 Hgl -,Mn,Te 2: 419 Hg7Mn2Zn10 3: 28 HgNa 1: 242, 288, 294, 320 Hg2Na 1: 673: 3: 26 Hg4Na 3: 27 WgNa2S2 3: 21 HgdNi 3: 1-52? Hg2(N03)2 3: 21 HgPb, 1: 629; 2: 522 HgPd 1: 630 HgPd2 3: 158 HgPt 2: 518 HgZPt 1: 408-9; 2: 518 HgS 1: 183-4, 242, 261; 3: 21 HgS-HgSe 3: 21 HgSe 1: 172, 1834, 191, 261; 3: 21 Hg2Se4Sn 1: 355 HnSnn 1: 3; 2 521, 576-8, 581-9 H&r$no,83 1: 404 HgdTb 3: 27 HiTe 1: 172, 183-4, 191,875,890-1,958, 1025; 2 326, 507; 3: 21, 31 HgTe2Tl 3: 28 HgTeZn 2: 419 HgxTeZnl -, 2: 423 HgTi 3: 158 HgTi, 3 158 Hg5TI2 1: 669 HgU 3: 158 Hg2U 3: 158 Hg3U 3: 158 Hg4U 3: 158 HoNi 1: 261 HoNi2 3: 526 WoPt, 2: 502 HoSi 1: 261 HoSi2 2: 230 I4Sn 1: 702
1032 InIrLi, 3: 239 InllKx % . 117, 118. 129 IniiK;Nazh 3: 123, 124 1nloKloN13 120 In,oK,oPd 3: 120 InloKIOPt3: 120 InloKgZ11 3: 120 InLi 1: 66, 2: 510 In3Lili I: 121 InL12Pd 3: 239 In2LiPd 3: 239 lnLi2Pt 3: 239 InzLiPt 3: 239 InLi2Rh 3 239 Ili,I,IRh 3: 239 3: u239 I~~LIR InMg3 I: 777,780 InMn, 2 444 InlMnTe4 1: 160 InN 1: 110, 3 780 IiiNa 3: 126 Inll 8Na7 3: 125, 125 In,74Na15 3: 125 Ing6Nay6 3: 126 Inq7NaP6NiZ8: 126, 127 InL79Nal72Ni23: 8 In192Na172Ni2 3: 126 Tn97NaP6Pd23: 126 In179Na172Pd23: 8 Inl,2Nt~,12Pd23: 126 Iny7Nag6Pt23: 126 In179Na172Pt23: 8 In192Na172Pt2 3: 126 Inl6 RNdgzIiz 3: 126 Inql;4Na23Zn4 3: 126, 127 InNh - 3 2: 352 InNi 1: 714, 718 InNi-, 1: 112. 364. 373 5. 404. 3: 5 InNi; 1: 725-6; 3 416. 417 In3Niz 1: 1I3 In(OH), 3: 12 InP 1: 110, 172. 179, 181-4, 707, 875; 2 3245, 327. 132-3, 335-8, 340 1, 346,418,421,4268,623,626; 3: 161, 669,670, 780.783 lnPd 3: 75, 231, 232, 233, 237. 238 In3Pr 2: 654 InPt 1: 631 ItlPtq 3: 72 I n 2 t 3: 238
In!3b'f: 16. 110. 121. 151. 178 81. 184. 343, 763, 875, 9 9 , 958, 1016, 1022, 1025; 2 253, 324,327, 332-3, 3445, 418,432, 639.642.654; 3: 28, 34,97, 158, 161. 667, 780. 794, 795-796 l n 4 h 3: 34 in&h7s 1:705 InSq 1: 935 InTe 1: 707; 2 642 In2Te31: 350, 354 InsoTeSo1: 705 InTl3: 55 1 n 4 ~ i I: 3 404 In2ZnSc4 1: 360 1~4C0Nb53: 59 IrLiMgSn 3: 75. 240 IiLiZSn 3: 239
Compound Index IrMn 1: 261 Ir22M0782 352 IrNb 1: 200, 205 7, 876, 885, 891; 2 245-6,249 50,252-3,3 55, 56.59, 78. 412 IrNb, 1: 619, 815 Ir7Nb 3 67, 68, 69, 72, 72 IrilNb, 3 78 Ir4Nb5Ni 3: 59 lrz8Nb722 352 Iro71Nh0212r008 3: 68 IrzSNi25TiZo3: 59 I r 2 5 N i ~ ~ V3~59 o IrSb, 3 105 Ir@c,, 1: 322 Ir4Sell1: 321 ItSi 2 230, 3 675 IrSil 3: 675 11SI? 2: 230, 3: 675 It2Si7 2 230 TriSn7 1: 328 IiTa 1: 64-5. 291 h T a 3: 69 Ir;Ta2 1: 65 IrTi I: 71417 IrTi? 2: 352 1 r 6 . 54 ~ 3:~ 56 ~ ~
Ir37V632: 352 IrZr 3: 56 IrZr, 1: 396 h3Zr 3: 68
KLi 1: 681 K,MgNal,TI,s 3: 120 KNa 1: 681 KNaz 1: 109 KuNtTTIt? 3: 255 K;Nai3Ti'i5 i3 3: 120 K2,Na2TIl93: 120, 255 K6Na14T11$h3 120 iRtn3: 120 KO 1: 242 KPZn 1: 375 KPb 1: 670; 3 28, 254 KPb:, 2: 432 K0,SPbo.s 1: 661 K4Pb4 1: 686 K4Ph9 3: 116, 117 KloPdTlj, 3: 120 KloPtTIlo 3: 120 KRb 1: 681 KSb 3: 158 KSh, 3: 158 K$b 3: 158 K,Sb 3: 158 K$b4 3: 158 K,,Si17 3: 117 KTI 3: 120, 255 KcTI?, 3: 123
La(A1,Cu)S 1: 1028 LaMnzSi2 3 177 La,Mn2Si,Y I-s 3: 170 LaNizSnz 1: 241 LaNi, 2: 475, 478-80,4834. 510; 3: 97. 107, 794
LaSNi 3: 30 Li"lo $12 5 I:417 L"203 1: 404 LaPtSb 1: 372-3. 375 LdRU3Si2 1: 415 LaSb 3: 161 LasSn3 3: 89
(La,Sr)zCu04 1: 266 (La,Y)Mn2Si23: 177, 178, 180 LaZnl? 1: 417 LiIrMgSn 3 75 LiMgPdSb 3: 240 LiMgPdSn 3: 75, 240 L I SMgo ~ SPtSb 3 240 LIQSMgPtSh 3: 240 LlMgo jPtSb 3: 240 LiMgPtSh 3: 75,240 LIMg2PtSb 3: 238 Lio ,MgPtSn 3 75, 240 LiMgPtSn 3: 75, 240 Li2-2xMg, + ,SI 3: 252 Li0 07MgZnl9l 1: 413 Lio llMgZni 89 1: 413 Lio2oMgZnl 80 1: 413 Lq,z7MgZn, 77 1: 413 L I Q ~ ~ 75M1:~413 Z ~ ~ Lio 50MgZnlSo 1: 413 LIQsbMgZnl 44 1: 413 Lio 77MgZnl21 3: 413 LIN& 1: 353, 359 LiNa I: 681 LISN,Si 3: 34 LI~NISI 3: 239 LINIZSII 3: 239 LIO$ 3 255 LiPb 2: 432, 648 L14Pb I: 670-1, 673 L17Pb2 2: 648 Ll2,Phs 1: 121 LiPb 2: 510 LiPhPdq 3: 239 Li,PbPd 3 75, 239 Li,PdSh 3: 239 LiPdlSn 3: 75, 239 Li2PdSn 3: 238, 239 LiZPtSb 3 239 Li2PtSn 3: 75,239 Li9Pt6Sn53 75 LiRh 1: 681 LiRh 1: 286, 288, 406 LiSb 2 510 L1,sb 2: 502 Li$bZn 1: 320 LISI2 510 Li12S173: 130 LI& I: 321, 325-6 LiSn 1: 95. 291, 3: 29 Li+n 1:667,669,6734, LiSSnz1: 95 Li7Sn21: 95 LI7SU3 1: 95 L113Sn~1: 95 I-i22Sns1: 95 Li Sn 2 510 L i b 1: 67 LiZn 1: 319 LuNi 1: 261 Lu,& 1: 299-300 M,(Al,Si), 3: 549 M,,R6 1: 447
C M3Si 3: 549,550,551,555 MSSi, 3: 549, 550, 550, 551, 555 Mg32(Al,Zn)4Y1: 329, 414, 475, 477 MgNi2 1: 108-9,410,412; 2: 524; 3: 6,158 Mg2Ni 2: 475, 478,483-4, 645; 3: 158 MgNiSb 3: 239 MgNiZn 1: 109 MgO 1: 523, 639; 3: 612 MgzPb 1: 9, 111, 121, 764; 2: 253, 642 MgPd 3: 236, 238 Mg6Pd 1: 321, 325-6, 329 MgPdSb 3 75, 239, 240 M~i.5PdSb3: 239 MgzPdSb 3: 239 MgPdSn 3: 75,239 MgPr 2: 654 MgPt 3: 237 MgPtSb 3: 75, 239 MgPtSn 3: 75, 238, 239 MgBRE 2 651 MgS 1: 111,422 Mg,Sb, 2: 642 MgSe 1: 11I, 422 MgSi2 21: 642 Mg2Si 1: 9, 111, 428, 764, 875, 891, 1027-8; 2 252-3, 524; 3 158, 757 MgzSn 1: 9, 11, 14, 111, 191, 319, 321, 764, 953, 1026-8; 2: 253, 642; 3: 27 MgSr2 1: 109 MgzSr 2: 432 MgTe 1: 110-11 M a T h 2: 651 M g T l 2 432 Mg17Y3 2: 518 MgZn2 1: 107-9, 121, 252, 4065,407, 409-14, 416-17, Plate I Chapter 18 Vol. 1, 537; 2: 238, 253,432,642; 3: 6 Mn2,Gd6 2: 444 Mn,LaNiSWx 2 509 MnMo 1: 417 Mn63M037 1: 409 Mn3N2 1: 320 MnN2Si 1: 240 MnN4Ta3 1: 304 Mn2NdSi2 3: 526 MnNi 1: 445, 720-1; 2: 444, 470 MnNi, 1: 11, 444, 498, 505, 508, 725-7, 763, 773, 779, 794, 803-6, 875, 897, 908, 924; 2: 28, 139, 444 MnNi2Sb 1: 950 MnNiV 1: 417 MnO 3 166, 267 MnO~(OH)2SiZn23: 9-10 MnP 1: 242, 261 MnPd, 3 78 MnllPd2, 1: 289 MnPdSb 2: 449 MnPdzSn 1: 71 MnPt 3: 412 MnPt, 2: 444; 3: 72, 137 MnPtSb 2: 435, 448-9 MnPtzSb 3: 950 MnPtSn 2: 449 MnPt,Sn 1: 950 (Mn,Re)A17Pd2 1: 435 MnRh 1: 261 Mn,Rh 1: 939 Mn12R 1: 448 NnS 1: 422 MnSb 2 444,449, 654 Mn2Sb 1: 935; 'E: 444
~
~ Index ~ ~
~
n
MnSbZn 1: 375 MnSe 1: 422 MnSi 2: 230; 3: 158, 675, 733 NnSi,.72 2: 468-9 MnSil 75-x 3 158 MnSi2'1: 67, 1001; 2: 505-6; 3: 675 MnSi2-x 2: 329 Mn3Si 1: 445 Mn5Si33: 158, 733 Mn9Si23: 158 Mn,,SilY 1: 404 MnSiTi 1: 464 Mn37Si2Ti611: 463, 465 Mn2Si2U 2 448 Mn2Si2Y3: 177 MnSn2 2: 432 Mn2Sn 2 444 Mn3Sn 1: 534, 938-9; 3: 417,450 Mn3Sn2 3: 759 MnTeZn 2: 426 Mn12Th 1: 329, 398,404-5, 413,416; 2: 311 Mn23Th6 1: 321-2, 396; 2: 479 Mn23Th3Y3 1: 397 MnTi 1: 42, 465 Mn2Ti 1: 706; 2: 475, 483, 524 Mnu6 1: 816 MnV 1: 417; 2 444 Mn12Y 1: 449 Mn23Y6 1: 447; 2: 444, 479 MnZnl, 1: 212, 214 Mn2Zs 2: 475, 479,481,483-4 Mo(A10.03Si0.97)2 3 459 Mo(A1,Si)z 3 458, 459 M o ~ B ~1:Y376 Mo3(Cr,Mo)5 1: 414 M O ~ ~ ( F ~ , M ~ , M O1:) 414 ~~F~,M Mo3(Mo,Ni)5Ni6 1: 414 MoN 1: 195 3: 459 M03Nd5016 1: 328 MoNi 1: 728 MoNi, 1: 775 MoNi, 1: 728; 3 149 MoNi4 1: 282, 288-9, 291, 2934, 328, 618,727-8,785,797,802-3,821,845; 2: 282; 3 271, 307, 418 MOO, 3: 718 MOO, 1: 597; 3: 487, 574, 718 M o ~ O S1: 803 Mo75Os25 2 352 MO6PbS8 1: 266 MoPt, 1: 36,282,288-9,291-4, 320,339; 3: 418 M082PtlS 2: 352 MOm65Re,35 2: 352 M o ~ R u2 ~408; 3: 78 Nos2 1: 428; 2: 417; 3: 711 MoSi2 1: 16, 60, 67, 88, 252, 261, 338-41, 386, 391, 4034, 539, 642-3, 649, 652, 874-6, 880-3, 885, 997-1001, 1003-10; 2 211-13, 219-21, 223, 225-7, 229-31, 240, 244, 246, 249, 253, 288-90, 294-5,470, 616, 652; 3: 305, 307, 317, 319, 319, 325, 342, 344-345,346,346,347,348,437,454, 455,458,485486,502,574,583,583, 584,645,657,669,675,717,731,733, 733,735, 735,737, 757,758,774,775, 776 Mo2Si 1: 597
d
1033
Mo,Si2 1: 1001; 2 253 Mo,Si 1: 579,638; 2: 21 1,243,253; 3 486, 487, 489, 491, 657, 774, 775, 776 Mo5Si33: 642,998-9, 1003-4; 2 225,244, 246, 249, 288; 3: 455, 485, 486487. 487,657,717-719, 731,733, 733,737, 757, 774, 775, 776 Mo77Si232: 352 2: 352 I~f040T~60 M o U ~1: 387, 390 M03U22 1: 320 (M00,97V0.03)Si23: 459 MOW 1: 41 M o ZS ~ 2: 523 N2.?Fe17Tm23: 168, 169 "a02 3: 30 NNb 1: 300, 301, 303 N3Nb4 1: 299-301 NSe 3: 786 N4Si3 2: 294; 3: 479, 485, 583, 657, 669, 670 NTlx 1: 300-2 NTl6 1: 302 N0.,Ti 1: 301-2, 579 NTi 3: 653, 663, 665, 669, 669 NZs 3 552, 665 NaO 1: 262-3 Na202 3: 30 Na204S2 3: 31 NaP 1: 242, 263 NaPb 1: 242, 667, 680; Na2Pb5 2 596 Na4Pb 1: 667,680-1 Na4Pb4 1: 686 ~Na5Pb2 ~ ~ 2: 642 Na,lPb, 1: 322 Nal6RbgSi136 3: 131 Na2S 3: 31 NaSb 3: 158 Na3Sb 2: 647; 3: 158 Na5Sb3Sn 3: 128 NasSb4Sn 3 128 NaSi 1: 242 Na,Si13, 3: 131 NaSn 3: 28, 254, 255-257, 257, 258 NaTl 1: 321, 325-6, 328; 3: 1I3 NaTi 1: 67, 110, 121, 242, 245, 319-20 NaZnl3 1: 252, 329, 397-8, 417 Nbg(A1,Ge) 2: 355 Nb,(Al,Ni) 1: 434 (Nb,HS), Si, 3: 550 NbNi 2 651; 3: 671,671 NbNi, 1: 763, 766; 2: 282, 650--1; 3: 415, 415, 461, 671, 671 NbNiP 3: 671 Nb205 2: 124 Nb750~252: 352 NbP 1: 289; 2: 506 NbPd3 1: 291 NbPt, 1: 291; 3: 70 Nb,Pt 1: 803; 2: 352 NbRh 3: 56 NbZRh, 1: 291 Nb75Rh25 2: 352 NbRu 1: 41, 67; 3 56 Nb,,S 3: 757 NbS2 3 711, 716 NbgSb 1: 158
1034 NbSi, 1: 67, 647, 649, 999, 1001; 2: 219, 227, 231, 253; 3 458, 459, 675, 731, 733, 757, 758 Nb,Si 1: 638 Nb,Si 3: 485, 541, 545, 547, 548,548,549, 550, 553, 757 NbsSi, 1: 647, 999, 1001; 2: 21 1, 224-5, 295-6; 3: 345-346,479,485,490,541, 545, 547, 548, 549, 550, 553, 554, 555-556, 658, 731. 733, 757 Nb82SiI8 2: 352 Nb3Sn 1: 16, 158, 162 3, 559, 579-80, 646, 701, 773, 788, 792, 803, 806, 808, 875, 890-1, 951,953, 957, 1017; :351-60,364,36434; 3: 97,158,258, 60, 731, 759 Nb5Sn 3: 794 (Nb,Ta),Sn 2: 355, 360 NbTi 3 346, 491 (Nb,Ti),Si3 3: 555-556, 658 (Nb,Ti),Sn 2: 355, 360, 369-70 NdOs4Sb1, 3: 106 Ni3(A1,Fe) 1: 788 Ni,(Al,Ti) 1: 778, 785, 970 3: 526 Ni,(Mno.gVg,2)Sn 3: 526 Ni,(Mno.9Vo.~)Sn3: 527 Ni80P,, 3: 681 Nig4P16 3: 701 Ni,,P,6Pd,o 3: 686, 686 N~4oP,oPd4o 3: 682, 682, 683, 685, 686, 687,689,689,690,691,692,692, 695, 697, 697, 698, 701, 703 Ni4.$9Pd,6 3: 687 Ni47P17Pd363: 687 Ni57P,gPd253: 686 Ni PloGx.-vPdx3: 686 Ni$Zr 3: 671 (Nio.5Pdo.s)iocxPx3: 688, 688, 689 Ni5Pr 3: 519,521, 522 NiPt 3: 223, 223 Ni3Pt 3: 223 Ni,,Pt2, 3: 223 NigoPtIo3: 223 Nil-,Ru,Ti 3: 60 NiS 3: 711 Ni,S, 3: 708, 711, 711, 713, 715, 757 NiSb 3: 97 Nisi 3: 675, 757 Nisi, 3: 233, 235, 238, 458n, 675, 757 Ni2Si 3: 675, 757 Ni,Si 1: 122, 498, 505, 508, 527-8, 540, 545-6,55 1-2,593-4,604-5,649,656, 725-6, 806, 897-900, 904, 908, 915, 924, 928, 999; 2: 27-9, 35, 137, 143, 212--14,233,253; 3: 66,502, 503, 507, 510, 513,645,675, 731, 744, 757, 759 Ni76(Al15Ti9) 1: 887 Ni(~eo.,Sio,5)2: 622 Ni*(Geo.sSio5 ) 2: 622 Ni20Ti, 1: 394 Ni3P 1: 404 NiSP, 1: 404 Ni40P,~Pd40Sis1: 744 NiP4S,, 1: 352 NiPci 1: 720-1 (Ni,Pd)Sb 1: 630
Compound Index Ni,Pr 2 655 NiPt 1: 41, 720-1, 777, 788,946-7 Ni,Pt 1: 725-6, 777 Ni,Pts-,Si 2: 615, 616 Ni,Pu 1: 414; 2: 316 NiS 1: 112 NiSb 1: 112, 566, 577, 764; 2 344, 654 NiSb, 1: 629 NiI2Sbs81: 705 Nisi 1: 261; 2: 230, 246, 253, 615-17, 619-20, 622, 624 Nisi, 1: 261; 2: 213, 228, 230-1, 253, 502, 505-6, 522, 614-16, 620, 654 Ni,Si 1: 261,373,999; 2: 230,253,615-17, 642 Ni,Si 1: 261, 498 Nis,Si, 1: 404 NisSiz I: 261, 649; 3: 675, 757 Ni3,Sis, 3: 744 Ni76.$32,.7 3: 443 NiSiTa 2: 619 Ni2SiTa3 2: 619 Ni2Si2Tb3: 177, 179 Ni Si Tb, 3: 104 Ni&i$i 1: 261 Ni,SiTi, 1: 261 Ni,(Si,Ti) 1: 656, 970; 2: 214-16 Ni4Si3Ti 1: 261 Ni4Sf7T42: 616 Ni5SiT161: 261 Ni7Si3Ti22: 6 16 Ni78.4s110.9Ti10.7 3: 443 Ni,Si,Y 1: 329 Nii6Si7Zr6 2: 61, 282 NizSrn 2: 479 NiSn 1: 767; 2: 506-7, 521; 3: 417, 671, 671 Ni,Sn 1: 286-9, 291, 293-4, 767, 843; 2: 505 Ni3Sn2 1: 112, 764, 767, 957; 2: 504-6, 642; 3: 671, 671, 731, 759 Ni3Sn4 1: 767; 2 505-6; 3: 29, 671, 671 NiTa, 3: 671, 671 Ni3Ta 1: 763; 3: 671 Ni,Ta 3: 671 Ni17Th2 1: 405, 416 NiTi 1: 41-2, 88, 261, 544, 564, 643, 694, 711-18,721-4,728,799,811,817-18, 820, 832, 916; 2 432, 475, 529-30, 532,536,538,540-1,5434548,556, 649-50; 3: 56,265, 266,267,27 1,404, 645, 731, 794 NiTi, 1: 328,332,3934,643,701; 2: 475, 616 Ni2Ti 1: 706, 860 Ni3Ti 1: 261, 291, 407, 643, 727, 763; 2 616 Ni40Ti601: 742 NiTiZr 2: 505 NiTm 1: 261 N N 1: 41 NiV, 3: 759 Ni2V 3: 759 Ni3V 1: 65,69,72,530-1,540,542-3,845, 915, 924; 3: 414, 446, 447, 448, 449, 464 Ni4W 1: 727-8; 2 282 Ni,jW 3: 307 NiY 1: 714 NiYb 1: 261 NiZn 1: 714, 720; 3: 28
NiZn3 3: 28 NiZn8 2 521 NiSZiizt 1: 768; 2 521 NiZr 1: 697-8, 817, 820; 2: 505, 647 NiZr2 1: 396, 701, 817-20; 2 649; 3: 265, 266,266, 270, 271 Ni3Zs 1: 820; 2: 647; 3 416, 417 NisZr 3: 330 Ni5Zr3 3: 611 NifoZr71: 701 Ni,,Zr67 3: 681 Ni35Zr651: 742 NpSn, 1: 219 OPb 1: 240, 242 03PbTi 3: 758 OSi 2: 227; 3: 718 0 2 % 1: 597; 2 82, 227; 3: 574, 578, 607, 736 0,SrTi 1: 173 O2Th 3: 645 OTi 3: 275 O2Ti 1: 328; 2: 81-2, 124; 3: 8, 576, 577, 578, 582, 716, 717, 751, 756 O,V, 1: 842; 3: 496, 583, 645, 656, 657 OZn 1: 172, 176,178 -9,182-3,261; 2 345 0 z r 3: 345 02Zr 1: 639, 842; 2: 295; 3: 583, 607, 608 Os7Sc4 3: 87 OsSi 1: 261 OsSi2 1: 388 Os2Si, 3: 675 OsTi 1: 714 O~soVzjo2: 352 O~l-~\ryT,2 613
P16Pd84 3: 701 PRQ 1: 375 PRR2 1: 428 P3Sn41: 304 PTa 2: 506 P4Th3 1: 399 PTi, 1: 328 PV 2 506 P2Zn 1: 240 PbPd, 1: 631 Pb2Pd3 1: 630 Pb,Pd, 1: 630 Pb9Pd13 1: 630 PbPt, 1: 6; 2 523 Pb4Rb4 1: 686 PbS 1: 261, 875; 2: 328,418, 464 PbS4T123: 827 PbSe 1: 261; 2 325, 328, 418,464 PbSeSn 2: 325 Pb(Se,Te) 2: 329 Pb,Sn 1: 6 PbSnTe 2: 325, 328-9, 418 Pb,Sr 1: 288, 294 PbTe 1: 121, 261; 2: 325, 328, 418, 458, 464-6, 507 PbU 1: 288-91, 294, 407 Pb,U 2 647 Pb4U 2 647 Pb,Zr, 2: 647-8 Pd(As,Pb), 9: 627 Pd,(As,Sb), 1: 627 Pd,tA~,sn)~ 1: 629 Pds(As,Sb), 1: 627, 629 Pd,(As,Sn), 1: 629 Pd,,(As,Sb)4 1: 629
Compound Index Pd(Bi,Pb), 1: 631 Pd(Bi,Pb) 1: 630 PdPt 1: 41 fPd,Pt),Sn 1: 627, 630 (Pd,Pt),Sb, 1: 628 pd1.54pt1.S9sn0.$7 630 PdRh 1: 41 PdRh, 1: 50 (Pd0.67Rh0.33)3Ta 1: 291 Pd2RSi 1: 399 (Pd0~7Ru0.3)Zr 2: 251; 3: 78 PdSb 1: 261, 630 Pd3Sb 1: 630 PdsSb2 1: 630 PdgSb, 1: 627-30 Pdl,Sbl2 1: 630 Pdl&bg3 5 1: 705 Pd2oSb7 1: 628 PdSi 1: 691; 2: 211, 505-6; 3: 675 Pd$i 1: 691, 694; 2: 230-3, 616-17, 619, 622; 3: 675 Pd3Si 1: 691; 3: 475, 757 Pd4Si 3: 675 PdsSi 1: 691; 3: 675 PdgSi? 1: 691; 3: 757 PdsqS120 3: 270 PdSiU 1: 454 pd58.8si20.6u20.6 480 PdSn4 2: 523 Pd2Sn 1: 629, 631 Pd3S11 1: 631 PdsSn, 1: 630 PdzoSn131: 630 PdTe 1: 261, 651; 3 780, 785 PdTe, 1: 651; 3: 785 Pd4-,Te 1: 321 PdgTe4 1: 651; 3: 785 PdTi 1: 714-17; 3: 56 Pd5Ti, 1: 320 Pd3U 1: 221 PdsU 2: 469 Pd88.9Ull.l 3: 78 PdV 1: 41 PdV-3 2: 352 Pd,V 1: 69 PdZn 3: 29 PdZr2 1: 396 PoZn 1: 261 Pr(Mg,Cd)g 2: 518 PrNiS 2: 654 PrPt 1: 261 PrPt2 3: 522 PrPt, 2: 655 Pr3Rh4Sn13 1: 329 PrSe 2: 654 PrT13 2 655; 3: 522 Ft(Bi,Sb), 1: 628 PtRh 1: 41 Pt5R 2 518 Pt2RSi 1: 399 Pt25Rh75 3: 213, 213 PtSb2 1: 628 Pt,SbdU, 1: 217 PtSi 2 211,213,230-3,505-6,622,624; 3 675 PtSiTi 2: 622 Pt2Si 2: 230, 615-16, 622; 3: 675 PtSi3U2 1: 215 PtSn 1: 629 Pt,Sn 1: 616; 2 649; 3: 72, 219 PtzTa 1: 286, 288, 294, 341
Pt3Ta 3: 70 PtlSTag52: 352 PtTi 1: 64, 714-17; 2 622; 3: 56 PtTi, 1: 64; 2: 352 Pt3Ti 1: 64, 616, 724-5; 3: 72, 72, 219 PtsTi 1: 64, 288-91, 294 Pt2U 1: 404 Pt3U 1: 130, 214-16, 219, 221, 936, 1018; 3: 416, 417 PtV 1: 289 PtV3 2: 352 Pt,V 3: 70, 72 PtZn 2: 506 PtSZnZl 1: 322 PuTe 1: 217 RbSb 3: 158 RbSb, 3: 158 Rb,Sb 3: 158 Rb3Sb7 3: 158 RbsSb2 3: 158 RbSSb4 3: 158 Rb12Si173: 117, 118 Rb,TI,, 3 119 RblsT127 3: 119, 123 RelsRu3STi501: 887 Res0Ru20Ti501: 887 ReSi 3: 733 ReSi2 1: 67, 339, 388; 3: 454, 675, 733 ResSi3 3: 733 Re7Sil;Ua 1: 328, 408 Re2,TiS 1: 322, 407, 891; 2: 245-6, 252-3 Re2U 1: 410 Re71V292 352 ReW 2: 613-14 Re3W 2: 613 Re_4oVV,60 2: 352 RhBi4 1: 328 RhSb, 3: 105 RhSc 1: 63 RhSi 1: 64, 261; 2: 230; 3: 675 Rh2Si 3: 675 Rh3Si43: 675 Rh4SiS3: 675 RhTa 1: 407; 3: 56 RhTi 1: 41, 714-17; 3 56 Rh3Ti 1: 724-5 Rh,,V,, 2 352 RhsZn2, 1: 322 RhZr 3: 56 RhZr2 1: 396 RhZr3 1: 305, 695 Rh3Zr 2: 479 RUSC1: 885, 891-2; 2: 245-6, 249-50, 252-3; 3: 55 RuSi 1: 64, 261, 694; 3 675 Ru2Si 3: 675 RqSi 3 675 Ru2Si2U 1: 216, 219-21; 2 229 Ru2Sn3 1: 340 RuTa 1: 887; 2 249-50. 253; 3: 55, 56, 78 RullTag 1: 885, 887, 891; 2: 245, 252 R u ~ 1:~885, T 887 ~ ~ ~ RuTi 1: 694, 714, 885, 887, 891; 2: 245-6, 252-3; 3: 56 RuV 3: 56 Ru~W 2 ~408; 3: 78 RuZr 1: 67, 204-7, 694, 876 S$b, 3 6, 7 SSc0.67 1: 299--300
1035 SSc0.83 3: 299-300 S0.07Se0.93Zn2 428 S0,08Se0.92Zn2: 429 SSeZn 2: 339,426 SSm 1: 217 SSn 2: 328 S2Sn 3: 757 SSr 1: 111 S2Ta 3: 711 S2Ti 3: 711, 716 S3V2 3: 711 S2VV 3: 711 S3Yz 3: 756 SZn 1: 172, 174-4, 179-82, 184, 191, 240, 242, 252,254, 261-2, 320, 343-5, 350, 353-4, 358-60,404, 875, 1025; 2: 325-6, 410, 426; 3: 780 S2Zr 3: 711 Sb,(As,Sb)2 1: 629 SbSn 3: 3, 7, 630; 2: 592-4; 3: 161 Sb70Sn301: 705 SbTa,, 2: 352 Sb2Te32: 329,462, 642 SbT13 2: 352 Sb2Ti71: 321-2, 324-8, 341 SbU 1: 154 Sb2V 1: 393 Sb24V76 2 352 SbY 1: 204-7, 874, 876, 885, 891-2; 2: 245-6, 252-3 SbZn 3: 28, 158,159, 161 Sb2Zn3 3: 159, 161 Sb3Zn2 1: 702 Sb3Zn4 3: 34, 159, 161 SeSn 2 328 SeSr 1: 111 SeTm0.7s 1: 299-301 SeTm 1: 217-18 SeZn 1: 172,175, 179, 181, 184,261, 1022, 1025; 2 325-6, 336, 339, 410-12, 416-17, 426,428-9,431; 3: 669 SiSr2 1: 111-12 SiTTa, 2: 253 SiTaS 1: 1008; 2 295 Si2Ta 1: 67, 647, 652, 999, 1005; 2 219, 227, 230-1, 253, 616, 618-19, 654; 3: 458, 459, 669, 675, 731, 733, 757 Si3Ta51: 647,999; 2: 618; 3: 731, 733, 757 Si2Te 1: 67 SiTh 1: 375 Si2Th 1: 364, 372-3, 375, 405, 417; Si3Th22: 21 1 SiTi 1: 261, 337; 2: 228, 253; 3: 574, 675, 733 SiTi, 3: 548, 549, 574 Si2Ti 1: 67,261,4034,641,652. 875,885, 999; 2: 213, 219, 223, 227-8, 230-1, 505,604,617,622,624-7; 3: 460,574, 575,664,668,669,673,674,675,675, 731, 733, 757, 759 Si3Ti 2: 253 Si3Tis 1: 261, 649,652,999; 2: 82,212-13, 224-5, 228, 240-1, 253; 3: 418, 430, 548,549,553, 574,583,645,664, 733, 740, 745, 757 SiTiV 1: 475 Si4Tis 3 574, 757 Si,TilOo-x 3: 740 Si3Ti2Zr32: 225 SiU 1: 816; 3: 1-59
1036 SiU, 1: 288,294, 320,791,815-16; 2 648; 3: 159 Si2U 1: 375; 3: 159
Si3U 2: 245; 3: 159 Si3U2 3 131 SiV, 1: 162, 638, 148, 792, 794-5, 797, 875, 885, 890-1, 953, 957, 1017; 2: 211, 213, 223-4, 2454, 252-3, 352, 357, 384; 3: 258, 266, 757 Si2V 1: 67, 875, 883, 885-6, 891; 2 193, 213, 231, 245-6, 252-3, 505-6, 616; 3: 458, 459, 675, 733, 757 Si,Vs 2 213, 245-6, 253; 3 733, 757 Si2(V,Fe) 2: 183 Si2W 1: 67, 643, 652, 874-5, 877, 879-80, 882, 8854,999, 1001; 2 213, 223, 230-1, 253, 616-19, 622, 629, 654; 3: 454, 458, 669, 675, 733, 757 SizW, 2 253 Si3W5 1: 252, 262, 999; 2: 629; 3: 733 Si2Y 2: 230 Si3Y52: 245-6, 253 SiZr 2: 253; 3: 733, 736, 737 SiZr, 2: 253; 3 736, 737 SiZr3 3: 736, 737 SiZZr 1: 67, 288-90,294, 305; 2 219, 227, 230-1, 253, 505-6, 614-15, 618, 620; 3: 475, 733, 736, 737 Si2Zr33: 736, 737 Si3Zr, 2 224-5; 3 733, 736, 737 Si4Zrs 1: 405
~ o ~ ~ Index o u ~ d Srn2(Co,Ce,Cu,Zr),, 2: 478 Sm(FeIITi) 1: 434 SnTaW32: 352 SnTe 1: 123, 261; 2: 328, 464 Sn2STe751: 705 SnTi, 1: 534, 875, 879, 881, 885, 891; 2: 244-6, 252-3, 650; 3 417 Sn2U3 3: 159 Sn3U 3: 159 Sn4US2: 647 Sn5U32 647; 3 159 Sn,21Vm79 2: 352 Sn20Zr802: 352 SrTe 1: 110-1 1 SrZn, 1: 413 Te,Ti, 1: 321 Te4Ti5 1: 328 TeTi2 1: 670 TeTm 3: 161 TeU 2 448 TeqYb3 3: 827 TeZn 1: 172, 179, 184, 1027; 2; 423,426; 3 671, 671, 780 ThZn2 3: 159 ThZn4 3: 159 Th2Zn 3: 159 Th2Zn15--172: 31 I Th2ZnI71: 252,405, 413; 3: 159 Ti(I~fn,\/')~ 2: 485 Ti(Pt0.89Ni0.1 1: 291 TiTmSi 1: 475 Index c o m ~ i ~ by~Pd. Nush
TiU, 1: 377 T N 1: 41-2 TixWI-, 3 666 TiZn 1: 714-15 TiZn3 1: 724-5 U2ZnI7 1: 214-15, 219-20,413 V(Al,Si),, 2: 183 1: 288-9 1, 294 V~ZIXS V2Zr 1: 953; 2: 475,479, 481
WSiz 2: 223 Y6(Fel -xMnx)23 1: 448 YZn 1: 875; 3: 159 YZn2 3: I S 9 YZn3 3: 159 YZnS 3 159 YZn12 3 159 Y2Zn17 3: 159 Y$?hll 3: 159 YI3Zn58 3: 159 Yb(A1,Cu)S 1: 1028 Zn2Zr 1: 935, 1023 Znz2Zr 1: 257 Zr(Cr,Fej2 1: 791, 813, 814; 2 649 Zr2(Cr,Fe) 1: 791 Zr2(Fe,Ni) 1: 791, 813-14
I INTRODUCTION 1 Historical Sketch - J. H. Westbrook
I1 BONDING AND STABILITY 2 Electronic Theories of Alloy Phase Stability - P. E. A. Turchi 3 Ab Initio Calculations - A. E. Carlsson and P. J. Meschter 4 The ~mbedded-AtomMethod - A. I?. Voter 5 Bond Characterization from Thermodynamic Properties - M. Ellner and B. Predel 6 Band ~tructuresand Their Interpretation - D. J. Singh 7 Phonon Dispersion Curves and Their Interpretation - L. Pintschavius 8 Bond Characterization from Crystal Morphalogy - G. A. Wolff -Principles Calculation of Elastic Properties - M. J. Mehl, . A. Papaconstanto~oulos 10 Heavy Ferinion Compounds - M. C. Aronson and B. R. Coles
HI CRYSTAL STRUCTURES 11 Factors Governing Crystal Structures - P. Villars 12 Close-Packed Structures - J. Hauck and K. Mika 13 Body-Centered Cubic Derivative Structures - E. E. Hellner and R. Schvvarz 14 Wurtzite and Zinc Blende Structures - E. Parthi: 15 Atomic Environments in Some Related Intermetallic Structure Types - J. L. C. Daams 16 Same Important Structures of Fixed Stoichiometry - M. V. Nevitt and C. C. Koch 17 Topologically Close-Packed Structures - E. Gladyshevskii and 0. Bodak 18 Structure ~ a p p i n g- D. G. Pettifor netic Structures - U'.L. Roth 20 Quasicrystals and Related Structures - K, F. Kelton
ZV DEFECT STRUCTURES 21 ~tructureof Antiphase oundaries and Domains - Y.-Q. Sun 22 Dislocations - P. Veyssikre and J. Douin 23 Point Defects - C. de Novion 24 Structure of Grain Boundaries - T. Takasugi 25 Free Surfaces - D. Farkas
V FORMATION AND C~NSTITUTION 26 Naturally Occurring Intermetallic Compounds - R. C. DeVries 27 Synthesis and Processing Techniques - P. L. Martin and D. A. Hardwick 28 Liquid and Vapor Species - R. N. Singh and N. H, March
~ o n t @ofn V ~ ~o l u ~ ~
29 Metastnble Phases - W. L. Johnson 30 Quasibinary and Quasiternary Intermetallic Compound Systems - N. M. Matveeva 31 Amorphous Phases - A. L. Greer VI ~ I N ~ T AND ~ C SPHASE TRA~SFORMATIONS 32 Diffusion - L, N. Larikov 33 Ordering and Disordering Processes - 0. Dimitrov 34 Irradiation Damage - L. M. Wowe 35 Crystallographie Transformations - C. M. Wayman and H. R. P. Inoue 36 Microstructur~lStability - J. H. Perepezko VI1 PROPERTY FUNDAMENTALS 37 Elastic Properties - M. Nakamura 38 I n t e r ~ r a n u l and ~ r Cleavage Fracture - C . L. l3riant 39 Plastic Deformation - G. Sauthoff 40 Magnetic Principles - J. S . Kouvel 41 Ele~tricaland ~lectronicBehavior - M. M. Braunovi~ 42 Corrosion of Intermetallic Compounds - D. J. Duquette 43 Oxidation Behavior of High-Temperature Intermetallics - J. Doychak 44 Thermal Behavior - 6. K. White INDEXES
I STRUCTURAL APPLICATIONS 1 Ni,A1 in Nickel- ased Superalloys - D. L. Anton 2 Ni,Al and its Alloys - C. T. Liu and D. P. Pope 3 NiAl and its Alloys - D. B. Miracle and R. Darolia 4 Gamma TiAl and its Alloys - S.-C. Huang and J. C. Chesnutt 5 Ti,A1 and its Alloys - D. Banerjee 6 Zr,Al: A Potential Nuclear Reactor Structural Material - E. M. Schulson 7 A1,Ti and its Ll, Variations - M. Yamaguchi and H. Inui 8 AI-Rich Intermetallics in Aluminum Alloys - S. K. Das 9 FeAl and Fe,Al - K. Vedula 10 Silicides: Science, Technology and Applications - K. S. Kumar 11 Miscellaneous Novel Intermetallics - R. L. Fleischer 12 Intermetallics as Precipitates and Dispersoids in High-Strength Alloys - A. J. Ardell 13 Intermetallic Composites - D. B. Miracle and M. G. Mendiratta 11 ELECTROMAGNETIC APPLICATIONS 14 Magnetic Applications - H. H. Stadelmeier and B. Reinsch 15 Se~iconductorApplications - IS.Masumoto, A. Katsui and T. Matsuoka 16 Superconductor Applications - Z. J. J, Stekly and E. Gregory 17 Magnetostriction: Materials and Applications - R. D. Greenough and M. P. Schulze 18 Optical Applications - M. W. England and E. T. Arakawa 19 Magneto-Optical Applications - W. A. McGahan 20 T h e r ~ o e l ~ c t rand i c Electrical Applications - M. V. Vedernikov
111 CHEMICAL AND METALLURGICAL APPLICATIONS 21 I~termetallicHydrides and their Applications - L. Schlapbach, F, Meli and A. Ziittel 22 High-Temperature Coatings for Gas Turbines - J. R. Nicholls and D. J. Stephenson 23 Electrochemical Applications - A. K. Vijh 24 Process ~ e t a l l u r Applications ~y - J. H. Westbrook
IV MISCELLANEO~SA P P L I ~ A T I ~ N S 25 Shape-Memory Alloy Applications - L. McD. Schetky 26 Applications in Gold Jewelry - W. S. Rapson 27 Dental Amalgam - RI, M. Waterstrat and T. Okabe 28 I ~ t e r ~ e t a l l iin c sTribology - W. A. Glaeser 29 Diffusion Barriers - R. de Reus 30 Heat Storage at Elevated Temperatures - C. E. Birchenall and D. Farkas 31 ~iscellaneousApplications - J. H. Westbrook
INDEXES