Advances in Imaging and Electron Physics, Volume 127 (Advances in Imaging and Electron Physics)

ADVANCES IN IMAGING AND ELECTRON PHYSICS VOLUME 127

EDITOR-IN-CHIEF

PETER W. HAWKES CEMES-CNRS Toulouse, France

ASSOCIATE EDITORS

BENJAMIN KAZAN Xerox Corporation Palo Alto Research Center Palo Alto, California

TOM MULVEY Department of Electronic Engineering and Applied Physics Aston University Birmingham, United Kingdom

Advances in

Imaging and Electron Physics Edited by

PETER W. HAWKES CEMES-CNRS Toulouse, France

VOLUME 127

Amsterdam Boston Heidelberg London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo

This book is printed on acid-free paper. Copyright ß 2003, Elsevier (USA).

All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2003 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 1076-5670/2003 $35.00 Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail: [email protected]. You may also complete your request on-line via the Elsevier homepage (http://elsevier.com), by selecting ‘‘Customer Support’’ and then ‘‘Obtaining Permissions.’’

Academic Press An Elsevier Imprint. 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com

Academic Press 84 Theobald’s Road, London WC1X 8RR, UK http://www.academicpress.com International Standard Book Number: 0-12-014769-6 PRINTED IN THE UNITED STATES OF AMERICA 03 04 05 06 07 08 9 8 7 6 5 4 3

2

1

CONTENTS

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Future Contributions . . . . . . . . . . . . . . . . . . . . . . . .

vii ix xi

Scanning Nonlinear Dielectric Microscopy Yasuo Cho I. II. III. IV. V. VI.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . Principle and Theory for SNDM . . . . . . . . . . . . . Quantitative Measurement . . . . . . . . . . . . . . . . Higher-Order Nonlinear Dielectric Microscopy . . . . . Three-Dimensional Measurement Technique . . . . . . . Application of SNDM Technique for High-Performance Ferroelectric Material and Devices . . . . . . . . . . . . VII. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

2 3 17 30 37

. . . . . . . . .

41 54 55

High-Order Accurate Methods in Time-Domain Computational Electromagnetics: A Review J. S. Hesthaven I. Introduction . . . . . . . . . . . . . . . . . . . . II. Maxwell’s Equations in the Time Domain . . . . III. Case for High-Order Methods in Computational Electromagnetics . . . . . . . . . . . . . . . . . . IV. High-Order Finite Difference Schemes . . . . . . V. Spectral Methods . . . . . . . . . . . . . . . . . VI. High-Order Finite Volume Schemes . . . . . . . VII. Finite Element Schemes . . . . . . . . . . . . . . VIII. Issues in Temporal Integration . . . . . . . . . . IX. Conclusions and Outlook . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . .

v

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

59 63

. 67 . 70 . 84 . 94 . 97 . 113 . 115 . 117

vi

CONTENTS

Prefiltering for Pattern Recognition Using Wavelet Transform and Neural Networks Fan Yang and Michel Paindavoine I. II. III. IV.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . Introduction to Neural Networks . . . . . . . . . . . . Introduction to Wavelet Transform . . . . . . . . . . . Pattern Recognition Using Wavelet and Neural Network for Signal and Image Processing Applications . . . . . . V. Wavelet Prefiltering Technique Applied to Handwriting Movement Analysis and Face Recognition . . . . . . . . Appendix . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

126 127 147

. . .

158

. . . . . . . . .

177 199 201

. . . . . . . .

. . . . . . . .

209 214 217 222 326 353 358 361

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

381

Electron Optics and Electron Microscopy: Conference Proceedings and Abstracts as Source Material P. W. Hawkes I. II. III. IV. V. VI. VII.

Introduction . . . . . . . . . . . . International Congresses (ICEM) . Regional Congresses . . . . . . . National Conferences . . . . . . . Thematic meetings . . . . . . . . Acknowledgements . . . . . . . . Acronyms . . . . . . . . . . . . . References . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

CONTRIBUTORS

Numbers in parentheses indicate the pages on which the authors’ contributions begins.

Yasuo Cho (1), Research Institute of Electrical Communication, Tokohu University, Sendai 980-8577 Japan P. W. Hawkes (207), CEMES-CNRS, F-31055 Toulouse cedex, France J. S. Hesthaven (59), Division of Applied Mathematics, Brown University, Providence, Rhode Island 02912 Michel Paindavoine (124), Laboratoire Electronique, Informatique et Image—CNRS FRE 2309, Aile des Sciences de l’Inge´nieur, Universite´ de Bourgogne, BP 47870-21078 Dijon cedex, France Fan Yang (124), Laboratoire Electronique, Informatique et Image—CNRS FRE 2309, Aile des Sciences de l’Inge´nieur, Universite´ de Bourgogne, BP 47870-21078 Dijon cedex, France

vii

This Page Intentionally Left Blank

PREFACE

The first three of the four contributions that make up this volume cover themes that I find particularly interesting and timely. We begin with an account by Yasuo Cho of a form of microscopy, introduced by the author and his colleagues, with the aid of which the microscopic distribution of polarization in ferromagnetic materials can be visualized directly. Readers of these Advances will not need to be reminded of the importance of ferroelectric materials in modern electronics applications and this technique for observing the essential characteristics of such materials is hence of the highest interest. The numerical calculation of electromagnetic fields has been the object of several articles here, but these have usually been biased toward static fields. The chapter by J.S. Hesthaven is therefore all the more welcome in that it is concerned with time-dependent fields. The author first points out the limitations of using second-order schemes for solving Maxwell’s equations. The solution proposed consists in going to higher order. The various ways of doing this are then discussed in turn: the finite-difference approach, spectral methods, finite-volume schemes, and the finite-element procedure. In the third chapter we return to wavelet transforms and neural networks, various aspects of which have been examined in earlier volumes of these Advances, since these are topics in rapid expansion. After a brief introduction to the subject, the authors describe the various wavelet transforms and show how many pattern recognition tasks can be addressed with the aid of wavelets and neural networks. Examples from several very different applications are presented. Finally, a wavelet prefiltering technique that has proved useful for handwriting and drawing analysis and for face recognition is described. The fourth chapter is a compilation of information about congresses and congress publications in the broad area of electron microscopy and microanalysis and electron optics. Much of this information had accumulated in my files over the years and, in view of the difficulty I had encountered in filling some of the gaps, I decided to make it permanently available by publishing it here. The various sections provide full information about many of the major series of congresses and the associated publications. First are the International, European, Asia-pacific, and South American congresses on (electron) microscopy. These are followed by all the information I have been able to collect about the many national congresses. ix

x

PREFACE

Here there are some major gaps. In some cases, I was not able to obtain any response from the national microscopy society of the country in question (Bulgaria, Latvia, Egypt, among others) while in others, Asia in particular, it was difficult to track down information about past events. In the final group are thematic meetings such as the High-voltage Electron Microscopy congresses and conferences on charged-particle optics, scanning electron microscopy, microanalysis, the ‘‘three-beam’’ and related meetings, and Compumag. A reasonably complete list of references of a historical or bibliographical character completes the article. I hope that this compilation will be of some service to future historians of the subject. In conclusion, I thank most warmly all the contributors for taking so much trouble to make their chapters accessible to non-specialists and list articles promised for future volumes. Peter W. Hawkes

FUTURE CONTRIBUTIONS

T. Aach (vol. 128) Lapped transforms S. van Aert, A. den Dekker, A. van den Bos, and D. van Dyck (vol. 130) Statistical experimental design for quantitative atomic-resolution transmission electon microscopy G. Abbate New developments in liquid-crystal-based photonic devices S. Ando Gradient operators and edge and corner detection C. Beeli Structure and microscopy of quasicrystals I. Bloch (vol. 128) Fuzzy distance measures in image processing G. Borgefors Distance transforms B. L. Breton, D. McMullan, and K. C. A. Smith (Eds) Sir Charles Oatley and the scanning electron microscope A. Bretto (vol. 130) Hypergraphs and their use in image modeling H. Delingette Surface reconstruction based on simplex meshes R. G. Forbes Liquid metal ion sources E. Fo¨rster and F. N. Chukhovsky X-ray optics A. Fox The critical-voltage effect L. Frank and I. Mu¨llerova´ (vol.128) Scanning low-energy electron microscopy

xi

xii

FUTURE CONTRIBUTIONS

L. Godo and V. Torra Aggregation operators A. Go¨lzha¨user Recent advances in electron holography with point sources A. M. Grigoryan and S. S. Agaian Transform-based image enhancement algorithms with performance measure. A. Hanbury (vol. 128) Morphology on a circle H. F. Harmuth and B. Meffert (vol. 129) Calculus of Finite Differences in Quantum Electrodynamics M. I. Herrera The development of electron microscopy in Spain D. Hitz Recent progress on HF ECR ion sources K. Ishizuka Contrast transfer and crystal images G. Ko¨gel Positron microscopy W. Krakow Sideband imaging N. Krueger (vol. 130) The application of statistical and deterministic regularities in biological and artificial vision systems B. Lahme Karhunen–Loeve decomposition B. Lencova´ Modern developments in electron optical calculations M. A. O’Keefe Electron image simulation N. Papamarkos and A. Kesidis The inverse Hough transform G. Mauro d’Ariano and M. G. A. Paris (vol. 128) Quantum tomography

FUTURE CONTRIBUTIONS

xiii

K. S. Pedersen, A. Lee, and M. Nielsen The scale-space properties of natural images M. Petrou Image registration M. Rainforth Recent developments in the microscopy of ceramics, ferroelectric materials, and glass E. Rau Energy analysers for electron microscopes H. Rauch The wave-particle dualism J. J. W. M. Rosink and N. van der Vaart HEC sources for the CRT O. Scherzer (vol. 128) Regularization techniques G. Schmahl X-ray microscopy S. Shirai CRT gun design methods T. Soma Focus-deflection systems and their applications J.-L. Starck The curvelet transform I. Talmon Study of complex fluids by transmission electron microscopy M. Tonouchi Terahertz radiation imaging N. M. Towghi Ip norm optimal filters Y. Uchikawa Electron gun optics D. van Dyck Very high resolution electron microscopy

xiv

FUTURE CONTRIBUTIONS

K. Vaeth and G. Rajeswaran Organic light-emitting arrays C. D. Wright and E. W. Hill Magnetic force microscopy M. Yeadon Instrumentation for surface studies

ADVANCES IN IMAGING AND ELECTRON PHYSICS, VOL. 127

Scanning Nonlinear Dielectric Microscopy YASUO CHO Research Institute of Electrical Communication, Tohoku University, Sendai 980-8577, Japan I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Principle and Theory for SNDM . . . . . . . . . . . . . . . . . A. Principle of SNDM . . . . . . . . . . . . . . . . . . . . . 1. Capacitance Variation with Alternating Electric Field . . . . . . 2. System Setup of SNDM . . . . . . . . . . . . . . . . . B. Nonlinear Dielectric Imaging with Subnanometer Resolution . . . . . 1. Microscopic Observation of Area Distribution of Ferroelectric Domain Using SNDM . . . . . . . . . . . . . . . . . . 2. Comparison between SNDM Imaging and Piezoresponse Imaging . C. Theory for Nonlinear Dielectric Imaging . . . . . . . . . . . . . 1. General Theorem for the Capacitance Variation under an Applied Electric Field . . . . . . . . . . . . . . . . . . . . . 2. Theoretical Calculation for SNDM Image . . . . . . . . . . . III. Quantitative Measurement. . . . . . . . . . . . . . . . . . . . A. Difference between Needle-Type SNDM and Cantilever-Type SNDM . . B. Quantitative Measurement Using Needle-Type SNDM . . . . . . . 1. Quantitative Measurement of Linear Dielectric Constants Using Needle-Type SNDM . . . . . . . . . . . . . . . . . . . 2. Quantitative Measurement of Nonlinear Dielectric Constants Using Needle-Type SNDM . . . . . . . . . . . . . . . . . . . C. Quantitative Measurement of Dielectric Properties Using Scanning Nonlinear Dielectric Microscopy with Electroconductive Cantilever. . . 1. Quantitative Measurement of Linear Dielectric Constant Using Cantilever-Type SNDM . . . . . . . . . . . . . . . . . 2. Quantitative Measurement of Nonlinear Dielectric Constant Using Cantilever-Type SNDM . . . . . . . . . . . . . . . . . IV. Higher-Order Nonlinear Dielectric Microscopy . . . . . . . . . . . . A. Theory for Higher-Order Nonlinear Dielectric Microscopy . . . . . . B. Experimental Details of Higher-Order Nonlinear Dielectric Microscopy . V. Three-Dimensional Measurement Technique . . . . . . . . . . . . . A. Principle and Measurement System . . . . . . . . . . . . . . . B. Experimental Results . . . . . . . . . . . . . . . . . . . . VI. Application of SNDM Technique for High-Performance Ferroelectric Material and Devices. . . . . . . . . . . . . . . . . . . . . . A. Determination of the Polarities of ZnO Thin Films on the Polar Substrate B. Scanning Nonlinear Dielectric Microscopy Study on Periodically Poled LiNbO3 for a High-Performance Quasi-phase Matching Device . . . . C. Tbit/inch2 Ferroelectric Data Storage Based on SNDM . . . . . . . VII. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

2

. . . . .

3 3 3 6 7

. . . . . .

7 10 11

. . . . .

. . . . .

11 14 17 17 20

. .

20

. .

22

. .

25

. .

25

. . . . . . .

. . . . . . .

28 30 30 33 37 37 39

. . . .

41 41

. . . .

44 49 54 55

. . . .

1 Copyright 2003, Elsevier (USA). All rights reserved. ISSN 1076-5670/03

2

CHO

I. Introduction Recently, ferroelectric materials, especially in thin film form, have attracted the attention of many researchers. Their large dielectric constants make them suitable as dielectric layers of microcapacitors in microelectronics. They are also investigated for application in nonvolatile memory using the switchable dielectric polarization of ferroelectric material. For the characterization of such ferroelectric materials, a high-resolution tool is required for observing the microscopic distribution of remanent (or spontaneous) polarization of ferroelectric materials. With this background, we have proposed and developed a new, purely electrical method for imaging the state of the polarizations in ferroelectric and piezoelectric materials and their crystal anisotropy. It involves the measurement of point-to-point variations of the nonlinear dielectric constant of a specimen and is termed ‘‘scanning nonlinear dielectric microscopy (SNDM)’’ (Cho, Kirihara, and Saeki, 1995a,b; Cho, Kirihara, and Saeki, 1996; Cho, Atsumi, and Nakamura, 1997; Cho, Kazuta, and Matsuura, 1999; Odagawa and Cho, 2000a,b). This is the first successful, purely electrical method for observing the ferroelectric polarization distribution without the influence of the screening effect from free charges. To date, the resolution of this microscope has been improved down to the subnanometer order. Moreover, we demonstrate that the resolution of SNDM is higher than that of a conventional piezoresponse imaging by using the scanning force microscopy (SFM) technique (Gruverman et al., 1997; Eng et al., 1999a). In this chapter, we describe the theory for detecting polarization and the technique for the nonlinear dielectric response and report the results of the imaging of the ferroelectric domains in single crystals and thin films using SNDM. Especially in a measurement of lead zirconate titanate (PZT) thin film, it was confirmed that the resolution was subnanometer order. We also describe the theoretical resolution of SNDM and the quantitative measurement technique using SNDM. Next, we report the new SNDM technique. In the preceding conventional SNDM technique, we measure the lowest-order nonlinear dielectric constant, 333, which is a third-rank tensor. To improve the performance and resolution of SNDM, we have modified the technique so that higher nonlinear dielectric constants 3333 (fourth-rank tensor) and 33333 (fifthrank tensor) are detected. It is expected that higher-order nonlinear dielectric imaging will provide higher lateral and depth resolution. We confirmed this improvement over conventional SNDM imaging through experimentation and use the technique to observe the growth of a surficial

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

3

paraelectric layer on periodically poled LiNbO3 (Cho et al., 2001; Ohara and Cho, 2001; Cho and Ohara, 2001). In addition to this technique, a new type of scanning nonlinear dielectric microscope probe, called the ‘‘311-type’’ probe, and a system to measure the ferroelectric polarization component parallel to the surface are developed. This is achieved by measuring the ferroelectric material’s nonlinear dielectric constant 311 instead of 333, which is measured in conventional SNDM. Experimental results show that the probe can satisfactorily detect the direction of the polarization parallel to the surface (Odagawa and Cho, 2002). Finally, some applications of the SNDM technique for actual problems to be solved for high-performance ferroelectric materials and devices are described. At first, the polarities of piezoelectric thin films on piezoelectric substrates are also determined. Next, studies of the domain distribution of periodically poled LiNbO3 (PPLN) are performed for high-quality, quasi-phase matching (QPM) devices, and some very important results are obtained. Finally, the formation of artificial small-inverted domain is reported to demonstrate that the SNDM system is very useful as a nanodomain engineering tool. The nanosize domain dots were successfully formed in a single crystal of LiTaO3. This means that we can obtain a very high density ferroelectric data storage with the density above Tbit/inch2.

II. Principle and Theory for SNDM A. Principle of SNDM 1. Capacitance Variation with Alternating Electric Field The theory for our proposed microscope is basically the same as that for our previously reported method (Cho and Matsuno, 1992), with the addition that we account for the spatial variation of the nonlinear dielectric constants. This enables us to obtain the capacitance variation due only to the nonlinear effect independently, separating it from the capacitance variation due to temperature change, and also enables us to determine the nonlinear dielectric constant of each order. For the nonlinear dielectric material with a spontaneous polarization Ps3 along the z-axis, the polynomial expansion form of electric displacement D3 as the function of electric field E3, is 1 1 D3 ¼ Ps3 þ 33 E3 þ 333 E32 þ 3333 E33 þ . . . 2 6

ð1Þ

4

CHO

Z Applied voltage Vp,w p

Ps 3

e

33

e

333

e

3333

d

Figure 1. Capacitance variation with alternating electric field.

where 33, 333, and 3333 are a linear, lowest-order nonlinear, and higherorder nonlinear dielectric constant, respectively (Cho, Kirihara, and Saeki, 1995a,b; Cho, Kirihara, and Saeki, 1996). The even rank tensors, including the linear second-order dielectric constant ij, are insensitive to the states of the spontaneous polarization. On the other hand, the lowest-order nonlinear dielectric constant, ijk, is very sensitive to the spontaneous polarization and other properties of the crystals. For example, there is no ijk in a material with a center of symmetry; the sign of ijk changes in accordance with the inversion of the spontaneous polarization. As shown in Figure 1, we consider the situation in which a strong alternating electric field Ep3 with amplitude E ¼ Vp =d (Vp: amplitude of the applied voltage) and angular frequency !p is applied to the capacitor Cs, producing a change of the capacitance resulting from the nonlinear dielectric response. We define the z-axis (three-direction) as the direction of the spontaneous polarization Ps and consider the capacitance variation along this z-direction. ~ 3 is We consider the state in which a small high-frequency electric field E ~ and an used for measuring the dielectric constants with an amplitude E angular frequency !0 as ~ cos !0 t ~3 ¼ E E

ð2Þ

is superposed on the applied alternating electric field Ep3 as Ep3 ¼ Ep cos !p t

ð3Þ

~3 E3 ¼ Ep3 þ E

ð4Þ

~ and !p<< !0. We assume Ep>>E Under the first order approximation all field components can be expressed as

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

~3 D3  Ps3 ¼ Dp3 þ D

5 ð5Þ

where the subscript p denotes the state caused by a large applied electric field and the tilde, the state of a small high-frequency field. Substituting Eq. (4) into Eq. (1) and using Eq. (5) to subtract the spontaneous polarization Ps3 from the equation, we obtained ~ 3 Þ2 þ 1 3333 ðEp3 þ E ~ 3 Þ3 þ . . . ~ 3 ¼ 33 ðEp3 þ E ~ 3 Þ þ 1 333 ðEp3 þ E Dp3 þ D 2 6 1 1 2 3 ð6Þ þ 3333 Ep3 þ ... ¼ 33 Ep3 þ 333 Ep3 2 6 1 2 ~3 þ . . .ÞE þð33 þ 333 Ep3 þ 3333 Ep3 2

In this calculation, we neglected very small higher-order terms associated ~ 3 , . . . to obtain the first-order variation of electric displacement ~ 2, E with E 3 3 ~3. ~ D3 associated with E From Eq. (6), we calculate the relationship between the small high~ 3 and the small high-frequency electric displacefrequency electric field E ~ 3 induced by E ~ 3 as ment D ~ 3 ¼ ð33 þ 333 Ep3 þ 1 3333 E 2 þ . . .ÞE ~3 D p3 2

ð7Þ

Eq. (7) indicates that the linear dielectric constant 33 (Ep3) is changed by an electric field Ep3 applied from outside the specimen. The dielectric ‘‘constant’’ becomes 1 2 þ ... 33 ! 33 ðEp3 Þ ¼ 33 þ 333 Ep3 þ 3333 Ep3 2

ð8Þ

Using Eq. (3) and Eq. (8), the parallel plate capacitance Cs (t) is given by S Cs ðtÞ ¼ 33 ðEp3 Þ   d  S 1 1 33 þ 3333 Ep2 þ 333 Ep cos !p t þ 3333 Ep2 cos 2!p t  d 4 4

ð9Þ

where S and d are the area and the thickness of the capacitor, respectively. The first term of Eq. (9) shows the original capacitance plus a small static capacitance change, and the other terms indicate the alternating variation of the capacitance. Thus, the static capacitance becomes   1 S S  33 ð10Þ Cs0 ¼ 33 þ 3333 Ep2 4 d d The final formula indicating the ratio of the alternating capacitance variation
Cs (t) to the static value of capacitance without time dependence is

6

CHO


CS ðtÞ 333 3333 2 ¼ Ep cos !p t þ E cos 2!p t 33 433 p CS0

ð11Þ

From Eq. (11), we find that the alternating capacitance variation related to the lowest-order nonlinear dielectric constant 333 is directly proportional to the applied electric field Ep cos!pt with the same angular frequency !p. On the other hand, the alternating capacitance related to the higher-order nonlinear dielectric constant 3333 varies with the frequency 2!p and its amplitude is proportional to the square of the amplitude of the applied electric field. 2. System Setup of SNDM Figure 2 shtows the system setup of the SNDM using the LC lumped constant resonator probe (Cho, Atsumi, and Nakamura, 1997). In the figure, Cs (t) denotes the capacitance of the specimen under the center conductor (the tip) of the probe. Cs (t) is a function of time because of the nonlinear dielectric response under an applied alternating electric field Ep3 (¼ Ep cos !pt, fp ¼ 5–200 kHz). As the E3 component is dominant in all three electric fields along x-, y-, and z-axis, the ratio of the alternating variation of capacitance
Cs (t) to the static value of capacitance Cs0 without time dependence is given in Eq. (11), even when we use tip with a very small pointed end. This LC resonator is connected to the oscillator tuned to the resonance frequency of the resonator. The previously mentioned electrical parts [i.e., tip (needle or cantilever), ring, inductance, and oscillator] are assembled into a small probe for the SNDM. The oscillating frequency of the probe (or oscillator) (around 1.3  2 GHz) is modulated by the change of capacitance

Figure 2. Schematic diagram of SNDM.

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

7


Cs(t) because of the nonlinear dielectric response under the applied electric field. As a result, the probe (oscillator) produces a frequency modulated (FM) signal. By detecting this FM signal using the FM demodulator and lock-in amplifier, we obtain a voltage signal proportional to the capacitance variation. Each signal corresponding to 333 and 3333 is obtained by setting the reference signal of the lock-in amplifier at the frequency !p of the applied electric field and at the doubled frequency 2!p, respectively. Thus we can detect the nonlinear dielectric constant just under the needle and can obtain the fine resolution determined by the diameter of the pointed end of the tip and the linear dielectric constant of specimens. The capacitance variation caused by the nonlinear dielectric response is quite small (
Cs(t)/ Cs0 is in the range of 103 to 108.). Therefore the sensitivity of the SNDM probe must be very high. The measured value of the sensitivity of the previously mentioned lumped constant probe is 1022F, which is much higher than that of the scanning capacitance microscope (SCM), whose typical sensitivity is 1018F. B. Nonlinear Dielectric Imaging with Subnanometer Resolution 1. Microscopic Observation of Area Distribution of Ferroelectric Domain Using SNDM For this study, the tip of the lumped constant resonator probe was fabricated using electrolytic polishing of a tungsten wire needle or a metalcoated conductive cantilever. The radius of curvature of the chip was 1 m–25 nm. To check the performance of the new SNDM, first, we measured the macroscopic domains in a multidomain BaTiO3 single crystal.  Figure 3 shows the two-dimensional image of the so-called 90 a-c domain, which is obtained by a coarse scanning over a large area. The sign of the nonlinear dielectric constant 333 of the +c-domain is negative, whereas it is positive in the c-domain. Moreover the magnitude of 111 ¼ 222 is zero in the a-domain because BaTiO3 belongs to the tetragonal system at room temperature. Thus, we can easily distinguish the type of the domains. To demonstrate that this microscopy is also useful for the domain measurement of thin ferroelectric films, we measured a PZT thin film. Figure 4 shows the SNDM (a) and atomic force microscopy (AFM) (b) images taken from a same location of PZT thin film deposited on an SrTiO3 (STO) substrate using metal organic chemical vapor deposition. From the figure, it is apparent that the film is polycrystalline (from Figure 4b) and that each grain in the film is composed of several domains (from Figure 4a). From X-ray diffraction analysis, this PZT film belongs to the tetragonal

8

CHO



Figure 3. A two-dimensional image of the 90 a-c domain in a BaTiO3 single crystal and the cross-sectional (one-dimensional) image along the line A-A0 .

Figure 4. Images of a PZT film on a SrTiO3 substrate. (a) Domain patterns by SNDM; (b) surface morphology by AFM.

phase, and the diffraction peaks, corresponding to both the c-axis and a-axis, were observed. Moreover, in Figure 4a, the observed signals were partially of zero amplitude and partially positive. Thus, the images show that we succeeded  in observing 90 a-c domain distributions in a single grain of the film. These images of the film were taken from a relatively large area. Therefore, we also tried to observe very small domains in the same PZT film on STO substrate. The results are shown in Figure 5. The bright area and

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

9

Figure 5. Nanoscale ferroelectric domain on PZT thin film, (a) domain image, (b) cross sectional image of nanoscale 180 c-c domain (b) cross-sectional (one-dimensional) image of 0 phase signal along the line A-A .

the dark area correspond to the negative polarization and the positive polarization, respectively. The figure shows that we can successfully observe  a nanoscale 180 c-c domain structure. Figure 5b shows a cross-sectional image taken along line A-A0 in Figure 5a. As shown in this figure, we measured the c-c domain with the width of 1.5 nm. Moreover we found that the resolution of the microscope is less than 0.5 nm. However, as the data shown in Figure 5 are phase images, some readers may think that the subnanometer resolution of SNDM is not convincingly proven in the references because phase profiles are invariably abrupt and cannot be considered as the definition of the resolution and amplitude signals show more realistic resolution. Therefore, here we show the amplitudd images in Figure 6 to demonstrate that the resolution of SNDM is really subnanometer order. These images were taken from an epitaxial ˚ )/La-Sr-Co-O/SrTiO3. Its macroscopic surface PZT thin film (4000 A topography and the domain pattern of this PZT thin film are shown in Figure 7. Square c-domains and their surrounding a-domain strip pattern are clearly observed.

10

CHO

Figure 6. (a) Amplitude image of nanoscale ferroelectric domain on PZT thin film. (b) Cross-sectional amplitude image taken along A-A0 .

Figure 7. Macroscopic surface topography and domain pattern taken from an epitaxial ˚ )/La-Sr-Co-O/SrTiO3. PZT thin (4000 A

The strip shape domain pattern is seen in Figure 6. Figure 6b is a crosssectional image taken along line A-A0 in Figure 6a. From the distance between the clearly distinguishable structures in the image, it is apparent that SNDM has subnanometer resolution. 2. Comparison between SNDM Imaging and Piezoresponse Imaging Another commonly reported high-resolution tool for observing ferroelectric domains is piezoelectric response imaging using SFM (Gruverman et al., 1997; Eng et al., 1999). From the viewpoint of resolution for ferroelectric

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

11

Figure 8. Simultaneously taken images of a PZT film. (a) Topography by AFM, (b) domain patterns by SNDM, and (c) domain patterns by SFM (piezoimaging).

domains, SNDM will surpass the piezoresponse imaging because SNDM measures the nonlinear response of a dielectric material, which is proportional to the square of the electric field, whereas the piezoelectric response is linearly proportional to the electric field. The concentration of the distribution of the square of the electric field in the specimen underneath the tip is much higher than that of the linear electric field. Thus, SNDM can resolve smaller domains than those measured by piezoimaging technique. To prove this fact experimentally, we also performed simultaneous measurements of the same location of the just mentioned PZT film sample by using AFM (topography) imaging, SNDM imaging, and piezoimaging. The results are shown in Figure 8. These images were taken under the same conditions using the same metal-coated cantilever. From the images, we can prove that SNDM can resolve greater detail than a conventional piezoresponse imaging by using SFM technique. C. Theory for Nonlinear Dielectric Imaging 1. General Theorem for the Capacitance Variation under an Applied Electric Field In this section, we derive a general theorem for the capacitance variation due to the nonlinear dielectric response of a material under an applied electric field. As shown in Figure 9, we consider that a metal conductor with a charge Q and electrostatic potential V exists in the space where the dielectric materials with respective linear and nonlinear dielectric constants of lij and lijk ðl ¼ 1; 2; 3; . . .Þ are distributed.

12

CHO

Figure 9. Capacitance variation under an applied electric field.

Taking the nonlinear dielectric response of the materials into account, the relationship between the stored charge Q in the metal and its potential V can be generally expressed by the polynomial expansion as   1 1 ð12Þ Q ¼ Cs0 V þ V 2 þ V 3 þ . . . 2 6 Therefore the stored energy W in this system is given by Z 1 1 W ¼ VdQ ¼ Cs0 V 2 þ Cs0 V 3 þ . . . 2 3

ð13Þ

On the other hand, with the linear and nonlinear dielectric constants, the energy density wl stored in the lth dielectric material with an area of l can be expressed by 1 1 wl ¼ lij Ei Ej þ lijk Ei Ej Ek þ . . . 2 3

ð14Þ

where Ei (i ¼ 1, 2, 3 . . .) is the electric field in the material. In this equation and in the following equations, we employed the Einstein convention that a repeated suffix represents a summation with respect to this suffix. Integrating Eq. (14) over the whole area in the outside of the metal conductor, then picking out the term proportional to the cube of the potential V (or the cube of the electric field strength E) and comparing the corresponding term in Eq. (12), we finally obtain the relationship

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

1 1X Cs0 V 3 ¼ 3 3 l

Z

l

lijk Ei Ej Ek dv

13 ð15Þ

Thus, using the electric field Ei and nonlinear dielectric constant lijk, we can express the second order coefficient in Eq. (12) as X Z lijk Ei Ej Ek Cs0  ¼ dv ð16Þ V3

l l Next, we also consider the situation in which a relatively large voltage V0 is applied to the metal, producing a change in its differential capacitance due to the nonlinear dielectric response. For measurement of the differential ~ is also superposed capacitance variation, a small high-frequency voltage V on V0 (Cho, Kirihara, and Saeki, 1996; Cho and Matsuno, 1992). Thus the total voltage applied to the metal is ~ V ¼ V0 þ V Substituting Eq. (17) into Eq. (12), we obtain   1 2 ~ ~ ~ Q ¼ Q0 þ Q ¼ Cs0 V0 þ V þ ðV0 þ V Þ þ . . . 2 1 ¼ Cs0 V0 þ Cs0 V02 þ . . . 2 ~ þ ... þ ðCs0 þ Cs0 V0 þ . . .ÞV

ð17Þ

ð18Þ

From Eq. (18), we can calculate the relationship between the small high~ induced by V ~ and the small high-frequency charge Q ~. frequency voltage V ~ to V ~ , we obtain ~ gives the differential capacitance C As the ratio of Q ~ ¼Q ~ =V ~ ¼ Cs0 þ Cs0 V0 þ : : : ¼ Cs0 þ
Cs þ : : : C

ð19Þ

where
Cs denotes the first-order differential capacitance variation and is given by
Cs ¼ Cs0 V0

ð20Þ

Thus, combining Eq. (16) and Eq. (20), we obtain the final formula that gives the first-order capacitance variation as Z l ijk Ei Ej Ek
Cs X ¼ dv ð21Þ V0 V3

l l This equation indicates that the first-order variation of differential capacitance per unit applied voltage in a system with arbitrarily shaped boundaries can be obtained exactly from the calculation of the stored energy attributable to the nonlinear dielectric constant lijk . Therefore, Eq. (21) gives

14

CHO

the general theorem for the capacitance variation under an applied electric field. 2. Theoretical Calculation for SNDM Image By applying Eq. (21) to a model of the tip of an SNDM probe as shown in Figure 10, we consider the SNDM image, theoretically. As in many papers on scanning probe microscopy, we also modeled the needle tip as a spherical conductor with radius a, assuming that the thickness of the sample is much larger than the diameter of the pointed end of the needle (Gao and Xiang, 1998; Gao et al., 1997). Generally, the nonlinear dielectric constant lijk , which is a third-rank tensor, exists in anisotropic material, and there is no lijk in a material with a center of symmetry. In this calculation, because the E3 component is dominant in all three electric fields along the x-, y-, and z-axis, we use an approximation for simplicity, assuming the relationship between the field strength of the electric displacement D3 and that of the electric field E3 is 1 D3 ¼ 33 E3 þ 333 E32 þ . . . 2

ð22Þ

Z

a

Y X

,

e33 e333

Figure 10. The model of the needle tip of an SNDM for calculating the nonlinear dielectric signal.

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

15

where 33 and 333 are the linear and nonlinear dielectric constants defined in the Z-direction. From Eqs. (21) and (22), the first-order capacitance variation in this model is obtained as Z  3
CS Ez ¼ 333 dv ¼ 333 Snl ð33 Þ ð23Þ V0 V where the parameter Snl (33) defined by the equation is the quantity that gives the capacitance variation per unit of nonlinear dielectric constant 333, and we term this Snl the ‘‘capacitance variation susceptibility.’’ The electric fields in this model combining the infinitely thick dielectric plate and the spherical conductor are given in the chapters describing the image-charge method found in the standard textbooks of electromagnetic theory (Goto and Yamasaki, 1970). Defining the normalized coordinates with respect to the radius a of the tip of the needle as x y z ¼ X ; ¼ Y ; and ¼ Z ð24Þ a a a and the three components of the electric field Ex, Ey, and Ez are given by Ei ðx; y; zÞ ¼

V E i ðX ; Y ; ZÞ a

ði ¼ 1; 2; 3Þ

ð25Þ

The newly defined normalized electric field components, E x ; E y , and E z in Eq. (25) are functions of the linear dielectric constant 33 and the normalized coordinates X, Y, and Z only and are independent of the radius a and the voltage V. Substituting Eq. (25) into Eq. (23), we obtain Z Z 0 Z 1Z 1
Cs Ez3 3 ¼ 333 dv ¼  E z dXdYdZ ð26Þ 333 3 V0 1 1 1 dielectric V material

Thus the capacitance variation susceptibility Snl (33) is given by Z 0 Z 1Z 1 3 E z dXdYdZ Snl ð33 Þ ¼ 1

1

1

ð27Þ

Snl (33) is a function of 33 only and does not depend on the tip radius a. This implies that the probe sensitivity or signal strength of SNDM does not change even if we choose a probe needle with a smaller tip radius to obtain a finer resolution. In other words, in principle, we can use an infinitely thin probe needle and obtain a clearly resolved image without degradation of the signal-to-noise ratio of the SNDM signal. This is a beneficial feature of SNDM for observing very small ferroelectric domains and local crystal anisotropy.

16

CHO

Capacitance variation susceptibility

0.3

0.2

0.1

0 1

100

200

300

400

500

Relative dielectric constant Figure 11. Capacitance variation susceptibility as a function of linear relative dielectric constant 33/0.

We calculated Snl (33) as a function of the relative dielectric constant of the specimen and obtained the result shown in Figure 11. The value of Snl (33) is almost constant (0.3) for 33 > 10. An application of Snl (33) to SNDM imaging is to estimate what depth information can be obtained using SNDM. We calculated the depth sensitivity of SNDM by integrating the region from the sample surface to the position Z ¼ H ¼ (h/a) shown in Figure 12. R0 R1 R1 3 E dXdYdZ ð28Þ DSnl ð33 ; HÞ ¼ H 1 1 z Snl ð33 Þ DSnl (33, H) gives the ratio of the signal arising from the region between the surface Z ¼ 0 and the position Z ¼ H to the whole signal strength of the SNDM (where H denotes the depth normalized to the tip radius a). From the calculated results shown in Figure 13, it is clear that the SNDM is sensitive in the very shallow areas, especially when the dielectric constant is large. These results are quite understandable because the electric field under the needle is more highly concentrated with a larger dielectric constant.  Next, we calculated one-dimensional images of the 180 c-c domain boundary lying at Y ¼ 0 as shown in Figure 14. These images are obtained by calculating the following equations.

17

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

Figure 12. Region of integration for obtaining depth sensitivity.

2 Im ð33 ; Y0 Þ ¼ Snl ð33 Þ

Z

Y0 1

f ðY ÞdY 

Z

1 1

f ðY ÞdY



ð29Þ

and f ð33 ; Y Þ ¼

Z

0 1

Z

0

1

3

E z dXdZ

ð30Þ

where Y0 is the tip position normalized to the tip radius a. Figure 15 shows the calculated results. The resolution of the SNDM image is heavily dependent on the dielectric constant of the specimen. For example, for the case of 33 =0 ¼ 1000 and a ¼ 10 nm (a needle tip with a radius of 10 nm is easily obtainable), the resolution is about 0.1 nm. Thus, we conclude that an atomic scale image can be taken by SNDM. III. Quantitative Measurement A. Difference between Needle-Type SNDM and Cantilever-Type SNDM There have been two types of SNDM developed using the LC lumped constant resonator probe. In one type of SNDM, we have used an electrolytic-polished tungsten wire as a probe tip (the needle-type) (Cho,

18

(a)

e33

(b)

= 10

D Snl (e33, H )

D Snl (e33, H )

0.8 0.6 0.4

0.8 0.6 0.4

0

0.5

1

1.5

2

2.5

0

3

0

0.5

Normalized depth H

e33

(d)

= 300

0.8

D Snl (e33, H )

D Snl (e33, H )

1

1

1.5

2

2.5

3

Normalized depth H

0.6 0.4

e33

1

CHO

(c)

= 1000

0.8 0.6 0.4 0.2

0.2 0

= 30

0.2

0.2 0

e33

1

1

0

0.1

0.2 0.3 Normalized depth H

0.4

0.5

0

0

0.005

0.01 0.015 Normalized depth H

Figure 13. Depth sensitivity of the SNDM. H denotes the depth normalized to the tip radius a.

0.02

0.025

19

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

Scanning a

Y = y /a Y0

P

P



Figure 14. Calculation for the one-dimensional image of the 180 c-c domain boundary.

er

Im(e33,Y 0)

1

= 10

(b)

0.5 0 −0.5

0 0.5

−2 −1.5 −1 −0.5

1

1.5 2

Im(e33,Y 0)

(a)

−1

Im(e33,Y 0)

−0.1

−0.05

0 −0.5

(d)

= 300

er

1

0.5 0

0.5

Y0( y/a)

0

−0.5 −1

0.05

0.1

Im(e33,Y 0)

er

1

= 30

−1

Y0( y/a) (c)

er

1

= 1000

0.5 0 −0.02 −0.01 −0.5 −1

0

0.01

0.02

Y0( y/a)

Y0( y/a) 

Figure 15. One-dimensional calculated images of the 180 c-c domain boundary. The abscissa denotes the normalized Y coordinate (= y/a).

Atsuni, and Nakamura, 1997; Cho, Kazuta, and Matsuura, 1999; Cho, Matsuura, and Kushibiki, 1998) and in the other type of SNDM, an electroconductive AFM cantilever has been used (the cantilever-type) (Odagawa and Cho, 2000a,b). The needle-type SNDM is useful for largearea measurement with a high scanning speed, and the cantilever-type

20

CHO

SNDM can measure linear and nonlinear dielectric properties in a very small area with quite a high (subnanometer order) resolution. In the case of the needle-type SNDM, the quantitative measurement of the linear and nonlinear dielectric constants can be relatively easily performed (Cho et al., 2000). On the other hand, if we use the same method for the linear and nonlinear dielectric constants using the cantilever-type SNDM, the quantitative measurement is difficult because of the existence of the beam of the cantilever. This capacitance between the beam and the specimen is quite large in comparison with that immediately under the tip so that this large stray capacitance masks the capacitance to be measured immediately under the tip and it prevents sensitivity calibration of the tip for the absolute value of a localized capacitance. Therefore, we have to develop a new quantitative measurement method using the cantilever-type SNDM (Ohara and Cho, 2002). In the following section, we first present a quantitative measurement method for the linear and nonlinear dielectric constant using a needle-type SNDM and then we describe the quantitative measurement method of the linear and nonlinear dielectric constants using cantilever-type SNDM. B. Quantitative Measurement Using Needle-Type SNDM 1. Quantitative Measurement of Linear Dielectric Constants Using Needle-Type SNDM In this section, we describe a quantitative method for measuring linear dielectric constants using needle-type SNDM with an LC lumped constant resonator probe. A conceptual figure describing the measurement of the linear dielectric constant is shown in Figure 16. In this figure, L and C0 show the inductance and the stray capacitance C0 (which inevitably exists between the needle and the electrical circuit of the probe), respectively. f0 is the carrier frequency of the probe when the probe needle is far from the specimen and is given by f0 ¼

1 pffiffiffiffiffiffiffiffiffi 2 LC0

ð31Þ


fs denotes the frequency shift from f0 when the needle makes contact with the specimen. Therefore f0+
fs is expressed by     1 1 1 Cs 1 Cs f0 þ
fs ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  pffiffiffiffiffiffiffiffiffi 1  ¼ f0 1  ð32Þ 2 C0 2 C0 2 LðC0 þ Cs Þ 2 LC0

where Cs is the static capacitance variation from C0 by the tip contact

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

21

L L Needle C0 C0 f0

f0 + ∆ fs a

∞ Dielectric substrate

Cs

Metal stage Figure 16. Conceptual figure describing the quantitative measurement of the linear dielectric constant by the needle-type SNDM.

and is much smaller than C0 (i.e., Cs << C0). Thus, the ratio of
fs to f0 is given by
fs 1 Cs ¼ f0 2 C0

ð33Þ

Cs can also be calculated using the image charge method (Gao et al., 1997; Goto and Yamasaki, 1970) and is given as  1  ln 1b 1 ð34Þ Cs ¼ 40 a b Thus, we obtain  
fs a lnð1  bÞ ¼ 20 þ1 f0 C0 b

ð35Þ

where b¼

33  0 33 þ 0

ð36Þ

Eq. (35) indicates that we can obtain the linear dielectric constant of the specimen by a relative measurement using a standard sample, even if the values of the stray capacitance C0 and the tip radius a are unknown. To check these results, we measured the carrier frequency shift
fs for several dielectric materials. The results are shown in Figure 17 with the theoretical curve adjusted to the data of the standard sample. In this

22

CHO

Frequency shift ∆ fs [MHz]

5

STO (Standard sample)

LiTaO3 LiNbO3 Sapphire SiO2 0

Relative dielectric constant

er

400

Figure 17. Carrier frequency shift
fs vs relative dielectric constant r. The solid line indicates the theoretical value which is adjusted to the data of the standard sample (STO).

measurement, for the standard sample we chose SrTiO3 (STO) single crystal. Given that both theoretical and experimental data agree, it is clear that we can determine the absolute value of the linear dielectric constant by a relative measurement using a standard sample. Therefore, as a trial, we performed a quantitative measurement of the linear dielectric constant  distribution of a 128 rotated Y-cut LiNbO3 substrate with a titaniumdiffused inversion layer. The results are shown in Figure 18a together with its polarization image (nonlinear dielectric phase image). In Figure 18b, the negative sign of the signal shows the original direction of the polarization (Cho et al., 1999), so that we can determine, from the data in Figure 18a, that the linear dielectric constant in the titanium-diffused area is smaller than the original constant. As is well known in the optical region, where the electrical polarization governs the dielectric phenomenon, the dielectric constant (the refractive index) in the titanium-diffused area is larger than the original constant. However, it is found from this experiment that in the microwave region, where ionic polarization is dominant, the dielectric constant decreases after titanium diffusion. 2. Quantitative Measurement of Nonlinear Dielectric Constants Using Needle-Type SNDM Using Snl (33), as defined in Section II.C.2, we can measure the nonlinear dielectric constant quantitatively using the needle-type SNDM. As shown in Figure 19, we consider a situation in which a relatively strong alternating voltage with amplitude V0 is applied to the capacitor Cs, producing a change

23

SCANNING NONLINEAR DIELECTRIC MICROSCOPY



Figure 18. Images of 128 rotated Y-cut LiNbO3 with a titanium-diffused inversion layer. (a) Linear dielectric image; (b) polarization image (nonlinear dielectric image).

Applied voltage

C0 f0 + ∆ fs

V a Dielectric substrate

Cs + ∆Cs e333

Metal stage Figure 19. Conceptual figure of the measurement of the nonlinear dielectric constant.

in the capacitance
CS resulting from the nonlinear dielectric response. In this situation, the ratio of the frequency deviation
fd caused by the applied voltage to the carrier frequency fs ð¼ f0 þ
fs Þ is given by the equation
fd 1
Cs 1
Cs  ¼ fs 2 ðC0 þ Cs Þ 2 C0

ð37Þ

24

CHO

From Eq. (37) and Eq. (23), it can be seen that if the tip radius a and the linear dielectric constant 33 are known, we can directly measure the nonlinear dielectric constant (absolute measurement) because C0 can be calculated using Eq. (35). Moreover, even if the radius a is unknown, we can determine the nonlinear dielectric constant from a relative measurement using a standard sample and the following equation.   ui st Snl st 33 fs
fd ui st  ui  ui ð38Þ 333 ¼ 333 Snl 33 fs
fdst

where the superscripts st and ui mean ‘‘standard sample’’ and ‘‘under investigation,’’ respectively.
fdui and
fdst are the measured frequency deviations under the same applied voltage condition of V0. Based on the results, we performed an absolute measurement of the nonlinear dielectric constants of Z-cut LiNbO3 and Z-cut LiTaO3 substrates. The radius of the tip needle was 12.5 m, and the data are summarized in Table 1. Unfortunately, there have been few reported values of nonlinear dielectric constants with which the measured data can be compared to check the accuracy of the preceding method. Therefore, we also determined the standard values of 333 for LiNbO3 and LiTaO3 by using another accurate standard method termed the ‘‘dynamic measuring method of capacitance variation with alternating electric field’’ (Cho and Matsuno, 1992). In this measurement of the standard values of 333, we measured the capacitance variation of parallel plate capacitors made of Z-cut LiNbO3 and LiTaO3 substrates with a size of 10 mm  10 mm  0.5 mm. The values of 333 obtained from the absolute measurement by using SNDM and the standard values of 333 are given in Table 2. The data show good agreement with each other, although the exact evaluation of the tip radius a is fairly difficult. Of course, the relative measurement for 333 is easier and more accurate to obtain than the absolute measurement. Thus, we also performed the relative measurement of the nonlinear dielectric constant 333 of Z-cut LiTaO3 using the Z-cut LiNbO3 as the standard sample. In this measurement, we used a probe needle with an unknown tip radius. The result is also shown in TABLE 1 Measured Data for Nonlinear Dielectric Constant by SNDM Sample

V0(V )

f0 + fs(MHz)

fd (Hz)

f0(MHz)

LiNbO3 LiTaO3

1.0 1.0

1069.164 1069.045

21.2 35.4

1070.239 1070.239

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

25

TABLE 2 Nonlinear Dielectric Constant of LiNbO3 and LiTaO3 Obtained from the Absolute Measurements of SNDM, with Their Standard Values Measured by the Dynamic Method

Sample

Nonlinear dielectric constant by SNDM (F/V)

Nonlinear dielectric constant by the dynamic method (F/V)

2.91  1019 5.01  1019

1.12  1019 2.26  1019

LiNbO3 LiTaO3

TABLE 3 Nonlinear Dielectric Constant of LiTaO3 Obtained from the Relative Measurement Using the Z-cut LiNbO3 as a Standard Sample and Its Standard Value

Sample LiNbO3 (standard data) LiTaO3

Nonlinear dielectric constant by SNDM (F/V)

Nonlinear dielectric constant by the dynamic method (F/V)

1.12  1019 1.93  1019

1.12  1019 2.26  1019

Table 3. From the relative measurement, the value of the nonlinear dielectric constant of LiTaO3 was 1.8 times larger than that of LiNbO3. In comparison with the standard value of 333 of LiTaO3 in Table 2, which is 2.0 times larger than that of LiNbO3, both values from the relative measurement and the standard values of the nonlinear dielectric constants show good agreement. C. Quantitative Measurement of Dielectric Properties Using Scanning Nonlinear Dielectric Microscopy with Electroconductive Cantilever 1. Quantitative Measurement of Linear Dielectric Constant Using Cantilever-Type SNDM We consider the model of the cantilever-type SNDM as shown in Figure 20. This figure shows that the capacitance as a function of the linear dielectric constant  is composed not only of the capacitance immediately under the tip CS but also of that between the beam of the cantilever and the specimen CL. Moreover, the capacitance CL is much larger than the capacitance CS, such that the capacitance CL masks the capacitance CS. Therefore, we cannot use the quantitative measurement method of the needle-type SNDM (which does not have the capacitance CL) without separating the

26

CHO

Figure 20. Model of the cantilever-type SNDM for measuring the average value of the linear dielectric constant.

capacitance CL from the capacitance CS for measuring the dielectric constant distribution quantitatively. Therefore, in this section, we describe the method of separating the capacitance CL from the capacitance CS. In Figure 20, we can approximate that the frequency shift
fL is due only to the capacitance CL because the capacitance CL is much larger (almost 1000 times larger) than the capacitance CS. Here, the capacitance CL is a function of the average value of the linear dielectric constant of the specimen because the area under the cantilever beam is much larger than the scanning area. Thus, we can determine the average value of the linear dielectric constant of the specimen  by detecting the frequency shift
fL. Figure 21 shows the relationship between the average value of relative linear dielectric constant r and the frequency shift
fL. In this figure, the solid line is the standard curve measured using standard samples. Using this curve and the measured data of
fL we can determine the average value of the linear dielectric constant of specimen. Next, we consider a situation wherein the tip of the cantilever scans on the specimen and SNDM detects the capacitance variation as a function of the microscopic linear dielectric constant distribution of the specimen, as shown in Figure 22. In this case, the frequency varies with the frequency deviation
f 0 s from fL as a function of the microscopic linear dielectric constant distribution. This
f 0 s is given by

27

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

Figure 21. Experimental result of the average value of the linear dielectric constant.

Figure 22. Model of the cantilever-type SNDM for measuring the linear dielectric constant distribution.

 

12
CS 1 þ C0 þC  1 fL L þCS !  12 2 1 fL  1 fL ¼ 1 þ C0 f0
CS


fs0 ¼

ð39Þ

Here, we used the relationship between the frequency and the capacitance as shown in Eq. (40).  2 fL C0 ¼ ð40Þ f0 C0 þ CL þ CS In Eq. (39), the capacitance variation
CS can be calculated similarly to Eq. (34) and is given by

28

CHO

Figure 23. Calibration data for tip sensitivity (linear dielectric constant distribution of titanium-diffused LiNbO3).

 lnð1  bÞ lnð1  b0 Þ ;
CS ¼ 40 a  b b0   0 0  þ
  0 where  !  þ
; b ¼ ; b ¼  þ 0  þ
 þ 0 

Thus, substituting Eq. (41) for Eq. (39), we obtain !12  2   
fs0 fL a lnð1  bÞ lnð1  b0 Þ þ1 1 ¼ 40  f0 fL C0 b b0

ð41Þ

ð42Þ

Eq. (42) shows that we can obtain the linear dielectric constant distribution of the specimen by calibrating the tip sensitivity (a/C0) using a standard sample with a microscopic distribution of linear dielectric constant. Figure 23 shows the linear dielectric constant distribution of the standard sample (titanium-diffused-LiNbO3) for calibrating the tip sensitivity. [We also measured the linear dielectric constant distribution of the same titaniumdiffused-LiNbO3 using the needle-type SNDM and calculated the tip sensitivity of this cantilever employing Eq. (42).] Using the tip sensitivity of this cantilever, we determined the linear dielectric constant distribution of a two-phase composite ceramic composed of TiO2 and Bi2Ti4O11(Cho, Jintsugawa and Yamanouchi, 2000) quantitatively, as shown in Figure 24a. Figure 24b shows the SEM image of this ceramic. In Figure 6b, the black area represents TiO2, and the white area represents Bi2 Ti4O11 and TiO2 and has a larger linear dielectric constant than Bi2Ti4O11. From Figure 24 we confirm that the cantilever-type SNDM can measure the linear dielectric constant distribution corresponding to the grain of the ceramics quantitatively. 2. Quantitative Measurement of Nonlinear Dielectric Constant Using Cantilever-Type SNDM Figure 25 shows the model of the cantilever-type SNDM for measuring the nonlinear dielectric constant. In this figure, the resonance frequency varies as a function of the alternating applied electric field Ep ¼ E0cos (!pt)

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

29

Figure 24. (a) Linear dielectric constant distribution of the TiO2-Bi2Ti4O11 ceramic. (b) SEM image of the TiO2-Bi2Ti4O11 ceramic.

L

C0

fL + ∆fd V CL Cs + ∆Cd

Ferroelectric material Metal stage

Figure 25. Model of the cantilever-type SNDM for measuring the nonlinear dielectric constant.

because the capacitance variation
Cd occurred by the nonlinear dielectric effect. Therefore, we can obtain the signal corresponding to the nonlinear dielectric constant by detecting the resonance frequency variation
fd. Moreover, in this case, the gap between the beam of the cantilever and the specimen is more than 15 m. Therefore, CL does not change with the alternating applied electric field Ep because SNDM has no sensitivity in case of the existence of the gap between the detector and the specimen. Then, we present the relationship between the frequency variation
fd and the capacitance variation
Cd as shown in Eq. (43).   1
Cd 1
Cd fL3 ð43Þ ¼
fd ¼  2 C0 þ CL þ CS 2 C0 f02

30

CHO TABLE 4 Measured Data of the Nonlinear Dielectric Constant by Cantilever-Type SNDM Sample

V (v)

fL (MHz)


fd (Hz)

LiNbO3 (standard sample) LiTaO3

1.0 1.0

1242.81 1242.68

7.78 13.45

where we use the relation as fL3 C0 ¼ f02 C0 þ CL þ CS

ð44Þ

On the other hand, from Eq. (43) the relationship between the capacitance variation
Cd and the nonlinear dielectric constant 33 is given by Z 0 Z 1Z 1
Cd 3 ¼ 333 ¼ E Z dXdYdZ ¼ 333 Snl ð33 Þ ð45Þ V 1 1 1 Thus, with the combination of Eqs. (43) and (45), the frequency variation
fd is represented by   1 V fL3 Snl ð33 Þ ð46Þ
fd ¼  333 2 C0 f02 This equation shows that we can measure the nonlinear dielectric constant quantitatively using a standard sample for calibrating the stray capacitance C0. Based on this result, we measure the nonlinear dielectric constant of Z-Cut LiTaO3 single crystal (LT) using the Z-Cut LiNbO3 single crystal (LN) as the standard sample. Table 4 shows the measured data by cantilever-type SNDM. From these results, we obtain the equation LN LT 333 ¼ 1:8  333 . We think this data is good agreement with the previously LN mentioned standard data, which is LT 333 ¼ 2:0  333 . IV. Higher-Order Nonlinear Dielectric Microscopy A. Theory for Higher-Order Nonlinear Dielectric Microscopy Eq. (47) is a polynomial expansion of the electric displacement D3 as a function of electric field E3. 1 1 1 D3 ¼ Ps3 þ 33 E3 þ 333 E32 þ 3333 E33 þ 33333 E34 þ . . . 2 6 24

ð47Þ

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

31

where 33, 333, 3333, and 33333 correspond to linear and nonlinear dielectric constants and are tensors of rank second, third, fourth, and fifth, respectively. Even-ranked tensors including linear dielectric constant 33 do not change with polarization inversion, whereas the sign of the oddranked tensors reverses. Therefore, information regarding polarization can be elucidated by measuring odd-ranked nonlinear dielectric constants such as 333 and 33333. Considering the effect up to E 4, the ratio of the alternating variation of capacitance
Cs underneath the tip to the static capacitance Cs0 is given by
Cs ðtÞ 333 1 3333 2 Ep cos !p t þ E cos 2!p t  33 Cs0 4 33 p 1 33333 3 E cos 3!p t þ . . . þ 24 33 p

ð48Þ

This equation shows that the alternating capacitance of different frequencies corresponds to each order of the nonlinear dielectric constant. Signals corresponding to 333, 3333, and 33333 were obtained by setting the reference signal of the lock-in amplifier in Figure 2 to frequency !p, 2!p, and 3!p of the applied electric field, respectively. Next, we consider the resolution of SNDM. From Eq. (48), the resolution of SNDM is found to be a function of electric field E. We note that the electric field under the tip is more highly concentrated with the increase of 33 (Matsuura, Cho, and Odagawa, 2001), and the distributions of E2, E3, and E 4 fields underneath the tip become much more concentrated in accordance with their power than those of the E field, as shown in Figure 26. From this figure, we find that higher-order nonlinear dielectric imaging has a higher resolution than lower-order nonlinear dielectric imaging. In addition, we show the theoretical results of the higher-order nonlinear dielectric imaging using the general theorem for capacitance variation due to

Figure 26. Distribution of E, E2, E3, and E 4 fields under the needle tip. (a, Denotes tip radius.)

32

CHO

the nonlinear dielectric response of a material under an applied electric field in Section II.C.
C ¼ 333 Snl V

ð49Þ


C 0 0 ¼ 3333 Snl V2

ð50Þ


C 00 00 ¼ 33333 Snl V3

ð51Þ

where V denotes applied voltage,
C,
C0 , and
C00 are the capacitance variation with angular frequencies !p, 2!p, and 3!p, respectively, and Snl, S0 nl, and S00 nl are the capacitance variation susceptibilities related to 333, 3333, and 33333, respectively. Snl, S0 nl, and S00 nl are the normalized sensitivity with respect to 333, 3333, and 33333 and are expressed as Z

Snl ð33 Þ ¼ 0 Snl ð33 Þ ¼

00 Snl ð33 Þ ¼

0

1 6a

1

0

Z

Z

1

Z

0

1

1

Z

1

1

Z

3

E z dXdYdZ

1

1

1

1 2a

Z

1

1

Z

1

1

Z

1

1

4

E z dXdYdZ 5

E z dXdYdZ

ð52Þ ð53Þ ð54Þ

Based on these equations, we calculated the theoretical one-dimensional image when the 180 c-c domain boundary lies at Y ¼ 0 as shown in Figure 14. Figure 27 shows the results for 333 and 3333 imaging where the linear dielectric constant 33/0 of the material is assumed to be 30 and 300, respectively. In the figure, the horizontal axis is the coordinate normalized by tip radius a. Comparing Figure 27a with Figure 27b, we understand that the resolution is higher when 33 is larger. In addition, 33333 imaging has a much higher resolution than 333 imaging. For example, in the case of 33/0 ¼ 300, the resolution of 333 imaging is approximately 0.03 times the tip radius a, and that of 33 imaging is approximately 0.005 times. Next, we calculated the depth sensitivity for 33, 333, 3333, and 33333 SNDM imaging. This sensitivity is obtained by integrating the SNDM signal that arises in the region from the sample surface to the position Z ¼ H (¼ h/a) shown in Figure 28a. Figure 28b and c show the results for

33

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

(a)

Signal strength [arb. units]

1 0.5

e333 e33333

−0.3

−0.2

−0.1

0

0

0.1

0.2

0.3

−0.5 −1 Coordinate normalized by tip radius a

Signal strength [arb. units]

(b)

1 0.5

e333 e33333

−0.03

−0.02

−0.01

0

0

0.01

0.02

0.03

−0.5 −1 Coordinate normalized by tip radius a 

Figure 27. Calculated 333 and 33333 images of 180 c-c domain boundary. (a) 33/0 = 30. (b) 33/0 = 300. The abscissa denotes the normalized coordinate.

the specimen with 33/0 of 30 and 300, respectively. In Figure 28b and c, the horizontal axis also denotes the depth normalized by tip radius a. It is clear that SNDM is sensitive in very shallow areas, especially when the linear dielectric constant 33 is large; as expected, higher-order nonlinear dielectric imaging can detect a much thinner layer than lower-order nonlinear dielectric imaging. From these calculated results, it is revealed that higher-order nonlinear dielectric imaging has a higher resolution and senses a much shallower area. Thus, higher-order imaging is very useful for observing a surface layer with unit cell scale thickness formed on ferroelectric material. B. Experimental Details of Higher-Order Nonlinear Dielectric Microscopy We confirmed through experiments that 33333 imaging has higher lateral resolution than 333 imaging using an electroconductive cantilever as a tip with a radius of 25 nm. Figure 29 shows 333 and 33333 images of the twodimensional distribution of lead zirconate titanate (PZT) thin film. The two

34

CHO

Figure 28. (a) Integration region for obtaining depth sensitivity. (b) Depth sensitivity for 333 and 33333 SNDM imaging (33/0 = 30). (c) Depth sensitivity for 333 and 33333 SNDM imaging (33/0= 300). H denotes the depth normalized by the tip radius a (H = h/a).

images can be correlated, and it is clear that the 33333 image resolves greater detail than the 333 image because of the higher lateral and depth resolutions. Next, we investigated the surface layer of periodically poled LiNbO3 (PPLN) by 333, 3333, and 33333 imaging. Figure 30a shows 333, 3333, and

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

35

Figure 29. (a) 333 and (b) 33333 images of PZT thin film.

33333 signals of virgin unpolished PPLN. In this figure, only 333 imaging detects the c-c domain boundary, whereas 33333 imaging does not. The 3333 signal shows weak peaks at domain boundaries. The reason is that 3333 and 33333 imaging is affected by the surface paraelectric layer. To prove the existence of a surface paraelectric layer, we polished and measured the PPLN. Figure 30b shows the images of it. In this figure, it is clear that 33333 imaging can detect the c-c domain boundary after removal of the paraelectric layer. Moreover, 3333 imaging can also detect periodic signals, in contrast to our expectation. The nonlinear dielectric signals of a positive area of PPLN are stronger than those of a negative area immediately after polishing, possibly because the negative area is more easily damaged than the positive area and has already been covered by a very thin surface paraelectric layer with weak nonlinearity even immediately after polishing. One hour after polishing, we conducted the 333, 3333, and 33333 imaging again, and the results are shown in Figure 30c. In this figure, the 33333 signal disappears and the 3333 signal becomes flat again, whereas 333 imaging clearly detects the c-c domain boundary. This implies that the entire surface area of PPLN is covered by the surface paraelectric layer again. From theoretical calculations on the LiNbO3 substrate 33333 and 3333 imaging has sensitivities down to 0.75-nm depth and 1.25 nm, respectively, whereas 333 imaging has sensitivity down to 2.75-nm depth when a tip of 25-nm radius is used. Thus, we conclude that the thickness of this surface paraelectric layer ranges between 0.75 and 2.75 nm. From these results, we succeed in observing the growth of the surface layer, and we confirm that the negative area of LiNbO3 can be more easily damaged than the positive area.

36

CHO

(a)

e

(5)

e

(4)

Paraelectric layer

(3)

e

0.1

2

100 333 Signal e 3333 Signal e 33333 Signal

Signal [Hz/V3]

Signal [Hz/V2] 0

0

e

e

0

33333

e333

50

3333

Signal [Hz/V]

e

−0.05

−1

−50 0

38

Distance [ m m] Paraelectric damaged layer

(b) e

(5)

e

(4)

(3)

e

333

−50 0

Distance [ m m] Figure 30. (Continued )

38

−1

0

e

Signal e 3333 Signal e 33333 Signal e

33333

Signal [Hz/V3]

Signal [Hz/V2] 0

0

3333

50

e

Signal [Hz/V] e333

0.1

2

100

−0.05

37

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

(c) e

(5)

e

(4)

Paraelectric layer

(3)

e

100

0.1

2 Signal Signal e 33333 Signal e

0

0

Signal [Hz/V3] 0

−1

−50 0

Distance [ m m]

e

33333

Signal [Hz/V2]

50

3333

3333

e

e333

Signal [Hz/V]

333

e

−0.05

38

Figure 30. (a) 333, 3333, and 33333 images of virgin PPLN; (b) immediately after polishing; and (c) 1 h after polishing.

V. Three-Dimensional Measurement Technique A. Principle and Measurement System A new type of scanning nonlinear dielectric microscope (SNDM) probe, named ‘‘311-type probe,’’ and a system to measure the ferroelectric polarization component parallel to the surface using SNDM have been developed (Odagawa and Cho, 2002). This is achieved by measuring a nonlinear dielectric constant of ferroelectric material 311, instead of 333, which is measured in conventional SNDM. Figure 31 shows parallel plate models of nonlinear dielectric constant measurements. Because precise descriptions of the 333 measurement have been mentioned we explain only the 311 measurement. We consider the situation in which a relatively large electric field E 3 with the amplitude Ep and angular frequency !p is applied to the capacitance Cs, producing a change of the capacitance resulting from the nonlinear dielectric response. We detect the capacitance variation
Cs, which is perpendicular to the polarization direction (z-axis) by a high-frequency electric field with small ~ 1) as shown in Figure 31b. (In the 333 amplitude along x-axis (E

38

CHO

measurement, we detect
Cs along the direction of spontaneous polariza~ . We call tion Ps3.) That is, in the 311 measurement, E is perpendicular to E this type of measurement, which uses the crossed electric field, an 311-type measurement. In this case, the final formula is given by
Cs ðtÞ 311 3311 2 Ep cos !p t þ E cos 2!p t ¼ 11 411 p Cs0

ð55Þ

where 11 is the linear dielectric constant and 311 and 3311 are nonlinear dielectric constants. From this equation, by detecting the component of capacitance variation with the angular frequency of the applied electric field !p, we can detect the nonlinear dielectric constant 311. According to this principle, we develop a 311-type probe for measuring the polarization direction parallel to the surface. Figure 32 shows a schematic diagram of the measurement system. We put four electro;des around the probe tip to supply the electric field E, which causes the nonlinear effect. Electrodes A and B supply E 3 , which is along the z-axis, and electrodes C and D supply (a)

Applied electric field

(b)

E3

e11

e33

Ps 3 E3

Applied electric field

~

e333 e3333

∆Cs

E3

e311 e3311

Ps 3

~

∆Cs

E1

Figure 31. Capacitance variation with alternating electric field. (a) 333 measurement and (b) 311 measurement.

Figure 32. Schematic configuration of (a) new 311 probe and (b) measurement system.

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

39

E 2 , which is along the y-axis. We apply voltages to the electrodes as satisfying the condition that E 3 and E 2 just under the tip become parallel to the surface without concentrating at the tip as shown in Figure 32b. (In Figure 32b the components related to y-direction are omitted for ~ for measuring simplification.) On the other hand, the electric field E the capacitance variation concentrates at the probe tip as the conventional ~ because measurement. It is sufficient to consider only the x-component of E ~ underneath the tip is perpendicular to the we confirmed that most of E surface. Moreover, we can obtain any electric field vector E with arbitrary rotation angle by combining the amplitude of E 2 and E 3 . Therefore, we need not rotate the specimen for detecting a lateral polarization with an arbitrary direction. B. Experimental Results Figure 33 shows the measurement result of PZT thin film as changing the direction of applied electric field E. In Figure 33a, where E is parallel to the polarization direction, the pattern corresponding to the polarization is observed, whereas no pattern is observed in Figure 33b because, in this case, E is perpendicular to the polarization direction. Figure 33c and d are the cases where E is applied along the intermediate direction. The pattern can be observed in Figure 33c and d because the vector E can be divided by the component along the polarization direction. However, the opposite contrast

Figure 33. Images of PZT thin film, (a)–(d) 311 images, (e) 333 image, and (f) topography.

40

CHO

was obtained because the signs of the component along the polarization are opposite. Moreover, the new probe can measure both 311 and 333 independently. Figure 33e shows an 333 image, which corresponds to the perpendicular component of the polarization. From the same position in Figure 33a, signals are observed. It means that this polarization has both parallel and perpendicular components, that is, the polarization tilts from the surface. Figure 33f is a topography, which is also measured simultaneously. From these results, we confirmed that the new probe and system can be applied to the three-dimensional polarization measurements. Based on this technique, we have developed a more advanced method to measure the distribution of polarization directions parallel to the surface.  We use a rotating electric field by applying a 90 phase shifted electric voltage between electrodes A, B, and C, D; that is, we apply the electric fields E 3 ¼ E cos !pt and E 2 ¼ E sin !pt just under the probe tip of Figure 32. Under this condition, the electric field rotates with an angular frequency of !p, and the amplitude of the capacitance variation is changed periodically. When the electric field is parallel to the polarization direction, the capacitance variation is at a maximum, and the electric field is perpendicular to the polarization direction, the capacitance variation is at a minimum. Consequently, if we detect the capacitance variation by a lock-in amplifier using a reference signal of angular frequency !p, we can obtain the angle of the polarization direction directly from the phase output of the lock-in amplifier. Figure 34 shows the measurement results on a PZT thin film using a rotating electric field. A histogram of the measured data clearly    shows three peaks, which correspond to 0 , 90 , and 180 . In Figure 34 we can see these three regions. One region is white and black. Within the white and black regions, the white parts and the black parts show the polarization   directions +180 and 180 , respectively. Therefore these regions can be considered to be the same domain. Another region is dark gray. The polarization direction in this region is 0 . The third region is bright gray,   where the polarization direction is 90 . We suppose that the reasons a 90 region does not exist are related to a film growth condition. The polarization direction for each region is shown by arrows. The domain structure of this   figure is not the typical 90 a-c domain and 180 c-c domain, which is usually seen in PZT ceramics. The reason is that the sample is a thin film. Because the grain size of this film is about 300 nm, Figure 34 shows the polarization distribution in a grain. We suppose that the domain structure in a grain depends on a film growth condition. As shown in this figure, we could successfully observe the domain structure, showing the different directions of the polarization parallel to the surface with high spatial resolution.

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

41

Figure 34. Image of a PZT thin film measured by SNDM using rotating electric field.

VI. Application of SNDM Technique for High-Performance Ferroelectric Material and Devices A. Determination of the Polarities of ZnO Thin Films on the Polar Substrate Recently, numerous types of polar dielectric thin films, including ferroelectric, pyroelectric, and piezoelectric thin films, have been developed. Some films are used for ferroelectric random access memory (RAM) and some are for piezoelectric applications including micromachining and surface acoustic wave (SAW) devices. Determination of the polarity of a film is very important not only from the scientific viewpoint of understanding a film growth mechanism but also from an engineering standpoint. For example, the electromechanical coupling property of SAW devices deeply depends on the combination of the polarities of a thin film and a substrate. However, it has been difficult to determine the polarity of thin films by conventional methods using pyroelectric or piezoelectric responses. In particular, in the

42

CHO

configuration where there is a thin film on a polar substrate, determination of the film polarity is quite difficult because, in the conventional method in which the piezoelectric or pyroelectric response is measured, the output signal is proportional to the thickness of the polar material so that the signal from a thin film is masked by that from the substrate. Therefore, we demonstrate that our SNDM technology is very useful for determining the polarity of thin films deposited on polar and nonpolar substrates. We chose ZnO thin films on several cuts of LiNbO3 and LiTaO3 substrates, which are typical piezoelectric materials, as samples and measured their polarities. From the theoretical results shown in Figure 13, we found that the tip radius of the probe needle and the dielectric constant of the specimen determine the sensitivity in the depth direction of SNDM. For example, when we measured the sample with a relative dielectric constant of 300, the output signal saturated at the point that the normalized depth was 0.1. This means that SNDM was insensitive to the material below a depth of 0.1 a. In other words, this means that SNDM was sensitive only in the region from the surface to a depth of 0.1 a. In the same way, in the measurement of ZnO thin films, whose relative dielectric constant is about 10, SNDM is sensitive in the region up to the depth of the tip radius of the probe needle. Accordingly, by using a probe needle whose tip radius is smaller than the thickness of the thin film, we can determine the polarity of the ZnO thin film only, even if the substrate is polarized too. To confirm that we can measure the polarities of a thin film and the substrate separately, we performed one-dimensional scanning, as shown in Figure 35, where the specimen was a ZnO thin film with thickness of 6.3 m on a Z-cut LiTaO3 substrate with a thickness of 480 m. The result is shown in Figure 36. From Figure 36, it is clear that the polarity of the substrate was positive and that of the thin film was negative. One-dimensional scanning

ZnO film LiTaO3 substrate

Figure 35. Measurement of polarity distribution using SNDM.

43

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

Frequencey deviation (Hz/V)

24 ZnO thin film

12

negative LiNbO3 substrate

0

positive −2

0

500

1000 Distance ( m m)

1500

2000

Figure 36. One-dimensional image of the film edge of a ZnO thin film on a LiTaO3 substrate.

Because we had asserted that SNDM has the ability to determine the polarity of polar materials clearly, we investigated the polarities of ZnO thin films deposited on LiNbO3 and LiTaO3 substrates varying the combinations of the polarization direction of the polar substrates and the thickness of the films. The samples were prepared by electron cyclotron resonance (ECR) sputtering and were c-axis oriented. Table 5 summarizes the results. From the data shown in Table 5, we found that positive ZnO thin film grew on the negative Z-cut substrates and negative films grew on the positive Z-cut substrates. In the case of Y-cut and rotated Y-cut substrates, the tendency is not clear. From these results, we can explain the polarity of ZnO on the LiNbO3 and LiTaO3 by assuming that the pyroelectric voltage causes this phenomenon. Figure 37 shows the film-producing process. When the ZnO thin film is deposited, the temperature of the substrate rises so that a voltage with the opposite sign to the polarity of substrate is generated at the substrate surface by the pyroelectric effect. Consequently, a ZnO film with the same polarity as the voltage at the surface grows because the dipole moment of ZnO is attracted by the pyroelectrically generated electric field. In other words, the negative dipole moment on the substrate is attracted by the negative electric field. However, we have to point out that ZnO thin films produced under the other film growth conditions may have the opposite polarity because the growth mechanism is governed by the electric field around the substrate, and the polarity of the electric field may be different under other conditions. Therefore, for different film-producing conditions, it would be necessary to determine the polarities of the polar thin films using our SNDM technique.

44

CHO TABLE 5 Polarities of ZnO Thin Films Deposited on LiNbO3 and LiTaO3 Substrates by ECR Sputtering

Substrate

Polarity of substrate

Thickness of film (m)

Polarity of film

LiNbO3 Y-cut 128-cut 41-cut Z-cut Z-cut Z-cut Z-cut Z-cut

+ + + +  +  +

1.6 1.7 1.7 6.3 6.3 1.48 1.48 1.7

    +  + 

+ +  +  + 

5 6.3 6.3 6.3 6.3 1.48 1.48

    +  +

LiNbO3 36-cut 36-cut 36-cut Z-cut Z-cut Z-cut Z-cut

Pyroelectric effect Negative voltage − +

− +

− +

− +

− +

− +

− +

− +

−

Ps

Substrate

Ps − +

− +

− +

− +

− +

+ − −

− +

Substrate +

− +

+

− +

− +

+ − Film + + − − − +

Ps − +

− +

Substrate +

− +

+

− +

Positive Temperature voltage rising Figure 37. Schematic illustration of the film growth process.

B. Scanning Nonlinear Dielectric Microscopy Study on Periodically Poled LiNbO3 for a High-Performance Quasi-phase Matching Device Recently, quasi-phase matching (QPM) devices for optical parametric oscillation (OPO) have been widely studied for applications to light wavelength conversion. Periodically poled LiNbO3 (PPLN) is one of the most

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

45

promising candidates for QPM-OPO devices (Hatanaka et al., 2000). A well-controlled and very accurate polling technique with submicron order and submicron precision is needed for high-efficiency OPO. Therefore, it is very important to evaluate precisely the domain distributions of PPLN to improve the polling technique and aim for higher efficiency. As previously mentioned, we have developed and reported scanning nonlinear dielectric microscopy (SNDM) for observing ferroelectric polarization distributions and for detecting local crystal anisotropy. Moreover, SNDM can measure a large area of the specimen. For example, very fast scanning can be performed even on a 3-inch  wafer. This is a highly preferable feature for the evaluation of PPLN, which requires a relatively large size to convert light wavelength sufficiently. In this section, we describe the experimental results of an SNDM study on a PPLN domain distribution. By observation of the domain distribution of PPLN using SNDM, we have made some very important findings for obtaining QPM devices with very high performance. First, we observed the domain distribution on PPLN using SNDM. The PPLN used in this study was fabricated by applying a high voltage between the periodic electrode (on +Z cut surface) and the uniform electrode (on Z cut surface) on homogeneously poled Z cut congruent LiNbO3 as shown in Figure 38. Positive voltage was applied on +Z cut surface. Here, we define the original +Z surface and Z surface of the substrate as the front and back surface of the PPLN. To prepare the sample for measuring the depth distribution of a single domain layer, we bevel polished both the front and back surfaces of the PPLN. Front surface (+Z )

Back surface (−Z ) Figure 38. PPLN fabricated by application of a high voltage.

46

CHO

Examples of results are shown in Figure 39 and Figure 40. Figure 39 shows an SNDM image of a virgin PPLN sample. We found that the SNDM signal was vertically asymmetric; in other words, the SNDM signal had shifted downward giving an offset. On the other hand, Figure 40 shows an SNDM image of a PPLN sample in which the surface layer of 10 m has been removed by polishing. In this case, the SNDM signal is vertically symmetric. These results suggest that a single domain surface layer exists on the virgin PPLN and that the uniform negative SNDM signal arising from this layer is superposed on the symmetric periodic signals. Therefore, with the removal of this layer, the offset signal disappeared. Next, to investigate the single domain surface layer precisely, we measured the change in offset of the SNDM signal as a function of depth

Frequency deviation (Hz / V)

10

5 Offset

0

−5

−10 0

20

40 60 Distance ( m m)

80

100

Figure 39. SNDM image of virgin PPLN (Z cut face).

Frequency deviation (Hz/V)

10 5 0 −5 −10 0

20

40 60 Distance ( m m)

80

100

Figure 40. SNDM image of polished PPLN in which the surface layer of 10-m thickness (Z cut surface) has been removed by polishing.

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

47

1

Offset (a.u)

0.8 0.6 0.4

9mm 9 mm

0.2 0 0

2

4 6 Polishing depth ( m m)

8

Figure 41. Offset change of SNDM signal on the front surface.

0 0

2

4

6

8

−0.2 Offset (a.u)

9 mm

−0.4

9mm

−0.6 −0.8 −1 Polishing depth ( m m) Figure 42. Offset change of SNDM signal on the back surface.

from both the front and back surfaces of the PPLN. The results are shown in Figure 41 and Figure 42. Figures 41 and 42 show the change of offset of the SNDM signal on the front and back surfaces, respectively, as a function of polishing depth. From these results, we found that the offsets of the SNDM signals become smaller with greater polishing depth from the surfaces of the PPLN. Therefore, we conclude that the domain structure of the PPLN is as shown in Figure 43 and that the inverted domain does not penetrate through the whole depth of the specimen but has a single domain layer with a thickness of 1020 m on both sides of it. We advise on the removal of the single domain layer on the virgin PPLN fabricated by applying a high voltage to obtain a highly efficient nonlinear optical surface wave device.

48

CHO

Figure 43. Domain distribution of PPLN in the depth direction.

Frequency deviation (Hz/V)

20 10 0

0

20

40

60

80

−10

−20 Distance ( m m) Figure 44. SNDM image of PPLN before annealing.

Although our expectations were that it would be rectangular, the waveform of the SNDM signal of PPLN was approximately sinusoidal, even after removing the surface layer, as shown in Figure 40. We consider this phenomenon to be caused by a strong residual stress (or an internal electric field). The PPLN we measured was fabricated from congruent LiNbO3. It is well known that congruent LiNbO3 includes many lithium vacancies, and these defects play the role of pinning points against domain inversion (Kitamura et al., 1998). We assume that a large stress (or an internal electric field) was generated between domains of opposite sign when domains were inverted because of the existence of many pinning sites. To release the residual stress (or the internal field), we annealed the PPLN at a  temperature of 1000 C for 1 hour in air. The SNDM signals of the same sample before and after annealing are shown in Figure 44 and Figure 45, respectively. Comparing Figure 44 with Figure 45, we found that the SNDM waveform of the annealed PPLN became rectangular and that the strength of the SNDM signal became about 3 times larger. Thus we confirmed that the much reduced nonlinear dielectric constant could be recovered by releasing the residual stress (or the internal electric field) through annealing. The nonlinear dielectric constant in the optical region is the SHG coefficient. Although the polarization mechanisms in the optical and

49

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

Frequency deviation (Hz/V)

50 25 0 0

10

20

30

40

50

60

70

80

−25 −50 Distance ( m m) Figure 45. SNDM image of PPLN after annealing.

microwave regions are different, an experimental proportional rule (Miller’s rule) (Miller, 1964) between the second harmonic generation (SHG) coefficient and the nonlinear dielectric constant has been reported. Therefore, we propose that annealing is a very effective means of obtaining a high-performance QPM device with a much higher efficiency in the case of congruent LiNbO3. C. Tbit/inch2 Ferroelectric Data Storage Based on SNDM Nanosized inverted domain dots in ferroelectrics are expected to play a major role in the storage of information in next-generation ultrahigh-density recording systems. As a high-density recording media, ferroelectric materials are considered to be superior to the ferromagnetic materials widely used at present because the domain wall thickness of typical ferroelectric materials is of the order of a few lattice spaces, far less than that of ferromagnetic materials (Jona and Shirane, 1962). Scanning probe microscopy (SPM) has been extensively investigated as a method of forming and detecting small inverted domain dots in ferroelectric thin films such as lead zirconate titanate, or PZT (Pauch, Tybell, and Triscone, 2001). In this technique, domain dots are switched by applying a relatively large DC pulse to the probe, creating an electric field at the tip of the probe cantilever. These dots can then be detected via the AC surface displacement (vibration) of the ferroelectric material based on the piezoelectric response to an AC electric field applied at the same tip (Gruverman et al., 1997). This technology has clear implications for bit storage in ultrahigh-density recording systems, with anticipated storage densities of the order of Tbit/inch2. Although current techniques are capable of forming and detecting dots of around 100 nm, this measurement falls far short of the 1 Tbit/inch2 thought

50

CHO

to be possible (Pauch, Tybell, and Triscone, 2001; Eng et al., 1999b). The limiting factors involved are the resolution of the piezo-imaging method and the physical properties of the ferroelectric medium. For example, even if very small nanodots of less than 10 nm could be formed successfully, a detection method with resolution of finer than 1 nm would be required to detect such dots with sufficient accuracy. Therefore, it is of primary importance to improve the resolution of the domain detection device such that the smallest formable domain sizes can be resolved. The resolution of our SNDM is in the subnanometer range, much higher than other SPM methods used for observing polarization distributions. Moreover, our SNDM technique is a purely electrical method, allowing domain information to be read at much higher speed than by piezo-imaging, the read speed of which is limited by the mechanical resonant frequency of the atomic force microscopy (AFM) cantilever (typically around 100 kHz). More recently, we have studied the formation of small domain inverted dots in PZT thin film using SNDM and have successfully produced and observed a domain inverted dot with a radius of 12.5 nm (Matsuura, Cho, and Odagawa, 2001a). However, single-crystal material is expected to be more suitable for studying nanodomain engineering quantitatively with good reproducibility because current thin films still suffer from atomic-scale nonuniformities that prevent switching in the nanodomains in which they occur. To date, BaTiO3 is the only material that has been studied as a singlecrystal material for nanodomain switching (Eng et al., 1999b). Although BaTiO3 is a typical ferroelectric material, it is not suitable as a recording medium for a number of reasons. Most important, BaTiO3 single-crystal belongs to the tetragonal system, with the result that it has two possible   domains in which to store bits, a 90 a-c domain and a 180 c-c domain. Such a structure introduces a level of complexity not desired at present. Furthermore, the phase transition from the tetragonal phase to the orthorhombic phase occurs at 5 C, which is dangerously close to room temperature for a storage medium, possibly resulting in data loss as a result of ambient temperature drift. Lastly, it is difficult to fabricate large, highquality, and practical usable BaTiO3 single-crystal at low cost. A ferroelectric nanodomain engineering material suitable for practical  application as a storage medium should have a single 180 c-c domain, have an adequately high Curie point without phase transitions below the Curie point, and be producible as large single-crystals with good homogeneity at low cost. Lithium tantalate (LiTaO3) single-crystal satisfies all these conditions and has been used widely in optical and piezoelectric devices. The development of nanodomain engineering techniques based on LiTaO3 single-crystal will have applications in not only ultrahigh-density

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

51

data storage but also in various electro-optical, integrated-optical, and piezoelectric devices. In this study, we investigated the formation of small inverted domain dots in LiTaO3 single-crystal using the SNDM nanodomain engineering system. A schematic diagram of the SNDM domain engineering system is shown in Figure 46. In the case of single-crystal LiTaO3, the signal-to-noise ratio is sufficiently high to eliminate the need for the lock-in amplifier, and nanodomains can be easily detected using a normal oscilloscope. The use of a lock-in amplifier seriously limits the reading speed, so discarding that requirement is expected to result in very high reading speeds. An SNDM ferroelectric data storage system using single-crystal LiTaO3 is therefore a strong candidate for ultrahigh-density storage. The ferroelectric domain dots are produced using a pulse generator newly installed in the SNDM system, similar to the case for ferroelectric data storage based on piezoimaging. The two types of single-crystal LiTaO3 available commercially, stoichiometric lithium tantalate (SLT) and a congruent lithium tantalate (CLT), were examined for potential application as a data storage medium. Both forms have a good rectangular hysteresis loop suitable for memory application (Kitamura et al., 1998). SLT has fewer point defects because of the lower lithium content compared with CLT, with the direct result that 1 that of CLT. On the other hand, the coercive electric field of SLT is about 13 the natural domain size of CLT is smaller than that of SLT. Therefore, SLT Oscilloscope A, q = q1 − q2 q

Lock-in amplifier Reference w p, 2 w p, 3 w p

Pulse generator

1

FM demodulator Probe

q

2

Voltage source

Oscillator L

wp

FM signal

LC Resonant circuit C

Specimen

Metal stage Insulator Figure 46. Ferroelectric nanodomain engineering system based on scanning nonlinear dielectric microscopy.

52

CHO

is expected to offer faster switching and lower switching voltage, whereas CLT is suitable for higher density storage with smaller domain dots and stable retention. CLT has the additional not-insignificant benefit of being much less expensive than SLT because almost all commercially available LiTaO3 crystal is grown from a congruent composition melt. For the formation of very small domain dots, we fabricated and used very thin SLT and CLT single-crystal plates of thickness 70  150 nm. These plates were fabricated by mechanically polishing a single-crystal wafer to a thickness of around 1 m, followed by electron cyclotron resonance (ECR) dry etching to the desired thickness. The SNDM tip was a platinum iridium (PrIr)-coated electrically conductive cantilever with a tip radius of 25 nm. As a preliminary investigation of this method of domain engineering, we clarified the variation in inverted domain size in SLT with the duration of voltage application. Figure 47 shows the typical shapes of nanodots formed in a 100-nm-thick SLT plate with respect to amplitude (A cos ) and phase (cos ) for three voltage application times at 15 V. The phase image shows only the sign of the domain and is used to define the size of the nanodots.

Figure 47. Typical shapes of nanodomain dots for voltage application times of (a) 500 ns, (b) 100 ns, and (c) 60 ns using a voltage of 15 V and a 100-nm-thick SLT plate. Upper figures are amplitude (A cos ) images, and lower figures are phase (cos ) images.

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

53

As expected, the smallest inverted domain dot among these examples, with a radius less than 10 nm, was obtained at the shortest voltage application time of 60 ns. The smallest SLT dot obtained in this study had a radius of 6 nm as shown in Figure 48, which corresponds to a memory density of 4 Tbit/ inch2 if the dots are close packed. These nanodomain dots in SLT were confirmed to be stable over time through continuous measurements over the first 24 h and further measurements 1 month later. The dots exhibited no measurable change during this period. Single-crystal CLT was examined as a practical data storage medium. CLT was employed for this application because lithium vacancy defects in CLT effectively pin the sites of domains, and the smaller size of the natural domains allows for higher memory densities. The controllability of nanosized domain inversion was demonstrated on a 70-nm-thick CLT plate by writing domains to form the words ‘‘TOHOKU UNIV.’’ as shown in Figure 49. Two sizes of characters were written; both are shown in the figure at the same magnification. The larger characters were written with 10-s

Figure 48. Smallest single nanodot (phase image) with a radius of 69 nm obtained in this study. This dot was formed by 15-V, 60-ns pulse application to a 100-nm-thick SLT plate.

Figure 49. Nanodomain characters written at 14 V on 70-nm-thick CLT film (normalized amplitude). Large characters were written with 10-s pulses, and small characters were written with 5-s pulses.

54

CHO

Figure 50. SNDM images of close-packed array of domain dots (normalized amplitude) at a data density of (a) 0.62 Tbit/inch2 (11 V, 10 s) and (b) 1.50 Tbit/inch2 (12 V, 80 ns) using 70-nm-thick single-crystal CLT.

pulses, and the smaller characters were written with 5-s pulses, both at 14 V. The average size of one of the smaller characters is about 120 nm. The pulse application time for CLT was necessarily much longer that in the case of SLT, reflecting the lower switching speed of CLT. Ultrahigh-density bit storage was then performed at a data density of 1 Tbit/inch2 using the 70-nm CLT plate. Figure 50 shows SNDM images of the close-packed array of positive and negative domain dots at a density of 0.62 Tbit/inch2 and 1.50 Tbit/inch2. The 0.62 Tbit/inch2 array was formed at 11 V with 10-s pulses, and the 1.50 Tbit/inch2 array was formed at 12 V with 80 ns pulses. The average radius of dots in the 1.50 Tbit/inch2 array is 10.4 nm. Although the dots in the 1.50 Tbit/inch2 array may not be resolvable with sufficient accuracy for practical data storage, this system is fully expected to become practically applicable as a storage system after further refinement. We have thus demonstrated, using a ferroelectric medium and nanodomain engineering, that rewritable bit storage at a data density of more than 1 Tbit/inch2 is achievable. To the best of our knowledge, this is the highest density reported for rewritable data storage and is expected to stimulated renewed interest in this approach to next-generation ultrahigh-density rewritable electric data storage systems. VII. Conclusion In this chapter, first, a subnanometer resolution scanning nonlinear dielectric microscope (SNDM) was developed for the observation of ferroelectric polarization. We demonstrated that the resolution of SNDM

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

55

is higher than that of a conventional piezoresponse imaging. Also, we described the theoretical resolution of SNDM and the quantitative measurement technique using SNDM. This theoretical result predicted that an atomic scale image can be taken by SNDM. Next, we reported new SNDM technique detecting higher nonlinear dielectric constants 3333 and 33333. It is expected that higher order nonlinear dielectric imaging will provide higher lateral and depth resolution. Using this higher order nonlinear dielectric microscopy technique, we successfully investigated the surface layer of ferroelectrics. Moreover, a new type of scanning nonlinear dielectric microscope probe, called the 311-type probe, and a system to measure the ferroelectric polarization component parallel to the surface were developed. Finally the some applications of SNDM technique to actual problem to be solved for high-performance ferroelectric devices were described. At first, the polarities of piezoelectric thin films on piezoelectric substrates were determined. Next, studies of the domain distribution of periodically poled LiNbO3 (PPLN) were performed for high-quality quasi-phase matching (QPM) devices and some very important results were obtained. Finally, the formation of artificial small inverted domain was reported to demonstrate that the SNDM system is very useful as a nanodomain engineering tool. The nanosize domain dots were successfully formed in LiTaO3 single crystal. This means that we can obtain a very high-density ferroelectric data storage with the density above T-bits/inch2. Therefore, we have concluded that the SNDM is very useful for observing ferroelectric nanodomain and local crystal anisotropy of dielectric material with subnanometer resolution and also has a high potential as a nanodomain engineering tool.

References Cho, Y., and Matsuno, F. (1992). Dynamic measuring method of capacitance variation of piezoelectric ceramics with alternating electric field. Jpn. J. Appl. Phys. 31, 3627–3631. Cho, Y., Kirihara, A., and Saeki, T. (1995a). Development of nonlinear dielectric microscope and its application to measurement of ferroelectric polarization. IEEE Ultrason. Symp. Proc. 1, 529–534. Cho, Y., Kirihara, A., and Saeki, T. (1995b). New microscope for measuring the distribution of nonlinear dielectric properties. Denshi Joho Tsushin Gakkai Ronbunshi. 78-c-1, 593–598 or (1996) Electronics and Communication in Japan, Part 2. 79, Scripta Technica, Inc., 68–74. Cho, Y., Kirihara, A., and Saeki, T. (1996). Scanning nonlinear dielectric microscope. Rev. sci. Instrum. 67, 2297–2303.

56

CHO

Cho, Y., Atsumi, S., and Nakamura, K. (1997). Scanning nonlinear dielectric microscope using a lumped constant resonator probe and its application to investigation of ferroelectric polarization distributions. Jpn. J. Appl. Phys. 36, 3152–3156. Cho, Y., Matsuura, K., and Kushibiki, J. (1998). Scanning nonlinear dielectric microscope with submicron resolution. Jpn. J. Appl. Phys. 37, 3132–3133. Cho, Y., Kazuta, S., and Matsuura, K. (1999). Scanning nonlinear dielectric microscopy with nanometer resolution. Appl. Phys. Lett. 72, 2833–2835. Cho, Y., Matsuura, K., Kazuta, S., Odagawa, H., and Yamanouchi, K. (1999). Observation of ultrathin single-domain layers formed on LiTaO3 and LiNbO3 surfaces using scanning nonlinear dielectric microscope with submicron resolution. Jpn. J. Appl. Phys. 38, 3279–3282. Cho, Y., Jintsugawa, O., and Yamanouchi, K. (2000). Scanning electron-beam dielectric microscopy for the investigation of the temperature coefficient distribution of dielectric ceramics. J. AM. Ceram. Soc. 83, 1299–1301. Cho, Y., Kazuta, S., Ohara, K., and Odagawa, H. (2000). Quantitative measurement of linear and nonlinear dielectric characteristic using scanning nonlinear dielectric microscopy. Jpn. J. Appl. Phys. 39, 3086–3089. Cho, Y., and Ohara, K. (2001). Higher order nonlinear dielectric microscopy. Appl. Phys. Lett. 79, 3842–3844. Cho, Y., Ohara, K., Koike, A., and Odagawa, H. (2001). New functions of scanning nonlinear dielectric microscopy higher-order measurement and vertical resolution. Jpn. J. of Appl. Phys. 40, 3544–3548. Eng, L. M., Bammerlin, M., Loppacher, C. H., Guggisberg, M., Bennewitz, R., Liithi, R., Meyer, E., Huser, T. H., Heinzelmann, H., and Guntherodt, H.-J. (1999a). Ferroelectric domain characterization and manipulation: a challenge for scanning probe microscopy. Ferroelectrics. 222, 153–162. Eng, L. M., Guntherodt, H.-J., Schneider, G. A., Kopke, U., and Saldan˜a, J. M. (1999b). Nanoscale reconstruction of surface crystallography from three-dimensional polarization distribution in ferroelectric barium-titanate ceramics. Appl. Phys. Lett. 74, 233–235. Gao, C., Wei, T., Duewer, F., Lu, Y., and Xiang, X.-D. (1997). High special resolution quantitative microwave impedance microscopy by a scanning tip microwave near-field microscope. Appl. Phys. Lett. 71, 1872–1874. Gao, C., and Xiang, X.-D. (1998). Quantitative microwave near-field microscopy of dielectric properties. Rev. Sci. Instrum. 69, 3846–3851. Goto, K., and Yamasaki, S. (1970). Denjikigaku Enshu. (Exercise in Electromagnetic Theory), p.123. Kyoritsu Shuppan, Tokyo. Gruverman, A., Auciello, O., Ramesh, R., and Tokumoto, H. (1997). Scanning force microscopy of domain structure in ferroelectric thin films: imaging and control. Nanotechnology 8, A38–A43. Hatanaka, T., Nakamura, K., Taniuchi, T., Ito, H., Furukawa, Y., and Kitamura, K. (2000). Quasi-phase-mached optical parametric oscillation with periodically poled stoichiometric LiTaO3. Opt. Lett. 25, 651–653. Jona, F., and Shirane, G. (1962). ‘‘Ferroelectric crystals,’’ p. 46. Pergamon Press, New York. Kitamura, K., Frukawa, Y., Niwa, K., Gopalan, V., and Mitchell, T. E. (1998). Crystal growth  and low coercive field 180 domain switching characteristics of stoichiometric LiTaO3. Appl. Phys. Lett. 73, 3073–3075. Matsuura, K., Cho, Y., and Odagawa, H. (2001a). Fundamental study on nano domain engineering using scanning nonlinear dielectric microscopy. Jpn. J. Appl. Phys. 40, 4354–4356. Matsuura, K., Cho, Y., and Odagawa, H. (2001b). Measurement of the ferroelectric domain distributions using nonlinear dielectric response and piezoelectric response. Jpn. J. of Appl. Phys. 40, 3534–3537.

SCANNING NONLINEAR DIELECTRIC MICROSCOPY

57

Miller, R. C. (1964). Optical second harmonic generation in piezoelectric crystals. Appl. Phys. Lett. 5, 17–19. Odagawa, H., and Cho, Y. (2000a). Simultaneous observation of nano-sized ferroelectric domains and surface morphology using scanning nonlinear dielectric microscopy. Surface Science 463, L621–L625. Odagawa, H., and Cho, Y. (2000b). Theoretical and experimental study on nanoscale ferroelectric domain measurement using scanning nonlinear dielectric microscopy. Jpn. J. Appl. Phys. 39, 5719–5722. Odagawa, H., and Cho, Y. (2002). Measuring ferroelectric polarization component parallel to the surface by scanning nonlinear dielectric microscopy. Appl. Phys. Lett. 80, 2159–2161. Ohara, K., and Cho, Y. (2001). Fundamental study of surface layer on ferroelectrics by scanning nonlinear dielectric microscopy. Jpn. J. Appl. Phys. 40, 5833–5836. Ohara, K., and Cho, Y. (2002). Quantitative measurement of linear dielectric constant using scanning nonlinear dielectric microscopy with electro-conductive cantilever. Jpn. J. Appl. Phys. 41, 4961–4964. Pauch, P., Tybell, T., and Triscone, J.-M. (2001). Nanoscale control of ferroelectric polarization and domain size in epitaxial Pb(Zr0.2 Ti0.8) O3 thin films. Appl. Phys. Lett. 79, 530–532.

This Page Intentionally Left Blank

ADVANCES IN IMAGING AND ELECTRON PHYSICS, VOL. 127

High-Order Accurate Methods in Time-Domain Computational Electromagnetics: A Review J. S. HESTHAVEN Division of Applied Mathematics, Brown University, Providence, Rhode Island 02912

I. Introduction . . . . . . . . . . . . . . . . . . . . . II. Maxwell’s Equations in the Time Domain . . . . . . . . . . A. Scattered Field Formulation . . . . . . . . . . . . . . B. Maxwell’s Equations in One and Two Dimensions . . . . . III. Case for High-Order Methods in Computational Electromagnetics . IV. High-Order Finite Difference Schemes. . . . . . . . . . . . A. Extensions of the Yee Scheme . . . . . . . . . . . . . B. Compact Schemes and SBP Schemes . . . . . . . . . . . C. Fictitious and Overlapping Grid Methods . . . . . . . . . V. Spectral Methods. . . . . . . . . . . . . . . . . . . . A. Global Methods. . . . . . . . . . . . . . . . . . . B. Multidomain Formulations . . . . . . . . . . . . . . 1. The Local Scheme. . . . . . . . . . . . . . . . . 2. Connecting the Elements . . . . . . . . . . . . . . 3. A Few Examples . . . . . . . . . . . . . . . . . VI. High-Order Finite Volume Schemes. . . . . . . . . . . . . VII. Finite Element Schemes . . . . . . . . . . . . . . . . . A. Continuous Finite Element Techniques . . . . . . . . . . B. Discontinuous Finite Element Techniques . . . . . . . . . VIII. Issues in Temporal Integration . . . . . . . . . . . . . . IX. Conclusions and Outlook . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

59 63 65 66 67 70 73 77 80 84 84 86 87 89 91 94 97 98 104 113 115 117

I. Introduction The insight of James Clerk Maxwell led to a simple system of equations, known to us as Maxwell’s equations, to describe the propagation of electromagnetic waves and, when combined with constitutive relations and boundary conditions, the interaction of electromagnetic energy with matter. As simple as these equations appear, their importance is tremendous and accurate, efficient, and robust methods for solving them are at the heart of the development of emerging technologies, such as very low observable 59 Copyright 2003, Elsevier (USA). All rights reserved. ISSN 1076-5670/03

60

HESTHAVEN

vehicles, ground/foliage penetrating radars, phase-sensitive components, and high-speed electronics and electrooptics. The simplicity of Maxwell’s equations is indeed deceptive, and the accurate and efficient solution of these for realistic applications remains a significant challenge that continues to attract attention among computational mathematicians, physicists, and engineers alike. What complicates the solution of Maxwell’s equations is the need to accurately model the wave– matter interaction (i.e., reflection, refraction, and diffraction processes) the vectorial nature of the boundary conditions, and the size and geometric complexity one often encounters in applications. This imposes requirements on the accuracy and performance of the computational tools well beyond that of existing standard techniques. The need to identify new approaches to electromagnetic modeling and design is further emphasized by the growing interest in very broad band signals and their interaction with large and geometrically complex objects, often including regions of inhomogeneous, anisotropic, lossy, or even nonlinear materials. As the frequency of the waves increases in applications and in modeling efforts, an additional complication enters through the need to model random surfaces and materials. The classical integral-based solution techniques (Chew et al., 2001), as unchallenged as they are for pure scattering problems are less appealing for broadband applications and problems including penetration, complex materials, and random effects. Finite element techniques (Jin, 1993; Volakis et al., 1998) can, at some cost, successfully address some of these concerns but does so by assuming monochromatic waves. This suggests that one turns the attention to time-domain methods for solving Maxwell’s equations. Indeed, the strength of this approach has been demonstrated successfully over the last few decades, beginning with the second-order accurate Yee (1966) scheme. As simple as this scheme is, it continues to be the main workhorse of computational electromagnetics in the time domain (Kunz and Luebbens, 1993; Taflove, 1995, 1998). It is easy to identify several reasons for the success of the Yee scheme, but its most appealing quality is perhaps its simplicity. Furthermore, use of the staggered grid improves the accuracy somewhat and can be shown to ensure that the divergence of the initial conditions in homogeneous regions is preserved exactly in agreement with Maxwell’s equations (Yee, 1966). The limitations of the Yee scheme are, however, equally straightforward to identify. Apart from second-order accuracy, limiting the electric size and duration of problems one can consider, the embedding of the computational geometry poses the most significant problem by requiring one to approximate boundaries and interfaces by a staircased curve. While this may seem adequate for many problems, it nevertheless affects the overall accuracy and essentially reduces accuracy of the scheme to the first order.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

61

Techniques attempting to overcome this are plentiful in the literature (e.g., Holland, 1983; Jurgens et al., 1992; Monk and Suli, 1994; Taflove, 1995; Yee et al., 1992). Most of these methods, however, sacrifice the simplicity of the original Yee scheme to achieve the improved accuracy, which remains, at best, second order. However, as the problems increase in size and the geometries in complexity, one begins to encounter the limits of the second-order scheme. In particular the accumulating dispersion errors becomes a major concern (e.g., Rao et al., 1999). Ways to overcome these errors are, however, few and well known—decrease the grid size or increase the order of the scheme. As the former quickly becomes impractical for large-scale problems, it is only natural to turn the attention to the development of high-order accurate methods for solving Maxwell’s equations in the time domain. As discussed in Section III, high-order methods are characterized by their ability to accurately represent wave propagation over very long distances using only a few points per wavelength. For three-dimensional large-scale computations, this translates into dramatic reductions in the required computational resources, i.e., memory and execution time, and promises to offer the ability to model problems of a realistic complexity and size. This comes at a price, however. The simplicity of the schemes is sacrificed somewhat for the accuracy, particularly when combined with a need for geometric flexibility. This increased complexity of the scheme is perhaps the main reason for the slow acceptance of high-order methods among practitioners of computational electromagnetics. Although the need for high-order accurate schemes was realized by some practitioners early on (Nachman, 1996), acceptance of this is still far from widespread. Evidence of this is the lack of contributions discussing high-order time-domain methods in recent overviews of state-of-the-art techniques in computational electromagnetics (Graglia et al., 1998; Lee and Lee, 1999). It is the purpose of this review to rectify this by offering an overview of a number of efforts aimed at the development of high-order accurate methods for the time-domain solution of Maxwell’s equations. By high order we shall refer to methods with a spatial convergence rate exceeding two. The question of which order of accuracy is suitable for large-scale applications is an interesting question in itself and can be analyzed as a cost–benefit question (Kreiss and Oliger, 1972; Fischer and Gottlieb, 1997; Wasberg and Gottlieb, 2000; Gottlieb and Hesthaven, 2001). While the answer naturally has some problem dependence, the general conclusion is that schemes of spatial order four to six offer an optimal balance between accuracy and computational work for a large class of applications solved with realistic accuracy tolerances. It is therefore natural to focus on methods that have the potential to reach this level of accuracy.

62

HESTHAVEN

Unavoidably, the discussion is colored by our own interests and experiences, and some less mature current developments have not been included in this discussion, most notably perhaps multiresolution timedomain methods (Taflove, 1998). These methods do display high-order accuracy under certain circumstances but are notoriously difficult to use for geometrically complex problems. Because this remains a major concern in many applications, we have chosen not to include a discussion. A good starting point for such methods is Taflove (1998). While more selective overviews are available, we shall strive to bring most current efforts into the discussion. It is hoped that this, one on hand, will be helpful as a starting point to the practitioner seeking alternatives to standard techniques and, on the other hand, can serve as an updated review of an emerging and rapidly evolving field to the interested computational mathematician. This review is organized as follows. Section II recalls Maxwell’s equations in the time domain and discusses boundary conditions, various simplifications, and standard normalizations. Section III is devoted to an overview of the by now classical phase-error analysis as a way of motivating the need to consider high-order accurate methods in time-domain electromagnetics, particularly as problems increase in size and complexity. This sets the stage for Section IV, which discusses extensions of the Yee scheme and other more complex finite difference schemes. It will become apparent that a major challenge in the development of high-order methods is in fact not to achieve high-order accuracy, but rather to do this in ways that enables geometric flexibility. An interesting development in this direction is the emerging embedding techniques, which are discussed in some detail. In Section III it is shown that higher-order schemes allow a significant reduction of the degrees of freedom without sacrificing accuracy. For some applications it may be natural to consider the ultimate limit, leading to global or spectral methods as discussed in Section V. As tempting as this approach is, the need for geometric flexibility again enters as a major concern. We discuss in some detail the elements of spectral multidomain methods, which combine the accuracy of global methods with the geometric flexibility of a multielement formulation. The need to decompose the computational domain into multiple elements to maintain accuracy and geometric flexibility is not unique to computational electromagnetics, and it is only natural that much work has focused on transferring successes from other branches of science into the timedomain solution of Maxwell’s equations. An example of this is discussed in Section VI where recent efforts on the development of high-order finite volume methods for the solution of Maxwell’s equations are outlined. A parallel and more extensive effort focuses on the development of finite

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

63

element methods for solving Maxwell’s equations in the time domain. This, as discussed in Section VII, is more involved and requires attention to a number of issues, e.g., the proper form of the equations, a suitable variational statement, and element types. We shall discuss some possibilities and recent developments before turning the attention to discontinuous element schemes, which are discussed in some detail due to their attractive properties for problems such as Maxwell’s equations. As we shall see, finite element formulations are in general the mathematically most complex approach but also result in formulations that appear most promising at this point in time, assuming—naively—that the associated grid generation is a minor issue. In Section VIII, we conclude with a brief discussion of issues related to high-order time stepping and discrete stability before offering a few concluding remarks in Section IX.

II. Maxwell’s Equations in the Time Domain We concern ourselves with the direct solution of Maxwell’s equations on differential form ~ @D ~ þ J~ ; ¼ r~  H @~t ~ ¼ ~; r~  D

~ @B ~ ¼ r~  E @~t ~ ¼0 r~  B

ð1Þ ð2Þ

in the three-dimensional domain, , with the charge distribution, ~ð~ x; ~tÞ. ~ ~ ~ ~ The electric field, E ð~ x; tÞ, and the electric flux density, Dð~ x; tÞ, as well as the ~ ð~ ~ ð~ magnetic field, H x; ~tÞ, and the magnetic flux density, B x; ~tÞ, are related through the constitutive relations ~ ¼ ~E ~; D

~ ¼ ~H ~: B

~ , are in general The permittivity tensor, ~, and the permeability tensor,  anisotropic and may depend on space and time as well as the strength of the fields themselves. The current, J~ , is typically assumed to be related to the ~ , through Ohms law, J~ ¼ ~E ~ , where ~ measures the finite electric field, E conductivity, although more complex relations are possible. In the subsequent discussion, we shall generally assume that the materials can be assumed to be isotropic, linear, and time invariant. In this case the constitutive relations become ~; ~ ¼ ~0 r E D

~ ¼ ~0 r H ~: B

64

HESTHAVEN

Here ~0 ¼ 8:854  1012 F=m and ~0 ¼ 4  107 H=m represent the vacuum permittivity and permeability, respectively, and r(x) and r(x) refer to the relative permittivity and permeability, respectively, of the materials. It is worthwhile pointing out, however, that most of the methods discussed in the following can be extended to include much more complex and even nonlinear materials with limited additional effort required. Taking the divergence of Eq. (1) and applying Eq. (2) in combination with Gauss’ law for charge conservation shows that if the initial conditions satisfy Eq. (2), and the fields are evolved according to Maxwell’s equations, Eq. (1), the solution will satisfy Eq. (2) at all times. Hence, one generally views Eq. (2) as a consistency relation on the initial conditions and limits the solution to the time-dependent part of Maxwell’s equations, Eq. (1), although the validity of doing so remains somewhat controversial (Jiang et al., 1996; Kangro and Nicolaides, 1997). To simplify matters further, we obtain the nondimensionalized equations by introducing the normalized quantities x¼

~ x ; ~ L

t¼

~t ; ~ =~c0 L 1

~ is a reference length and ~c0 ¼ ð~ 0 ~0 Þ2 represents the dimensional where L vacuum speed of light. The fields themselves are normalized as E¼

~ ~ 1 E Z 0 ; ~0 H

H¼

~ H ; ~0 H

J¼

~ J~ L ; ~0 H

pffiffiffiffiffiffiffiffiffiffiffi ~ 0 ¼ ~0 =~ 0 refers to the dimensional free space intrinsic impedance where Z ~ 0 is a dimensional reference magnetic field strength. and H With this normalization, Eq. (1) takes the form r

@E ¼ r  H þ J; @t

r

@H ¼ r  E; @t

ð3Þ

which is the form of the equations we shall consider. The components of the fields are subsequently referred to as E ¼ ðE x ; E y ; E z ÞT ; and likewise for H and J. To solve Maxwell’s equations in the vicinity of boundaries, penetrable or not, we shall need boundary conditions relating the field components on either side of the boundary. Assuming that a normal unit vector, ^n, to the boundary is given, the boundary conditions on the electric field components take the form ^ n  ðE 1  E 2 Þ ¼ 0;

^ n  ðD1  D2 Þ ¼ s ;

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

65

where Ei and Di ; i ¼ ð1; 2Þ, represent the fields on either side of the interface and s signifies a surface charge. Equivalently, the conditions on the magnetic fields are given as ^ n  ðH 1  H 2 Þ ¼ J s ;

^ n  ðB1  B 2 Þ ¼ 0;

where Js represents a surface current density. In the general case of materials with finite conductivity, no surface charges and currents can exist, and the relevant conditions become ^ n  ðE 1  E 2 Þ ¼ 0;

^ n  ðH 1  H 2 Þ ¼ 0;

ð4Þ

expressing continuity of the tangential field components. The normal components of the flux densities must likewise satisfy ^ n  ðD1  D2 Þ ¼ 0;

^ n  ðB1  B2 Þ ¼ 0;

ð5Þ

i.e., while they are continuous, the normal components of the fields themselves are discontinuous. For the important special case of a perfect conductor, the conditions take a special form as the perfect conductor supports surface charges and currents, whereas the fields are unable to penetrate into the body, i.e., ^ n  E ¼ 0;

^ n  B ¼ 0:

ð6Þ

A. Scattered Field Formulation For scattering and penetration problems involving linear materials it is often advantageous to exploit the linearity of Maxwell’s equations and solve for the scattered field, (E s , H s ), rather than for the total field, (E, H ). These are trivially related as E ¼ Ei þ Es;

H ¼ H i þ H s;

where (E i , H i ) represents the incident field, illuminating the scattering object. A particularly useful illumination is the vacuum plane wave of the form 

L ^ i ^  Ei: E ¼ A exp i2 k  x  vt ; H i ¼ k l

^x ; k ^y ; k ^z ÞT is the normalized wave vector and  the normalized ^ ¼ ðk Here k frequency. One can think of L/l as a normalized inverse wavelength of the illuminating wave. For monochromatic plane wave illumination, it is customary to take L ¼ l to simplify matters. Assuming that (E i , H i ) represents a particular solution, e.g., the plane wave solution given earlier to Maxwell’s equations, one recovers the scattered field formulation

66

HESTHAVEN

r

@E s @E i ¼ r  H s þ E s  ðr  ir Þ þ ð  i ÞE i ; @t @t

ð7Þ

@H s @H i ¼ r  E s  ðr  ir Þ ; @t @t

ð8Þ

r

where ir ðxÞ; ir ðxÞ; and i ðxÞ represent relative permittivity, permeability, and conductivity of the media in which the incident field is a solution to Maxwell’s equations, e.g., in the aforementioned case of a plane wave vacuum field illuminating the object we have ir ¼ ir ¼ 1 and i ¼ 0: To simplify matters we assume Ohms law, J ¼ E. In this formulation, the boundary conditions along a dielectric interface are inchanged, i.e., ^ n  ðE s1  E s2 Þ ¼ 0;

^ n  ðH s1  H s2 Þ ¼ 0

ð9Þ

for the tangential components, whereas conditions on the scattered field components become ^ n  Ei; n  E s ¼ ^

^ n  B s ¼ r ^n  H i

ð10Þ

in the case of a perfectly conducting boundary. The general conditions on normal components can likewise be derived directly from Eq. (5).

B. Maxwell’s Equations in One and Two Dimensions For completeness, let us also state Maxwell’s equations in the one- and twodimensional cases. In the former case we simply have r

@E z @H z ¼ þ J z; @t @x

r

@H z @E z ¼ : @t @x

ð11Þ

Both field components are tangential to a material interface and, thus, always continuous—but not smoother than that. At a metallic boundary, Ez vanishes. This set of equations is well suited for testing new schemes as it captures essential features of Maxwell’s equations, e.g., two-way wave propagation and loss of smoothness across material interfaces. To model effects of polarization, reflection/refraction at interfaces, diffraction, and so on, we need to consider two-dimensional problems. In this case, Maxwell’s equations separate into two independent cases— polarizations—with the transverse electric (TE) form being

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

r

@E x @H z ¼ þ J x; @t @y

r

@E y @H z ¼ þ J y; @t @x

r

@H z @E x @E y ¼  ; @t @y @x

67

ð12Þ

@ by assuming that E z ¼ 0 and @z ¼ 0: The other polarization, known as the transverse magnetic (TM) form, is given as

r

@H x @E z ¼ ; @t @y

r

@H y @H z ¼ ; @t @x

r

@E z @H y @H x ¼  þ J z; @t @x @y

ð13Þ

by taking H z ¼ 0: Boundary conditions and scattered field forms can be recovered from the general case discussed above.

III. Case for High-Order Methods in Computational Electromagnetics To come to an appreciation of the need for high-order methods in timedomain electromagnetics, let us briefly recall the question of phase errors associated with finite-difference methods, as first presented in the pioneering work of Kreiss and Oliger (1972), see also Gustafsson et al. (1995). Consider, as the fundamental component of Maxwell’s equations, the scalar wave equation @u @u ¼ c ; @t @x

uðx; 0Þ ¼ eikx

in the domain x 2 [0, 2] and subject to periodic boundary conditions. Here k ¼ 2/l is the wave number. To begin with, we consider only the effect of the spatial approximation and restrict the discussion to finite difference methods. One should keep in mind, however, that the conclusions reached remain qualitatively true also for other high-order accurate schemes discussed later.

68

HESTHAVEN

We introduce an equidistant grid xj ¼

2j ¼ jh; N

j 2 ½0; N  1;

such that uðxj ; tÞ ¼ uj . Using a 2m’th order, explicit central difference approximation to the spatial derivative of u(x, t) yields the semidiscrete scheme !

m X du

2ð1Þn ðm!Þ2 1 Dn uj ; ¼ c ðm  nÞ!ðm þ nÞ! dt xj 2n n¼1

where

Dn ¼

E n  En ; h

En uj ¼ ujþn

ð14Þ

represents the central difference and shift operator, respectively. The exact solution to this semidiscrete equation is uðx; tÞ ¼ eikðxcm ðkÞtÞ : Here cm (k) is termed the numerical wave speed. Clearly we wish that c ’ cm (k) over as large a range of the wave number, k, as possible. A measure of this, the phase error, is defined as em ðkÞ ¼ jkðc  cm ðkÞÞtj: The analysis of the phase error allows us to answer questions about the proper choice of schemes for a specified phase error and the overall efficiency of high-order methods. Let us introduce nondimensional measures of the actual scheme. In particular, we introduce p¼

l 2 ¼ ; h kh

v¼

ct ; l

which can be recognized as the number of points per wavelength, p, and the number of wave periods, , we wish to advance the wave. The phase error thus becomes   v 2 2m ; em ð p; vÞ ’ m p where m is a constant specific to the truncation error of the different schemes, e.g., 1 ¼ 3; 2 ¼ 15; 3 ¼ 70 (Kreiss and Oliger, 1972), etc. If we term the maximal acceptable phase error, p, we recover the lower bounds

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

rffiffiffiffiffiffiffiffiffiffi rffiffiffiffi v v pm ðv; p Þ 2 2m / 2m m p p

69 ð15Þ

on the number of points per wavelengths, pm (, p), required to ensure a specified error, p, after  periods of propagation. We note that the required number of points per wavelength depends not only on the acceptable accuracy, p, but also on the number of periods, , needed to complete the computation, i.e., the effect of the phase error accumulates over time. Assume that we wish to propagate a wave in a d-dimensional box with side lengths l. Clearly, considering general problems of size Ll simply scales all results with Ld. The memory needed to store the fields is proportional to  2md v : Memory / ð pm Þ / p d

Furthermore, the work needed to advance the solution to the final time, t, scales as  dþ1 t v 2m d : / ð2mÞ v Work / ð2mpm Þ
t p d

The strong dependence on 2m, i.e., the order of the scheme, suggests that the use of high-order schemes (m > 1) is advantageous when measured in memory usage and/or required computational work in the following situations:





p  1, i.e., when high accuracy is required.  1, i.e., when long-time integration is needed. d > 1, i.e., for multidimensional problems. pm < 10, i.e., efficient discretizations of large problems.

These are clearly situations of relevance to the modeling of electromagnetic phenomena. While this analysis does not include effects of grid anisotropy on wave propagation, this is only to the benefit of low-order schemes, which will suffer most from such phenomena. Furthermore, the popular use of staggered grids will not improve the efficiency of the low-order methods qualitatively (Yang and Gottlieb, 1996). Thus, the use of high-order accurate methods promises to enable the accurate and efficient modeling of transient electrically large problems over long times. It is the purpose of what remains to discuss a number of recently developed computational methods that aim at fulfilling these promises.

70

HESTHAVEN

IV. High-Order Finite Difference Schemes The most widely used computational technique for solving Maxwell’s equations in the time domain, the finite-difference time-domain (FDTD) method, can be traced back to a scheme introduced by Yee (1966). It utilizes the special structure of Maxwell’s equations and introduces a spatially staggered equidistant grid in which the problem of interest is embedded. For a thorough discussion of such schemes, we refer to the texts by Kunz and Luebbens (1993) and Taflove (1995, 1998). Let us introduce ui ¼ uðxi Þ as a grid function defined on an equidistant grid, xi, with grid size, h. Using the notation of Eq. (14), the familiar second-order central finite difference scheme is dui 1 ¼ D 1 ui : dx 2 To recover a semidiscrete approximation to Eq. (11), we define a set of staggered grids, xi and xi+12, shifted in space by h/2, on which E and H are collocated, respectively. This yields ðxi Þ ðxiþ1 Þ

dEiz ¼ D1 Hiz ; 2 dt z dHiþ 1 2

dt

2

z ¼ D1 Eiþ 1: 2

2

z

For simplicity, we assume no current i.e., J ¼ 0. Approximating the temporal integration by a staggered-in-time leap-frog scheme yields ðxi Þ

Einþ1  Ein nþ1 ¼ D12 Hi 2 ;
t nþ1

ðxiþ1 Þ 2

n1

Hiþ12  Hiþ12 2

2


t

n ¼ D1 Eiþ 1; 2

2

which is indeed the classic Yee scheme, proposed in 1966. Here E ni ¼ E z ðxi ; n
tÞ and similarly for H z. In regions with smoothly varying materials, this explicit scheme is second-order accurate in space and time (Monk and Suli, 1994). The success of the Yee scheme, combined with the realization that secondorder accuracy may well be insufficient for many applications, has spawned much recent work in the development of higher order accurate schemes of a similar nature. To highlight the problems associated with such extensions, let us consider a simple example.

71

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

Consider the one-dimensional problem, r ðxÞ

@E z @H z ¼ ; @t @x

@H z @E z ¼ ; @t @x

defined in the domain x 2 ½L; L and with a material interface positioned at x ¼ a; jaj < L and metallic walls at jxj ¼ L; i:e:; E z ð L; tÞ ¼ 0. The permittivity is assumed to be piecewise constant as  ð1Þ r L  x  a r ðxÞ ¼ : ð2Þ r a<x
h

N X i¼0

u2i

!12

;

h¼

2L ; N

and u is the difference between the computed and the exact solution. (b)

(a) 2

10−1

1.5

10−2 10−3

1 0.5

10−4 Ez

0

10−6

−0.5

10−7

−1

10−8

−1.5

10−9

−2 −1 −0.75 −0.5 −0.25 0 x

0.25 0.5 0.75

||d Hz||h

10−5

1

10−10 101

||d Ez||h

102

N −4

103 N

Figure 1. Metallic cavity problem, L ¼ 1, r ¼ 1, and the final time for computation is T ¼ 2. (a) The solution at T ¼ 2 and (b) the expected fourth-order global convergence as a function of the number of points, N.

72

HESTHAVEN

While such straightforward extensions of the Yee scheme perform well for homogeneous problems with grid-conforming geometries, these schemes also inherit the problems associated with the Yee scheme, i.e., the need to staircase general geometries and the inability to correctly enforce physical jump conditions, Eqs. (4) and (5), at material interfaces. While a consequence of such staircasing is an accuracy reduction, i.e., one is solving a problem that is O(h) different, is well established in the literature (see Ditkowski et al., 2001) it appears less appreciated that the physical interface conditions at a material interface are equally important. To emphasize this point, we show in Figure 2 results for the cavity problem discussed earlier, assuming, however, that for x 2 [0, L] the cavity is filled ð2Þ with an r ¼ 2:25 material. While the solution remains continuous across the material interface, it does not remain smooth, i.e., using a difference scheme across the interface implies a reduced accuracy as is also confirmed in Figure 2. The popular use of averaging of the material coefficients (Taflove, 1998; Cohen, 2002) restores O(h2) accuracy at best. One should keep in mind that the situation may well be worse for multidimensional problems where the averaging technique is much less effective due to the likely existence of discontinuous fields. Indeed, one can construct simple tests where even the Yee scheme fails to converge due to this (Ditkowski et al., 2001). Thus, the formulation of high-order finite-difference methods entails not only the derivation of the high-order accurate finite-difference stencils, but (a)

(b) 2.5 2 1.5

100 e (1)

e (2)

= 1.0

= 2.25

1 0.5 0

N −1

10−1 10−2

|||d Hz||h

EZ

HZ

10−3 e (0)

−0.5

= (e (1) +

10−4

−1 −1.5

|||d Ez||h

e (2))/2

|||d Ez||h

10−5

N−2

−2 −2.5 −1 −0.75 −0.5 −0.25

0 x

0.25 0.5 0.75

1

10−6 101

102

103 N

ð1Þ

ð2Þ

Figure 2. Metallic cavity problem, L = 1, r ¼ 1:0; r ¼ 2:25, and the final time for the computation is T ¼ 2. (a) The solution at T ¼ 2 and (b) the global convergence as a function of the number of points, N, using a straightforward fourth-order scheme as well as one making use of an averaged material parameter.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

73

also techniques to treat the embedded geometries to the order of the scheme. The latter is considerably more complex than the former as the subsequent discussion illustrates.

A. Extensions of the Yee Scheme It is a simple matter to derive a direct higher order accurate finite-difference stencil on a staggered grid, i.e., we have the explicit fourth-order scheme dEiz 1 ¼ ð27D1=2  D3=2 Þ Hiz ; dt 24 z dHiþ1=2 1 z ¼ ðxiþ1=2 Þ ð27D1=2  D3=2 Þ Eiþ1=2 : dt 24 ðxi Þ

ð16Þ

This appears to have been considered first in the context of electromagnetics in Fang (1989) as a direct extension of the Yee scheme, i.e., using a secondorder accurate scheme in time. Subsequent works using this approach include Petropoulous (1994), Petropoulous et al. (1998), and Taflove (1995). Combining this with the Yee scheme in subgridded areas, results are reported in Georgakopoulos et al. (2002a,b). Close to metallic boundaries one can use third-order closures of the form z

ðxi Þ

z

z

z

dEiz 23Hi1=2 þ 21Hiþ1=2 þ 3Hiþ3=2  Hiþ5=2 ¼ ; dt 24h

ð17Þ

which suffices to ensure global fourth-order accuracy (Gustafsson, 1975). This is the scheme used on the examples shown in Figures 1 and 2. A stable 4th order closure is proposed in Petropoulos and Yefet (2001). While one may continue such developments and define stencils of arbitrary order, such methods have little practical value because the corresponding one-sided closures tend to be unstable (Strang, 1964; Gustafsson et al., 1995). We shall therefore restrict attention to the fourth-order scheme described above, as has been done in most of the current literature. Attempting to overcome the accuracy problems the solution escapes the obvious, e.g., using a high-order approximation to the material properties (Taflove, 1998; Yefet and Turkel, 2000), may improve matters quantitatively but does not make a qualitative difference, i.e., the convergence rate typically remains second order (Tornberg and Engquist, 2002). Furthermore, the extension of such techniques to multidimensional problems, with higher order geometric information, e.g., curvature, would need to enter the model to maintain design accuracy, remains elusive.

74

HESTHAVEN

e

H

(1) 1/2

H (1)

(1)

E1

E0

h

(1) 3/2

Emat

g Lh

(1)

(1)

e

(1)

H N−3/2

H N−1/2 E

(1) N −1

Hmat

H (2)

E0

(2) 1/2

(2)

H 3/2 (2)

E1

(2)

(2)

(2)

H N −3/2 H N −1/2 (2)

E N −1

g Rh

Figure 3. Definition of grid, numbering, and various parameters for solving onedimensional Maxwell’s equations in a PEC cavity filled with two materials.

Initiated in Ditkowski et al. (2001) in the context of Maxwell’s equations, steps in a different direction have been taken. The central idea is to use the staggered grid scheme, Eqs. (16) and (17), in homogeneous regions away from boundaries and then locally modify the scheme close to boundaries and interfaces. This latter part must be done in a geometryconforming way to overcome the staircasing problem and must include the physically correct jump conditions. As shown in Ditkowski et al. (2001), such schemes, termed embedding schemes, allow one to fully restore second-order accuracy in a modified Yee scheme, thus overcoming problems of staircasing and the effect of internal boundaries in a unified way. As the scheme is modified locally only, it maintains the simplicity and computational efficiency of the original formulation because most of the additional work, i.e., computing the local stencils, is done in a preprocessing stage. The extension of these ideas to fourth-order embedding methods is far from trivial and questions remain unanswered. To illustrate the potential of such methods, however, let us return to the cavity problem described earlier but allow the material interface to be positioned anywhere inside the cavity, i.e., we do not require geometric conformity. We shall use Figure 3 to highlight the elements of the scheme. Everywhere away from the internal material boundary we use the fourth-order staggered grid method given in Eqs. (16) and (17). Also, grid points not directly ð1Þ ð2Þ adjacent to the interface, e.g., E N1 and H 1=2 , are updated using the onesided third-order scheme, Eq. (17), reaching into the homogeneous region. The critical question is naturally the update of the points directly next to the ð1Þ ð2Þ interface, i.e., H N1=2 and E 0 . The idea put forward in Ditkowski et al. (2001) is to form extrapolated values, Hmat and Emat, from the left and right, respectively, and use these in combination with the physical jump conditions to complete the scheme.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

75

Using the notation of Figure 3, we define the extrapolated fields as Hmat ¼

ð7  2gL Þð5  2gL Þð3  2gL Þ ð1Þ HN1 2 48 ð7  2gL Þð5  2gL Þð1  2gL Þ ð1Þ HN3  2 16 ð7  2gL Þð3  2gL Þð1  2gL Þ ð1Þ þ HN5 2 16 ð5  2gL Þð3  2gL Þð1  2gL Þ ð1Þ HN7  2 48

and Emat ¼

ð7  2gR Þð5  2gR Þð3  2gR Þ ð2Þ E0 48 ð7  2gR Þð5  2gR Þð1  2gR Þ ð2Þ  E1 16 ð7  2gR Þð3  2gR Þð1  2gR Þ ð2Þ E2 þ 16 ð5  2gR Þð3  2gR Þð1  2gR Þ ð2Þ  E3 : 48

Note that due to the geometry of the problem, L þ R ¼ 12, the schemes to ð2Þ ð1Þ update HN1 and E0 are then given as 2

ð1Þ dHN1 2

dt

¼

46 15  16gL ð1Þ E Emat  4hð1 þ 2gL Þ N1 hð1 þ 2gL Þð3 þ 2gL Þð5 þ 2gL Þ þ

5  12gL 3  8gL ð1Þ ð1Þ E E  ; 2hð3 þ 2gL Þ N2 4hð5 þ 2gL Þ N3

and ð2Þ

ð2Þ

dE0 46 15  16gR ð2Þ ¼  H1 Hmat þ dt 4hð1 þ 2gR Þ 2 hð1 þ 2gR Þð3 þ 2gR Þð5 þ 2gR Þ 

5  12gR 3  8gR ð2Þ ð2Þ H3 þ H5 : 2hð3 þ 2gR Þ 2 4hð5 þ 2gR Þ 2

It is worth emphasizing that the stencils do not collapse even if the interface is positioned very close to or at a grid point. This is a consequence of the staggered grid, which is an essential component of the scheme to ensure a uniformly bounded time-step restriction.

76

HESTHAVEN (b)

(a)

100

100 10−1

g L = 0.50

10−2

10−1 10−2

g L = 0.25 ||d Hz||h

||dEz||h

10−3 10−4

g L = 0.00

N−4

10−5

10−4

g L = 0.00

N−4

10−5 10−6

10−7

10−7 102 N

g L = 0.25

10−3

10−6

10−8 101

g L = 0.50

103

10−8 101

102 N

103

Figure 4. Same problem as in Figure 2, however, solved using the fourth-order embedding scheme. (a) The global convergence of E z while (b) the same for H z is illustrated.

As an illustration of the performance of the scheme, Figure 4 shows results obtained for the problem discussed in relation to Figure 2, although it allows the interface to be positioned away from a grid point also. In such a situation the unmodified scheme would yield only O(h) convergence due to staircasing. However, as shown in Figure 4, the embedded scheme recovers full accuracy regardless of the position of the material interface. Albeit less general, similar ideas exploiting locally modified explicit schemes have also been developed in Yefet and Petropoulos (2001). There the position of the interface is restricted to coincide with the grid points but the physical jump conditions are enforced as described earlier. A slight generalization along similar lines is found in Xie et al. (2002), where such ideas are combined with smooth curvilinear mappings. Petropoulos and Yefet (2001) discussed how the embedding can be utilized as a separator between different grids rather than different materials, thus allowing for subgridding. While the embedding schemes are appealing and appear to offer a good balance between computational complexity and obtainable accuracy, much development remains to be done to make these methods mature alternatives. In particular, the stable and accurate treatment of curved interfaces and metallic boundaries remains a challenge. Wang et al. (2002, 2003) took a related yet slightly different approach. Motivated by Tam and Webb (1993), the authors apply dispersionrelation-preserving (DRP) fourth-order explicit schemes to solve Maxwell’s equations in two (Wang et al., 2001) and three (Wang et al., 2002) spatial dimensions. Such schemes are derived by extending the stencil beyond the

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

77

minimum five points. The additional degrees of freedom to define the stencil are used to optimize its wave propagation characteristics, e.g., by minimizing the phase error. While such an approach is highly accurate for wave propagation, the wide stencil makes it difficult to terminate the stencil and, thus, deal with complex geometries. Comparative studies of DRP and classic high-order finite difference methods are reported by Jurgens and Zing (2000). B. Compact Schemes and SBP Schemes The problem with stability and accuracy of the straightforward fourth-order extension of the Yee scheme, Eq. (16), discussed above has led to a number of alternative developments. These have mostly focused on implicit computations of the derivatives, i.e., P

d u ¼ Qu: dx

ð18Þ

Here u represents the grid vector, and the two matrices, P and Q, are constructed to ensure accuracy and/or stability of the approximation. A classical example of such methods are compact schemes [see Lele (1992) for an introduction]. These were introduced in the context of Maxwell’s equations in Taflove (1998), Shang (1999), and Turkel and Yefet (2000). For the purpose of illustration, let us continue the use of a staggered grid as described earlier. Then, the classical fourth-order compact scheme for computing derivatives is   1 dui ¼ 12D1=2 ui ;
xD1 þ 11 dx 2 i.e., it is an implicit scheme, involving the solution of a tridiagonal matrix. Its main appeal lies in a very compact stencil, using only nearest neighbor values while more accurate than the explicit schemes discussed earlier. Furthermore, away from boundaries and interfaces, the scheme conserves divergence due to the staggered grid. Close to boundaries, special stencils are needed as for the explicit scheme. In Turkel and Yefet (2000) and Taflove (1998), a fully implicit closures is proposed on the form 26

du1=2 du3=2 du5=2 du7=2 5 þ4  ¼ 24D1=2 u1=2 : dx dx dx dx

78

HESTHAVEN

In combination, these expressions yield 2

5 4 22 1 1 22 : : : : : : : : : :

2

1 0 1 1 0 1 : : : : : : : : : :

26 6 1 6 6 0 6 16 : P¼ 6 246 6 : 6 0 6 4 0 0

1 : 0 : 1 0 : : : : : 1 : 0 : 1

: : : : : : : : : : 22 1 1 22 4 5

3 0 07 7 07 7 : 7 7; : 7 7 07 7 15 26

and 1 6 0 6 6 0 6 16 6 : Q¼
x6 6 : 6 0 6 4 0 0

0 0 1 : : 0 : :

: : 0 : : 1 0 0

: : : : : 1 1 0

: : : : : 0 1 1

3 0 07 7 07 7 :7 7: :7 7 07 7 05 1

We recover the fourth-order semidiscrete compact scheme for the onedimensional Maxwell’s equations as r

dE zh ¼ P1 QH zh ; dt

mr

dH zh ¼ P1 QE zh : dt

We have introduced the vectors of grid functions z E zh ¼ ½E0z ðtÞ; E1z ðtÞ; . . . ; EN1 ðtÞ; ENz ðtÞT ; z z z z ðtÞ; H3=2 ðtÞ; . . . ; HN3=2 ðtÞ; HN1=2 ðtÞT ; H zh ¼ ½H1=2

and similarly for the vectors of materials er ¼ ½r ðx0 Þ; r ðx1 Þ; . . . ; r ðxN1 Þ; r ðxN ÞT ; mr ¼ ½r ðx1=2 Þ; r ðx3=2 Þ; . . . ; r ðxN3=2 Þ; r ðxN1=2 ÞT : Because P is banded, its inversion is cheap. Results in Taflove (1998) and Turkel and Yefet (2000) confirm the expected accuracy and stability of the scheme for the one-dimensional Maxwell equations and the two-dimensional TM form, Eq. (13), assuming simple grid-conforming

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

79

boundaries and homogeneous materials. Dispersion relation-preserving compact schemes are discussed in Lele (1992). Although the compact scheme achieves higher order spatial accuracy using a narrow stencil, it suffers from the same problems as the Yee scheme and its straighforward extensions discussed earlier, i.e., difficulties with the accurate representation of the boundaries and material interfaces. The implicit nature of the compact scheme, however, makes it difficult to utilize local remedies as for the explicit scheme because any such local adjustment has a global impact. Initial work in this direction is reported in Turkel and Yefet (2000), in which the compact stencil is modified locally to allow for a nonconforming Dirichlet boundary condition as required in the twodimensional TM form, Eq. (13). The scheme, however, requires one to physically move the grid points, thus introducing severe stiffness for cases in which the boundary is close to a grid point of the equidistant grid. More general types of boundary conditions, e.g., magnetic boundaries, are not treated. In related work (Taflove, 1998; Turkel and Yefet, 2000), the problem of material interfaces is addressed by using high-order smooth approximations to material parameters. While this visually improves on the accuracy, a rigorous analysis was not done and the computational results were restricted to cases where all field components are continuous. Using a nonstaggered grid, it was proposed in Shang (1999) to terminate the compact stencils with explicit schemes. While it is found experimentally that one needs to use a filter to avoid instabilities, full three-dimensional scattering results have been reported. The accuracy and stability of this approach is not known. The formulation of the compact schemes, leading to the operators P and Q given earlier, is done with accuracy in mind. The equally important question of stability must then be addressed subsequently. This is known to be a task of considerable complexity and often requires special techniques to impose boundary conditions (see Carpenter et al., 1993, 1994). The complementary approach to this is the direct construction of stable high-order schemes. Such schemes, known as summation by parts (SBP) schemes, were originally proposed in Kreiss and Scherer (1974) and were developed further in Strand (1994) and Olsson (1995a,b). The discrete operators, P and Q, are derived to mimic the integration by parts property of the divergence operator, leading to the conditions that P be symmetric, positive, definite, and Q almost skew symmetric, i.e., Q þ QT ¼ diag ½1; 0 . . . ; 0; 1. Both P and Q are typically banded, with examples given in Kreiss and Scherer (1974) and Strand (1994). Imposing boundary conditions in these types of schemes is a bit more complex, as directly modifying the operators may destroy the SBP property.

80

HESTHAVEN

The standard approach is thus to impose the conditions weakly through a simultaneous approximation term (SAT) as du du ; uð1Þ ¼ g ) P1 ¼ Qu  T ½uð1Þ  g: dx dx Here T ¼ diag ½0; 0; . . . ; 0;  where  1 ensures stability. Because the boundary conditions are imposed as an additional term, more complex boundary operators can be imposed in a similar way. SBP schemes for Maxwell’s equations are discussed in Nordstro¨m and Gustafsson (2003), showing the expected accuracy and stability for the two-dimensional TE form, Eq. (12), in simple grid-conforming geometries. The scheme preserves divergence in regions of homogeneous materials. Treatment of material interfaces is done in a way similar to that discussed in Section IV.A, i.e., by treating the different regions separately and using the physical jump conditions to connect the regions. As for the compact scheme, SBP methods have problems treating geometrically complex problems due to the implicit nature of the schemes. Furthermore, the SBP property is delicate and even the use of simple curvilinear mappings may destroy this property, thus ruining the stability. It is worthwhile mentioning that a second-order accurate scheme, using the SAT approach, for arbitrary-embedded metalic boundaries has been proposed in Abarbanel et al. (2003) for the wave equation. It is conceivable that similar ideas can be adapted to a fourth-order scheme, although the analysis promises to be complex.

C. Fictitious and Overlapping Grid Methods In the straightforward extensions of the Yee scheme discussed in Section IV.A, it was proposed to use extrapolations and strongly enforce the jump conditions. Methods taking this approach one step further by using the equation repeatedly at the interface also were proposed in Driscoll and Fornberg (1998, 1999) for one- and two-dimensional problems in electromagnetics. Similar ideas have been proposed previously in the context of acoustics and elasticity but apparently were never implemented (Zhang and Symes, 1998a,b). These schemes employ a standard high-order explicit finite-difference scheme on a nonstaggered grid in regions with homogeneous materials. Close to boundaries and interfaces, however, a different procedure is taken, much in the spirit of Section IV.A, albeit using a different approach. To illustrate the central idea, consider again the one-dimensional Maxwell’s equations, Eq. (11), on the form

81

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

x0(1)

x(1) N−1

x1(1)

(1)

xN

x(1) N+1

x(1) N+M (e(2), m (2))

(e(1), m (1)) x (2) −1

x (2) −M

x(2) 0

x1(2)

x(2) N−1

x(2) N

Figure 5. Illustration of ghost grids and numbering used in overlapping grid methods.

@q @q ¼ AðxÞ ; @t @x

Ez ; q¼ Hz 

0 1 ðxÞ r : A¼ 1 r ðxÞ 0 

For simplicity, we restrict attention to the case of a material interface at x ¼ xmat across which we have that q is continuous, i.e., þ qðx mat ; tÞ ¼ qðxmat ; tÞ:

Using the equations themselves, however, we also have that ð1Þ  þ ð1Þ þ Aðx mat Þq ðxmat ; tÞ ¼ Aðxmat Þq ðxmat ; tÞ;

ð19Þ

i.e., we have conditions on the first spatial derivatives, qð1Þ , of q across the interface. One can of course repeat this argument as often as needed to obtain p ð pÞ  p ð pÞ þ þ Aðx mat Þ q ðxmat ; tÞ ¼ Aðxmat Þ q ðxmat ; tÞ:

We assume that we solve Maxwell’s equations on a simple equidistant grid, xj, although it could also be staggered. Consider the situation in Figure 5, where the two regions of different materials are separated at xmat, which do not have to coincide with a grid point. Everywhere away from the interface, we shall use whatever explicit finite-difference is preferred (cf. Section IV.A). To update the values of q ð1Þ ð2Þ at points close to the interface, e.g., xN and x0 , we shall assume the ð1Þ ð2Þ existence of ghost points, xNþm and xm ; m ¼ 1 . . . M. Clearly, if the ð1Þ ð2Þ values of q were known at these points, one could update q at xN and x0 using standard finite-difference stencils. We can, however, use the additional constraints, Eq. (19). One can approximate the one-sided derivatives as central differences qð pÞ ðx mat Þ ’ ð pÞ j

NþM X

j¼NM

ð pÞ

ð1Þ

vj qðxj Þ;

qð pÞ ðxþ mat Þ ’

M X

j¼M

ð pÞ

ð2Þ

wj qðxj Þ;

where are the weights associated with the computation of derivatives ð pÞ using values left of the interface while wj reflects the use of values from right of the interface. These can be found on closed form using Lagrange

82 1

1

0.5

0.5

0

0

E

E

HESTHAVEN

−0.5

−0.5

0 x

0.5

−1 −1

1

1

1

0.5

0.5

0

0

E

E

−1 −1

−0.5

Yee, t = 4

−0.5 −1 −1

0 x

0.5

−0.5

0 x

−0.5

CSE, t = 100 −0.5

FD4, t = 10

1

−1 −1

0.5

1

BPS, t = 1000 −0.5

0 x

0.5

1

Figure 6. Computational results for a pulse undergoing multiple reflections at a material ð1Þ ð2Þ interface (r ¼ 1:0 and r ¼ 4:0) as obtained using different schemes. The computations are terminated where the results are visibly bad. While the Yee scheme quickly loses the correct solution, the standard fourth-order finite difference performs poorly after only 10 periods. The overlapping scheme (BPS) uses a global scheme in each domain and performs very well after a long time. Results marked CSE are obtained using a spectral multidomain scheme (Section V.B). Figure courtesy of T. Driscoll and B. Fornberg.

polynomials as in Section IV.A or computed as discussed in Fornberg (1998). Note that p ¼ 0 corresponds to interpolation at xmat. There are a total of 2M unknown ghost values, implying that we will need 2M constraints, Eq. (19), to recover these, typically resulting in a scheme of O (h2M) close to the interface (e.g., if a fourth-order scheme is used in the interior, one needs four additional constraints to compute the four ghost values). Clearly, one can initialize all operators in a preprocessing stage because they depend on the weights but these also depend on the order of accuracy and the position of the interface. In the original work of Driscoll and Fornberg (1998), this is taken to the limit by using maximal accuracy, i.e., a global spectral method, everywhere in each region of homogeneous material. This requires additional attention to positions of the grids close to the interfaces. We refer to Driscoll and Fornberg (1998) for details. To illustrate the performance of such an approach, Figure 6 shows computational results obtained by solving the one-dimensional Maxwell’s equations, Eq. (11). The problem is very similar to that considered earlier, although the domain is considered periodic rather than truncated by a metallic cavity and the initial condition is a Gaussian pulse in one domain. As the pulse propagates, it experiences multiple reflections and

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

83

Figure 7. Example of an overlapping grid approach used to extend the ghost point approach to two-dimensional problems. Figure courtesy of T. Driscoll and B. Fornberg.

transmissions at the interfaces. Figure 6 clearly illustrates the importance of correctly treating the material interfaces, particularly for problems requiring long-time integration. In Driscoll and Fornberg (1999), these ideas are extended to twodimensional problems, simplified by assuming that the material interface can be mapped smoothly to align with a coordinate axis. In that case, modifications needed to maintain accuracy remain essentially one dimensional. The only additional complication is that deriving conditions, Eq. (19), for the multidimensional case introduces cross-derivatives for M > 1. Thus, only one ghost point is used and the stencils become one sided. For smooth interfaces, it is proposed to use an overlapping patch or grid, conforming to the interface and employing the ghost point approach to update the solution at the interface. The solution at the patch is blended smoothly, using a partition of unity approach, with the solution at an underlying equidistant grid to obtain the global solution. An example of a grid is shown in Figure 7. Computational examples of this and other simple grids can be found in Driscoll and Fornberg (1999). While the use of fictitious (or ghost) points shows promise, many issues remain open, particularly related to the extension of such techniques to more

84

HESTHAVEN

general two- and three-dimensional problems, as well as problems involving nonsmooth geometries. Furthermore, the stability of these methods has not been analyzed.

V. Spectral Methods The classical phase error analysis (Section III) and the results discussed so far suggest advantages in going to even higher order accurate schemes to further reduce work and memory requirements while maintaining accuracy. A straightforward execution of such ideas, however, introduces issues related to computational efficiency when computing with very wide stencils, as well as difficulties associated with finite computational domains and complex geometries. In the following we discuss techniques proposed to overcome these concerns while maintaining the accuracy and efficiency of very high-order schemes. A. Global Methods If we maintain the typical scenario from high-order finite difference schemes and assume that we have a simple equidistant grid, one can imagine using a stencil spanning the whole computational grid, i.e., a global method. Problems with this straightforward approach are several, e.g., the computational cost and the development of stable and accurate means of dealing with the ends of the computational domain. The benefits of overcoming such problems in order are, however, quite substantial, as can be realized by recalling Eq. (15). Letting m increase, we see that one could expect that the required number of points per wavelength becomes independent of accuracy and integration time. In other words, once this requirement is fulfilled, the scheme solves the wave propagation problem exactly. As was shown in Kreiss and Oliger (1972), this intuition holds for periodic problems with only two points per wavelength. One way of overcoming some of the problems in order to harvest the advantages of using a global scheme was first proposed in Liu (1997) in the context of Maxwell’s equations. At first, one assumes that the solution is spatially periodic to overcome the problems with terminating the computational domain and designing large, one-sided stencils. A further advantage of this assumption is the well-known result (Kreiss and Oliger, 1972; Gustafsson et al., 1995; Fornberg, 1996) that the infinite-order

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

85

finite-difference scheme for a periodic problem is nothing else than a pseudospectral Fourier method. In other words, the O(N2) computation of derivatives

N X du

¼ Djk uðxk Þ; dx xj k¼0

where D is a dense differentiation matrix, can be done through a Fourier series as

N N X du

1 X ~ uðxj Þ expðinxj Þ; ¼ ðinÞ~ u expðinx Þ; u ¼ n j n dx xj N þ 1 j¼0 n¼0

where xj ¼ 2j=ðN þ 1Þ represents the equidistant grid points. The benefit of this formulation is that both summations can be done in O(N log N) operations by using the fast Fourier transform. The assumption of periodic solutions may, at first, seem to severely limit the use of such methods. The central idea in Liu (1997), however, was to surround the computational domain with an absorbing layer, a perfectly matched layer (PML) (Berenger, 1994a,b; Taflove, 1998). Assuming that the absorption of waves is sufficiently efficient, the solution on the outer boundary almost vanishes, thus achieving the periodicity. This approach has been used successfully to model large-scale three-dimensional wave propagation and scattering problems (see Liu, 1999) using as little as two points per wavelength. See also Li et al. (2000) for a comparison of PSTD and classical Yee schemes for scattering problems. As efficient and simple as this approach is, it has a number of limitations. The need to completely surround the computational problem with an absorbing layer essentially limits the attention to open space problems, although one could deal with simple interior problems by choosing a particular basis. However, the most severe limitation is the very same as that of the simple extensions of the Yee scheme, i.e., an inability to handle interior interfaces and boundaries. This is emphasized by the simple approximation result that ku  uN k  N q kuðqÞ k;

where uN represents the Fourier approximation of u, and u(q) reflects the q’th derivative. Clearly, if u is very smooth, i.e., kuðqÞ k is bounded for high values of q, the convergence is very fast and the function is well represented with only few points per wavelength. Unfortunately, it is the other limit that is relevant regarding the solution of Maxwell’s equations for problems involving interior boundaries and interfaces. In such cases a best case

86

HESTHAVEN

scenario is that q  1, i.e., one cannot expect better than local first- and global second-order accuracy even for problems where material interfaces are aligned with the grid. For curvilinear interfaces, where the fields may be discontinuous, the situation is worse and the combined impact of a lack of smoothness and staircasing will be significant. Due to the global nature of the approximation and the need to use fast Fourier transform for computational efficiency, it is difficult to see how to overcome these shortcomings, e.g., straightforward local modifications of the stencils as for the finite-difference schemes are not possible, and the benefits of using local smooth mappings are limited to problems with moderate geometric complexity (Canuto et al., 1988; Bayliss and Turkel, 1992).

B. Multidomain Formulations The most significant restriction of the global methods discussed above is the inability to correctly deal with problems in complex geometries. While several techniques were discussed for the fourth-order finite-difference schemes in Section IV, these methods are only now emerging and much work is still needed. Furthermore, it is unclear whether such techniques allow one to formulate schemes with accuracy fourth-order. Thus, it seems natural to consider alternatives, allowing one to maintain global high-order accuracy even in situations with geometric complexity. A key observation is that the efficiency of a high-order method is closely related to the smoothness of the solution. When internal interfaces and boundaries are present, the global smoothness is generally reduced and one does not benefit as much from using high-order methods as one could expect. However, the solution often remains smooth in regions of smoothly varying or constant material parameters, with these regions separated by well-defined geometric features. The only practical way to take advantage of this is to leave the simple equidistant grids behind and consider the formulation of high-order accurate schemes using body-conforming grids. However, for general geometries, one cannot hope to accomplish this with simple globally mapped grids but must consider a multielement or multidomain formulation in which the computational domain is composed as a union of elements. Such an approach introduces a couple of issues that need careful attention (i.e., how does one compute derivatives is the individual elements to high order and how does one connect the local element-wise solutions to form the global solution in a stable manner?). The resolution of these questions has been the topic of recent research (Kabakian, 1996;

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

87

Yang et al., 1997a,b; Yang and Hesthaven, 1999, 2000), paving the way for a high-order accurate scheme without the problems of the finitedifference scheme. The elements of this formulation are discussed in more detail. 1. The Local Scheme One assumes that the computational domain, , is split into K nonoverlapping logical rectangular elements. This is done in a way such that interfaces are aligned with the geometry, i.e., returning to the one-dimensional cavity problem discussed previously, a straightforward splitting is into two elements, corresponding to each of the two regions of different materials. As we will now need to represent solutions and derivatives of solutions on finite domains, it is well known that we must abandon the use of a simple equidistant grid in each domain. Indeed, we must use a grid that clusters close to the ends of the element. A suitable choice could be the mapped Chebyshev Gauss Lobatto nodes (see, e.g., Fornberg, 1996; Gottlieb and Hesthaven, 2001). i ¼ 0 . . . N : xi ¼ a þ

1  cosði=NÞ ðb  aÞ; 2

where the element spans [a, b] and N + 1 are the number of grid points in the domain. Following the basic approach of a finite-difference method, one can now form element-wise Lagrange interpolation polynomials on the form ð1ÞNþ1þj ð1  ðxÞ2 ÞTN ððxÞÞ ; N 2 ci ððxÞ  ðxi ÞÞ 0

li ðxÞ ¼

where T n ðÞ ¼ cosðn arccos Þ represents the n’th order Chebyshev polynomial, c0 ¼ cN ¼ 2; and ci ¼ 1 otherwise. The scaled variable, (x), is given as ðxÞ ¼ 2

xa  1: ba

With this, we can represent the local element-wise solutions as uN ðxÞ ¼

N X i¼0

uðxi Þli ðxÞ;

and compute the point-wise derivatives in a fashion similiar to that of as for finite-difference schemes, i.e., by a matrix multiply as

88

HESTHAVEN

N X du

duN

uðxi ÞDji ;

xj ’

xj ¼ dx dx i¼0

where the differentiation matrix, D, has the entries (Gottlieb et al., 1986) 8 2N 2 þ 1 > > >  i¼j¼0 > > 6 > > iþj > cj ð1Þ > > i 6¼ j : dli ðxj Þ < ci xj  xi ¼ Dji ¼ xi > dx > 0 > > 2ð1  x2i Þ > > > 2 > > 2N þ 1 : i¼j¼N 6

Thus, we can represent solutions and evaluate derivatives with spectral accuracy, provided the solution is sufficiently smooth on the element (Canuto et al., 1988). The extension of this approach to multidimensional problems utilizes tensor products, i.e., a two-dimensional function is represented as uN ðx; yÞ ¼

N X N X i¼0 j¼0

uðxi ; yi Þli ðxÞlj ðyÞ;

and likewise for a three-dimensional field. The computation of derivatives follows the one-dimensional approach given earlier. While this allows the accurate computation of spatial derivatives, it also assumes that u(x, y) is defined on a rectangular grid. This restriction can be overcome by considering a curvilinear representation. In other words, we assume the existence of a smooth nonsingular mapping function, , relating the local (x, y, z) coordinate system to the general curvilinear coordinate system (, , ) as illustrated in Figure 8. To establish a one-to-one

h h

(x, h, z ) = Ψ (x, y, z)

∆

∆ z

x

(x, y, z) = Ψ −1(x, h, z )

x z

∆ y

z

x

Figure 8. Illustration of the curvilinear mapping used in the multidomain formulation.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

89

correspondence between the unit cube, I  R3, and the general curvilinear hexahedral, D, we construct the local map for each subdomain using transfinite-blending functions (Gordon and Hall, 1973). We refer to Hesthaven (1999) for a thorough account of this procedure within the present context. Thus, we have Cartesian coordinates, (x, y, z) 2 D, and the general curvilinear coordinates, (, , ) 2 I. On curvilinear form, Maxwell’s equations become

Q

@q @q @q @q þ AðrÞ þ AðrÞ þ AðrÞ ¼ 0; @t @ @ @

ð20Þ

with the state vector, q ¼ ðE; HÞT , and the material matrix, Q ¼ diag ðr , r , r , r , r , r Þ. The general operator, A(n), depending on the local normal vector, n ¼ ðnx , ny , nz Þ, obtained from the metric through the mapping, , is given as 2 3 0 0 0 0 nz ny 6 0 0 0 nz 0 nx 7 6 7 6 0 0 0 n n 07 y x 6 7: AðnÞ ¼ 6 ny 0 0 07 6 0 nz 7 4 nz 0 nx 0 0 05 nx 0 0 0 0 ny Figure 9 shows a simple two-dimensional holographic waveguide coupler and the geometry-conforming multidomain grid. The mapped Chebyshev grid in each element allows an accurate computation of derivatives, whereas the body-fitted grid ensures that the solution is smooth inside each element, hence enabling one to take advantage of the accuracy of the high-order scheme. 2. Connecting the Elements Having the ability to accurately and efficiently compute derivatives in a general curvilinear hexahedral and, thus, solve Maxwell’s equations in such a domain, we must now focus on the question of how to assemble these local solutions to recover a global solution in a time-stable and accurate manner. Clearly, care has to be exercised here as Maxwell’s equations support counter propagating waves, consisting of both electric and magnetic fields, i.e., one cannot simply enforce continuity across the interfaces. The central observation to make, utilized in the context of gas dynamics also (Kopriva, 1986; 1989; Hesthaven, 1999) is that Maxwell’s equation,

90

HESTHAVEN

(ES, HS) n0 = 1

(E i, H i )

x

n1

d1

n2

d2

n3

d3

x

z 8 7 6 5 4 3 2 1 0 −1 −2 −3 0

5

10 z

15

20

Figure 9. (Top) A sketch of a diffractive waveguide coupler and (bottom) a multidomain spectral grid used to model such a geometry.

written as in Eq. (20), is a strongly hyperbolic can diagonalize the matrix Q1 AðnÞ as 2 1 0 6 0 1 6 6 0 0 ST Q1 AðnÞS ¼ cr jnj 6 6 0 0 6 4 0 0 0 0

system. In other words, we 0 0 0 0 0 0

0 0 0 0 0 0

0 0 0 0 1 0

3 0 07 7 07 7: 07 7 05 1

Here cr ¼ ðr r Þ1=2 is the local speed of light and |n| the length of the normal. The entries of S can be found in Yang and Hesthaven (2000) and a simplified two-dimensional form in Yang et al. (1999). Let us first consider the case where two neighboring elements can be assumed to have smoothly varying materials. If we compute the characteristic functions, R ¼ ST q, then the entries in the aforementioned diagonal matrix tell exactly how these functions are propagating, e.g., R1 and R2 propagate antiparallel to n, R5 and R6 propagate parallel to n, while

91

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

4

4

2

2

S

0 x

x

0

−2

−4

−6 −2

0

2

z

4

6

8

−2

S

S

S

S

S

S

S

S

S

S

S

S

S

−4

S

−6

S

−2

S

S S 0

S

S 2

z

4

S 6

8

Figure 10. Illustration of a plane waveguide test case. The grid shows the general layout with the high-index waveguide just below x ¼ 0 and N ¼ 16 modes in each domain. On the right is a snapshot of the H z component at an arbitrary time illustrating the total field region, as well as the surrounding scattered field region (marked by an S).

R3 and R4 signifies nonpropagating DC components. With this one knows exactly which information propagates where at any point of the boundary of an element. Furthermore, what leaves one element, i.e., R5 and R6, must correspond exactly to what enters the neighboring element through R1 and R2. Thus, R5 and R6 provide the boundary conditions needed to advance the neighboring element. The nonpropagating characteristic waves can be required to be continuous. At a material interface, the situation can be dealt with in two different ways. One can either rescale the characteristic variables to account for the abrupt change in the materials or one can abandon the characteristic variables and simply enforce the physical jump conditions on the fields, e.g., continuity of the tangential fields. 3. A Few Examples To illustrate the performance of the multidomain spectral scheme discussed earlier, let us consider a few examples. As a first one, consider simple two-dimensional TM-polarized wave propagation in a planar multilayer waveguide, as illustrated in Figure 10. The waveguide is 6l long, where the core layer has a thickness of d2 ¼ l and an index of refraction n2 ¼ 1.45, the surrounding layers both have n1 ¼ n3 ¼ 1.4, while the thickness of these two layers are d1 ¼ l and d3 ¼ 4l,

92

HESTHAVEN

respectively. The total field region in which the computation is conducted, as well as the surrounding scattered field region with the absorbing layers, is shown in Figure 10. A fourth-order Runge–Kutta scheme is used to advance in time and a PML to truncate the computational domain (for details, see Hesthaven et al., 1999). As a validation of the expected spectral accuracy, Table 1 lists the global L1 error measured after 10 periods. Not only do we find spectral convergence but also that less than six points per wavelength (Nppw) yield an acceptable accuracy for many applications. As a second example, discussed in more detail in Yang and Hesthaven (2000), we consider scattering by an axisymmetric three-dimensional metallic scatterer, in this case a rocket-shaped nonsmooth object. Figure 11 illustrates the body-conforming grid, and Figure 12 shows a comparison of the bistatic radar cross section (RCS) for different polarizations compared with results obtained using a contemporary integral equation solver. The results agree well over a 40 db dynamic range. As a final example, consider a three-dimensional problem, in this case plane wave scattering by a ka ¼ 5.3 dielectric sphere. The sphere consists of a nonmagnetic material with r ¼ 3 (Yang and Hesthaven, 2000). Excellent results for the radar cross section, obtained with about eight points per wavelength on the surface of the sphere, are shown in Figure 13 along with a segment of the grid. TABLE 1 Error in Computation of the Plane Waveguide Solution at t ¼ 10 N

Nppw


t

L1(H z )

L1(H x )

L1(E y )

12 16 20 24

4.3 5.7 7.1 8.5

3.1E-2 2.1E-2 1.4E-2 1.1E-2

5.0E-2 1.1E-3 6.9E-6 2.2E-6

3.6E-1 8.5E-3 4.8E-5 1.5E-5

2.5E-1 6.0E-3 3.9E-5 1.1E-5

Figure 11. Typical multidomain grid for the solution of scattering by a three-dimensional axisymmetric missile.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

93

The multidomain scheme has by now been implemented and tested for a variety of problems, including three-dimensional waveguide and diffractive optics (Dinesen et al., 2000), quasi three-dimensional (Zhao and Liu, 2002) and fully three-dimensional scattering (Yang and Hesthaven, 2000; Zhao et al., 2002) and propagation in lossy media (Yang and Hesthaven, 2000; Fan et al., 2002). Excellent parallel performance is demonstrated in Dinesen et al. (2000).

25

Horizontal polarization

20

20

15

15

10

10

RCS (dB)

RCS (dB)

25

5 0

Vertical polarization

5 0 −5

−5

−10 −10 −15 −15 −20 −20 0

20

40

60

0

80 100 120 140 160 180 Theta (degree)

20

40

60

80

100 120 140 160 180

Theta (degree)

Figure 12. (Left) The RCS (, 0) for a missile subject to axial illumination by a horizontally polarized plane wave and (right) results under vertical polarization. A reference solution is marked by ‘‘þ’’. Z X Y

10

3

1

Z

0 −1 −2

RCS (dB)

5

2

0

−2 0

Y 2 3

2

1

−2 0 −1

X

−3 −3

−5

−10 0

20

40

60

80 100 120 140 160 Theta (degree)

180

Figure 13. (Left) An example of a three-dimensional curvilinear grid for scattering by a ka ¼ 5.3 dielectric sphere with r ¼ 3, and r ¼ 1. (Right) The computed bistatic radar cross section (RCS) (full lines) as compared with the exact solution (dashed line) computed using the exact Mie series (Bowman et al., 1987).

94

HESTHAVEN

As flexible and versatile as the multidomain spectral approach is, these benefits do come at a price, most notably the complexity of constructing a high-order body-conforming block-structured grid. Furthermore, for highly curved elements, one has to be careful to avoid instabilities caused by aliasing and to resolve both the solution and the geometry with sufficient accuracy. For nontrivial problems it is often advantageous to use a high-order filter (Yang et al., 1997b; Hesthaven et al., 1999, Gottlieb and Hesthaven, 2001) to improve robustness, although care has to be taken not to adversely impact the accuracy.

VI. High-Order Finite Volume Schemes The need for geometric flexibility is shared with many other diciplines and it is tempting to try to take advantage of such related developments. Given the wave nature of the solutions, it is natural to turn the attention toward methods from the gas dynamic where one of the most remarkable and successful developments has been finite volume methods, combining the geometric flexibility of an unstructured grid with the ability to handle nonsmooth solutions. The finite volume method is based on a discretization of the conservation law @u þ r  f ðuÞ ¼ 0; @t where u is the solution and f(u) represents a flux, possibly of a nonlinear nature. Introducing a grid with grid points, xi 2 , centered in the individual control volumes, D, we integrate over the control volume and invoke Gauss’ theorem to recover I ui d ^ þ n  f ðuÞdx ¼ 0; AðDÞ dt @D where A(D) represents the area/volume of D, ui the cell-averaged solution, i.e., Z  uðxÞdx; ð21Þ u ¼ D

n an outward pointing normal vector at the boundary of D. and ^ To put this into the context of Maxwell’s equations, one need only realize that Eqs. (7) and (8) can be written as

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

Q

@q þ r  F ðqÞ ¼ S; @t

95 ð22Þ

where Q represents the materials, q ¼ ½E; HT , and the flux, F ¼ ½F 1 ; F 2 ; F 3 T has the components  ^ei  H ; ð23Þ F i ðqÞ ¼ ^ei  E where ^e ,  ¼ ðx; y; zÞ, signifies the three Cartesian unit vectors. The close connection between gas dynamics and electromagnetics has been explored in a series of papers (Shang and Gaitonde, 1995, 1996; Shang and Fithen, 1996; Gaitonde and Shang, 1996; Shang et al., 1996; Gaitonde et al., 1997) devoted to the development of high-order accurate finite volume methods on structured and locally orthogonal but globally unstructured grids. So far, everything in the aforementioned discussion remains exact. However, because we only have cell-centered solution values, ui , evaluating the fluxes, f(u), along the circumference of the element cannot be done in a straightforward manner. This problem, being one of reconstruction in contrast to the approximation of derivatives as discussed so far, is at the heart of the finite volume method and is where the approximation enters. As shown in Harten et al. (1987), if one can evaluate the local fluxes to O(hn), then the truncation error of the cell-averaged solutions, u, is also O(hn), i.e., we can focus on schemes for reconstruction of the local fluxes. Borrowing directly from successes in computational fluid dynamics, one could use the notion of characteristic waves, discussed in Section V.B, and form the edge-based solution by upwinding from both sides of the edge. Assuming for simplicity a locally Cartesian grid, as done in Shang and Fithen (1996) and Gaitonde and Shang (1996), one expresses the edge fluxes as f ðxiþ1=2 Þ ¼ F ðuL ; uR Þ ¼ F þ ðuL Þ þ F  ðuR Þ; where F þ ðuÞ and F  ðuÞ correspond to the downwind, i.e., positive eigenvalues, and F þ ðuÞ to the upwind, i.e., negative eigenvalues, components of the characteristic waves, see Section V.B. This flux splitting is nonunique with suggestions given in Shang and Fithen (1996) and Shang and Gaitonde (1996) for a general curvilinear formulation. Given the linearity of the fluxes, the accuracy of the reconstructed solution values, i.e., uL and uR reconstructed from the left and right of the edge, determines the overall accuracy. Assuming a locally equidistant grid, it is proposed in Shang and Gaitonde (1995) and Gaitonde and Shang (1996) to use the MUSCL fluxes

96

HESTHAVEN

uLiþ1=2 ¼



 1 i ; 1 þ ðr þ 2
Þ u 6

uR iþ1=2 ¼

  1 1  ð2r 
Þ uiþ1 ; 6

where r ¼ E0  E1 and
¼ E1  E0 where Ei is the shift operator defined in Eq. (14). This approach is based on local Taylor expansions and is accurate to O(h3), i.e., the scheme can be expected to be third-order accurate on a locally uniform grid. Alternatives to the upwinded reconstructions are discussed in Shang and Gaitonde (1995). The numerical dispersion and grid anisotropy for this method are discussed in Shang and Fithen (1996), Shang and Gaitonde (1996), and Gaitonde and Shang, (1996), and simulations using curvilinear, orthogonal grids are shown in Shang et al. (1996). As discussed earlier, at the heart of the finite volume scheme is the need to reconstruct the local solution using only cell-averaged values. The approach discussed earlier is essentially limited to third-order accuracy by the MUSCL flux. An alternative is discussed in Gaitonde and Shang (1996) and introduces the new one-dimensional variable Z x dV ¼ uðxÞ; uðsÞ ds; V ðxÞ ¼ dx 0 i.e., if one can evaluate the point-wise derivative of V ðxÞ accurately, one can reconstruct the local point-wise value of uðx) accurately. However, from the definition of u, Eq. (21), it follows directly that V1=2 ¼ 0; Viþ1=2 ¼ Vi1=2 þ hui ; assuming a simple one-dimensional equidistant grid. The extension to multiple dimensions involves tensor product grids. With the grid function Viþ1/2 computed, we can now use any of the finite difference techniques discussed in Section IV to compute the local derivative of Viþ1/2 to recover uiþ1/2 and, consequently, the local flux. Clearly, the order of this approach will depend on the scheme chosen to evaluate the derivative of V(x). In Gaitonde and Shang (1997) and Gaitonde et al. (1997), it is suggested to use an implicit compact stencil, similar to the ones discussed in Section IV.B. Other techniques discussed in Section IV could equally well be used. Dispersion errors of the compact schemes are discussed in Gaitonde and Shang, and errors associated with stretched grids are addressed in Gaitonde et al. (1997) Dispersion-optimized compact reconstructions are introduced in Gaitonde and Shang (1997). As appealing and simple as the finite volume schemes are, they suffer from a shortcoming similar to those of the finite difference schemes discussed previously, e.g., an inability to deal accurately with material

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

97

interfaces and complex geometries. This is caused by the high-order reconstructions based on logical Cartesian grids. Furthermore, the compact reconstruction essentially assumes local smoothness of the solutions, which may not be the case across material interfaces. Exploiting embedding techniques may be a way of overcoming this.

VII. Finite Element Schemes Through the aforementioned discussions it has become clear that the need to handle geometrically complex problems accurately and systematically is perhaps the most significant challenge when developing new methods for solving Maxwell’s equations. This realization is, however, not unique to electromagnetics, and much work has been done to address this problem in other areas of computational science. The ability to handle this effectively and accurately remains one of the main reasons for the remarkable success of finite element methods in solid and fluid mechanics [see Hughes (2000) and references therein], leading to its widespread use and availability of the numerous commercial software environments. The use of finite elements for solving Maxwell’s equations has, however, been relatively slow, despite early efforts (Silvester, 1969; Cendes and Silvester, 1971; Silvester and Ferrari, 1983). This can be attributed partly to the need to address numerous technical questions, e.g., element types, equation form, and correct variational statements, and partly to the failure of the most straightforward formulations. The success of finite difference methods for many problems, combined with its simplicity, also made the finite element formulation less attractive. With the growing need to solve geometrically complex large-scale problems, the last decade has seen an increasing interest in the flexibility offered by finite element schemes, although most of the developments have been for problems formulated in the frequency domain (Jin, 1993; Volakis et al., 1998). Only more recently have finite element schemes for the time-domain solution of Maxwell’s equations received more attention (Lee et al., 1997), focusing almost exclusively on low-order formulations. The development of high-order accurate finite element methods for the time-domain solution of Maxwell’s equation remains an emerging field at this point in time, although some of the results discussed subsequently illustrate its potential.

98

HESTHAVEN

A. Continuous Finite Element Techniques When formulating a finite element scheme for solving Maxwell’s equations, one encounters a number of questions, the first one being on which form to consider the equations themselves. One could consider solving the equations on first-order form, Eq. (3), r

@E ¼ r  H þ J; @t

r

@H ¼ r  E: @t

ð24Þ

The treatment of these first-order nonself-adjoint operators is, however, often pose significant problems for classical finite element formulations. An attractive alternative, and one that is used most often, is recovered by combining the two first-order equations to recover the self-adjoint curl–curl form r

@2E 1 @J : þr rE ¼ 2 @t r @t

ð25Þ

Both equations are subject to appropriate boundary conditions, i.e., continuity of tangential field components at material boundaries, Eq. (4), and vanishing tangential electric fields at conductors, Eq. (6). For both Eq. (24) and Eq. (25), some condition at the far field is also needed if the domain is open (Lee et al., 1997). The latter formulation is often preferred, partly because of the self-adjoint operator, natural for the formulation of standard finite element schemes, and partly because of the decoupling between the fields, thus reducing the number of unknowns. However, this formulation also comes with a number of pitfalls, as discussed later. Let us first, however, consider schemes for the first-order form and introduce the inner product Z u  v dx: ðu; vÞ ¼

The variational form of Eq. (3) then follows as d ðr E; Þ ¼ ðr  H; Þ þ ðJ; Þ ; dt d ðr H; Þ ¼ ðr  E; Þ ; dt for all  a test function, which can be a scalar or a vector-valued function. To seek the semidiscrete numerical scheme, assume that the computational domain, , is partitioned into K nonoverlapping elements, D, on which the test functions have support.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

99

Let us first consider the simplest case in which the test function is a scalar nodal element, much as is done in classical finite elements (Hughes, 2000). Thus, we assume that the numerical solutions are given as X X Ei ðtÞi ðxÞ; Hh ðx; tÞ ¼ Hi ðtÞi ðxÞ; Eh ðx; tÞ ¼ i

i

ðx; y; zÞ; ðE i ; H i Þ

where  ¼ represents the unknowns, being nodal values or expansion coefficients, and i ðx) are the locally defined basis functions, which are assumed continuous. Although not generally necessary, in the Galerkin form considered here, the trial and test functions are the same. Inserting the numerical solutions into the variational statement yields the semidiscrete form as d x E ¼ Sy H zh  Sz H yh þ MJ xh dt h d M E yh ¼ Sz H xh  Sx H zh þ MJ yh dt d M E zh ¼ Sx H yh  Sy H xh þ MJ zh dt d M H xh ¼ Sz E yh  Sy E zh dt d M H yh ¼ Sx E zh  Sz E xh dt d M H zh ¼ Sy E xh  Sx E yh ; dt M

ð26Þ

where ðE h ; H h Þ represents the global degrees of freedom. We likewise have the globally defined mass matrices Mij ¼ ði ; r j Þ ;

Mij ¼ ði ; r j Þ ;

Mij ¼ ði ; j Þ ;

as well as the differentiation matrix Sij

  @j ¼ i ; : @

For the harmonic case, it was shown in Lynch and Paulsen (1990), however, that this most obvious form harbors spurious vector modes, which may lead to convergence to wrong solutions. This was attributed to a lack of enforcing the constraint of divergence-free fields. Another interpretation of this is the inability to properly represent the null space of the curl operator (Sun et al., 1995). This topic of spurious solutions to Maxwell’s equations has received significant attention in the literature (Lynch and Paulsen, 1991; Paulsen and Lynch, 1991; Jiang et al., 1996), primarily in the context of frequency domain solutions. An introductory overview is given in Sun et al. (1995). In

100

HESTHAVEN

the time domain these problems appear to be much less significant and somewhat controllable through smoothness of the initial conditions (Jiang et al., 1996; Kangro and Nicolaides, 1997). Nevertheless, the solutions proposed to overcome problems of spurious modes in frequency domain schemes generally have also been used in the development of schemes for the time domain. While several solutions have been proposed, by far the most popular is the use of a vectorial basis in the formulation of finite element schemes, i.e., X X Hi ðtÞN i ðxÞ; ð27Þ Ei ðtÞN i ðxÞ; Hðx; tÞ ¼ Eðx; tÞ ¼ i

i

where (Ei, Hi) are scalars and Ni ðx) represents the vectorial basis. The main motivation for seeking vector basis functions is the observation that the boundary conditions for Maxwell’s equations are vectorial, i.e., it is natural when seeking a conforming discretization to utilize vector basis functions. Such basis functions, often known as curl-conforming elements, should satisfy fundamental properties of the solutions to Maxwell’s equations, e.g., support tangential continuity of the solutions. This allows one to impose tangential continuity between elements with different materials, as well as impose boundary conditions in a natural way. Furthermore, the use of such elements guarantees the absence of spurious modes in frequency domain finite element schemes (Bossavit, 1990). An introduction to vector elements and how they address the concern of spurious modes is given by Sun et al. (1995). Such vector elements, known as edge elements (Bossavit, 1988), Nedelec elements (Nedelec, 1980, 1986), Whitney forms (Bossavit, 1988; Hiptmair, 1999, 2001), and curl/div conforming vector elements (Graglia et al., 1997; Ainsworth and Coyle, 2003), have a number of interesting properties. In particular, they are constructed to provide a discrete analog to the continuous vector algebra and to enforce only minimal continuity across element boundaries, i.e., curl-conforming elements enforce tangential continuity whereas div-conforming elements enforce normal continuity. Albeit at considerable technical effort, edge elements can be constructed to an arbitrary high order of modal/hierarchic (Nedelec, 1986; Webb, 1999; Ainsworth and Coyle, 2003), as well as an interpolatory type (Graglia et al., 1997) and for simplices as well as quadrilaterals and hexahedrals. A general abstract construction is discussed in Hiptmair (1999, 2001, 2002), and elements suitable for a nonuniform order are derived in Demkowicz and Vardapetyan (1998). Using curl-conforming elements, the semidiscrete form of Eq. (24) becomes

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

M

d Eh ¼ SH h þ MJ h ; dt

M

d H h ¼ SE h ; dt

101 ð28Þ

where (Eh, Hh) again represents the global degrees of freedom. The globally defined mass matrices are given as Mij ¼ ðN i ; r N j Þ ;

Mij ¼ ðN i ; r N j Þ ;

Mij ¼ ðN i ; N j Þ ;

ð29Þ

as well as the stiffness matrix Sij ¼ ðN i ; r  N j Þ : While the use of these elements effectively eliminates the spurious modes and adds a lot of structure to the solutions, it does not overcome another impact of conforming finite element scheme, i.e., the need to invert a global, albeit sparse, mass matrix, even if explicit time stepping is used. As the order of the scheme increases, more degrees of freedom are needed on each element, quickly rendering this inversion expensive. An approach to circumvent this has been developed in Cohen and Monk (1999), where it was demonstrated that one can use mass lumping to diagonalize the mass matrices without sacrificing accuracy, even on curvilinear elements. This makes the scheme fully explicit at the semidiscrete level and competitive with alternative methods. Unfortunately, this approach is successful only when using quadrilateral and hexahedral Nedelec-type elements, as discussed in depth in Cohen (2002) and Cohen and Monk (1999). The computational results are limited to two-dimensional problems. A dispersion analysis of the semidiscrete scheme is also included in Cohen and Monk (1999), displaying properties similiar to those of the finite difference scheme discussed in Section III. While the development of curl-conforming Nedelec elements presents a major advancement, it comes at a price. Not only are these families of elements complex, but they also have a significantly higher number of degrees of freedom as compared with the classical nodal elements. This is summarized in Table 2, illustrating that the curl-conforming elements typically have d times more degrees of freedom, with d being the dimension of the problem. However, as one needs d scalar fields, the differences are significant for low-order elements only. An alternative to the use of curl-conforming elements, essentially overcoming the problem of spurious modes, is to change the variation statement to account for the divergence constraint, e.g., as a penalty term

102

HESTHAVEN

TABLE 2 Degrees of Freedom for Nodal and Curl Elements of Order n Nodal element Quadrilateral Hexahedral Triangle Tetrahedron

2

(n þ 1) (n þ 1)3 1 2 ðn þ 1Þðn þ 2Þ 1 6 ðn þ 1Þðn þ 2Þðn þ 3Þ

Curl element 2(n þ 1)(n þ 2) 3(n þ 1)(n þ 2)2 (n þ 1)(n þ 3) 1 2 ðn þ 1Þðn þ 3Þðn þ 4Þ

d ðr E; Þ ¼ ðr  H; Þ þ ðJ; Þ þ ðr  r E; Þ ; dt d ðr H; Þ ¼ ðr  E; Þ þ ðr  H; Þ : dt Similar forms have been shown to successfully eliminate the spurious modes (Jiang, 1998; Jiang et al., 1996) using the general language of leastsquares stabilized low-order finite element schemes. As promising as this approach appears, we are unaware of any high-order results. While the developments of high-order finite element schemes for the firstorder system remain limited, there has been more recent activity regarding the development of finite element schemes for Maxwell’s equations on the curl–curl form, Eq. (25). Assuming again the use of scalar nodal finite elements, the strong variational form for Eq. (25) is   d2 1 d ð E; Þ þ r  r  E;  ¼ ðJ; Þ ; r

2 dt r dt

yielding the semidiscrete Galerkin form d2 x d E þ Sy;x E yh  Sy;y E xh  Sz;z E xh þ Sz;x E zh ¼ M J xh dt2 h dt d2 d M 2 E yh þ Sz;y E zh  Sz;z E yh  Sx;x E yh þ Sx;y E xh ¼ M J yh dt dt 2 d x;z x x;x z y;y z y;z y z  d M 2 E h þ S E h  S E h  S E h þ S E h ¼ M J zh ; dt dt M

ð30Þ

where S; ij ¼



i ;

@ 1 @ j @ r @



;



and the remaining operators are defined as described earlier. It is, however, more common to balance the smoothness between trial and test functions and consider the weak form

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

d2 ðr E; Þ þ dt2

I

103

  Z 1 1 d ^ n  r  E  dx  r  r  Edx ¼ ðJ; Þ

r r dt @



with a semidiscrete form very similar to Eq. (30). As for the first-order schemes discussed earlier, much attention has been paid to the problem of spurious modes in the frequency-domain form of the curl–curl equations. Indeed, it was in such schemes that problems with spurious solutions were first observed (Silvester, 1969). This has led to several different approaches to overcome this, following ideas similar to those discussed previously. The straightforward approach is to employ high-order curl-conforming elements to eliminate the possibility of spurious modes. Assuming solutions of the form in Eq. (27), this yields the semidiscrete scheme d2 d M 2 E h  SE h ¼ M J h ; dt dt where Eh and Jh represent vectors of global electric fields and currents, the global mass matrices are defined in Eq. (29), and the stiffness matrix has the entries   1 Sij ¼ r  N i ; r  N j : r

As demonstrated in Jiao et al. (2003a) this formulation allows for the development of high-order accurate schemes for the time-domain solution of the curl–curl equations. The effort demonstrates the viability of such an approach for solving full three-dimensional, time-dependent problems in combination with perfectly matched layers (Jiao and Jin, 2003) or a global boundary elements technique (Jiao et al., 2002). Although the available results remain limited, they nevertheless demonstrate the potential of such an approach. The alternative approach, modifying the variational statement to include the divergence constraint, takes the from (Lynch and Paulsen, 1990)  I  Z d2 1 1 ^ n  r  E  dx  ðr E; Þ þ r   E dx 2 dt r r

I Z @

1 d ^ n r  r E dx ¼ ðJ; Þ : þ ðr  r EÞr dx  dt

@ r r A related approach, as discussed in Jiang (1998) and Jiang et al. (1996), is the use of a least-squares stabilized finite element scheme, thus avoiding the direct penalization. Boyse et al. (1992), Paulsen et al. (1992), and Boyse and Paulsen (1997) proposed solving Maxwell’s equations using vector and scalar potentials, likewise eliminating spurious modes. We are unaware of attempts to combine such formulations with high-order elements.

104

HESTHAVEN

B. Discontinuous Finite Element Techniques As promising as the continuous finite element formulation is, it suffers from a number of problems that are not easily overcome. As already discussed, the need for a conforming discretization not only complicates matters but also results in the need to invert a global mass matrix at every time step. While this mass matrix is sparse and typically well conditioned, work associated with this inversion increases for higher-order methods and may become a bottleneck for large-scale parallel computations. Recently, however, formulations that eliminate these issues have appeared. While they can be derived for Maxwell’s equations on the firstorder form, Eq. (24), as well as for the curl–curl form, Eq. (25), recent work has focused on the former. We shall thus seek solutions to Eq. (24) in a general domain, with

considered as the union of nonoverlapping body-conforming elements, D. To simplify the derivation, we shall furthermore consider Maxwell’s equations conservation form, Eq. (31), as QðxÞ

@q þ r  FðqÞ ¼ Sðqi ; xÞ: @t

ð31Þ

Recall that q represents the state vector, the flux F is given in Eq. (23), Q reflects a diagonal matrix with material parameters, and S signifies the sources, e.g., the incoming fields and/or the current. To formulate the scheme we assume that there exists an approximate solution, qh , on the form X qh ðx; tÞ ¼ qi ðtÞi ðxÞ ð32Þ i

within each element. Similarly, we assume that F h and S h are polynomial representations of the flux and of the source, respectively. Note that we do not place any global constraints on the basis, i, i.e., it is in general discontinuous and nonconforming. To seek equations for the unknowns, we require the approximate solution to Maxwell’s equations, qh , to satisfy  Z  @qh þ r  F h  S h i ðxÞ dx Q @t D ð33Þ I  ¼ i ðxÞ^ n  ½F h  F  dx: @D

We emphasize that integration is over the local element, D, and not the full domain, , in contrast to the continuous finite element schemes discussed in Section VII.A.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

105

Here Fi and i represent sequences of N test functions, F * signifies a numerical flux, and ^ n is an outward-pointing unit vector defined at the boundary of the element. If the numerical flux is consistent, the scheme is clearly consistent. However, boundary/interface conditions are not imposed exactly but rather weakly through the penalizing surface integral. Thus, the formulation is inherently discontinuous and yields, through its very construction, a highly parallel local scheme. Let us define the local inner product Z u  vdx; ðu; vÞD ¼ D

the local mass matrices Mij ¼ ði ; r j ÞD ;

Mij ¼ ði ; r j ÞD ;

Mij ¼ ði ; j ÞD ;

and the discrete differentiation operator   @  Sij ¼ i ; j ; @ D where  ¼ ðx; y; zÞ. The boundary integration operator is defined as I i j dx: Fij ¼ @D

ð34Þ

ð35Þ

ð36Þ

With this, we can write the semidiscrete form of Maxwell’s equations as d x E  Sy H zh þ Sz H yh  MS hE;x ¼ FPhE;x dt h d E;y M E yh  Sz H xh þ Sx H zh  MS E;y h ¼ FP h dt

M

d z E  Sx H yh þ Sy H xh  MS hE;z ¼ FPhE;z dt h d M H xh  Sz E yh þ Sy E zh  MS hH;x ¼ FPhH;x dt d M H yh  Sx E zh þ Sz E xh  MS hH;y ¼ FPhH;y dt d M H zh  Sy E xh þ Sx E yh  MS hH;z ¼ FPhH;z : dt M

ð37Þ

Here ðE h ; H h Þ;  ¼ ðx; y; zÞ, represents the local degrees of freedom, SE,, and SH, represents the components of the sources discussed in Section II, and PH; for and we have introduced the penalizing boundary fluxes, PE; h h E h and H h , respectively. We shall define these shortly.

106

HESTHAVEN

One notes immediately that relaxing the continuity of the elements decouples the elements and results in a block-diagonal global mass matrix that can be inverted in preprocessing. The price paid for this is the additional degrees of freedom needed to support the local basis functions. However, for high-order elements, this is only a small fraction of the total number of degrees of freedom. The coupling of the local solutions to recover the global solution is accomplished through the numerical fluxes, F *. In this regard, one can view these methods as a high-order generalization of the finite volume schemes discussed in Section VI, albeit without the complications of wide stencils and complex procedures for the reconstruction of the point-wise solution. Given the linearity of Maxwell’s equations, it is natural to use upwinding, similar to the patching through characteristics discussed for the spectral multidomain schemes in Section V. This is given on the form (Mohammadian et al., 1991)  1 ^ n  ð^ n  ½E h   Zþ ½H h Þ; PEh ¼ Z

ð38Þ

 1 ^ n  ð^ n  ½H h  þ Y þ ½E h Þ: PH h ¼Y

ð39Þ

Here ½q ¼ q  qþ measures the jump in the field values across an interface. Superscript ‘‘þ’’ refers to field values from the neighbor element, whereas superscript ‘‘’’ refers to field values local to the element. To account for the potential differences in material properties in the two elements, the local impedance, Z , and conductance, Y , is defined as rffiffiffiffiffiffi 1 

Z ¼ ¼  Y and the sums  ¼ Zþ þ Z ; Y ¼ Y þ þ Y  Z of the local impedance and conductance, respectively. As a possible alternative to the upwind fluxes in Eqs. (38)–(39), one could use a simple central flux, resulting in  1 Z þ ^ n  ½H h ; PEh ¼ Z  1 Y þ ^ PH n  ½Eh : h ¼Y This ensures semi-discrete energy conservation but appears to suffer from spurious modes and problems of robustness. For further details, see Hesthaven and Warburton (2001), Piperno and Fezoui (2003), and Hesthaven and Warburton (2003b).

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

107

Choosing the test functions, Fi, Ci and the numerical flux, F  , one has a large degree of freedom when designing different schemes. Focusing on Galerkin schemes, in which case Ci ðxÞ ¼ Fi ðxÞ ¼ i ðxÞ, it is worth realizing that following integration by parts in Eq. (33), this becomes the much studied discontinuous Galerkin method (Cockburn and Shu, 2001; Cockburn et al., 2000; Atkins and Shu, 1998). This is, however, only one among many different formulations in the same family of discontinuous element/penalty methods. We refer to Hesthaven (2000), Hesthaven and Gottlieb (1999), and Hesthaven and Teng (2000), where other choices are studied in the general context of conservation laws and problems of wave propagation. To complete the scheme, one needs to specify the element type and an associated basis i(x), most often of polynomial nature, and define the unknown coefficients, qh, for functions defined on the elements. Using general curvilinear quadrilaterals, as in Kopriva et al. (2000, 2002), it is natural to use a tensor-product interpolating basis as is done for the spectral multidomain schemes discussed in Section V.B. The advantage of this is, apart from its simplicity, that one recovers a diagonal local mass matrix by using polynomials defined at quadrature points. This results in schemes that are very similar to those in Section V.B, the main difference being whether the characteristic conditions on the boundary fluxes are imposed weakly or strongly. A nonconforming extension of such schemes is discussed in Kopriva et al. (2002), and the dispersion characteristics of such schemes are discussed in Hu et al. (1999). Extensions to problems with nonuniform grids are analyzed in Hu and Atkins (2001, 2002), confirming that such discontinuous formulations are well suited for wave propagation. In Warburton (2000) and Hesthaven and Warburton (2002, 2003), the development of a Galerkin scheme on nodal tetrahedral elements is initiated, aimed at demonstrating the potential of using a discontinuous element formulation for solving very large geometrically complex three-dimensional problems in time-domain computational electromagnetics. Choosing the appropriate form of the local basis on the tetrahedron is less a question of formulation and more a question of performance as measured by efficiency and accuracy of the final scheme. An immediate candidate is the monomial basis, i ðxÞ ¼ x1 y2 z3 with |a|  n. As is well known, however, this will lead to extremely ill-conditioned operators as the basis becomes almost linearly dependent for a high polynomial order and prohibits the stable and accurate computation at high order. To overcome such conditioning problems, we first follow the approach of Section V.B and introduce a smooth curvilinear mapping, : D ! I, between the general element, D, and a canonical tetrahedron, I, on which we seek an orthonormal basis. Such a basis has been known for a long time

108

HESTHAVEN

(Proriol, 1957; Koornwinder, 1975; Dubiner, 1991). This leaves the question of how to compute the expansion coefficients, q. Clearly, with an orthonormal basis at hand, it may seem natural to use this as the local basis. The impact of doing so, however, is that all modes are needed to evaluate qh point wise. This lack of separation between inner modes and boundary modes is not optimal for the discontinuous formulation where the flux term depends on fluxes at the boundary of D only. To overcome this issue, one could seek to give up the strict orthonormality of the basis to achieve a separation between inner and boundary modes. Such a basis is developed in Karniadakis and Sherwin (1999) and provides an approach, albeit rather complex, to achieve arbitrarily high-order accuracy. Using a nodal element, however, one can define qh as an interpolating polynomial, i.e., we require that X 8i : qh ðxðj i Þ; tÞ ¼ qj ðtÞj ðj i Þ; j

where j ðjÞ is the orthonormal basis on I and j i are predefined grid points in I. The number of nodes, N, is simply that required for completeness, as listed in Table 2. On vector form this yields the requirement that qh ¼ Vq; Vij ¼ j ðji Þ;

ð40Þ

where V is a multidimensional Vandermonde matrix. The genuine multivariate Lagrangian polynomials are qh ðxðjÞ; tÞ ¼

N X i¼1

qh ðxðj i Þ; tÞLi ðjÞ; VT L ¼ ;

where the latter expression for evaluation of the Lagrange polynomials follows from the interpolation property. Here L ¼ [L1 ðjÞ; ::; LN ðjÞ]T and the basis is given as f = [1 (j) , . . ., N (j)]T: The final issue in need of attention is the choice of nodal points, ji, within I. As is well known, the success of high-order Lagrangian interpolation is critically dependent on the correct distribution of the nodes. This is a problem that has received some attention recently and nodal distributions, enabling the construction of well-behaved unique Lagrange polynomials. Up to order 10 on the tetrahedron (Chen and Babus˜ka, 1996; Hesthaven and Teng, 2000) has been constructed. The nodal distributions are characterized by having exactly N nodes. Furthermore, the nodal set includes vertices, edges, and faces of the tetrahedron. The number of nodes on each face is exactly that required to support a two-dimensional multivariate polynomial, i.e., N2d ¼

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

109

ðn þ 1Þðn þ 2Þ=2 nodes on each face. The same characteristics are shared by nodes on the triangles (Hesthaven, 1998). In this setting it is more natural to recast the scheme in physical space. The only difference with Eq. (37) is that (E h , H h ) then represents the N-long vectors of nodal values in each element, S h the nodal values of the source function, and PEh and PH h the nodal values of the numerical flux as defined in Eqs. (38) and (39). The discrete, point-wise operators are given as Z Z @Lj dx: ð41Þ Li Li Lj dx; Sij ¼ M ij ¼ @ D D The form of the boundary operator, F, is simplified as a consequence of the uniqueness of the Lagrange polynomial and the structure of the nodal points, i.e., integration of the three-dimensional Li over the surface is equivalent to the sum of the integration of the two-dimensional Lagrange polynomials defined by the nodal distribution on the faces. This implies that I face li2D lj2D dx; Fij ¼ face X ð42Þ T face 1 F¼ RTface ðV1 V2D Rface : 2D Þ F faces

Here l 2D represents the unique two-dimensional Lagrange polynomials i defined by the nodes on each of the four faces, V2D is the associated Vandermonde matrix similar to the three-dimensional form, Eq. (40), and Rface is an N2d  N that serves to extract those nodes situated at each face of the element. To reiterate the importance of this separation between internal and boundary nodes, we note that the operation count for evaluating the scheme, Eq. (37), assuming no separation, is O(6N2) for each variable. For the nodal scheme, or a modal scheme with a similar separation, the work scales like O(2N2 þ 4N N2d). Hence, the relative saving in operations scales as work with nodal basis 1 2 ¼ þ : work with simple modal basis 3 n þ 3

This clearly becomes increasingly important as the order of the approximation, n, increases, although even for n ¼ 3 do we find a one-third reduction. One of the main advantages of the nodal element is the ease with which one can relax the restriction on tetrahedra having straight faces only. Clearly, this will impact the evaluation of the discrete operators, Eqs. (41) and (42), by requiring specific operators for each element and sufficient accuracy in integration to evaluate entries in the operators. However, the

110

HESTHAVEN

TABLE 3 Relative Time for a 245,000 Element Grid with Sixth-Order Elements as a Function of the Number of Processors Number of processors

64

128

256

512

Relative time

1.00

0.48

0.24

0.14

evaluation of boundary fluxes is straightforward in a nodal representation n, vary along the faces. even as the normal vectors, ^ Details of the nodal-based discontinuous element scheme and its efficient implementation can be found in Hesthaven and Warburton (2002, 2003), including a complete convergence analysis and alternative divergencepreserving formulations. The discontinuous element formulation can be expected to allow a highly efficient parallel implementation on contemporary large-scale distributed memory machines. As a verification of this, Table 3 lists the relative parallel speedup for a single large-scale application, demonstrating superlinear scaling. Similar and more extensive studies, given in Hesthaven and Warburton (2002), confirm this high parallel efficiency for a variety of applications. Let us conclude this discussion with a few examples. Advancement in time is done using a low-storage, fourth-order explicit Runge–Kutta method (Carpenter and Kennedy, 1994), and the computational domain is terminated with a combination of stretching of the grid and characteristic boundary conditions at the outer boundaries (Yang et al., 1997b). As a first, simple two-dimensional problem, we consider TM-polarized plane wave scattering by a ka ¼ 15 metallic cylinder. Figure 14 shows both a fraction of the grid, illustrating the body-conforming high-order nodal grid, and the bistatic RCS computed using a fixed, very coarse grid and achieving convergence by increasing the order of the scheme. As an example of a more challenging three-dimensional problem, consider plane wave scattering by a perfectly conducting conesphere, consisting of a  60.5-cm-long cone with a half angle of 7 , capped smoothly with a 7.49-cm spherical cap radius. Illuminated by a 9-GHz plane wave, the object is approximately 21 wavelengths long. What makes the problem challenging though is not only its electric size but also the very sharp apex and the long shadow region. Figure 15 shows a detail of the grid near the apex, as well as the full bistatic cross section for axial plane wave illumination of the conesphere, showing excellent agreement with high-fidelity results obtained using a CFIE integral equation solver. The computation utilizes about 270,000

111

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

9 35

8

30 RCS (dbm)

Y/l

7 6 5 4 3

25 20

n =8 Exact series solution n =4

15 10

2 2

3

4

5

6

7

8

5

9

0

30

60

X/l

90 120 q (deg.)

150

180

Figure 14. (Left) Details of the body-conforming grid used to compute scattering by a two-dimensional PEC cylinder. (Right) The rapidly converging bistatic radar cross section when increasing the order, n, of the scheme. Z

7 deg Cone-Sphere − 9GHz 40

X

Y

35

TM-Polarization

30

j inc = 0 deg. j inc = 180 deg.

RCS (q, 0) (dbm)

25 20 15 10 5 0 −5

10

−10

CFIE/EFIE

−15

3D Time-Domain (USEMe)

−20 0

15

30

60

90 q

120

150

180

Figure 15. (Left) Details of the body-conforming grid used to compute scattering by a large PEC conesphere. The surfaces are triangulated for visualization based on the nodes of high-order elements. (Right) Computed bistatic radar cross section (RCS) for vertically polarized plane wave illumination at the tip and compared with results using an integral equation-based frequency domain solver (CFIE).

tetrahedral elements at third order with a resolution at the surface of up to 20 points/wavelength. We note the excellent agreement and a dynamic range exceeding 50 db. Similar results and agreement have been found for TE-polarized illumination. As an example of a problem involving penetration, we consider plane wave scattering by a dielectric cylinder, 5l long, with a radius of 1l and made of a

112 30

50

20

40 30

10

20 0 10 −10

0

−20

RCS (q, 90) (dBm)

RCS (q, 0) (dBm)

HESTHAVEN

−10

−30 0

30

60

90 q

120

150

−20 180

Figure 16. Scattering by a finite length dielectric cylinder with er ¼ 2.25. We show the RCS (, 0) for vertical polarization ( ) of the illuminating field and RCS (, 90) for horizontal polarization ( ) compared with results obtained using a spectral multidomain axisymmetric code (full line).

Figure 17. Application of an unstructured grid discontinuous element high-order method to the solution of electromagnetic scattering a military aircraft. The frequency of the incoming plane wave is 600 MHz. (Left) A part of the triangulated surface grid and (right) one of the magnetic field components on the surface of the plane are shown. Computation is performed with fourth-order elements and approximately 245,000 tetrahedra to fill the computational volume.

nonmagnetic material with a permittivity of r ¼ 2.25, similar to that of glass. We find that using approximately 67,000 elements, supporting a fourthorder approximation and with an average vacuum edge length at the cylinder of l/3 suffices to accurately predict the far field scattering (see Figure 16). As a final example, illustrating the level of complexity that can currently be considered with this formulation, Figure 17 shows part of a 245,000

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

113

element grid for 600 MHz scattering by a military aircraft. One of the field components on the surface of the aircraft is also shown. Many more examples of the geometric flexibility and high-order accuracy of this approach for the time-domain solution of Maxwell’s equations can be found in Hesthaven and Warburton (2002, 2003a), including scattering by aircraft configurations in the GHz regime. VIII. Issues in Temporal Integration The discussion of the various schemes has so far focused on the spatial discretization of Maxwell’s equations, leading to systems of ordinary differential equations. To solve these, a number of techniques are available in the literature, and the majority of high-order spatial discretization schemes are combined with standard methods, such as explicit third- or fourth-order Runge–Kutta methods (Butcher, 1987; Hairer et al., 1993). Interesting alternatives to these classical approaches are low-storage Runge– Kutta methods (Butcher, 1987; Carpenter and Kennedy, 1994), limiting the need for additional stages, and dispersion-optimized Runge–Kutta schemes (Hu et al., 1996), designed for propagating waves over long distances. The use of 3rd or 4th order temporal schemes, while striving to achieve higher spatial order, may appear slightly counter productive. This does, however, represent a tradeoff between several things, e.g., higher-order temporal schemes tend to be rather complex and expensive in terms of memory and function evaluations. Furthermore, in many practical applications is it often observed that using an explicit 4th order accurate temporal integration suffices to ensure that errors are dominated by spatial errors. Using a spatial high-order finite difference scheme, many practitioners continue to use the second-order-accurate leap frog scheme, also used in the classical Yee (1966) scheme often choosing the time step under error constraints rather than stability constraints. This approach is used in Turkel and Yefet (2000), Yefet and Turkel (2000), and Yefet and Petropoulos (2001). In Xie et al. (2002), a deferred correction technique using a backward differentiation method is proposed to achieve fourth order. The situation is very similar when using finite volume or finite element discretizations of the first-order Maxwell’s equations where second-order leap frog schemes (Cohen, 2002; Cohen and Monk, 1999) or explicit Runge– Kutta methods (Gaitonde and Shang, 1997; Gaitonde et al., 1997) remain the main workhorses. For finite element discretizations of the curl–curl equations, leading to an equation of second order in time, the standard choice is the Newmark scheme (Hughes, 2000), generally chosen to be

114

HESTHAVEN

second-order accurate and either implicit or explicit (Jiao and Jin, 2002; Jiao et al., 2002). Interesting alternatives could be Nystro¨m methods to enable a higher order accuracy. As the finite element discretization of the curl–curl form always requires a matrix inversion, implicit schemes seem most attractive as they come at limited additional cost. Conditions for discrete stability naturally depend on the details of spatial and temporal discretization, as well as the form in which Maxwell’s equations are stated. However, combining any of the semidiscrete schemes discussed here with an explicit time integration scheme generally yields a condition for discrete stability as
t  C min

pffiffiffiffiffiffiffiffi r r h:

What separates the different schemes is partly the value of the constant C, typically of O(1), but most importantly what the grid size, h, means. Naturally, for the extensions of the Yee scheme discussed in Section IV or the high-order finite volume schemes in Section V, h maintains its simple meaning from the equidistant grid. However, for the more complicated multidomain/multielement schemes, the geometric flexibility comes at a price, as typically one has h/

l ; n2

where n represents the order of the approximation and l the smallest edge length of the elements. This illustrates that one should strive to use as large elements as possible to avoid prohibitively small time steps and, thus, very long computing times. Some attempts to slightly improve on this are discussed in Driscoll and Fornberg (1998), Fornberg (1996), and Hesthaven et al. (1999), although one has to be careful not to increase the time step at the expense of accuracy. Ultimately, this emphasizes the need to support curvilinear body-conforming elements in the formulation because one must aim to resolve the solutions and not the geometry, as the latter may result in unnecessarily small stable time steps. As applications become increasingly complex, the geometries themselves often require small cells and, thus, small time steps. Techniques to overcome this remain active research areas. Fully implicit time stepping is of course an option, but may be prohibitive for large-scale problems in which the stiffness is localized to small regions of the grid. More interesting alternatives include the use of nonconforming discretizations (Demkowicz and Vordapetyan, 1998; Kopriva et al., 2002), explicit-implicit Runge–Kutta methods (Kennedy and Carpenter, 2002) enabling splitting on the grid, and time-accurate local time-stepping methods (Dawson and Kirby, 2001).

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

115

IX. Conclusions and Outlook Looking through the list of references accompanying this review, one quickly realizes that most references directly related to the high-order accurate timedomain solution of Maxwell’s equation are less that 5 years old. This is perhaps a testament to the timeliness of this review, but certainly reflects the activity experienced in this research area over the last few years. However, learning about the various efforts also emphasizes that much work remains to be done. The simplicity of the finite difference bases embedding schemes, avoiding grid generation and allowing the treatment of complex, even moving, boundaries in a simple manner is also its Achilles heel. It is indeed difficult to imagine higher than fourth-order accuracy, and many issues related to the stability of general interfaces remain open. However, fourth order may well suffice for many problems of moderate size and complexity. Indeed, if stable and robust versions of such methods could be developed, they may well have the potential to succeed the current golden standard—the Yee scheme. Currently, however, there seems to be no robust alternative to multielement schemes, be they spectral multidomain schemes, high-order timedomain finite element schemes, or discontinuous finite element methods. Each of these formulations has its own advantages and disadvantages, although at this particular point in time, the development of discontinuous element formulations, (Section VII.B) appears to be the most advanced. Many issues continue to require serious attention. Apart from the plentiful theoretical questions, e.g., semidiscrete and fully discrete stability, smoothness of the solutions around nonsmooth geometries and its impact on the convergence rate, the importance of the divergence constraints in time-domain schemes etc, many issues with a potential for immediate impact remains open. Perhaps most evident is the need to consider alternatives to the widely used explicit time-stepping schemes. For large-scale geometrically complex problems, this is becoming a bottleneck. Another area that continues to require attention is the development of accurate and efficient means to truncate the computational domain. This becomes of increasing importance as accuracy requirements increase. Perfectly Matched Layer methods (Berenger, 1994, 1996) have received much attention in the last decade and continue to be a viable solution. Their cost for large-scale problems is, however, a concern. Global boundary conditions (e.g., Hagstrom, 1999; Grote and Keller, 1998; Ryaben’kii et al., 2001) deserve serious attention, as does the recently demonstrated use of time-domain integral equations (Jiao et al., 2002) as a means to truncate the computational domain.

116

HESTHAVEN

High-order accurate multielement techniques are currently limited by low-order grid generation, i.e., most commercial grid generation software does not support higher order descriptions of boundaries and interfaces. To fully ripe the benefits of high-order accuracy, this must be overcome, e.g., through a more dynamic interface between the model description and the grid generation. This problem is, however, not unique to electromagnetics, and there is currently significant research activity to overcome this restriction and enable high-order model description and grid generation. Adaptive solution techniques, as well as accurate and efficient means to treat randomness in geometries, materials, and sources, are areas that have received only very limited attention in the past. Nevertheless, advances in these areas have the potential for a dramatic impact as applications continue to emphasize higher frequencies and more complex signal forms and materials. While it took the insight of Maxwell to realize the beautiful simplicity of electromagnetic wave propagation, recent advances in high-order accurate methods for such phenomena suggest that less can do when it comes to solving them computationally. As complex as these problems are, advances over the last decade are substantial and encouraging, although the applications continue to surpass the computational capabilities in complexity and size. Nevertheless, the gap is slowly narrowing, and the continued development of high-order accurate methods for the time-domain solution of Maxwell’s equations may eventually enable the development of robust, accurate, and efficient computational tools, powerful and versatile enough to address the electromagnetic problems of tomorrow.

Acknowledgments Much of the work contained here was done in collaboration with many collaborators over an extended period of time. Special thanks goes to Cedric Chauviere, Adi Ditkowski, Palle Dinesen, David Gottlieb, Jiaming Jin, Eli Turkel, Tim Warburton, Daniel White, and Baolin Yang for comments, suggestions, and fruitful collaborations over the years. Thanks also to Bengt Fornberg and Tobin Driscoll for providing graphical material. This work was partly supported by NSF under contract DMS-0074257, ARO under contract DAAD19-01-1-0631, by AFOSR/DARPA under contract F496201-0426, and by the Alfred P. Sloan Foundation through a Sloan Research Fellowship.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

117

References Abarbanel, S., Ditkowski, A., and Yefet, A. (2003). Bounded error schemes for the wave equation on complex domains. Submitted for publication. Ainsworth, M., and Coyle, J. (2003). Hierarchic finite element bases on unstructured tetrahedral meshes. Int. J. Numer. Meth. Eng. (In press). Atkins, H., and Shu, C. W. (1998). Quadrature-free implementation of discontinuous Galerkin methods for hyperbolic equations. AIAA J. 36, 775–782. Bayliss, A., and Turkel, E. (1992). Mapping and accuracy for Chebyshev pseudospectral approximations. J. Comput. Phys. 101, 349–359. Berenger, J. P. (1994). A perfectly matched layer for the absorption of electromagnetic waves. J. Comput. Phys. 114, 185–200. Berenger, J. P. (1996). Three-dimensional perfectly matched layer for the absorption of electromagnetic waves. J. Comput. Phys. 127, 363–379. Bossavit, A. (1988). Rationale for ‘‘edge-elements’’ in 3-D fields computations. IEEE Trans. Magnet. 24, 74–79. Bossavit, A. (1990). Solving Maxwell’s equations in a closed cavity, and the question of spurious modes. IEEE Trans. Magnet. 26, 702–705. Bowman, J. J., Senior, T. B. A., and Uslenghi, P. L. E. (1987). ‘‘Electromagnetic and acoustic scattering by simple shapes.’’ Hemisphere Publishing Corporation, New York. Boyse, W. E., Lynch, D. R., Paulsen, K. D., and Minerbo, G. N. (1992). Nodal-based finite element modeling of Maxwell’s equations. IEEE Trans. Antennas Propagat. 40, 42–651. Boyse, W. E., and Paulsen, K. D. (1997). Accurate solutions of Maxwell’s equations around PEC corners and highly curved surfaces using nodal finite elements. IEEE Trans. Antennas Propagat. 45, 758–1767. Butcher, J. C. (1987). ‘‘The Numerical Analysis of Ordinary Differential Equations. RungeKutta and General Linear Methods.’’ Wiley, New York. Canuto, C., Hussaini, M. Y., Quarteroni, A., and Zang, T. A. (1988). ‘‘Spectral Methods in Fluid Mechanic.’’ Springer Verlag, New York. Carpenter, M. H., Gottlieb, D., and Abarbanel, S. (1993). Stable and accurate boundary treatments for compact, high-order finite-difference schemes. Appl. Numer. Math. 12, 55–87. Carpenter, M. H., Gottlieb, D., and Abarbanel, S. (1994). Time-stable boundary conditions for finite-difference schemes solving hyperbolic systems: Methodology and applications to highorder compact schemes. J. Comput. Phys. 111, 220–236. Carpenter, M. H., and Kennedy, C. A. (1994). Fourth order 2N-storage Runge-Kutta scheme. NASA-TM-109112. NASA Langley Research Center, VA. Cendes, Z. J., and Silvester, P. (1971). Numerical solution of dielectric loaded waveguides. I Finite element analysis. IEEE Trans. Microwave Theo. Tech. 18, 1124–1131. Chen, Q., and Babus˘ka, I. (1996). The optimal symmetrical points for polynomial interpolation of real functions in a tetrahedron. Comput. Methods Appl. Mech. Engrg. 137, 89–94. Chew, W. C., Jin, J., Michielssen, E., and Song, J. (eds.). (2001). ‘‘Fast and Efficient Algorithms in Computational Electromagnetics’’. Artech House, Boston. Cockburn, B., and Shu, C. W. (2001). Runge-Kutta discontinuous Galerkin methods for convection dominated problems. J. Sci. Comput. 16, 173–261. Cockburn, B., Karniadakis, G. E., and Shu, C. W. (2002). ‘‘Discontinuous Galerkin methods. Theory, computations, and applications.’’ Springer Verlag, New York. Cohen, G. C. (2002). ‘‘Higher-Order Numerical Methods for Transient Wave Equations.’’ Springer-Verlag, Berlin.

118

HESTHAVEN

Cohen, G., and Monk, P. (1999). Mur-Nedelec finite element schemes for Maxwell’s equations. Comput. Methods Appl. Mech. Engrg. 169, 197–217. Dawson, C., and Kirby, R. (2001). High resolution schemes for conservation laws with locally varying time steps. SIAM J. Sci. Comput. 22, 2256–2281. Demkowicz, L., and Vardapetyan, L. (1998). Modeling of electromagnetic absorption/ scattering using hp-adaptive finite elements. Comput. Methods Appl. Mech. Engrg. 152, 103–124. Dinesen, P. G., Hesthaven, J. S., and Lynov, J. P. (2000). Rigorous Analysis of Focusing Grating Couplers Using a Time-Domain Spectral Collocation Method. InDi_ractive/ Holographic Technologies and Spatial Light Modulators 7. (I. Cindrich, S. H. Lee, and R. L. Sutherland, eds.), (International Society for Optical Engineering, Bellingham, WA, 2000), Proceedings of SPIE 3951, 11–19, San Gose, CA. Ditkowski, A., Dridi, K., and Hesthaven, J. S. (2001). Convergent cartesian grid methods for Maxwell’s equations in complex geometries. J. Comput. Phys. 170, 39–80. Driscoll, T. A., and Fornberg, B. (1998). A block pseudospectral method for Maxwell’s equations. I. One-dimensional case. J. Comput. Phys. 140, 47–65. Driscoll, T. A., and Fornberg, B. (1999). Block pseudospectral methods for Maxwell’s equations. II. Two-dimensional, discontinuous-coefficient case. SIAM J. Sci. Comput. 21, 1146–1167. Dubiner, M. (1991). Spectral methods on triangles and other domains. J. Sci. Comput. 6, 345–390. Fan, G. X., Liu, Q. H., and Hesthaven, J. S. (2002). Multi-domain pseudospectral time-domain simulations of scattering by objects buried in Lossy Media. IEEE Trans. Geosci. Remote Sens. 40, 1366–1373. Fang, J. (1989). ‘‘Time-Domain Finite-Difference Computation for Maxwell’s Equations.’’ Ph.D. thesis, U.C. Berkeley. Fischer, P., and Gottlieb, D. (1997). On the optimal number of subdomains for hyperbolic problems on parallel computers. Int. J. Supercomput. Appl. High Perform. Comput. 11, 65–76. Fornberg, B. (1996). ‘‘A Practical Guide to Pseudospectral Methods.’’ Cambridge University Press, Cambridge. Fornberg, B. (1998). Calculation of weights in finite difference formulas. SIAM Rev. 40, 685–691. Gaitonde, D., and Shang, J. S. (1996). ‘‘High-Order Finite-Volume Schemes in Wave Propagation Phenomena.’’ AIAA Paper 96–2335. Gaitonde, D., and Shang, J. S. (1997). Optimized compact-difference-based finite volume schemes for linear wave phenomena. J. Comput. Phys. 138, 617–643. Gaitonde, D., Shang, J. S., and Young, J. L. (1997). ‘‘Practical Aspects of High-Order Accurate Finite-Volume Schemes for Electromagnetics.’’ AIAA Paper 97–0363. Georgakopoulos, S. V., Birtcher, C. R., Balanis, C. A., and Renaut, R. A. (2002a). Higherorder finite-difference schemes for electromagnetic radiation, scattering, and penetration I. Theory. IEEE Antennas Propagat. Mag. 44, 134–142. Georgakopoulos, S. V., Birtcher, C. R., Balanis, C. A., and Renaut, R. A. (2002b). Higherorder finite-difference schemes for electromagnetic radiation, scattering, and penetration, 2. Applications. IEEE Antennas Propagat. Mag. 44, 92–101. Gordon, W. J., and Hall, C. A. (1973). Transfinite element methods: Blending-function interpolation over arbitrary curved element domains. Numer. Math. 21, 109–129. Gottlieb, D., Hussaini, M. Y., and Orszag, S. A. (1984). In ‘‘Theory and applications of spectral methods’’ (R. G. Voigt, D. Gottlieb, and M. Y. Hussaini, eds.), Spectral Methods for Partial Differential Equations. SIAM Publications, pp. 1–53. Gottlieb, D., and Hesthaven, J. S. (2001). Spectral methods for hyperbolic problems. J. Comp. Appl. Math. 128, 83–131.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

119

Graglia, R. D., Lubbers, R. J., and Wilton, D. R. (eds.) (1998). Special issue on advanced numerical techniques in electromagnetics. IEEE Trans. Antennas Propagat. 45. Graglia, R. D., Wilton, D. R., and Peterson, A. F. (1997). Higher order interpolatory vector bases for computational electromagnetics. IEEE Trans. Antennas Propagat. 45, 329–342. Grote, M. J., and Keller, J. B. (1998). Non-reflecting boundary conditions for Maxwell’s equations. J. Comput. Phys. 127, 327–342. Gustafsson, B. (1975). The convergence rate for difference approximations to mixed initial boundary value problems. Math. Comp. 29, 396–406. Gustafsson, B., Kreiss, H. O., and Oliger, J. (1995). ‘‘Time-Dependent Problems and Difference Methods.’’ Wiley, New York. Hagstrom, T. (1999). Radiation boundary conditions for the numerical simulation of waves. Acta Numerica 8, 47–106. Hairer, E., Nørsett, S. P., and Wanner, G. (1993). ‘‘Solving Ordinary Differential Equations I.’’ Springer Verlag, New York. Harten, A., Engquist, B., Osher, S., and Chakravarthy, S. R. (1987). Uniformly high order accurate essentially non-oscillatory schemes III. J. Comput. Phys. 71, 231–303. Hesthaven, J. S. (1998). From electrostatics to almost optimal nodal sets for polynomial interpolation in a simplex. SIAM J. Numer. Anal. 35, 655–676. Hesthaven, J. S. (1999). A stable penalty method for the compressible navier-stokes equations III. Multidimensional domain decomposition schemes. SIAM J. Sci. Comp. 20, 62–93. Hesthaven, J. S. (2000). Spectral penalty methods. Appl. Numer. Math. 33, 23–41. Hesthaven, J. S., Dinesen, P. G., and Lynov, J. P. (1999). Spectral collocation time-domain modeling of diffractive optical elements. J. Comput. Phys. 155, 287–306. Hesthaven, J. S., and Gottlieb, D. (1999). Stable spectral methods for conservation laws on triangles with unstructured grids. Comput. Methods Appl. Mech. Engin. 175, 361–381. Hesthaven, J. S., and Teng, C. H. (2000). Stable spectral methods on tetrahedral elements. SIAM J. Sci. Comput. 21, 2352–2380. Hesthaven, J. S., and Warburton, T. (2002). High-order nodal methods on unstructured grids. I. Time-domain solution of Maxwell’s equations. J. Comput. Phys. 181, 1–34. Hesthaven, J. S., and Warburton, T. (2003a). High-order accurate methods for time-domain electromagnetics. Comput. Model. Engin. Sci. (In press). Hesthaven, J. S., and Warburton, T. (2003b). High order nodal discontinuous Galerkin Methods for the Maxwell eigenvalue problem. Philo. Trans. Royal Soc. Hiptmair, R. (1999). Canonical construction of finite elements. Math. Comp. 68, 1325–1346. Hiptmair, R. (2001). Higher order Whitney forms. Progr. Electromagn. Research PIER 32, 271–299. Hiptmair, R. (2002). Finite elements in computational electromagnetics. Acta Numerica 2002 237–339. Holland, R. (1983). Finite difference solutions of Maxwell’s equations in generalized nonorthogonal coordinates. IEEE Trans. Nuclear Sci. 30, 4589–4591. Hu, F. Q., and Atkins, H. L. (2001). Eigensolution analysis of the discontinuous Galerkin method with non-uniform grids I. One space dimension. ICASE Report No. 2001–40, pp. 35. Hu, F. Q., and Atkins, H. L. (2002). ‘‘Two-dimensional wave analysis of the discontinuous Galerkin Method with non-uniform grids and boundary conditions.’’ AIAA-paper, pp. 2002–2514. Hu, F. Q., Hussaini, M. Y., and Manthey, J. L. (1996). Low-dissipation and low-dispersion Runge-Kutta schemes for computational acoustics. J. Comput. Phys. 124, 177–191. Hu, F. Q., Hussaini, M. Y., and Rasetarinera, P. (1999). An analysis of the discontinuous Galerkin method for wave problems. J. Comput. Phys. 151, 921–946.

120

HESTHAVEN

Hughes, T. J. R. (2000). ‘‘The Finite Element Method: Linear and Dynamic Finite Element Analysis.’’ Dover, New York. Jiang, B.-N. (1998). ‘‘Least Squares Finite Element Methods: Theory and Applications in Computational Fluid Dynamics and Electromagnetics.’’ Springer-Verlag, New York. Jiang, B.-N., Wu, J., and Povinelli, L. A. (1996). The origin of spurious solutions in computational electromagnetics. J. Comput. Phys. 125, 104–123. Jiao, D., and Jin, J. M. (2003). Three-dimensional orthogonal vector basis functions for time-domain finite element solution of vector wave equations. IEEE Trans. Antennas Propagat. 51, 59–66. Jiao, D., Jin, J. M., Michielssen, E., and Riley, D. (2002). Time-domain finite-element simulation of three-dimensional scattering and radiation problems using perfectly matched layers. IEEE Trans. Antennas Propagat. 51, 286–305. Jiao, D., Ergin, A., Shanker, B., Michielssen, E., and Jin, J. M. (2002). A fast time-domain higher-order finite-element boundary-integral method for 3-D electromagnetic scattering analysis. IEEE Trans. Antennas Propagat. 50, 1192–1202. Jin, J. M. (1993). ‘‘The Finite Element Method in Electromagnetics.’’ Wiley, New York. Jurgens, H. M., and Zingg, D. W. (2000). Numerical solution of the time-domain Maxwell Equations using high-accuracy finite difference methods. SIAM J. Sci. Comput. 22, 1675–1696. Jurgens, T. G., Taflove, A., Umashaankar, K., and Moore, T. G. (1992). Finite-difference timedomain modeling of curved surfaces. IEEE Trans. Antennas Propagat. 40, 357–366. Kabakian, A. (1996). A spectral algorithm for electromagnetic wave scattering in the timedomain: Applications to RCS computations. AIAA Paper 96–2334. Kangro, U., and Nicolaides, R. A. (1997). Spurious fields in time-domain computations of scattering problems. IEEE Trans. Antennas Propagat. 45, 228–234. Karniadakis, G. E., and Sherwin, S. J. (1999). ‘‘Spectral/hp Element Methods for CFD: Numerical Mathematics and Scientific Computation.’’ Clarendon Press, Oxford. Kennedy, C. A., and Carpenter, M. H. (2002). Additive Runge-Kutta schemes for convectiondiffusion-reaction equations. Appl. Numer. Math. Koornwinder, T. (1975). Two-variable analogues of the classical orthogonal polynomials. In ‘‘Theory and Application of Special Functions’’ (R. A. Askey, ed.), pp. 435–495. Academic Press, San Diego. Kopriva, D. A. (1986). A spectral multidomain method for the solution of hyperbolic systems. Appl. Numer. Math. 2, 221–241. Kopriva, D. A. (1989). Computation of hyperbolic equations on complicated somains with patched and overset Chebyshev grids. SIAM J. Sci. Stat. Comput. 10, 120–132. Kopriva, D. A., Woodruff, S. L., and Hussaini, M. Y. (2000). In ‘‘Discontinuous Spectral Element Approximation of Maxwell’s Equations, in Discontinuous Galerkin Methods: Theory, Computation and Applications,’’ (B. Cockburn, G. E. Karniadakis, and C. W. Shu, eds.), Lecture Notes in Computational Science and Engineering 11. Springer Verlag, New York. Kopriva, D. A., Woodruff, S. L., and Hussaini, M. Y. (2002). Computation of electromagnetic scattering with a non-conforming discontinuous spectral element method. Int. J. Num. Meth. Engin. 53, 105–122. Kreiss, H. O., and Oliger, J. (1972). Comparison of accurate methods for the integration of hyperbolic problems. Tellus 24, 199–215. Kreiss, H. O., and Scherer, G. (1974). Finite element and finite difference methods for hyperbolic partial differential equations. In ‘‘Mathematical Aspects of Finite Elements in Partial Differential Equations.’’ Academic Press, New York. Kunz, K. S., and Luebbers, R. J. (1993). ‘‘The finite difference time domain method in electromagnetics.’’ CRC Press Inc.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

121

Lee, J. F., Lee, R., and Cangellaris, A. (1997). Time-domain finite-element methods. IEEE Trans. Antennas Propagat. 45, 430–441. Lee, J. F., and Lee, R. (eds.) (1999). Special issue on computational electromagnetics. Comput. Methods Appl. Mech. Engrg. 169. Lele, S. K. (1992). Compact finite-difference schemes with spectral-like resolution. J. Comput. Phys. 103, 16–42. Li, Q., Chen, Y., and Ge, D. (2000). Comparison study of the PSTD and FDTD methods for scattering analysis. Microwave Opt. Techno. Lett. 25, 220–226. Liu, Q. H. (1997). The PSTD algorithm: A time-domain method requiring only two cells per wavelength. Microwave Opt. Techno. Lett. 15, 158–165. Liu, Q. H. (1999). Large-scale simulations of electromagnetic and acoustic measurements using the pseudospectral time-domain (PSTD) algorithm. IEEE Trans. Geosci. Remote Sens. 37, 917–926. Lynch, D. R., and Paulsen, K. D. (1990). Time-domain integration of the Maxwell equations on finite elements. IEEE Trans. Antennas Propagat. 38, 1933–1942. Lynch, D. R., and Paulsen, K. D. (1991). Origin of vector parasites in numerical Maxwell solutions. IEEE Trans. Microwave Theo. Techniq. 39, 383–394. Mohammadian, A. H., Shankar, V., and Hall, W. F. (1991). Computation of electromagnetic scattering and radiation using a time-domain finite-volume discretization procedure. Comput. Phys. Comm. 68, 175–196. Monk, P., and Suli, E. (1994). Error estimates of Yee’s method on non-uniform grids. IEEE Trans. Magnet. 30, 393–412. Nachman, A. (1996). A brief perspective on computational electromagnetics. J. Comput. Phys. 126, 237–239. Nedelec, J. C. (1980). Mixed finite elements in R3. Numer. Math. 35, 315–341. Nedelec, J. C. (1986). A new family of mixed finite elements in R3. Numer. Math. 50, 57–81. Nordstro¨m, J., and Gustafsson, R. (2003). High-order finite difference approximations of electromagnetic wave propagation close to material discontinuities. J. Sci. Comput. 18, 215–234. Olsson, P. (1995a). Summations by parts, projections, and stability I. Math. Comp. 64, 1035–1065. Olsson, P. (1995b). Summations by parts, projections, and stability II. Math. Comp. 64, 1473–1493. Paulsen, K. D., Boyse, W. E., and Lynch, D. R. (1992). Continuous potential Maxwell solutions on nodal-based finite elements. IEEE Trans. Antennas Propagat. 40, 1192–1200. Paulsen, K. D., and Lynch, D. R. (1991). Elimination of vector parasites in finite element Maxwell solutions. IEEE Trans. Microwave Theo. Techniq. 39, 395–404. Petropoulous, P. G. (1994). Phase error control for FD-TD methods of second and fourth order accuracy. IEEE Trans. Anten. Prop. 42, 859–862. Petropoulos, P. G., Zhao, L., and Cangellaris, A. C. (1998). A reflectionless sponge layer absorbing boundary condition for the solution of Maxwell’s equations with high-order staggered finite difference schemes. J. Comput. Phys. 139, 184–208. Petropoulos, P. G., and Yefet, A. (2001). Subgridding a fourth-order FD-TD scheme for Maxwell’s equations. In ‘‘Proceedings of Fourth International Workshop on Computational Electromagnetics in the Time-Domain: TLM/FDTD and Related Techniques (CEM-TD).’’ (C. Christopoulos, eds.), pp. 39–45, Nottingham, UK. Piperno, S., and Fezoui, L. (2003). ‘‘A discontinuous Galerkin FVTD method for 3D Maxwell Equations.’’ CERMICS research report, pp. 2003–4733. Proriol, J. (1957). Sur une famille de polynomes a´ deux variables orthogonaux dans un triangle. C. R. Acad. Sci. Paris 257, 2459–2461.

122

HESTHAVEN

Rao, K. R., Nehrbass, J., and Lee, R. (1999). Discretization errors in finite methods: Issues and possible solutions. Comput. Methods Appl. Mech. Engrg. 169, 219–236. Ryaben’kii, V. S., Tsynkov, S. V., and Turchaninov, V. I. (2001). Global discrete artificial boundary condition for time-dependent wave propagation. J. Comput. Phys. 174, 712–758. Shang, J. S. (1999). High-order compact difference schemes for time-dependent Maxwell equations. J. Comput. Phys. 153, 312–333. Shang, J. S., and Fithen, R. M. (1996). A comparative study of characteristic based algorithms for the Maxwell’s equations. J. Comput. Phys. 125, 378–394. Shang, J. S., and Gaitonde, D. (1995). Characteristic-based time-dependent Maxwell equation solvers on a general curvilinear frame. AIAA J. 33, 491–498. Shang, J. S., and Gaitonde, D. (1996). On high-resolution schemes for time-dependent Maxwell equations. AIAA Paper 96–0832. Shang, J. S., Gaitonde, D., and Wurtzler, K. (1996). Scattering simulations of computational electromagnetics. AIAA Paper 96–2337. Silvester, P. (1969). Finite element solution of homogeneous waveguide problems. 38, 313–317. Silvester, P. P., and Ferrari, R. L. (1983). ‘‘Finite Elements for Electrical Engineers.’’ Cambridge University Press, Cambridge. Strand, B. (1994). Summation by parts for finite difference approximations of d/dx. J. Comput. Phys. 110, 47–67. Strang, G. (1964). Accurate partial difference methods. II. Non-linear problems. Numer. Math. 6, 37–46. Sun, D., Manges, J., Yuan, X., and Cendes, Z. (1995). Spurious modes in finite element methods. IEEE Antennas Propagat. Mag. 37, 12–24. Taflove, A. (1995). ‘‘Computational Electrodynamics: The Finite-Difference Time-Domain Method.’’ Aztech House, Boston. Taflove, A. (ed.) (1998). ‘‘Advances in Computational Electrodynamics: The Finite-Difference Time-Domain Method.’’ Aztech House, Boston. Tam, C. K. W., and Webb, J. C. (1993). Dispersion-relation-preserving finite difference schemes for computational acoustics. J. Comput. Phys. 107, 262–281. Tornberg, A. K., and Engquist, B. (2002). Regularization techniques for numerical approximation of PDEs with singularities. UCLA Computational and Applied Mathematics Report 02–37. Turkel, E., and Yefet, A. (2000). On the construction of a high-order difference scheme for complex domains in a Cartesian grid. Appl. Numer. Math. 33, 113–124. Volakis, J. L., Chatterjee, A., and Kempel, L. (1998). Finite element methods for electromagnetics: Antennas, microwave circuits and scattering applications IEEE Press. Wang, S., Teixeira, F. L., Lee, R., and Lee, J.-F. (2002a). Dispersion-relation-preserving (DRP) 2-D finite difference time-domain schemes. Antennas Propagat. Soc. Intern. Symp. 3, 264–267. Wang, S., Teixeira, F. L., Lee, R., and Lee, J.-F. (2002b). Dispersion-relation-preserving FDTD algorithms for large-scale three-dimensional problems. IRRR Trans. Antennas Propagat. Warburton, T. (2000). Application of the discontinuous Galerkin method to Maxwell’s equations using unstructured polymorphic hp-finite elements. In ‘‘Discontinuous Galerkin Methods: Theory, Computation and Applications’’ (B. Cockburn, G. E. Karniadakis, and C. W. Shu, eds.), pp. 451–458. Springer Verlag, New York. Wasberg, C. E., and Gottlieb, D. (2000). Optimal strategy in domain decomposition spectral methods for wave-like phenomena. SIAM J. Sci. Comput. 22, 617–632. Webb, J. P. (1999). Hierarchal vector basis functions of arbitrary order for triangular and tetrahedral finite elements. IEEE Trans. Antennas Propagat. 47, 1244–1253.

HIGH-ORDER ACCURATE METHODS IN TIME-DOMAIN CEM

123

Xie, Z., Chan, C.-H., and Zhang, B. (2002). An explicit fourth-order orthogonal curvilinear staggered-grid FDTD method for Maxwell’s equations. J. Comput. Phys. 175, 739–763. Yang, B., and Gottlieb, D. (1996). Comparisons of staggered and non-staggered schemes for Maxwell’s equations. In ‘‘Proceedings of the 12th Annual Review of Progress in Applied Computational Electromagnetics,’’ Vol. II, pp. 1122–1131. Yang, B., Gottlieb, D., and Hesthaven, J. S. (1997a). On the use of PML ABC’s in spectral time-domain simulations of electromagnetic scattering. In Proceedings of the 13th Annual Review of Progress in Applied Computational Electromagnetics, pp. 926–933. Yang, B., Gottlieb, D., and Hesthaven, J. S. (1997b). Spectral simulations of electromagnetic wave scattering. J. Comput. Phys. 134, 216–230. Yang, B., and Hesthaven, J. S. (1999). A pseudospectral method for time-domain computation of electromagnetic scattering by bodies of revolution. IEEE Trans. Antennas Propagat. 47, 132–141. Yang, B., and Hesthaven, J. S. (2000). Multidomain pseudospectral computation of Maxwell’s equations in 3-D general curvilinear coordinates. Appl. Numer. Math. 33, 281–289. Yee, K. S. (1966). Numerical solution of initial boundary value problems involving Maxwell’s in isotropic media. IEEE Trans. Antennas Propagat. 14, 302–307. Yee, K. S., Chen, J. S., and Chang, A. H. (1992). Conformal finite-difference time domain (FDTD) with overlapping grids. IEEE Trans. Antennas Propagat. 40, 1068–1075. Yefet, A., and Petropoulos, P. G. (2001). A staggered fourth-order accurate explicit finite difference scheme for the time-domain Maxwell’s equations. J. Comput. Phys. 168, 286–315. Yefet, A., and Turkel, E. (2000). Fourth-order compact implicit method for the Maxwell equations with discontinuous coefficients. Appl. Numer. Math. 33, 125–134. Zhang, C., and Symes, W. W. (1998a). Fourth-order methods for acoustic waves with discontinuous material. In ‘‘Proc. of 4th International Conference on Mathematical and Numerical Aspects of Wave Propagation’’ (J. de Santos, et al. eds.). Philadelphia. Zhang, C., and Symes, W. W. (1998b). Fourth order, full stencil immersed interface method for elastic waves with discontinuous material. In ‘‘Proc. of 4th International Conference on Mathematical and Numerical Aspects of Wave Propagation’’ (J. de Santos, et al., eds.) Philadelphia. Zhao, G., and Liu, Q. H. (2002). The 2.5-D multidomain pseudospectral time-domain algorithm. Antennas Propagat. Soc. Intern. Symp. 3, 236–238. Zhao, G., Zeng, Y. Q., and Liu, Q. H. (2003). The 3-D multidomain pseudospectral timedomain method. IEEE Microwave Wireless Comp. Lett. 13, 184–186.

This Page Intentionally Left Blank

ADVANCES IN IMAGING AND ELECTRON PHYSICS, VOL. 127

Prefiltering for Pattern Recognition Using Wavelet Transform and Neural Networks FAN YANG AND MICHEL PAINDAVOINE Laboratoire Electronique, Informatique et Image—CNRS FRE 2309, Aile des Sciences de l’Inge´nieur, Universite´ de Bourgogne, BP 47870-21078 Dijon Cedex, France

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Introduction to Neural Networks . . . . . . . . . . . . . . . . A. Perceptron and Adaline Networks . . . . . . . . . . . . . . . 1. The Perceptron . . . . . . . . . . . . . . . . . . . . 2. The Adaline Network . . . . . . . . . . . . . . . . . 3. An Example of Pattern Recognition . . . . . . . . . . . . B. Linear Associative Memories. . . . . . . . . . . . . . . . . 1. Principle and Architecture of Linear Autoassociative Memories . . 2. Linear Autoassociative Memories Training with Widrow–Hoff Learning Rule . . . . . . . . . . . . . . . . . . . . 3. Linear Heteroassociative Memories . . . . . . . . . . . . C. Multilayer Neural Networks and Back-Propagation Learning Rule . . 1. Architecture and Notation of Multilayer Feed-Forward Networks . 2. The Generalized Widrow–Hoff Rule . . . . . . . . . . . . D. Radial Basis Function Networks . . . . . . . . . . . . . . . 1. Exact Interpolation . . . . . . . . . . . . . . . . . . 2. Radial Basis Function Networks . . . . . . . . . . . . . 3. RBF Neural Network Training . . . . . . . . . . . . . . III. Introduction to Wavelet Transform . . . . . . . . . . . . . . . A. Continuous and Discrete Wavelet Transform . . . . . . . . . . 1. Definition of Continuous Wavelet Transform . . . . . . . . . 2. Discrete Wavelet Transform . . . . . . . . . . . . . . . 3. Wavelet Packet Transform . . . . . . . . . . . . . . . . B. Wavelet Functions. . . . . . . . . . . . . . . . . . . . . 1. Haar Wavelet Function . . . . . . . . . . . . . . . . . 2. Daubechies Wavelet Function and Other Wavelet Functions . . . 3. Wavelet Analysis: Parallelism and Applications . . . . . . . . IV. Pattern Recognition Using Wavelet and Neural Network for Signal and Image Processing Applications . . . . . . . . . . . . . . . . . A. Different Techniques of Preprocessing for Pattern Recognition . . . . 1. Principal Component Analysis and Fourier Descriptors . . . . . 2. Wavelet Transform and Feature Extraction . . . . . . . . . 3. Wavelet Packet Transform (WPT) . . . . . . . . . . . . . 4. Feature Extraction Using Invariance Properties . . . . . . . . B. Audio, Underwater, and Eddy Current Signal Processing Applications . 1. Audio Signals Identification . . . . . . . . . . . . . . . 2. Underwater Signals Classification . . . . . . . . . . . . . 3. Neural Network Classifier for Eddy Current Signals . . . . . .

126

. . . . . . .

. . . . . . .

. . . . . . .

127 128 128 130 133 136 136

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

138 138 139 139 141 143 143 144 145 147 147 147 149 152 152 153 154 156

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

158 159 159 161 161 162 164 164 166 168

125 Copyright 2003, Elsevier (USA). All rights reserved. ISSN 1076-5670/03

126

YANG AND PAINDAVOINE

C. Classification of Electroencephalogram and Myoelectric Signals Using Wavelet Transform and Neural Networks . . . . . . . . . . . . 1. Sleep Stage Classification . . . . . . . . . . . . . . . . . . 2. Brain Disorder Classification . . . . . . . . . . . . . . . . 3. Classification of the Myoelectric Signal Using the Wavelet Packet Transform . . . . . . . . . . . . . . . . . . . . . D. Shape Recognition Applications and Image Classification by Content . . 1. Recognition and Classification of Blemishes . . . . . . . . . . . 2. A New Multiscale-Based Shape Recognition Method Applied to Images of Industrial Tools . . . . . . . . . . . . . . . . . . . . 3. Wavelet Index of Texture for Artificial Neural Network Classification of Landsat Images . . . . . . . . . . . . . . . . . . . . V. Wavelet Prefiltering Technique Applied to Handwriting Movement Analysis and Face Recognition . . . . . . . . . . . . . . . . . . . . . A. Filtering Techniques, Time–Frequency Representations, and Generalized Canny Deriche Filter . . . . . . . . . . . . . . . . . . . . 1. Filtering Techniques and Time–Frequency Representations . . . . . 2. Wavelet Transform and Generalized Canny Deriche Filter . . . . . B. Wavelet Filtering Technique and Movement Analysis in Handwriting and Drawing . . . . . . . . . . . . . . . . . . . . . . . 1. Handwriting and Drawing Signals . . . . . . . . . . . . . . 2. Filtering and Signal Segmentation Using Multiresolution Analysis . . 3. Discussion. . . . . . . . . . . . . . . . . . . . . . . . C. An Image Filtering Technique Combining a Wavelet Transform with an Autoassociative Memory: Application to Face Recognition . . . . 1. Linear Autoassociators and Eigenvalue Decomposition . . . . . . 2. Preprocessing Using Multiscale Edges . . . . . . . . . . . . . 3. Pattern Completion of Noisy Patterns . . . . . . . . . . . . . Appendix . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

170 170 171

. . . . . .

173 174 174

. .

174

. .

175

. .

177

. . . . . .

177 177 178

. . . .

. . . .

180 181 182 187

. . . . . .

. . . . . .

188 189 192 197 199 201

I. Introduction Pattern recognition, regarding engineering, would imply the study of automatic or semiautomatic systems capable of recognizing patterns. This concerns several disciplines, such as statistics, operational research, computer science, biology, and electronics. This chapter limits itself to signal and image processing where neural networks are the main tools in realization of such a system. Neural networks are built from simple units interlinked by a set of weighted connections. Generally, these units are organized in layers. Each unit of the first layer (input layer) corresponds to a feature of a pattern that we want to analyze. The units of the last layer (output layer) produce a decision after the propagation of information.

PREFILTERING FOR PATTERN RECOGNITION

127

In general, before feeding the computational data to neural networks, the signal must undergo a preprocessing in order to (1) define the initial transformation to represent the measured signal, (2) retain important features for class discrimination and discard that which is irrelevant, and (3) reduce the volume of data to be processed, e.g., data compression. This stage of preprocessing can be realized using many techniques: principal component analysis (PCA), Fourier transform, and other algorithms that allow the selection of the best parameters. These techniques either use or do not use knowledge from the analyzed signal. Often, the best representation remains the wavelet transform for nonstationary signals such as seismic tremors, human speech, engine vibrations, medical images, financial data, music, and many other types of signals. Pattern recognition using wavelet transform and neural networks is the theme discussed in this chapter. Sections II and III present, respectively, neural networks and wavelet transform. These correspond to the theory part of the manuscript. We describe basic concepts and give small examples to illustrate some basic models. These two sections provide only an introduction to neural networks and wavelet transform. During our arguments, we take the point of view of a computer scientist. In signal processing, as well as in other disciplines, experimentation and apprenticeship in conjunction with reality are the only efficient mean for comprehension of the concept. Sections IV and V focus on applications of pattern recognition using wavelet transform and neural networks. We will first discuss audio, underwater, and eddy current signals identification and classification in Section IV. Then we will analyze medical signal processing applications concerning the classification of EEG and myoelectric signals. One of these applications consists of classification of sleep stages. After that, we illustrate applications such as shape recognition and image classification by content. In Section V, wavelet transform is used as a prefiltering technique for movement analysis in handwriting and face recognition. Through this presentation, detailed with theory and complemented by many illustrated applications, it is hoped that the scientific population finds more interest in pattern recognition using wavelet transform and neural networks. II. Introduction to Neural Networks This section illustrates some basic neural network models. First we introduce single layer neural networks, including some of the classical approaches to the neural computing and learning problem. Two classical models are described: the perceptron, proposed by Rosenblatt in the late 1950s, and the adaline, presented in the early 1960s by Widrow and Hoff. These models are

128

YANG AND PAINDAVOINE

very easy to analyze and allow a straightforward introduction of important concepts, such as the Widrow–Hoff learning rule, generalization, and linear separability. Then, we give an example of pattern recognition using the single layer neural network. Section II.B focuses on linear autoassociative memory and linear heteroassociative memory, which is a generalization of the perceptron and corresponds to multiple linear regression. A single layer linear network has severe restrictions because it can only solve linear separable problems. Minsky and Papert (1969) showed that twolayer feed-forward networks can overcome many restrictions, but did not present a solution to the problem of how to adjust the weights from input to hidden units. This problem was probably a major cause of the lack of interest in neural networks at the end of the sixties. In 1985, Parker presented an answer to this question, as did Rumelhart and colleagues in 1986. The main idea behind the solution is that the error for the units of the hidden layer is determined using back-propagation of the error of the output layer units. These multilayer neural networks and the back-propagation learning rule are presented in Section II.C. Another nonlinear neural network model, radial basis function (RBF), is also described. Note that the area of neural networks is very vast. The presentation in this section is relatively general. We limit our description to only basic models in order to favor the comprehension of pattern recognition applications using the wavelet transform (WT) and neural networks given in Sections IV and V. Readers are invited to refer to Fausett (1994), Gurney (1997), Haykin (1999), Ripley (1996), Rojas (1996), Abdi et al. (1999), Bishop (2000), and Wermter and Sun (2000) for more complete information.

A. Perceptron and Adaline Networks 1. The Perceptron Constructed by Rosenblatt (1959) in the 1950s, the perceptron can be considered as the first neural network capable of learning. As its name indicates, the perceptron was intended as a model of perceptual activity. The aim of the perceptron was to associate input patterns with responses. The main idea behind the design of the perceptron was that an object is first registered by the cells of the retina (input neurons) and is then recognized by cells located in the brain (output neurons). Figure 1 shows a biological neuron and an artificial neuron of the perceptron. a. Building Block Networks with Threshold Activation Functions. A single layer feed-forward network consists of one or more output neurons j, each of which is connected with a weighting factor wij to all of the input i. In the

129

PREFILTERING FOR PATTERN RECOGNITION

Figure 1. Schematic of a biological neuron (left) (cf. http://vv.carleton.ca/neil/neural/ neuron.html) and an artificial neuron (right).

Bias neuron x0 = 1 x1

w1

w0 = −q Response o(a) a = Σ wi xi i

x2

w2

Thresholding

Output response 0 or 1

Computation of the activation

Figure 2. A basic threshold unit computes its activation a as the weighted sum of its inputs. The output o(a) is equal to 0 when a  0 and to 1 when a > 0.

simplest case, the network has only two inputs and a single output, as sketched in Figure 2. The input of the neuron is the weighted sum of the inputs. The output of the network is formed by the activation of the output neuron, which is a function of the inputs. a¼

2 X i¼1

w i xi

ð1Þ

The activation function can be linear so that we have a linear or nonlinear network. Here, we consider the threshold (or sign) function.  1 a> ð2Þ response ¼ oðaÞ ¼ 0 a The threshold can be seen as a weight. In this case, thresholding is implemented by keeping one input neuron always active (the 0th input

130

YANG AND PAINDAVOINE

neuron x0 ¼ 1) such that weight w0 is equal to . This is equivalent to rewriting Eq. 1 (2) as 8 2 X > > > w i xi > 0 1 a ¼ > < i¼0 ð3Þ oðaÞ ¼ 2 X > > > > a¼ wi xi  0: :0 i¼0

Such a simple device can now decide whether an input pattern belongs to one of two classes. The separation between the two classes in this case is a straight line given by the equation w1 x1 þ w2 x2 þ w0 ¼ 0:

ð4Þ

The single layer network represents a linear discriminant function. Now we show how perceptrons can use a systematic procedure to find (learn) the value of the weights appropriate for implementing specific association between inputs and output. b. Perceptron Learning Rule. Neural network learning consists of an adjustment of the weights of the connections between units according to some modification rule. Virtually all learning rules can be considered as a variant of the Hebbian learning rule (Hebb, 1949). The basic idea is that if two units, i and j, are active simultaneously, their interconnection must be strengthened. If j receives input from i, the simplest version of Hebbian learning prescribes a modification of the weight, wij, with
wij ¼ ai aj ,

ð5Þ

where  is a positive constant of proportionality representing the learning rate. The perceptron learning rule is very simple and can be stated as follows. 1. Start with random weights for the connections. 2. Select an input vector x from the set of training samples. 3. If o <> t(x) (the perceptron gives an incorrect response), modify all connections wi according to
wi ¼ t(x)xi. 4. Go back to step 2. 2. The Adaline Network a. Adaptive Linear Element. A model called Adaline was developed independently in a signal processing framework by Widrow and Hoff (1960) around the same time as perceptron. In a simple physical implementation (see Figure 3), this device comprises a set of controllable resistors connected to a circuit that can sum up currents caused by the input voltage signals. Although

131

PREFILTERING FOR PATTERN RECOGNITION

+1

level

w1j x1

w0j w2j

Σ

x2

oj

wIj xI Figure 3. Adaline neural network architecture.

here the adaptive process is exemplified in a case when there is only one output, it may be clear that a system with many parallel outputs is directly implementable by multiple units of the same kind. In the case of Adaline, the activation function is linear and the output signal is continuous oj ¼

I X

wij xi ,

i¼0

ð6Þ

where oj is the output signal of jth unit, j ¼ 1, . . ., J; xi is the input signal, i ¼ 0, 1, . . ., I; and wij is the input conductances. The purpose of this device is to yield a given value, oj, at its output when the set of values, xi, is applied at the input. The problem is to determine the coefficients wij in such a way that the input–output is correct for a large number of arbitrarily chosen signal sets. b. Widrow–Hoff Learning Rule. For the Adaline neural network, Widrow and Hoff (1960) introduced the delta rule in order to adjust the weights
wij ¼ ðtj  oj Þxi ,

ð7Þ

where  is a constant of proportionality and tj is the target (desired) response of the jth unit. Suppose we want to train the network such that a set of training samples consisting of the input value x is associated with a target output value t. For every given input sample, the output of the network differs from the target value t by t  o, where o is the actual output for this pattern. The delta rule uses an error function based on these differences in order to adjust the weights.

132

YANG AND PAINDAVOINE

The error function is the summed square error (the Widrow–Hoff rule is also called the least mean square learning procedure). The total error Etotal is defined as K X Ek k ¼ 1, 2, . . . , K, Etotal ¼ k¼1

J 1X Ek ¼ ðtj  oj Þ2 2 j¼1

j ¼ 1, 2, . . . , J,

ð8Þ

where Ek represents the error on pattern k. The least mean square procedure finds the values of all the weights that minimize the error function by a method called gradient descent. The idea is to make a change in the weight proportional to the negative of the derivative of the error as measured on the current pattern with respect to each weight @Ek : @wij

ð9Þ

@Ek @Ek @oj ¼ @wij @oj @wij

ð10Þ


wij ¼  The derivative is

Because of the linear units [Eq. (6)] @oj ¼ xi @wij

ð11Þ

@Ek ¼ ðtj  oj Þ @oj

ð12Þ

and

such that
wij ¼ (tj  oj)xi. Learning is achieved by adding, usually iteratively, to the present weights a small quantity,
wij. The Widrow–Hoff learning rule specifies the correction to apply at time, t, to the weight connecting the ith input unit to the jth output unit as ðt þ 1Þ

wij

ðtÞ

¼ wij þ ðtj  oj Þxi

ð13Þ

where t is the iteration number (i.e., this is iteration t), wij(t) is the weight of the connection at iteration t (the initial value wij(0) is generally chosen randomly), and  is a positive constant, generally between 0 and 1. It corresponds to the small positive constant to be added or subtracted to the weights of connection. Eq. (13) is very general and can be extended easily to more complex neural network architectures.

PREFILTERING FOR PATTERN RECOGNITION

133

3. An Example of Pattern Recognition We want to recognize the follow patterns 0 to 9 (Fig. 4). The purpose is to associate each pattern with its class corresponding to

?

Class = 0, 1, ..., 9

x : 5 rows * 3 columns = 15 input units

We initiate the matrix of weights with values chosen randomly 2

3 6 84 6 6 65 6 6 87 6 6 17 6 6 92 6 6 12 6 W¼6 6 75 6 55 6 6 35 6 6 69 6 6 13 6 6 21 6 4 58 49

65 22 70 74 7 95 83 15 51 47 44 43 14 85 24

38 7 90 45 27 97 25 37 31 74 35 52 96 75 58 36 52 52 15 77 14 28 24 92 32 54 41 88 54 68 73 73 50 61 98 50 75 8 53 5 70 51 4 5 81

3 66 41 55 1 75 7 53 29 54 56 7 7 84 4 20 21 57 7 7 16 22 29 60 66 7 7 92 48 18 56 52 7 7 18 92 92 36 89 7 7 50 1 53 5 19 7 7 68 5 40 81 20 7 7: 39 78 93 30 62 7 7 94 56 14 35 56 7 7 7 59 72 41 61 96 7 7 14 51 73 79 9 7 98 31 21 52 60 7 7 89 23 98 5 56 5 19 31 72 56 42

The number patterns 0, 1, . . ., 9 are presented to input units, and the responses of output units are calculated with ! 14 X wij xi : ð14Þ oj ¼ f ðaj Þ ¼ f i¼0

For the pattern of number 0, we calculate o0, o1, and o5: a0 ¼ 3  84 þ 65 þ 87 þ 92  12 þ 55 þ 35 þ 13  21 þ 58 þ 49 ¼ 340 o0 ¼ 1 because a0 > 0 a1 ¼ 65  22  70 þ 74  95 þ 83  51 þ 47 þ 43  14  85  24 ¼  179

134 YANG AND PAINDAVOINE

Figure 4. Number patterns 0–9.

135

PREFILTERING FOR PATTERN RECOGNITION

o1 ¼ 0 because a1  0 a5 ¼  66 þ 7 þ 84 þ 16 þ 18  50  39 þ 94  14 þ 98 þ 89 þ 19 ¼ 256 o5 ¼ 1 because a5 > 0 In the same manner, we obtain the matrix of responses as follows for patterns 0–9. 1 0 0 0 0 O¼ 1 1 0 1 1 Pattern 0

1 0 1 0 0 1 1 1 1 1 1

1 0 0 0 0 1 1 0 1 1 2

1 0 0 0 0 1 1 1 1 1 3

1 0 0 1 0 0 1 0 0 0 4

1 0 0 0 0 1 1 0 1 1 5

1 0 0 0 0 1 1 0 1 1 6

1 0 0 0 1 1 1 0 1 1 7

1 0 0 0 0 1 1 1 1 1 8

1 0 0 0 0 1 1 1 1 1 9

The total error, E, of incorrect classification is error ¼ 54%. The Widrow– Hoff learning rule is applied to modify the weights ðt þ 1Þ

wij

ðtÞ

¼ wij þ ðtj  oj Þxi :

with  ¼ 0.9, we obtain ð1Þ

ð0Þ

w00 ¼ w00 þ 0:9ð1  1Þ  1 ¼ 3 ð1Þ ð0Þ w01 ¼ w01 þ 0:9ð0  0Þ  1 ¼ 65 ð1Þ ð0Þ w05 ¼ w05 þ 0:9ð0  1Þ  1 ¼ 66:9: After 51 iterations, a new matrix of weights is obtained (see Figure 5), and the matrix of responses, O, is shown as follows. Figure 6 gives some results of pattern recognition for number 8. 2

1 60 6 60 6 60 6 60 O¼6 60 6 60 6 60 6 40 0

0 1 0 0 0 0 0 0 0 0

0 0 1 0 0 0 0 0 0 0

0 0 0 1 0 0 0 0 0 0

0 0 0 0 1 0 0 0 0 0

0 0 0 0 0 1 0 0 0 0

0 0 0 0 0 0 1 0 0 0

0 0 0 0 0 0 0 1 0 0

0 0 0 0 0 0 0 0 1 0

3 0 07 7 07 7 07 7 07 7 07 7 07 7 07 7 05 1

136

YANG AND PAINDAVOINE

Error 54%

18% 10

10% 20

7% 30

40

0% Iterations 50

Figure 5. Convergence of the network.

network

o = 0000000010

network

o = 0000000010

network

o = 0000001010

Figure 6. Decision of the network with some samples of number pattern recognition: a perfect pattern of number 8 is categorized correctly (top), a nearly perfect pattern of number 8 is categorized correctly (middle), and an imperfect pattern of number 8 is not categorized correctly (bottom).

B. Linear Associative Memories The class of models presented in this section are known as linear associators (Abdi et al., 1999; Valentin et al., 1994). They come in two forms: heteroand autoassociators. The heteroassociator can be used to learn arbitrary associations between input and output patterns. The autoassociator is a special case of the heteroassociator in which the association between an input pattern and itself is learned. 1. Principle and Architecture of Linear Autoassociative Memories Autoassociative memories are single-layer networks made of fully interconnected linear units (the output is a linear function of the activation).

PREFILTERING FOR PATTERN RECOGNITION

137

The goal of an autoassociative memory is to store a set of stimuli and to retrieve the stored stimuli when cued with partial or degraded versions of these stimuli. In contrast to a classical computer, which uses the exact memory address to retrieve stored information, an autoassociative memory is content addressable and is able to retrieve a whole pattern of information given one part of this information. Because of this property, autoassociative memories are widely used in many pattern recognition applications and for modeling human perceptual learning and memory. Linear autoassociative memories are built from fully interconnected linear units. A linear unit computes its activation as the sum of weights of its inputs. A linear transfer function transforms its activation into its output. If a unit has I inputs noted x1, . . . , xi, . . . , xI, each of them connected with a weight denoted wi, the activation of the unit will be a¼

I X

wi xi :

i¼1

ð15Þ

The output of the unit will be a with being a constant. Usually, for convenience, is set to 1. Using vector notation, if x is the input vector and w the weight vector, the output of the unit is response ¼ wT x

ð16Þ

Formally, an autoassociative memory is a network of I linear units fully interconnected by modifiable connections (see Figure 7). The set of connections is represented by a square symmetrical matrix W, 2 3 w1, 1 w1, 2 . . . w1, I 6 w2, 1 w2, 2 . . . w2, I 7 7: W¼6 ð17Þ 4: 5 : ... : wI , 1 wI , 2 . . . wI , I

Figure 7. An autoassociative memory composed of five fully interconnected units.

138

YANG AND PAINDAVOINE

The wi,i0 represents the strength of the connection between units i and i0. Each K pattern stored in the memory is represented by an I – dimensional vector denoted xk in which a given element, xi,k, corresponds to a feature describing the pattern k. Each element in these vectors is used as input to a unit of the autoassociative memory. The set of K input patterns is represented by an I  K matrix X whose kth column is xk. 2. Linear Autoassociative Memories Training with Widrow–Hoff Learning Rule Training results from modification of the value of the connections following the presentation of a set of patterns. The Widrow–Hoff learning rule seen previously iteratively adjusts the weight of the connection matrix in order to minimize the difference of the output of the target response. Eq. 13 describing Widrow–Hoff learning can be rewritten using matrix notation to specify the correction to apply to the complete set of weights after one iteration (after presentation of a given pattern k) (Valentin et al., 1994) WTðtþ1Þ ¼ WTðtÞ þ ðtk  ok ÞxTk ,

ð18Þ

where t represents the iteration number, xk is the I  1 kth input vector, tk is the J  1 kth target vector, ok is the J  1 kth response vector, and W is the I  J of weight of connection. In the case of autoassociative memories, we have that t k ¼ xk (autoassociative memories) and W is a square symmetrical matrix, WT ¼ W:

ð19Þ

Wðtþ1Þ ¼ WðtÞ þ ðxk  ok ÞxTk ¼ WðtÞ þ ðxk  WTðtÞ xk ÞxTk ;

ð20Þ

Therefore,

where k is a random integer (1  k  K ). This algorithm can be expressed in a more practical way by using I  K stimulus matrix X. In this case, the difference xk  ok, k ¼ 1, 2, . . . , K is computed for the complete set of patterns before implementing the correction Wðtþ1Þ ¼ WðtÞ þ ðX  WðtÞ XÞXT :

ð21Þ

3. Linear Heteroassociative Memories We have seen that the goal of autoassociative memories is to associate an input pattern to itself such that when part of the input is presented as a

PREFILTERING FOR PATTERN RECOGNITION

139

memory key, the memory gives back the complete pattern. Here, we introduce a more general family of models called heteroassociative memories. They differ from autoassociative memories in that an input pattern is associated to an output pattern instead of being associated to itself. The goal of a heteroassocaitive memory is to learn mapping between input– output pairs such that the memory produces the appropriate output in response to a given pattern. Heteroassociative memories are generally used to solve pattern identification and categorization problems. We rewrite the Widrow–Hoff learning rule for heteroassociative memories using matrix notation [see Eq. (21)] WTðtþ1Þ ¼ WTðtÞ þ ðT  WTðtÞ XÞXT :

ð22Þ

where T is the J  K matrix of target responses. C. Multilayer Neural Networks and Back-Propagation Learning Rule The basic models of single-layer linear networks presented previously have severe restrictions because they can only solve linear separable problems. Minsky and Papert (1969) showed that two-layer feed-forward networks can overcome many restrictions but did not present a solution to the problem of how to adjust the weights from input to hidden units. An answer to this question was presented by Rumelhart and colleagues in 1986. The main idea behind this solution is that the error for the hidden layer units is determined by back-propagation of the error of the output layer units. Backpropagation can be considered as a generalization of the Widrow–Hoff rule for nonlinear activation functions and multilayer networks. It adjusts the weights in any feed-forward network trained to associate pairs of patterns. 1. Architecture and Notation of Multilayer Feed-Forward Networks Feed-forward networks are made of several layers of nonlinear units. Each unit in a layer is connected to all the units of the preceding and following layers: these units receive their input from units from a layer directly below and send their output to units in a layer directly above the unit. There are no connections within a layer. The inputs are fed into the first hidden layer. The output of the hidden units is distributed over the next hidden layer, up until the last hidden layer, where the outputs are fed into a layer of output units (see Figure 8). Each nonlinear unit computes its activation by adding all the weighted inputs it receives. It transforms this activation into a response using a

140

YANG AND PAINDAVOINE

Hidden layers

Output layer

Input layer

Nonlinear function Figure 8. A multilayer network with several layers of nonlinear units: each layer may have a different number of units and a different transfer function.

1

0.25

0.9 0.8

0.2

Logistic derivative

Logistic function

0.7 0.6 0.5 0.4

0.15

0.1

0.3 0.05

0.2 0.1 0 −10

−5

0 x

5

10

0 −10

−5

0 x

5

10

Figure 9. Sigmoid (logistic) transfer function and her derivative—this function maps the set of real numbers into the [0, 1] domain.

nonlinear transfer function. Several transfer functions can be used. The most popular one is the sigmoid (logistic) function (see Figure 9) f ðaÞ ¼

1 : 1 þ ea

ð23Þ

This function has the interesting property of having an easy to compute derivative.

PREFILTERING FOR PATTERN RECOGNITION

f 0 ðaÞ ¼ f ðaÞ½1  f ðaÞ:

141 ð24Þ

Although back-propagation can be applied to networks with any number of layers, it has been shown that one layer of hidden units suffices to approximate any function. Therefore, in most applications, a feed-forward network with a single layer of hidden units is used with a sigmoid activation function for the units. 2. The Generalized Widrow–Hoff Rule We generalize the Widrow–Hoff (delta) learning rule to the set of nonlinear activation functions. The response of the jth output unit is a differentiable function of the total input xi, (i ¼ 0, 1, 2, . . . , I), given by oj ¼ f ðaj Þ j ¼ 1, 2, . . . , J I X aj ¼ wij xi :

ð25Þ

i¼0

To get the correct generalization of the Widrow–Hoff rule as presented in the previous sections, we must set
wij ¼ 

@E : @wij

ð26Þ

The error measure E is defined as the total quadratic error for the current pattern at the output units, E¼

J 1X ðtj  oj Þ2 , 2 j¼1

ð27Þ

where tj is the target response of the jth unit. We can write @E @E @aj ¼ : @wij @aj @wij

ð28Þ

By Eq. (25) we see that the second factor is @aj ¼ xi : @wij

ð29Þ

@E @aj

ð30Þ

When we define j ¼ 

142

YANG AND PAINDAVOINE

we will get an update rule, which is equivalent to the Widrow–Hoff rule, resulting in a gradient descent on the error surface if we make the weight changes according to
wij ¼ j xi :

ð31Þ

To compute j we apply the chain rule to write this partial derivative as the product of two factors, j ¼ 

@E @E @oj : ¼ @aj @oj @aj

ð32Þ

Let us compute the second factor. By Eq. (25) we see that @oj ¼ f 0 ðaj Þ, @aj

ð33Þ

which is simply the derivative of the function f for the jth unit. In order to compute the first factor of Eq. (32), we consider two cases. First, assume that unit j is an output unit of the network. In this case, it follows from the definition of E that @E ¼ ðtj  oj Þ: @oj

ð34Þ

Substituting this and Eq. (33) in Eq. (32), we get j ¼ ðtj  oj Þ f 0 ðaj Þ 8j:

ð35Þ

Second, if unit j is not an output unit (it is a hidden unit), we do not readily know the contribution of the unit to the output error of the network. However, the error measure can be written as a function of the inputs from hidden to output layer, and we use the chain rule to write J X @E @E @aj ¼ @hl @a j @hl j¼1

¼ ¼

J L X @E @ X zlj hl @aj @hl l ¼ 1 j¼1

J X @E j¼1

¼ 

@aj

J X j¼1

zlj

j zlj

,

ð36Þ

PREFILTERING FOR PATTERN RECOGNITION

143

where hl is the response of the lth hidden unit of network, l ¼ 1, 2, . . . , L, aj is the activation of the jth output unit of network, j ¼ 1, 2, . . . , J, and zlj is the weight between the lth hidden unit and the jth output unit of network. Substituting this in Eq. (32) yields l ¼ f 0 ðal Þ

J X

j zlj :

j¼1

ð37Þ

Eq. (35) and (37) give a recursive procedure for computing the  for all units in the network, which are then used to compute the weight changes according to Eq. (31). This procedure constitutes the generalized Widrow– Hoff rule for a feed-forward network of nonlinear units. D. Radial Basis Function Networks The network models discussed in the previous sections are based on units that compute a function of the scalar product of the input vector and a weight vector. This section considers the other major class of neural network models, in which the activation of a hidden unit is determined by the distance between the input vector and a prototype vector. 1. Exact Interpolation Radial basis function methods have their origins in techniques for performing the exact interpolation of a set of points in a multidimensional space (Powell, 1987). The exact interpolation problem requires that every input vector is mapped exactly onto the corresponding target vector, Consider a mapping from a d-dimensional input space x to a onedimensional target output space t. The data set consists of K input vectors, xk, together with corresponding targets tk. The goal is to find a function f(x) such that f ðxk Þ ¼ tk

k ¼ 1, 2, . . . , K:

ð38Þ

The radial basis function approach introduces a set of K basis functions, one for each data point, which take the form (|x  xk|), where  is some nonlinear function. Thus the kth such function depends on the distance |x  xk|, usually taken to be Euclidean, between x and xk. The output of the mapping is then taken to be a linear combination of the basis functions, f ðxÞ ¼

K X

k¼1

ðjx  xk jÞwk ,

where xk is the center of the basis function.

ð39Þ

144

YANG AND PAINDAVOINE

The interpolation conditions [Eq. (38)] can then be written in matrix form as

Provided the inverse matrix F

1

Fw ¼ t:

ð40Þ

exists, we can solve Eq. (40) to given w ¼ F1 t:

ð41Þ

When the weights in Eq. (39) are set to the values given by Eq. (41), the function f (x) represents a continuous differentiable surface, which passes through each data point. Several forms of basis function have been considered, the most common being the Gaussian   x2 ð42Þ ðxÞ ¼ exp  2 : 2 where  is a parameter whose value controls the smoothness properties of the interpolating function. The generalization to several output variables is straightforward. Each input vector xk must be mapped exactly onto an output vector tk. 2. Radial Basis Function Networks The exact interpolating function for noisy data is typically a highly oscillatory function. In this case, the interpolating function that gives the best generalization is one that is typically much smoother and gives averages over the noise on data. An additional limitation of the exact interpolation procedure is that the number of basis functions is equal to the number of patterns in the data set, and so for large data sets the mapping function can become very costly to evaluate. The architecture of a typical radial basis function network (Broomhead and Lowe, 1988; Moody and Darken, 1989) is shown in Figure 10. Modifications required to the exact interpolation procedure are as follows. 1. The number M of basis functions need not equal the number K of data points and is typically much less than K. 2. The centers of the basis functions are not constrained to be given by input data vectors. Instead, the determination of centers becomes part of the training process. In brief, the main idea of radial basis function networks is to replace the input vectors by a nonlinear transformation such that each input is represented by a vector coding its similarity to several centers and then to use a standard linear heteroassociative memory. RBF neural networks can be used in pattern recognition applications as kernel classifiers. They use overlapping simple functions in order to cover

145

PREFILTERING FOR PATTERN RECOGNITION

Hidden layer centers c1, c2, ..., cm Input layer

f

Output layer

f x

o f

f(x) = exp{− ||x − ci ||/ si2}

Heteroassociator

Figure 10. Architecture of a typical radial basis function network. The hidden layer computes the distance from the input to each of the centers (each center corresponds to a unit of the hidden layer). Units of the hidden layer transform their activation (i.e., the distance from the input) into a response using a nonlinear transformation (typically a Gaussian function). Units of the output layer behave like a standard linear heteroassociative memory.

complex decision regions that separate different categories of patterns (Musavi et al., 1992). The main advantages of these RBF classifiers are computational simplicity and robust generalization. 3. RBF Neural Network Training Radial basis function network training is processed following two phases. In the first phase, only the input data set xk is used to determine the parameters of the basis functions; number of centers, center impositions, and parameters, , are determined by unsupervised training techniques (i.e., methods that use only input data and not target data). In fact, the choice of centers and  determines in part the quality of the estimation given by the network. The set of centers is sometimes learned using learning techniques such as K – means (Moody and Darken, 1989; Musavi et al., 1992). The  of the  function can be approximated similarly from the samples. Here we present the algorithm inspired by a clustering method proposed by Musavi et al. (1992). Initially, we have K training (input) points in a d-dimensional space and each training point corresponds to a cluster (ck ¼ xk). 1. Take any point, ck, and its associated parameter, k (initially, k ¼ 0). 2. Find the nearest point, cl, of the same category by using the Euclidean distance. 3. Compute the mean of these two points. We obtain a new point with its associated parameter  ¼ kck2, cl k þ k .

146

YANG AND PAINDAVOINE

4. Compute the distance, D, from the new mean to the nearest point of all other categories. 5. If D > l, then accept the merge of ck and cl and start again from step 2. If the condition is not satisfied, reject the merge and recover the two original points and their parameter, , and then restart from step 1. 6. Repeat steps 1 to 5 until all points of each category are used.  is the ‘‘clustering parameter’’ with 1  l  3 (Musavi et al., 1992). Finally, we obtain the centers, cm, and their parameters, m (m ¼ 1, . . . , M  K ) of the hidden neurons of the RBF neural network. We can see (Figure 11) the result after using this clustering algorithm for two categories in a two-dimensional space. Note that we obtain two nonlinear decision regions. In fact, RBF can map spaces of any shape (e.g., nonlinear, convex, disjoint spaces). The basis functions are kept fixed while the final layer weights are found in the second phase of training, which requires the solution of a linear problem and is very fast. The procedures for training radial basis function networks can be substantially faster than the back-propagation algorithm. Because the parameters governing the basis functions are determined using relatively quick, unsupervised methods, the final-layer weights subsequently are found by fast linear supervise methods (see Section V.C.1).

CATEGORY A CATEGORY B Figure 11. Illustration of RBF neural network training: determination of hidden neurons.

PREFILTERING FOR PATTERN RECOGNITION

147

III. Introduction to Wavelet Transform Traditionally, Fourier transforms are utilized for performing signal representation and analysis. Although it is straightforward to reconstruct a signal from its Fourier transform, no local description of the signal is included in its Fourier representation. Therefore, Fourier transforms are powerful tools for analyzing the components of a stationary signal (a stationary signal is a signal that repeats itself). It is, however, less useful in analyzing nonstationary signals. The short-time Fourier transform (STFT) and, as a special case, the Gabor (1946) transform have been introduced for signal local description; because the signal is analyzed after filtering by a fixed window function, these transforms have the localization property that traditional Fourier transforms lack. However, because the envelope of the signal is the same for all frequencies, they do not provide enough time detail for high frequencies. The most frequently adapted representation to the nonstationary signal is the wavelet transform, which gives a better localization than the traditional Fourier transform and a better division of the time–frequency (or space–frequency) plane than the STFT. The first known example of a wavelet function is the Haar wavelet, and the wavelet transform was introduced by Grosmann and Morlet (1984) in order to study seismic reflection signals in geophysics applications. Many wavelet functions were constructed in the 1980s, e.g., Meyer (1993), Lemarie and Meyer (1986), and Daubechies (1989). Mallat and Meyer developed the multiresolution analysis in 1989, which is widely used for signal and image processing. This section first presents briefly the wavelet theory: wavelet transform, scaling function, discrete wavelet transform (DWT), pyramidal algorithm, filter bank, and multiresolution analysis. Then, we expose in detail the Haar wavelet transform in order to illustrate basic concepts of wavelet analysis. We also describe some other frequently used Wavelet functions: Daubechies, Mexican hat, Morlet, Meyer, and Coiflet. This introduction to wavelet transforms is finished by wavelet analysis applications. A. Continuous and Discrete Wavelet Transform 1. Definition of Continuous Wavelet Transform The wavelet family can be obtained by dilating and shifting a basic function, , the ‘‘mother Wavelet’’   xb 12 , ð43Þ a, b ðxÞ ¼ a a

148

YANG AND PAINDAVOINE

where a, b 2 R, a > 0. If a ¼ 1, then a, b represents versions of (x) shifted by b. If b ¼ 0, then a, b represents dilated versions of (x) with different frequencies controlled 1 by the parameter a. The scalar a2 also determines the amplitude of the wavelet functions. Larger values are assigned to higher frequency functions (i.e., with a < 1), and smaller values are assigned to lower frequency functions (i.e., with a > 1). The continuous wavelet transform, Wa,b f, of a function, f, is defined by the scalar product:   Z 1 1 xb dx: ð44Þ f ðxÞ Wa, b f ¼ h f , a, b i ¼ pffiffiffi a a 1

The admissibility condition denoted C ensures that the inverse Wavelet transform is applicable (Daubechies, 1989). This condition is satisfied as Z þ1 ðxÞ ¼ 0: ð45Þ 1

Figure 12 shows that the wavelet transform corresponds to an enlargement of the signal f (x) on the part centered in b with a dilation factor a f (x)

b1 y(

x

b2

x ) a y(

x − b1 a )

y(

x − b2 a )

x

Figure 12. Wavelet transform illustration.

PREFILTERING FOR PATTERN RECOGNITION

149

If f (x) is constant around time b, both the integral of the product between f and [Eq. (44)] and the wavelet coefficient Wa,b (case b ¼ b1) are very small.

If f (x) presents some variations at time b that are in phase with ðxaÞ, both the integral of the product between f and and the wavelet coefficient (case b ¼ b2) become important. Although the parameters a and b can be chosen with any real values, for ease of computation one usually sets a ¼ 2j and b ¼ ka, j, k 2 Z. The simplest type of wavelet basis is obtained as   j j x  2j k 2 ¼ 22 ð2j x  kÞ: ð46Þ j; k ðxÞ ¼ 2 j 2 This choice of parameters a,b naturally connects multiresolution analysis in signal processing with the world of wavelets, and sampling in the a–b plane introduces the discrete wavelet transform, which can be performed using a fast, pyramidal algorithm related to multiresolution analysis (Mallat, 1989; Strang and Nguyen, 1997). 2. Discrete Wavelet Transform a. Orthogonal Wavelets. The orthogonal Wavelet function (mother wavelet) is orthogonal to all functions obtained by shifting right or left by an integer amount. Furthermore, the mother wavelet is orthogonal to all functions obtained by dilating the mother by a factor of 2 j and shifting by a multiple of 2 j units (Daubechies, 1989). The orthogonality property means that the inner product of the mother wavelet with itself is unity, and the inner product between the mother wavelet and the shifts and dilates of the mother is zero. The collection of these shifted and dilated wavelet functions is called a wavelet basis, which refers only to an orthogonal set of functions, whereas the term wavelet function is used generically to refer to either orthogonal or nonorthogonal wavelets. A biorthogonal wavelet system consists of two sets of wavelets generated by a mother wavelet and a dual wavelet ~ (see examples of biorthogonal wavelets implemented in the Matlab toolbox). The use of an orthogonal basis implies the use of the DWT, which has a very important mathematical and engineering consequence: any continuous function may be uniquely projected onto the wavelet basis functions and expressed as a linear combination of the basis functions. Decomposition of a function in terms of orthogonal basis functions is, in fact, old news. What is new and exciting about the wavelet decomposition methodology is that wavelet basis functions have what is called compact support. This means that the basis functions are nonzero only on a finite interval, thus

150

YANG AND PAINDAVOINE

allowing the DWT to efficiently represent signals that have localized features. b. Scaling Function, Filter Bank, and Multiresolution Analysis. The DWT employs two sets of functions: scaling functions, , and the wavelet functions, . Often, prior to the construction of , one constructs  such that the functions (x  n), n 2 Z constitute an orthonormal system [see Daubechies (1989) for properties of orthogonormal wavelet functions]. Two-scale relations, (2x  n), are the fundamental components with which to construct the scaling function, , and the wavelet function . In fact, the different properties cause the functions  and to be linked via relations as follows (Daubechies, 1989) ðxÞ ¼

¼1 pffiffiffi nX hn ð2x  nÞ, 2 n ¼ 1

¼1 pffiffiffi nX gn ð2x  nÞ, ðxÞ ¼ 2

ð47Þ

n ¼ 1

where the numbers, hn, are called the filter coefficients of the function , and gn are called the filter coefficients of the function, . In digital signal processing terms, the scaling function is a low-pass filter that suppresses the high-frequency components of a signal and passes the low-frequency components (the scaling function is also often called the smoothing function). The wavelet function is a high-pass filter that passes the high-frequency components while the low-frequency components are suppressed. Thus, the DWT analysis can be performed using a fast dyadic pyramidal algorithm related to filter banks. Mallat (1989) showed that computation of wavelet representation can be accomplished by successive convolutions of the time-domain signal with quadrature mirror filters. The original signal x[n] is first passed through a half-band low-pass filter h[n] and a high-pass filter g[n]. The analysis simplifies if the length (dimension) N of the signal is a power of two. After filtering, the signal is decomposed into a coarse approximation (ylow) and detailed information (yhigh) ylow ¼ yhigh ¼

N X n

N X n

x½nh½2k  n

x½ng½2k  n;

ð48Þ

where ylow and yhigh of length N2 are the output of the low-pass and high-pass filters, respectively, after subsampling by 2. This decomposition halves the

151

PREFILTERING FOR PATTERN RECOGNITION

time resolution because the entire signal is now characterized by half the number of samples. However, it doubles the frequency resolution, as the frequency band of the signal now spans only half the previous frequency band. The result, yhigh, constitutes wavelet coefficients at the scale one. The ylow is then further decomposed using the same wavelet decomposition step. This procedure is repeated for further decomposition. At every scale, filtering and subsampling result in half the time resolution and double the frequency resolution. Each scale of decomposition analyzes the signal at different frequency ranges and at different resolutions, hence the name multiresolution analysis. c. Image Processing by Wavelet Transform. The discrete wavelet transform can be easily developed for images, which are two-dimensional signals and thus can be treated as a two-dimensional array. Because every row (column) of the array corresponds to a one-dimensional signal, one can use the previously described method to process it and, hence, an image can be processed in two directions sequentially. Figure 13 shows an interpretation of the process given by Wu et al. (2000). Figure 14 illustrates the wavelet decomposition of an image.

A

A L-col L-row H-col

A

H

V

D

Scale 0

H

D V Scale 1

A H V D

Scale 2

Decomposition process

Original image H-row

H-col A H V D

L-col Scale 2

A

H

V

D

A Scale 1

Scale 0

Reconstruction process Figure 13. Block diagram of the multiresolution wavelet decomposition of an image (right) and Flow diagram of the filter bank to decompose an image (left): L denotes a low-pass filtering operation and H denotes a high-pass filtering operation. The notation L-row (L-col) means that the L operation is applied to the row (column). The output from the L-low and L-col operations is the A sub-image. The H, V, and D sub-images, which represent the detail information of the decomposition, are similarly constructed. A, H, V, and D denote average, horizontal details, vertical details, and diagonal details, respectively.

152

YANG AND PAINDAVOINE

Figure 14. Wavelet decomposition of an image: original image (left) and Wavelet decomposition at scale 1 using Daubechies-4 function (right). From the upper left corner of the image we have: low-pass decomposition, horizontal details, vertical details and diagonal details.

3. Wavelet Packet Transform Wavelet packet transform (WPT), first introduced by Coifman et al. (1992), is a generalization of the dyadic wavelets transform. Wavelet packet functions offer more flexibility than wavelets in representing different types of signals. Dyadic wavelet decomposition is achieved by iterative filter bank operation over the low-frequency band of each scale. Wavelet packet decomposition, however, is achieved when the filter bank operation is iterated over all frequency bands at each scale. The WPT offers a rich decomposition structure set. The WPT is often associated with a best basis selection algorithm, which decides on a decomposition structure among the library of possible bases by measuring a data-dependent cost function. The best basis algorithm automatically adapts the transform to best match the characteristics of the analyzed signal. More details concerning the WPT will be given in Section IV (see Section IV.A.3 for the wavelet packet tree and Section IV.B.1 for an example of WPT application). B. Wavelet Functions One criticism of wavelet analysis is the arbitrary choice of the wavelet function, one of the important decisions on the user’s part. It is similar to the choice of instruments of observation. In choosing the wavelet function, there are several factors (Chui, 1992; Farge, 1992): orthogonality, complex

153

PREFILTERING FOR PATTERN RECOGNITION

or real function, width, and shape of the wavelet function. We describe the Haar wavelet function in order to illustrate some fundamental concepts and multiresolution theory introduced previously. 1. Haar Wavelet Function a. Definition. Figure 15 displays Haar scaling and Haar wavelet functions. The scaling function is defined by  1 x 2 ð0, 1Þ ð49Þ ðxÞ ¼ 0 x2 = ð0, 1Þ and Haar wavelet function is defined by 8 1 > > x 2 ð0, Þ > <1 2 1 ðxÞ ¼ 1 x 2 ð , 1Þ > > > 2 : 0 x2 = ð0, 1Þ:

ð50Þ

ðxÞ ¼ ð2xÞ þ ð2x  1Þ ðxÞ ¼ ð2xÞ  ð2x  1Þ:

ð51Þ

Functions  and

of the Haar wavelet family can be constructed, thus

From Eq. (50), the filter coefficients of the scaling and wavelet functions can be found as h0 ¼ h1 ¼ p1ffiffi2 and g0 ¼ g1 ¼ p1ffiffi2. Each step in a Haar transform calculates a set of wavelet coefficients and a set of averages. If a signal, s = s0, s1, . . . , sN  1, contains N elements, there will be N2 average and N2 coefficient values. The averages are stored in the lower half of the N elements array, and the coefficients are stored in the upper half. The averages become the input for the next iteration of the wavelet calculation, where for each iteration i þ 1, Ni þ 1 ¼ Ni/2. The f (x)

y (x)

1

1

0

−1

0.5

1

x

0

0.5

1

−1

Figure 15. Haar scaling function (left) and Haar wavelet function (right).

x

154

YANG AND PAINDAVOINE

recursive iterations continue until a single average and a single coefficient are calculated. This replaces the original signal of N elements with an average, followed by a set of wavelet coefficients whose size is an increasing power of two (e.g., 20, 21, 22, . . . , N2 ). The Haar wavelet transforms that calculate an average ai and a wavelet coefficient, ci, are shown as si þ siþ1 ai ¼ pffiffiffi 2 ð52Þ si  siþ1 ci ¼ pffiffiffi : 2 b. Wavelet Vector and Wavelet Matrix. In the linear algebra view of the Haar transform, the first average is calculated by the inner product of the signal [s0, s1, . . . , sN1] and the vector of the same size [p1ffiffi2 , p1ffiffi2 , 0, 0, . . . , 0]. This is the scaling vector. The first coefficient is calculated by the inner product of the signal and the vector [p1ffiffi2 ,  p1ffiffi2 , 0, 0, . . . , 0]. This is the wavelet vector. The next average and coefficient are calculated by shifting the scaling and wavelet vectors by two and calculating the inner product. The first step of the Haar wavelet for an eight-element signal can be described as the matrix–vector multiplication W s of the wavelet matrix and the signal, where 2 3 1 1 0 0 0 0 0 0 6 0 0 1 1 0 0 0 07 6 7 6 0 0 0 0 1 1 0 07 6 7 1 6 0 0 0 0 0 0 1 17 6 7: W ¼ pffiffiffi 6 0 0 0 0 0 07 2 6 1 1 7 6 0 0 1 1 0 0 0 07 6 7 4 0 0 0 0 1 1 0 05 0 0 0 0 0 0 1 1

The Haar wavelet transform has a number of advantages, e.g., it is fast and conceptually simple. It is exactly reversible without the edge effects that are a problem for other wavelet transforms. Of course, the Haar transform also has limitations, which can be a problem for some applications. The Haar window is only two elements wide, with shifts over by two values of wavelet transform algorithm at each scale, and thus some change in the signal is not reflected in the wavelet coefficients. 2. Daubechies Wavelet Function and Other Wavelet Functions The Daubechies-4 (D4) is, probably, the most used function of the Daubechies family (Daubechies, 1989). As indicated by its name, this

PREFILTERING FOR PATTERN RECOGNITION

155

wavelet set has four coefficients. Figure 16 displays the scaling and wavelet functions of D4. The scaling function, , and the wavelet function, , can be expressed as pffiffiffi ðxÞ ¼ 2½h0 ð2xÞ þ h1 ð2x  1Þ þ h2 ð2x  2Þ þ h3 ð2x  3Þ ð53Þ pffiffiffi ðxÞ ¼ 2½g0 ð2xÞ þ g1 ð2x  1Þ þ g2 ð2x  2Þ þ g3 ð2x  3Þ

The scaling function coefficients values are pffiffiffi pffiffiffi 1þ 3 3þ 3 h1 ¼ pffiffiffi h0 ¼ pffiffiffi 4 p 2 ffiffiffi 4 p 2 ffiffiffi 3 3 1 3 h3 ¼ pffiffiffi : h2 ¼ pffiffiffi 4 2 4 2

ð54Þ

The wavelet function coefficient values are g0 ¼ h3, g1 ¼ h2, g2 ¼ h1, and g3 ¼ h0. As with the Haar transform discussed earlier, the Daubechies scaling and wavelet function coefficients shift from left to right by two places in each iteration of a wavelet transform step. The result of the wavelet transform produces a ‘‘down sampled’’ smoothed version of the signal and a ‘‘down sampled’’ version of the signal that reflects changes between signal elements. If we compare the Haar wavelet and Daubechies-4 wavelet, we see that the Haar high-pass filter produces a result that reflects the difference between an even element and an odd element. The difference between an even element and its odd element successor will not be reflected in the coefficient calculated by a single step of the Haar high-pass filter. In contrast, there is overlap between successive Daubechies-4 high-pass filters, so change between any two elements will be reflected in the result. The wavelet literature covers a wide variety of wavelet functions. Figures 17–20

Figure 16. Scaling function of Daubechies-4 (left) and wavelet function of Daubechies-4 (right).

156

YANG AND PAINDAVOINE

Figure 17. Scaling function of Coiflet-4 (left) and wavelet function of Coiflet-4 (right).

Figure 18. Wavelet function of Mexican hat.

display, respectively, Coiflet, Mexican hat, Meyer, and Morlet wavelet functions (Daubechies, 1994). These figures are obtained using the Matlab 6.0 Toolbox. Readers are invited to refer to the Matlab Toolbox for properties of these wavelet functions. 3. Wavelet Analysis: Parallelism and Applications As with many algorithms in computer vision and image processing, the twodimensional DWT is computationally intensive and operates on a large data set. These factors, coupled with the demand for real-time operations in many image processing tasks, necessitate the use of parallel processing in order to

PREFILTERING FOR PATTERN RECOGNITION

157

Figure 19. Scaling function of Meyer (left) and wavelet function of Meyer (right).

Figure 20. Wavelet function of Morlet.

provide high performance at a reasonable cost. A number of parallel solutions of the DWT have been proposed (Lu, 1993; Koc et al., 1994; Misra and Nichols, 1994). The intrinsic parallelism in the wavelet transform makes it attractive for hardware implementation (Chen et al., 2001).

158

YANG AND PAINDAVOINE

The wavelet has been applied to a wide ranges of fields. For example, it can be used in a variety of statistical applications, including signal variance estimation, frequency analysis, and kernel regression. In digital signal processing, it provides powerful tools for analyzing, encoding, compressing, reconstructing, and modeling signals and images. Some important applications follow.

Enhance edge detection in image processing.

Achieve high rates of signal or image compression (e.g., the JPEG 2000 standard is based on the wavelet lifting scheme).

Restore noisy signal and degraded images.

Study the fractal properties of signals and images.

Extract information-rich features for use in classification and patternrecognition applications. Section IV gives many applications using the wavelet transform technique for feature extraction, and in Section V the wavelet transform is used as a prefiltering technique for movement analysis in handwriting and face recognition.

IV. Pattern Recognition Using Wavelet and Neural Network for Signal and Image Processing Applications A system of pattern recognition for signal or image processing applications is generally composed of two modules: a preprocessing module and a classifier. The before recognition preprocessing task usually consists of two roles: defining the initial transformation in order to represent the measured signal and retaining information that is important for class discrimination and discarding that which is irrelevant. This preprocessing stage is sometimes called ‘‘features extraction.’’ The goal is to extract the relevant features that enable the recognition of different patterns. It is also desirable to keep the dimension of the feature vector as low as possible because this results in a simpler classifier. However, sometimes it happens that the dimension of the feature vector is greater than the initial dimension of the data set because complementary information is integrated as textures or contours in order to better represent the measured signal. The structure of this section is as follows. We first briefly describe the different techniques of preprocessing commonly used for pattern recognition. Then we give some examples of audio, underwater, and eddy current signal processing for identification and classification purposes using the wavelet transform. We present the systems by describing the preprocessing

PREFILTERING FOR PATTERN RECOGNITION

159

technique used, the pattern recognition model, and the obtained results. We also introduce some medical signal processing applications concerning the classification of electroencephalogram (EEG) and myoelectric signals (MES). One of these applications consists of classification of sleep stages using the neural network and wavelet transform. After that, we illustrate applications of shape recognition and image classification by content. Aside from the examples described in this section, we can find many other recognition pattern applications using the wavelet transform and neural network for signal processing as ground fault discrimination, flaw echo location, short-term electrical load forecasting, particle shape classification, and so on. Readers are invited to refer to Tungkanawanich et al., 2000; Liu et al., 1997; Yao et al., 2000; Drolon et al., 2000; and Szu et al., 1996 for more details. A. Different Techniques of Preprocessing for Pattern Recognition 1. Principal Component Analysis and Fourier Descriptors a. Principal Component Analysis. The PCA approach consists of constructing an orthogonal basis from a data set that yields the best compression. This results in the smallest least square error if we use only M of the N coefficients for representation of a signal, x, where M  N. Thus, the basic idea behind the PCA approach for pattern recognition is that response data can be expressed as a superposition of a set of basis functions determined (learned) from examples. The learned basis functions are eigenvectors of the covariance matrix of the example data vector regarded as a stochastic column vector. Collecting the basis functions (eigenvectors of the covariance matrix in descending order) as columns of a matrix, the signal vector denoted by x can also be expressed using the new basis as y ¼ UT x,

ð55Þ

where y denotes the data vector expressed in the new basis and U is the eigenvector matrix. Through the PCA approach, an ‘‘intelligent’’ system has been obtained, one that extracts a suitable orthogonal set of basis functions from data examples (Bishop, 2000). The PCA produces an uncorrelated feature set by projecting data onto eigenvectors of the covariance matrix. The PCA provides a means of unsupervised dimensionality reduction because no class membership qualifies data when specifying eigenvectors of maximum variance.

160

YANG AND PAINDAVOINE

Using a geometrical interpretation, the coordinate system is rotated such that the axes in the new system point in the directions of the largest variance of the analyzed data. The axes are referred to as principal axes. The idea behind compression is to represent data using only a limited number of axes for which the variance is sufficiently large (see Figure 21). Another method of extracting basic functions using knowledge about analyzed data consists of independent component analysis (ICA). Readers will find a wellillustrated comparison between PCA and ICA by Bartlett (1998). PCA encodes dependencies in data by rotating the axes in order to correspond to directions of maximum covariance. ICA does not constrain the axes to be orthogonal and attempts to place them in the directions of statistical dependencies in the data. b. Fourier Descriptors (FDs). The Fourier transform decomposes a function into a basis set of sinusoidal components that are only localized in the frequency domain. The Fourier descriptor is a standard technique used in image processing for the recognition of different contours in digital images. The idea is to expand the contour of an object in a Fourier series and use a limited number of Fourier coefficients, called FDs, as feature for recognition of the object. Discrete Fourier transform (DFT) and fast Fourier transform (FFT) are widely used in real-time applications. The Fourier transform is a powerful tool for studying the frequency content of signals, but it does not provide any localization in time. In order to overcome this problem, the windowed Fourier transform, or short-time Fourier transform (STFT), was suggested Z ð56Þ STFT ½ f ðw; sÞ ¼ g ðt  sÞf ðtÞeiwt dt, where g is a given time window, which can be shifted in time. The STFT divides the time–frequency domain using a fixed tiling: once a window PCA

ICA

Figure 21. Example of two-dimensional data distribution and the corresponding principal component and independent component axes (cf. Bartlett, 1998).

PREFILTERING FOR PATTERN RECOGNITION

161

function is specified, each basis function has an identical aspect ration of time–frequency. If g is a Gaussian function, then the STFT is called a Gabor transform (Gabor, 1946). 2. Wavelet Transform and Feature Extraction A more elegant version of time–frequency analysis is wavelet analysis. The wavelet transform uses a sequence of small, compactly supported waves as a basis for the representation of the signal under study. This basis consists of local functions with different positions and scales (Mallat, 1989). Different from the PCA and the ICA, the basic functions used for WT are constructed without using any particular knowledge about the analyzed data. Unlike sines and cosines, wavelets can be constructed in many forms as long as certain requirements are satisfied. In pattern recognition, feature extraction denotes a procedure for mapping the original measurement into a relatively short vector representing features relevant for the classification. The multiscale aspect of the wavelet transform is particularly important for recognition and diagnostic purposes because if one can make some decision concerning the underlying time– frequency components of the signal, one may choose the appropriate scales in the wavelet transform while ignoring the contribution of the other scales. This allows one to achieve both relevant feature extraction and data compression (Mallet et al., 1997). The wavelet transform provides very general techniques that can be applied to many tasks in nonstationary signal processing. Like the FFT, the wavelet transform can be implemented with fast algorithms. DWT allows a better study of short-time high-frequency structures. Low frequencies are sampled with large time steps, whereas high frequencies are sampled with small-time steps. The tiling of the time–frequency domain is variable using WT: the aspect ratio of the wavelets varies such that the frequency resolution is proportional to the center frequency. This tiling has been shown to be appropriate for many physical signals, but the partition is nonetheless still fixed. 3. Wavelet Packet Transform (WPT) We use the multiresolution analysis proposed by Mallat in order to illustrate differences between wavelet transform and wavelet packet transform (Coifman and Wickerhauser, 1992; Coifman et al., 1992; Bachman et al., 2000; Benedetto and Frazier, 1994). In Mallat (1989), the wavelet transform is seen as generated by a pair of quadrature mirror filters (h, g), where h is a low-pass filter and g a high-pass filter. h is related to the scaling function  with the scale equal to 2 j, j 2 Z + and g to the mother wavelet by

162

YANG AND PAINDAVOINE

pffiffiffi ðxÞ ¼ n hðnÞ 2ð2x  nÞ pffiffiffi ðxÞ ¼ n gðnÞ 2ð2x  nÞ:

ð57Þ

ðHx Þk ¼ n hðn  2kÞxðnÞ ðGx Þk ¼ n gðn  2kÞxðnÞ:

ð58Þ

For a discrete signal, x, of finite length, we define the following operators H, G:

H, G acting on x are convolutions with both filters h, g and a downsampling by two. This transforms the signal, x, into two subbands of equal length, (Hx) contains the low-pass band and (Gx) the high-pass band. Recursive application of H and G on the low-pass band defines the wavelet transform; recursive application of the operators H, G on both bands defines the ‘‘wavelet packets transform’’ as shown in Figure 22. The WPT provides an adaptive tiling in the time–frequency domain: an overcomplete set of time– frequency tilings are provided and the best can be selected for the particular signal under study. 4. Feature Extraction Using Invariance Properties The sinusoidal basis functions used by Fourier transform are well localized in the frequency domain but not in the time domain. For image processing applications, this implies that it may be difficult to modify a particular region in the image by adding or deleting frequency components generated by the Fourier transform. However, the wavelet transform decomposes the image into a set of subimages called shapes with different resolutions

0−Fn

0−Fn H

G

H

G

0−Fn /2

Fn /2−Fn

0−Fn /2

Fn /2−Fn

H 0−Fn /4

G Fn /4−Fn /2

H 0−Fn /4

G

H

G

Fn /4−Fn /2 Fn /2−3Fn /4 3Fn /4−Fn

Figure 22. Wavelet tree (left) and Wavelet packet tree (right). The top subband contains the signal, x, with Nyquist frequency, Fn. The time resolution decreases by a factor of 2 in each iteration; the frequency resolution doubles as ones goes from the top to the bottom in the WP and WPT trees.

PREFILTERING FOR PATTERN RECOGNITION

163

corresponding to different frequency bands. In pattern recognition, the Fourier transform and the wavelet transform have been used for shape analysis, e.g., in the discrimination of closed curves so as to classify different present objects in an image. The difficulty of shape recognition is in classifying objects correctly, and this is independent of their position, orientation, and size. This property may be obtained in three basic ways (Chen and Hewit, 2000). The first is to train the classifier to be invariant. The training regime will explicitly tell the classifier that shapes that are to be classified are similar. Although this classification procedure is theoretically possible, in practice it may require an infeasibly large amount of training data and a correspondingly long training time. The second method is to build a degree of invariance into the classifier itself. For example, higher order neural networks (HONNs) (Reid et al., 1989) can have this type of inherent invariance, but the number of interconnections in a fully interconnected HONN may be too large to be practical for high-resolution image processing. This severely limits the application of HONNs in cases where the number of inputs is large and consequently increases the processing time. The third method is to extract invariant features from a given pattern during the preprocessing stage and then use a neural network as a secondary classifier. The extracted feature vectors must be invariant to any similarity transformation of a pattern, such as translation, rotation, and scale. Many algorithms have been developed for this (Sekita et al., 1992; Das et al., 1990) and are based on intrinsically describing the boundary of a pattern because the shape recognition is considered equivalent to the detection of the pattern boundary. One of these methods consists of representing the boundary using the polar coordinate system. By setting the center of the pattern as the origin, sampling of the boundary can be performed by angular sampling methods and repeated every 2 radians. Thus, the boundary shape can be represented by the distance between the sampling points and the pattern center, and so identification of a two-dimensional pattern can be transformed to that of the one-dimensional distance function, which is periodical. Another advantage of this transformation is that the distance sequence is already invariant to pattern translation in a two-dimensional domain. It should be noted that the pattern rotation has been transferred to the distance sequence shift. Thus, it only remains to derive algorithms to generate coefficients that are invariant to scaling and shifting of the distance sequence. Scaling invariance can be achieved by normalizing the distance sequence. The new distance sequence can be obtained by dividing the original distance sequence by its average value.

164

YANG AND PAINDAVOINE

Rotation invariance of the shape is equivalent to shift invariance of the distance function. Shift invariance is easy to achieve using the Fourier transform but more difficult using the wavelet transform. This problem was tackled by Mallat (1991), who studied the zero crossings of a wavelet transform. Simoncelli et al. (1992) studied the invariance property of a wavelet further and proposed shiftable multiscale transforms. More recently, Unser (1995) used overcomplete wavelet transforms in order to derive shift invariance of a wavelet. A similar method has been developed and used by Chen and Hewit (2000). B. Audio, Underwater, and Eddy Current Signal Processing Applications Time–frequency representations have received considerable attention in such diverse fields as speech recognition (Tzanetakis et al., 2001), radar classification (Ayrulu and Barshan, 2001), underwater acoustics, and geoacoustical signals. In many real-time applications, linear time–frequency representations (STFT, DWT, and WPT) are preferable to quadratic time– frequency representations and the continuous wavelet transform because firsts are accompanied by efficient discrete transforms. This section describes three examples of audio, underwater, and eddy current signal processing applications for identification and classification purposes using the wavelet transform. 1. Audio Signals Identification In the last decade, there has been an increasing demand for systems able to classify different types of auditive signals, especially speech recognition systems. Sound classification systems can be designed to identify different types of audio signals from a set of training data with robust performance (Renals and Rohwer, 1989; Arslan and Hansen, 1999; Kermit et al., 1999). We describe a general audio classification system proposed by Kermit and Eide (2000) that uses the Haar wavelet transform (Chui, 1992) in the preprocessing stage, a neural network comparing small signal segments present in specific types of sound from candidate audio signals against a predefined template. The classification performances are tested with three different applications concerning steady-state vowels, speech signals, and instrument recognition. a. Preprocessing Stage and Identification Procedure. A key to successful classification may be to isolate some kind of uniqueness that identifies the sound to be classified. Such uniqueness is sometimes present during a very short period of time in audio signals and is often repeated with some degree

PREFILTERING FOR PATTERN RECOGNITION

165

of perturbation for multiple succeeding periods throughout a signal. The system proposed by Kermit and Eide (2000) selects small audio segments that represent characteristic regions as features of identification. In order to compensate for some perturbations of these segments, the Haar wavelets (Chui, 1992), known to have a dampening effect on rich amplitudal variations often present in audio signals, are performed in the preprocessing stage. The application of Haar wavelets to a characteristic segment of a audio signal produces a set of Haar wavelet coefficients that give a template for identification of a specific audio signal. The identification process is performed by a neural network known as the O algorithm (Eide and Lindblad, 1992; Lindbald et al., 1997), which performs a similarity match between two sets of data. The O algorithm is similar to an RBF neural network. Here, the Haar wavelet coefficients are used as a predefined characteristic template representing some class of audio signals. A candidate template captured from an audio signal can be classified by comparing its similarity to the predefined templates. b. System Evaluation Using Three Different Data Sets. Three data sets are used in the evaluation of the proposed system. The first data set is a database consisting of 90 data files representing 10 separate trials of nine vowels in the Norwegian tongue. Each file contains 1024 mono samples at a 16-bit sampling rate of 8 kHZ. In this application of classification of steadystate vowels, windows with widths of N = 8, 16, and 32 samples were applied as a unique representation of vowel periods. About 90,000 tests were performed, and the results are presented in a confusion matrix table (Kermit and Eide, 2000). The maximum number of obtainable identifications for each vowel is about 100 because there are about 10 periods in each file from the vowel database. The second data set consists of 25 files representing five separate trials of five different words (cat, pit, fed, apple, and tall ). The vowel search from five different words is the application to speech signals. If a word contains six vowel periods where four of these are identified correctly and two periods are identified incorrectly, the vowel is still considered to be classified correctly. The classification performance is presented in Kermit and Eide (2000). The last data set differs slightly from the two sets of speech sound just described. A prerecorded musical composition on a compact disc was chosen such that some musical instruments could be isolated and recorded without interference from other instruments. This application consists of instrument recognition from a complex musical composition. The system test with three different instruments (bass guitar, keyboard, and snare drum) present during the 5-s musical composition gives, respectively, a correct classification of 75%, 62.5%, and 87.5%.

166

YANG AND PAINDAVOINE

2. Underwater Signals Classification Automatic classification of underwater passive sonar signals radiated by ships is a complex problem because of the large variability in both temporal and spectral characteristics in signal even when obtained from the same source. Each type of signal has its own characteristics and conventionally is identified by human experts either by listening to or by looking at the spectrograms of the processed sonar signals. Identification by human experts is usually not objective and is a very heavy workload. Chen et al. (1998) have proposed an approach using the wavelet transform and neural networks to classify the passive sonar signals of a ship in order to identify the class of ships. a. Underwater Signal and Pattern Classification System. A passive sonar underwater system is an acoustic receiving system used to receive underwater signals radiated by various ships and other marine life. Thus characteristics of radiated signals from ships are of particular importance in identifying the ships (Chen et al., 1998). Research on the models of the underwater signals sources radiated by ships shows that the signal sources can be divided into three headings: machinery signals, propeller signals, and hydrodynamic signals (Urick, 1983). Each signal produces its own component spectrum, and the tonal content of a the radiated signals of a ship is very characteristic of that ship. Some of the frequency components and harmonics vary with the speed of the engine, but others stay fixed. The pattern classification system proposed by Chen et al. (1998) takes into account these particular properties of underwater signals (see Figure 23). b. Preprocessing and Feature Extraction. Chen et al. (1998) obtained the spectral envelope by averaging over each spectrum in a spectrogram. Tonals, composed of high frequency, are important features for the spectrum of a ship and are usually buried in the spectrum. The wavelet transform has been used to extract tonal features from the spectrum. In fact, tonals have the same property as edges or corners, which are often characterized by the local variation. Measurement of the local variation of a signal must be defined with respect to a reference scale. This scale characterizes the neighborhood size, where the variations are computed. For signals including meaningful structures of different sizes, one needs to modify this scale parameter. By coupling scale and spatial variables of a spectrum signal, the local maxima of the wavelet transform allows one to find tonal features easily. The first-order derivation of the Gaussian function (x) is chosen as a mother wavelet (x). The wavelet transform of f(x) at scale 2 j ( j is an integer) and position x is given by

PREFILTERING FOR PATTERN RECOGNITION

167

Sonar data

Time-frequency distribution (spectrogram)

Average power spectral density

Feature extraction

Neural network classifiers

Output decisison Figure 23. Architecture of the pattern classification system (cf. Chen et al., 1998).



 d2 j d ðxÞ ¼ 2j ð f  2 j ÞðxÞ, W2 j f ðxÞ ¼ f  2 j ðxÞ ¼ f  2 dx dx j

ð59Þ

where 2 j ðxÞ ¼ 12 ð2xj Þ and 2 j ðxÞ ¼ 21j ð2xj Þ. W2 j f (x) is proportional to the first derivative of f(x) smoothed by 2 j (x). The maxima of |W2 j f(x)| corresponds to the maxima of the derivative of f  2 j (x), which are the sharp variation points at scale 2 j (Lee et al., 1993). Thus, tonal features can be extracted using the wavelet transform by choosing a small scale. The tonal features extracted are then used as inputs to the classifier. c. Neural Network Classifiers and Experimental Results. Two neural networks classifiers have been employed: multilayer perceptron (MLP) and radial basis function (RBF). The MLP has been trained by the backpropagation algorithm and its variants. Two sigmoid functions, unipolar and bipolar, have been tested. The data set is collected from three fishing boats. After signal preprocessing, a total of 200 average power spectrum density (APSD) patterns are included in the data set: 60 patterns belonging to the first class, 60 patterns belonging to the second class, and the rest belonging to the third class. The input vector has 256 components. Experimental results obtained from MLP and RBF with different configurations and learned algorithms show that networks trained with tonal features extracted by wavelet transform have 96% or 94% correction rates, but training with original APSDs has only an 80% correction rate.

168

YANG AND PAINDAVOINE

3. Neural Network Classifier for Eddy Current Signals A modern civil aircraft has a large number of lap joints and a number of rivets on the order of 104. The manual inspection of these rivets is not only tedious but time-consuming and therefore expensive. Moreover, considerable human error risk exists. Lingvall and Stepinski (2000) have proposed an approach for automatically detecting and classifying defects or riveted lap joints during the analysis of eddy current (EC) signals. a. Eddy Current Signals Inspection. This method employs a probe (coil with current) placed close to the inspected material. The primary flux in Figure 24 induces eddy currents in the material, which gives rise to a secondary flux. The secondary flux is then coupled back to the coil, which affects its impedance. Therefore, the magnitude and phase of the eddy currents will affect the loading on the coil and thus its impedance. If a discontinuity (defect) is present in the material, the eddy current density will be changed, which can be observed as a change in coil impedance. The EC inspection of riveted lap joints was performed using a simple mechanical scanner guiding the probe along the rivet line. The resulting signals are complex valued vectors, which correspond to the probe impedance as a function of its location. Three preprocessing steps of this signal must be performed before feature extraction and classification: three-point median filtering, rotation of EC signals, and amplitude normalization. ~ Primary flux

Coil Air

Conductor Eddy currents

Secondary flux Figure 24. Eddy current inspection (cf. Lingvall and Stepinski, 1999).

PREFILTERING FOR PATTERN RECOGNITION

169

b. Feature Extraction and Classification. The goal here is to extract the relevant features that enable detecting different classes of defects. Examples of classes are large defects, defects on the left (or right) side of the rivet, and so on. Large defects are detected easily by thresholding, but the features that preserve information about location must be used in order to distinguish between left or right defects. The feature extraction was applied in a window of 128 samples centered precisely around each rivet. Four types of feature extractors, applied to a windowed signal, were investigated: block mean values, FDs, DWT, and PCA. Block Mean Values. These consist of dividing the window positioned around each rivet into a number of blocks, and a mean value is calculated in each block. Fourier Descriptors. In the case of the complex valued EC Lissajous patterns, the Fourier coefficients were performed in the analyzing window using discrete Fourier transform. Discrete Wavelet Transform. Time–frequency distribution is useful for analyzing nonstationary signals such as EC signals. One of the most important features of DWT is the fact that the basis consists of local functions with different positions and scales. This feature permits the determination of particular scales where the EC signal has significant energy. For instance, the energy at small scales is mostly due to noise; therefore by removing small-scale coefficients, both noise reduction and data compression have been achieved. The mother wavelet chosen by Lingall and Stepinski is the Coiflet-2 wavelet, which is a rather smooth wavelet suitable for modeling EC signals. Principal Component Analysis. The basic functions used for DWT (here the Coiflet-2 family) are constructed without using any particular knowledge about the analyzed data. Through the PCA approach, an ‘‘intelligent’’ system can be obtained, which extracts a suitable orthogonal set of basis functions from examples of EC data. The classifier used is an MLP neural network with two outputs: one for defects on the right side of the rivets and one for defects on the left side. Only three neurons in the hidden layer have been implemented. EC signals were analyzed in windows of 128 samples after three preprocessing steps. Among the four types of feature extractors, block mean and FDs needed 12 coefficients in order to achieve a satisfactory classification performance; the DWT used the two largest scales, which resulted in 15 basis functions; and the best compression performance was obtained with the PCA; only 6 coefficients were required. All the extractors resulted in two to three missed detections (out of 68) and five to eight false detections (out of 640).

170

YANG AND PAINDAVOINE

C. Classification of Electroencephalogram and Myoelectric Signals Using Wavelet Transform and Neural Networks EEG analysis is a very useful technique for investigating the activity of the central nervous system. It provides information related to the brain activity based on electrical measurements taken from the scalp of a subject. Inference and studies about a subject’s health and effective treatment of some diseases can be carried out by analyzing the information obtained from an EEG. Readers are invited to refer to Nunez (1981) for an exposition on the biophysics of the EEG and the enormous complexities involved. Processing of the recorded information of EEG signals consists mainly of spectral analysis. When the FFT is applied to successive segments of an EEG signal, the frequency spectrum is observed to vary over time as the Fourier coefficients vary (Zhu et al., 1994), indicating that the EEG signal is a nonstationary signal. If the feature extraction method could include modeling of possible nonstationary effects in the signal, better results may be obtained for the classification of EEG signals. This section presents two applications related to the classification of EEG signals: sleep stage classification and brain disorder classification (Polikar et al., 1997). 1. Sleep Stage Classification Sleep scoring by a human expert is a very time-consuming task. In this last decade, several works introduced the use of neural networks as a tool for automated sleep scoring. Most of the systems, used spectral information of the signal using Fourier transformation (Robert et al., 1998). Jobert et al. (1994) illustrated some advantages of the wavelet analysis over the Fourier analysis in sleep research. Oropesa et al. (1999) designed a system based on a specific wavelet packet transform and a neural network structure for a classification task using time–frequency patterns. a. EEG Sleep, Sleep Stage Structure, and Feature Generator. In humans, five sleep stages and an awake stage are defined. Each sleep stage is characterized by a specific pattern of frequency content. For example, stage 1 is defined by no presence of alpha activity (8–13 Hz), low beta1 activity (13– 22 Hz), and theta activity (4–8 Hz) (Oropesa et al., 1999). The classification of EEG sleep is usually made by a visual scorer, which takes 30-s epochs and gives a classification according to the rules of Rechtschaffen and Kales (1968). Not every epoch has 100% of the properties of a specific stage. The decision is made according to which stage properties are present the most. Motivated by the adaptive time–frequency localization property of the wavelet transform and the fact that some structures in sleep recording have

171

PREFILTERING FOR PATTERN RECOGNITION

a well-defined time–frequency pattern, Oropesa et al. (1999) used multiresolution analysis with wavelets proposed by Mallat (1989). The EEG was sampled at 128 Hz, and the db20 wavelet from the Daubechies family of orthogonal wavelets was used. A WPT of eight levels was designed. Out of the family of generated subbands, Oropesa et al. (1999) selected those containing frequency information of the seven bands that correspond to K  complexes þ delta, delta, theta, alpha, spindle, beta1, and beta2 (see Figure 25). For every 30-s epoch taken from the central EEG electrode C3, the mean quadratic value (called energy) of the wavelet packet coefficients for each of seven bands was calculated. These seven numbers were used as features for the epoch. Additionally, six features based on total energy and the ratio of different energy values have been defined. b. Classification and Results. An MLP network trained by the Levenberg–Marquardt function (Bishop, 2000) is used for the classification task with the following structure: 13 neurons in the input layer, each one getting 1 of the 13 features, and 10 neurons in the hidden layer fully connected to 6 neurons in the output layer. The data set consisted of 1630 30-s epoch, half of which were used as the training set and the rest to test the performance of the network. The classification results compared to those of a human expert reached a 97.5% of agreement with the learning set and a 77.6% of agreement for the test set. 2. Brain Disorder Classification We present the application realized by Hazarika et al. (1997) for the classification of three classes of EEG signals: normal, schizophrenia (SCH), and obsessive compulsive disorder (OCD), which use the wavelet transform as a feature extraction technique. Hazarika et al. (1997) decomposed a typical segment of EEG signals using an orthogonal wavelet basis: the

7 6

1 2

3

4

5

Figure 25. WPT and selected subbands (cf. Oropesa et al., 1999).

172

YANG AND PAINDAVOINE

Lemarie wavelet (Lemarie and Meyer, 1986). The Fourier transform of the Lemarie mother wavelet (x) has the shape of a band-pass filter (see Figure 26). Thus, the wavelet function can be interpreted as the impulse response of a band-pass filter. The ‘‘upper’’ levels of the wavelet transform are considered ‘‘noise’’ components and can be ignored for the decomposition. Each EEG signal was divided into segments, with each segment comprising 27 = 128 samples, i.e., each segment was 1-s in duration. One hundred and twenty such segments were taken from each subject. A total of 41, 60, and 35 EEG files were obtained respectively from normal, SCH, and OCD subjects. The 26, 39, and 24 EEG files, respectively, were used for training, and the rest of the files were used for testing purposes. To reduce the number of coefficients used as features describing each segment, only four decomposition levels of the Lemarie wavelet transform have be executed. At each level of decomposition, Hazarika et al. (1997) measured the absolute value of the detail signals and retained the two coefficients with the highest magnitude. Thus, the 4  2 = 8 coefficients for each segment of the EEG signals have to be extracted as features. The classification of three classes is realized by an MLP neural network using eight signal feature parameters as the input, with a single hidden layer containing 50 hidden units and 3 output units. After training, the network with wavelet coefficients was able to correctly classify, respectively, over 66% of the normal class and 71% of the schizophrenia class of EEG signals.

(a)

1.0

(b)

1.0

0.6

y (x) 0.1

y (x) 0.5

−0.4

−0.8 −2.5

−1.2

0.0 x

1.2

2.5

0.0 −10.0

−5.0

0.0 x

5.0

10.0

Figure 26. (a) The Lemarie wavelet (x) and (b) modulus of the Fourier transform of (x).

PREFILTERING FOR PATTERN RECOGNITION

173

The wavelet transform thus provides a potentially powerful technique for preprocessing EEG signals prior to classification. 3. Classification of the Myoelectric Signal Using the Wavelet Packet Transform The MES, collected at the skin surface, has become an important tool in rehabilitation due to the ease with which it may be acquired. The MES provides information about the neuromuscular activity from which it originates, which has been fundamental as a source of control for assistive devices and schemes of functional electrical stimulation (Englehart et al., 1999). One of the MES applications consists of the control of powered prosthetic limbs, realized by pattern recognition. The control information extracted from the MES is based either on an estimate of the amplitude or the rate of change or on the time-domain feature (zero crossing, mean absolute value . . .) (Hudgins et al., 1993). This information is used to specify the function to be performed—the state of the device. Once the state is selected, it may be driven at a constant speed or its speed may be controlled in a manner proportional to the myoelectric activity. Englehart et al. (1999) proposed a means to improve the accuracy of classification in the form of a time-frequency–based signal representation. This classification system task was decomposed in a multistage process illustrated in Figure 27. In the feature extraction stage, the Hamming (width = 64 ms) for STFT was preferred to other windows. The mother wavelet functions used are Coiflet-4 for the wavelet transform and Symmelet-5 for the wavelet packet transform (WPT). Feature selection methods attempt to determine the best subset of the original feature set. Feature selection has been performed using a Euclidean distance class separability (CS) criterion (Fukunage, 1990). Feature projection methods attempt to determine the best combination of the original features. Feature projection was performed using PCA. The performance of each form of signal representation has been evaluated in the context of a linear discriminant analysis (LDA) classifier and

Measured signal

Feature extraction -Time domain -STFT

Dimensionality reduction -Feature selection -Feature projection

Classification

Class labels

-LDA -MLP

-Wavelet -Wavelet packet Figure 27. A breakdown of the classification and the subject methods investigated by Englehart et al. (1999).

174

YANG AND PAINDAVOINE

an MLP classifier. The LDA is implemented easily and is considered a statistical classifier (Fukunage, 1990). A roster of 16 healthy subjects participated in this study. Four classes of myoelectric signal patterns were collected from the biceps and triceps, corresponding to flexion and extension of the elbow and pronation and supination of the forearm. Each pattern consists of two channels of 256 points, sampled at 1000 Hz. The best performance is exhibited when using a WPT–PCA–LDA combination, yielding an average classification error of 6.25%. This represents a significant improvement in comparison with the method of Hudgin et al. (1993) (9.25% of average error). D. Shape Recognition Applications and Image Classification by Content 1. Recognition and Classification of Blemishes A blemish is a stain or a damage mark on the surface of a product that renders the product unacceptable. Recognition and classification of blemishes permit analysis of the cause of the blemishes and the development of ways to prevent them in the future. The system proposed by Chen and Hewit (2000) is based on a new approach to the derivation of shift invariance by using an overcomplete wavelet transform. The patterns were digitized as binary images using a linear CCD scanner. The shape recognition of the blemish mark is equivalent to the detection of the boundary of the mark. By setting a polar coordinate system and angular sampling of the boundary, the two-dimensional patterns were converted into one-dimensional distance functions and then normalized as scaling invariant sequences. The sequences were then processed by the FT, and then the WT, to generate vectors with shift invariance. A new method of generating shift invariance using an overcomplete wavelet transform is described in Chen and Hewit (2000). The complex orthogonal estimation (COE) (Tsang and Billings, 1992) algorithm was used along with the FT and WT for comparison. Two types of wavelet, Haar and Daubechies, are used. Recognition of Cshape and Sshape was implemented using an RBF neural network. The performances of the FT and WT are superior to the COE in terms of the recognition correctness and execution time. 2. A New Multiscale-Based Shape Recognition Method Applied to Images of Industrial Tools Lin et al. (1998) proposed a new shape recognition system based on three stages.

PREFILTERING FOR PATTERN RECOGNITION

175

a. Feature Vector Extraction. Multiscale wavelet-transformed extremal evolution contains information of the contour primitives in a multiscale manner. Lin et al. (1998) have shown that wavelet-transformed extremal values at small scales contain information about the individual contour primitives (a corner or an arc: individual dominant points), whereas wavelet-transformed extremal positions and values at large scales contain information about the neighboring primitives (neighboring dominant points). For this reason, the multiscale wavelet-transformed extremal evolution of contour orientation was used to form the feature vector for discriminating shape-dominant points. The first derivative of a normalized Gaussian function is chosen as the mother wavelet, namely  2 x : ð60Þ ðxÞ ¼ x exp  2 b. Wavelet Function Normalization. This permits creation of the method scale invariant and reduction of the distortion resulting from normalization of the object contours. c. Shape Recognition. Observing the transformed wavelet orientation in the test images, most of the significant structures can persist in three consecutive scales, s1, s2, and s3. Thus, an extremum at s1 is regarded as a feature point if it can also appear at s2 and s3. The feature position is decided at the first scale because the location resolution is better at a small scale. Hopfield neural networks are applied on images of industrial tools in order to test the performance. Experimental results have shown that this method can achieve satisfactory recognition under noisy, occluded, and affine conditions using only three scales. 3. Wavelet Index of Texture for Artificial Neural Network Classification of Landsat Images In remote sensing imagery, very large amounts of data are available for research programs concerning land, atmospheric, oceanic, and climate studies with the goal of developing new models and monitoring local and global changes, as well as predicting natural and unnatural phenomena, such as volcanic eruptions, earthquakes, and the effect of human activity on our planet (Szu et al., 1997). In order to analyze all of the available data and to optimize the amount of information, remote sensing imagery needs to be organized in ways that facilitate its mining, retrieval, and future analysis. Extracting image content appears to be one of the most powerful tools that can be used for such purposes, and the most promising ways to extract

176

YANG AND PAINDAVOINE

image content start with image classification. It is in this context that Szu et al. (1997) have proposed a system of Landsat image classification integrating texture information obtained by the wavelet transform. a. Integrating Texture Information. In fact, most of the classification errors occurring at the boundary between classes is observed when using a straightforward classification method. Szu and colleagues thought that these errors are due to the fact that in pixel-based classification methods, such as neural networks, no local spatial information is integrated in the classification decision. Thus, their study consists of an initial step in assessing the impact of integrating local information, particularly texture information, in the decision. b. Co-occurrence Matrices and Wavelet Transform. In order to study aspects of texture concerned with spatial distribution and spatial dependence among local gray tones, one often calculates several cooccurrence matrices with various distances, d, and angles, Q, and the gray tone co-occurrence matrix can be defined as a matrix of relative frequencies of two gray tones separated by a given distance, d, and at a given angles, Q (Szu et al., 1997). By using the multiresolution aspect of a wavelet transform, computations of the co-occurrence matrix can be performed at different resolutions, d, while the information from different angles, Q, are integrated within the same filter. More generally, computing texture with wavelets allows one to raise their localization properties and provides a spatial density function of the co-occurrence texture. c. Wavelet Transform Used and Results. Wavelet information has to be computed by the composite edge/texture wavelet transform: the Mexican hat filter m(x) is chosen as an edge-oriented wavelet and the Morlet wavelet M(x) is used as a texture-oriented wavelet. The two-dimensional composite wavelet filter is defined by  2  x þ y2 ð1  x2 Þð1  y2 Þ W ðx, yÞ ¼ exp  2  2  2 x y ð61Þ exp  cosðk0 xÞcosðk0 yÞ þ exp  2 2 ¼ mðxÞmðyÞ þ MðxÞMðyÞ: If f (x, y) is the image to be classified, its multiresolution wavelet decomposition is defined by   Z Z 1 t  b 1 u  b2 f ðt, uÞW , MW ð f Þða, bÞ ¼ pffiffiffiffiffiffi dt du ð62Þ a a jaj with multiple values of a and b = (b1, b2).

PREFILTERING FOR PATTERN RECOGNITION

177

Wavelet transform (texture) values associated with the original values of the multiband image will be introduced on a probabilistic neural network (PNN) in order to perform the final classification of the image. Preliminary results show that it is an encouraging exploitation track for better discriminates of Landsat imagery. V. Wavelet Prefiltering Technique Applied to Handwriting Movement Analysis and Face Recognition Wavelet transform theory has found many interesting applications in digital signal processing. It provides very general techniques that can be applied to many tasks in nonstationary signal processing. This section presents two applications using the wavelet transform as a prefiltering technique: handwriting movement analysis and face recognition. We applied the multiresolution analysis using the generalized Canny Deriche (GCD) filter as wavelet function, first to filter the original signal and second to detect the sharp variations in the filtered signal. Comparisons with other methods, which explored the kernel estimates technique, show that the GCD filter provided better filtering performance. We briefly discuss the main filtering methods currently used in the literature and raise the interest of a time–frequency representation of a signal. We describe the GCD filter. Then we detail how to apply this filter with a multiresolution manner in movement analysis of handwriting and drawing. An application of this wavelet prefiltering technique for face recognition concludes this section. A. Filtering Techniques, Time–Frequency Representations, and Generalized Canny Deriche Filter 1. Filtering Techniques and Time–Frequency Representations The existing standard signal processing methods for filtering are based primarily on two approaches: linear filtering (finite impulse response, FIR filters) and statistical methods. The noise to be eliminated is present in highfrequency components, and the use of FIR filters such as second-order Butterworth filters leads to a noticeable reduction of this noise. Determination of parameters such as cutoff frequency or transition band frequencies requires a frequency analysis of the noisy signal. Statistical methods include nonparametric regression techniques such as estimates established from spline functions or kernel estimates (Marquardt and Mai, 1994). However, the signal frequency characteristics are

178

YANG AND PAINDAVOINE

completely ignored in this method, which may turn out to be prejudiced in the case of complex nonstationary signals. The statistical approach neglects the frequency representation of the signal in favor of a temporal representation of the signal. Direct extraction of significant information (duration of phases, changes in rhythm, discontinuities, etc.) from a temporal representation of the signal is quite acceptable when a pseudodeterminist model of the signal can be applied, i.e., when the variations across time are not too random. If such a model cannot be a priori assumed, smoothing and differentiating the signal while considering its frequency components also become necessary. For nonperiodic signals, the Fourier frequency analysis (more precisely, the integral of Fourier) represents the signal as a superposition of sinusoidal waves of all possible frequencies of which the contribution is coded through their respective amplitudes. This method is powerful for stationary signals but is limited for nonstationary signals such as speech, music signals, and handwriting signals. In this case, the wavelet transform allows decomposition of the signal into functions of both time and frequency. 2. Wavelet Transform and Generalized Canny Deriche Filter The wavelet transform can be applied in order to extract the characteristics of a signal presenting sharp variations. For this scope, a multiscale analysis first introduced by Mallat and Zhong (1992) is usually used. Most multiscale sharp variation detectors smooth the signal at various scales and detect sharp variation points from their first or second derivative. The extrema of the first derivative corresponds to the inflection points of the smoothed signal. We call a smoothing function any function (t) whose integral is equal to 1 and converges to 0 at infinity. We suppose that (t) is differentiable and define the wavelet function (t) as ðtÞ ¼

dðtÞ : dt

ð63Þ

We denote s ðtÞ ¼ 1s ðstÞ as the dilation of the function (t) by a scaling factor s. Thus the wavelet transform of the signal x (t) at scale s and time b can be defined as Z 1 xðtÞ s ðt  bÞdt ¼ xðtÞ  s ðtÞ, ð64Þ Ws x ¼ X ðb, sÞ ¼ 1

from which it can be derived that   ds dðx  s Þ Ws x ¼ x  s ðtÞ ¼ s ðtÞ, dt dt where * represents the convolution operator.

ð65Þ

PREFILTERING FOR PATTERN RECOGNITION

179

Eq. (45) shows that the wavelet transform Wsx is the first derivative of the signal smoothed at scale s. For must purposes, wavelet function is not required to keep a continuous scale parameter s. In order to allow fast numerical implementations, we impose the condition that the scale varies along the dyadic sequence (2 j ) with j 2 Z. We denote 2 j (t) the dilation of (t) by a factor 2j 1 t

: ð66Þ 2 j ðtÞ ¼ j 2j 2

Using Eq. (64), the discrete wavelet transform of the signal x(t) at scale 2 j is defined by the following convolution product, W2 j x ¼ x 

2 j ðtÞ:

ð67Þ

In order to introduce the multiresolution analysis applied to signal processing, Mallat and Zhong (1992) defined (t) with a quadratic spline function, which is the derivative of the cubic spline function (t). In our application of wavelet transforms to handwriting or drawing signals and face recognition, we used multiresolution analysis using the ‘‘GCD filter’’ as wavelet function. The GCD filter, introduced by Bourennane and Paindavoine in 1993, possesses a good signal-to-noise ratio (SNR). The wavelet and smoothing  functions using the GCD filter are defined as follows:  x0 Cs ðk0 sxemsx þ emsx  esx Þ ð68Þ s ðxÞ ¼ x0 Cs ðk0 sxemsx  emsx  esx Þ s ðxÞ ¼ As ½ðk0 ms2 jxj  k0 s þ msÞemsjxj  ms2 esjxj ;

ð69Þ

where k0 and m are adjustable parameters for good detection and good localization of the sharp variations of the signal. These filters give an optimum SNR and localization for k0 ¼ 0.564 and m ¼ 0.215 (Bourennane et al., 1993). The wavelet and smoothing  functions are represented in Figure 28. Eqs. (68) and (69) represent the continuous wavelet function and the smoothing function, respectively. The corresponding wavelet discrete transforms (dyadic wavelet function 2 j and discrete smoothing function 2 j) are obtained using the Z transform applied to Eqs. (68) and (69). These Z transforms derive two third-order recursive filters moving in opposite directions. Details for the computing of these numerical filters are given in the Appendix. In order to illustrate multiresolution analysis, a noisy signal presenting slow and fast variations can be used. Figure 29 illustrates the advantages of the multiscale decomposition: (1) sharp variations on the smoothed signal can be eliminated if only the scale j ¼ 0 or j ¼ 1 is considered and (2) a sharp variation can be precisely detected if the scale j = 3 is considered.

180

YANG AND PAINDAVOINE

0.4 0.3 0.2 0.1 0 −0.1 −0.2 −0.3 −0.4 −10 −8 −6 −4 −2 0

2

4

6

Figure 28. Wavelet

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 8 10 −10 −8 −6 −4 −2 0

2

4

6

8

10

(left) and smoothing  (right) filters.

Original noisy signal

Smoothed signal using scale j = 3

Wavelet component at scale j = 0

Wavelet component at scale j = 1

Wavelet component at scale j = 2

Wavelet component at scale j = 3

Figure 29. Illustration of the multiresolution analysis.

B. Wavelet Filtering Technique and Movement Analysis in Handwriting and Drawing This section shows how the the wavelet transform can be fruitfully adapted to movement analysis in drawing and handwriting. In order to facilitate the comparison of results in this domain, we adopted the same procedure as

PREFILTERING FOR PATTERN RECOGNITION

181

used by Marquardt and Mai (1994), who explored the kernel estimates technique for smoothing handwriting signals. We applied multiresolution analysis using the GCD filter as a wavelet function (1) to filter the original signal (use of the 2 j filters) and (2) to precisely detect the sharp variations in the filtered signal (use of the 2 j filters). An optimal calibration of the filter parameters was made for a synthetic signal, and the filtering performance was assessed by estimating the residual errors in the position, velocity, and acceleration signals. The GCD filter provided better performance than other currently used filtering techniques. We also present how this technique can be applied to the problem of signal segmentation in the case of drawing movements. 1. Handwriting and Drawing Signals As for language and perception, the study of handwriting and drawing performances has progressively become a stimulating object of research, attracting scientists from very different horizons. For instance, neurophysiologists, experimental psychologists, and neurologists are concerned with handwriting and drawing in the context of an understanding of human movement planning, programming, and execution, including their respective disturbances. The interests of electronic engineers and computer scientists in handwriting and drawing are related to image processing in general, such as automatic pattern recognition or signature authentification, as well as to signal processing, through, for instance, the improvement of the technical performances of digitizers or the refinement of valid techniques for the analysis of handwriting signals. Handwriting and drawing can be studied both within a task-oriented approach, which focuses on the analysis of the products, i.e., the static traces or images forming handwritten characters or drawings, and within a processoriented approach, which deals with the analysis of the actual movements being performed during the handwriting or drawing task, usually recorded by means of a digitizer. The analysis of pen-tip displacements over a digitizer provides very valuable information for the study of human motor control [e.g., discovery of motor invariant and laws (Viviani and Terzuolo, 1980)] and movement disturbances [e.g., significant progress in the understanding of Parkinson’s disease (Teulings and Stelmach, 1991; Vinter et al., 1996)]. Globally, this approach requires computing different parameters from the first (velocity), second (acceleration), and sometimes third (jerk) derivatives of the pen-tip displacements as a function of time. Clearly, the validity of results obtained in handwriting movement analysis is partly dependent on the digitizer’s technical performances in the spatial and temporal domains and on the precision of the filtering techniques used

182

YANG AND PAINDAVOINE

to reduce the noise contained in the data. Different types of errors can be introduced by the very process of digitizing the pen-tip displacements. Ward and Phillips (1987) have distinguished several of them, such as missing coordinates and nonlinearity as examples of spatial errors or temporal asynchronisms in the sampling of x and y coordinates as an example of a temporal error. As pointed out by Marquardt and Mai (1994), when a kinematics movement analysis is desired, the greatest difficulty emerges from the random errors introducing stochastic noise in the signal because this noise will necessarily be largely amplified as successive derivatives of the signal are computed, with differentiation acting as a high-pass filter. In addition to the performances of the digitizer, noise can emerge from different sources when human movement is recorded, such as noise caused by stretch reflexes, a physiological tremor, or mechanical oscillations due to the spring-like characteristics of limbs (Van Galen et al., 1990). Having at one’s disposal highly efficient filtering methods appears crucial, and several solutions are used in the literature. Most of them deal with standard linear filter methods for digital low-pass filtering. Teulings and Maarse (1984) used a filter cutoff frequency of 10 Hz, but other parameters have been adopted in the literature [a transition band from 8 to 24 Hz in Teulings et al. (1986) and from 10 to 30 Hz in Schomaker and Thomassen (1986), for instance]. Clearly no agreement for the determination of these filter parameters is established in the literature. The main problem here is related to the representativity of the handwriting samples used for the determination of cutoff frequencies. Some solutions consist of nonparametric regression methods, such as kernel estimates, performant methods for an automatic definition of the filter parameters having been proposed (Amico and Ferrigno, 1992). The kernel estimate approach has been tested directly for handwriting signals and provides better results than FIR filters or second-order Butterworth filters for the first and second derivatives (Marquardt and Mai, 1994). The present section suggests that the wavelet filtering method provides an optimal solution for the movement analysis of handwriting and drawing signals. We will show that the solution is optimal both for the filtering procedure and for the precise location of significant positions occurring in the course of movement trajectory, these allowing a segmentation of the movement into meaningful units. 2. Filtering and Signal Segmentation Using Multiresolution Analysis a. Choice of Optimal Parameters for Filtering. In order to select the optimal scales for the smoothing of handwriting signals, synthetic data sets have to be studied. As mentioned previously, Marquardt and Mai (1994)

PREFILTERING FOR PATTERN RECOGNITION

183

made a similar study for the adaptation of the kernel estimate approach to handwriting signals, providing comparison tests with other smoothing methods. We therefore decided to consider the same synthetic data sets as those studied by Marquardt and Mai in order to facilitate a further comparative assessment of our method. In order to simulate hand-written loops as optimally as possible, these authors used sine waves at 3, 5, and 7 Hz, computed with an amplitude of 10 mm sampled at 165 Hz, adding a Gaussian noise of (x) ¼ 0.058 mm to the sine waves in order to simulate positional errors (see justifications of the approach in Marquardt and Mai, 1994). The determination of optimal filtering parameters is done with position, velocity, and acceleration signals. In order to study position, the original noisy signal is filtered with the smoothing GCD filter and then the positional error is measured in the filtered signal. In order to study velocity and acceleration, the first and second derivatives of the filtered signal are computed and, subsequently, the velocity and acceleration errors are measured. These computations are done with a scale ranging from j ¼ 5 to j ¼ 5 (the scale factor is s ¼ 2 j). The optimal smoothing scale can thus be obtained by minimization of the different mean square errors (MSE ) over the data points N: MSE ¼

N 1X 2 N i¼1 i

ð70Þ

with i, which is the difference between estimated and measured values. Figure 30 displays the results, showing the relation among MSE, scale, and signal frequency. Figure 30 reveals the existence of an optimal smoothing scale with the presence of minima on the different curves. For each class of signal, position, velocity, and acceleration, the scale that gives the best filtering result is a function of signal frequency. In order to improve the quality of filtering, the choice of the smoothing scale necessarily lies in a compromise among the three available values (at 3, 5, and 7 Hz), which can be achieved by determining the best minimization of the three errors. For this scope, the following equation was computed: Error ¼ Min½error23Hz þ error25Hz þ error27Hz :

ð71Þ

The final choice of the scales corresponding to sampled signals with a 167Hz frequency is as follows: 1.1 for position, 0.3 for velocity, and 0.4 for acceleration. b. Comparison Tests. We compared the results obtained with the GCD filter with those reported by Marquardt and Mai (1994). The evaluation

184

YANG AND PAINDAVOINE

Position

0.1 5 Hz 0.05

7 Hz

3 Hz

0 −2

0

−1

1 Scale

2

3

4

Velocity

2

1

0 −2.5

−2

−1.5

−1

−0.5

0 Scale

0.5

1

1.5

2

2.5

Acceleration

100

50

0 −2.5

−2

−1.5

−1

−0.5 Scale

0

0.5

1

1.5

Figure 30. Relation among MSE, scale, and frequency.

method used by these authors has been applied successively to three filters: a low-pass 13-Hz second-order Butterworth filter, a kernel estimation filtering described by Marquard and Mai, and the GCD filter. We therefore computed the standard deviations of the residual errors in the filtered signals and their respective derivative for a 1-Hz sine wave sampled at 167 Hz. Table 1 reports the obtained results. Results obtained with the GCD filter show a significant improvement of the smoothing performances in comparison to the other filters. The GCD filter offers a particularly good performance for the acceleration signal. However, these performances are available only after a calibration of the filter parameters. Optimal scale values depend on the sampled frequency of the noisy signal. For the sake of illustration, a 100-Hz sampled signal should be filtered using 2, 1.1, and 0.7 as scale values for position, velocity, and acceleration, respectively. A few technical details may be useful. The software discussed in this study is written in Matlab 5.0 and is PC compatible. The time for filtering 1000 equidistant data points at a sampled rate of 100 Hz is 4.5 ms using a

185

PREFILTERING FOR PATTERN RECOGNITION TABLE 1 Standard Deviations of Residual Errors for Position (pos.), Velocity (veloc.), and Acceleration (acc.) as a Function of the Filter

Pos. (mm) Veloc. (mm/s) Acc. (mm/s2)

Noisy signal

Butter worth 13 Hz

Kernel estimation

GCD opt.

0.058 13.78 3850

0.025 1.19 84

0.021 0.89 21.70

0.015 0.27 5.04

Pentium 600 MHz, and it is the same for position, velocity, and acceleration. This time is independent from the scale because we use the same third-order filters for these three signals, with different coefficients adapted to the chosen scale as described in the Appendix. This fast filtering time allows filtering of the original signal in almost real time. c. Segmentation. Because the essence of multiscale resolution analysis lies in the detection of variations in the signal, it can also be fruitfully used for resolving one of the main problems in handwriting and drawing research, that of automatic segmentation. Several algorithms and methods have been developed for recovering meaningful pieces of information from the recorded signal, such as letters (handwriting) or geometrical drawing segments (see Wesolkowski, 1996; Bontempi and Marcelli, 1996). We now briefly report how we adapted the multiresolution analysis to this question. The method was applied to the segmentation of angular figures, made of three segments (see Bontempi and Marcelli, 1996; Deshief et al., 1996). The scope was to resituate the three segments and the pauses that possibly occur at angles (Figure 31), which display the tangential velocity profile (Figure 31B) when drawing an obtuse figure (Figure 31A) delimited by the circles on the trace (Figure 31E). The primary interest of our method was in combining both spatial detection of the relevant local curvature maxima (the angles) and detection of the relevant local minima of velocity. Detecting only the velocity minima appeared too restrictive because as soon as the movement departs from ballistic-like movements, such as in children or in patients, several velocity minima are found between two relevant spatial segmentation points. The algorithm of spatial detection was aimed at providing candidate positions for a segmentation realized further in the velocity domain, eliminating minor accidents appearing along the trace. For our application, only positions associated with the angles had to be detected. These angular positions were characterized by important changes occurring in an angular direction, which was defined as

186

YANG AND PAINDAVOINE

A

B

1200

100

120

1400

1000

80

800 60 600 40

400

20

200 0 8600 8800 9000 9200 9400 9600 9800 10,000 10,200

C

0

0

20

40

60

80 100 120 140 160 180 200

20

40

60

80 100 120 140 160 180 200

D

1.5

60 50 1 40 30 0.5

20 10

0 20

40

60

80

100

120

140

0

E 1400 1200 1000 800 600 400 200 0 8600 8800 9000 9200 9400 9600 9800 10,000 10,200

Figure 31. Segmentation of angular figures.

PREFILTERING FOR PATTERN RECOGNITION

  dX : dðtÞ ¼ arctan dY

187 ð72Þ

More precisely, the critical angular positions corresponded to the local maxima in d(t), the variation of angular direction. In reference to Lee et al. (1995), we used the GCD smoothing filter at two different scales, a and b with a < b, in such a way that two objectives were aimed at (1) a precise localization of the critical spatial positions (with da) and (2) a strong reduction of the minor accidents in the trace (with db). Then, for each maxima xi detected with da, its value yi with db was determined and the ratio xyii was computed. Only the xi of whose ratios xyii were above a given threshold were considered as candidates for segmentation points. Figures 31C and 31D show the da, db smoothing and the resulting spatial segmentation points for the drawing of an obtuse figure. Each maximal curvature position obtained from the previous spatial detection defined only the area where a segmentation key position was present, not its precise location, because the optimal values of the smoothing scales obtained for the spatial detection (parameters a and b, estimated at 0.8 and 0.2 in our application) did not correspond to the optimal values for the smoothing of velocity signal (0.3). Furthermore, when a pause was present at angles, the spatial procedure did not provide two segmentation points, respectively, for the beginning and the end of the pause. Refinement of results issued from spatial detection using the velocity signal was necessary. This analysis was done with the tangential speed smoothed by the GCD filter at an optimal scale (0.3). Each area given by the spatial procedure was projected onto the smoothed velocity signal, and a search was conducted in the x < 0 and in the x > 0 directions until the half-height of the first peak located in the direction of the search was attained. A gradient descent search was then carried out until the first local velocity minimum. In the case of a pause, two successive minima were obtained with this procedure. 3. Discussion Most of the criteria used for the assessment of handwriting and drawing movements are derived from velocity and acceleration signals. The degree of fluency of movements, for instance, is often considered as an indicator of the level of coordination in movement production (Van Galen, 1991) and requires the analysis of the velocity profile, if not the acceleration profile (Hogan and Flash, 1987) of the recorded signal. Smoothing the original recorded signal with a filter resistant to the differentiation process is crucial to the validity of further analyses. In this perspective, considerable improvements in available filtering techniques have been made (Woltring, 1985; Marquardt and Mai, 1994).

188

YANG AND PAINDAVOINE

Here we explored a new alternative, the wavelet transform, and the reported study concludes there are very good performances for this filtering technique. As indicated clearly by residual errors measurement, when compared to other currently used filters, the best performances are offered by the wavelet transform, both for velocity and acceleration signals. However, one must consider that these optimal performances require a calibration of the filter parameters on the basis of the sampled frequency of the recorded signal, but this limit appears as a minor drawback because the variety of sampling frequencies used by researchers in the domain is not large (100 to 200 Hz essentially). The very good results obtained with wavelet analysis in different applications such as speech, music, and image analysis are confirmed in the application for handwriting and drawing signals (see also Lee et al., 1996; Sasi et al., 1997). A second part of our study was devoted to the question of segmentation of the recorded signals into meaningful segments for movement analysis. Again, the wavelet transform was used for this scope and provided interesting results, as also obtained by Lee et al. (1995) for a similar problem. Of course, further analysis is still needed here for an assessment of the power of the method if an automatic algorithm would be desired. To what extent such an approach can bring some solutions for the recognition of drawn or written characters is also still an open question. C. An Image Filtering Technique Combining a Wavelet Transform with an Autoassociative Memory: Application to Face Recognition Linear autoassociative memories are one of the most simple and wellstudied neural network models (Kohonen, 1977; Anderson et al., 1977) (see Section II.B). They are widely used as models for cognitive tasks, as well as pattern recognition, or digital signal processing, in part because they are formally equivalent to well-known techniques such as the Karhunen–Loe`ve transform or principal component analysis (Valentin et al., 1994). Even though linear autoassociators are known to be quite robust when noise is added to the patterns to be recognized, their performance is rather poor when a lot of noise is added to the stimulus. One of the ways to improve performance could be to use some pre- and postprocessing on the patterns to be recognized. This section evaluates the performance of a preprocessing technique using the wavelet transform applied to face images. In order to improve the performance of a linear autoassociator, we examined the use of several preprocessing techniques. The gist of our approach is to represent each pattern by one or several preprocessed (i.e., filtered) versions of the original pattern (plus the original pattern). First, we

PREFILTERING FOR PATTERN RECOGNITION

189

compared the performance of several preprocessing techniques (a plain vanilla version of the autoassociator as a control, a Sobel operator, a GCD operator, and a multiscale GCD operator) and a Wiener filter on a pattern completion task using noise-degraded versions of stored faces. We found that the multiscale Canny-Deriche operator gives the best performance of all models. Second, we compared the performance of the multiscale CannyDeriche operator with the control condition on a pattern completion task of noise-degraded versions (with several levels of noise) of learned faces and new faces of the same or another race than the learned faces. In all cases, the multiscale Canny–Deriche operator performs significantly better than the control. This section is organized as follows. First, we describe the linear autoassociator model applied to face images. Second, we compare the wavelet approach with other preprocessing or filtering techniques. Third, we look at the performance of the wavelet approach under various conditions of noise degradation. 1. Linear Autoassociators and Eigenvalue Decomposition a. Linear Autoassociator Description. The advantage of linear associators in comparison with nonlinear models is that they provide integration of a very large number of cells in the network. Their implementation is quite easy because they can be analyzed in terms of the singular value decomposition of a matrix (Valentin et al., 1994; Abdi, 1994). In addition, linear models constitute a first processing stage for numerous applications using more sophisticated approaches (for reviews, see Valentin et al., 1994). For example, Kohonen (1984) showed that an autoassociative memory could act as a content addressable memory for face images. Figure 32 gives an illustration. When linear autoassociative memories are applied to images, the first step is to transform each digitized image into a (image) vector by concatenating the columns of the matrix of the pixel values of the image. Images are ‘‘stored’’ into a connection-weight matrix, which models neural synaptic connections between neural cells associated with the image pixels (see Figure 33). In our description, we follow closely the formulation detailed in Abdi (1994). The patterns to be learned are represented by L  1 vectors ak where k is the stimulus number. The components of ak specify the values of the pattern to be applied to the L cells of the input layer for the kth stimulus. The responses of the network are given by L  1 vectors ok. The complete set of K stimuli is represented by an L  K matrix noted A (i.e., ak is the kth column of A. The set of K responses is represented by an L  K matrix noted O. The L  L synaptic weight connection matrix between the L input cells is

190

YANG AND PAINDAVOINE

Figure 32. Illustration of a content-addressable memory for faces. Images of faces were stored using an autoassociative memory. (Top) Two stimuli given as a key to probe the memory. (Bottom) Responses of the memory. The memory is able to reconstruct a face from an incomplete input (cf. Abdi, 1994).

3

Cellules 1

5

2

1

3

2

4

10

5

9

6

2 3

Matrice de connexion W

w23

1

w23

6

4 5

Figure 33. The linear autoassociator is applied to images: transform each digitized image into a vector by concatenating the columns of the matrix of the pixel values of the image.

denoted W. Learning occurs by modifying values of the connection weight between cells as explained later. The response (or estimation) of the model to a pattern x (which may or may not have been learned) is obtained as ^ ¼ Wx: x

ð73Þ

Because autoassociators are generally interpreted as content addressable memories, their performance is evaluated by comparing the output of the system with a test pattern that can be a copy or a degraded version of one of the patterns learned previously by the system. This is achieved by computing

PREFILTERING FOR PATTERN RECOGNITION

191

similarity measures (usually a cosine) between input and output. The ^ and x is an index of the quality of coefficient of correlation between x the estimation; hence the larger the similarity between input and output, the better the performance. In order to achieve a high level of performance, several iterative learning rules have been proposed. The most popular one is clearly the Widrow–Hoff learning rule. This is an iterative procedure that corrects the connection matrix W using the difference between the target response and the actual response of the network. In matrix notation, the Widrow–Hoff rule is written as Wðtþ1Þ ¼ WðtÞ þ ðak  WðtÞ ak ÞaTk ,

ð74Þ

with W[t] being the weight matrix at step t,  being a small positive constant, and the index k being chosen randomly. b. Eigen and Sigular Value Decomposition: The PCA Approach. Abdi and colleagues developed a simple method of implementing the Widrow– Hoff algorithm by using the eigen decomposition of W or singular decomposition of matrix A. These decompositions give rise to PCA. Eigenvectors of a matrix are vectors that have the property that, when multiplied by the matrix, their length is changed but their direction remains unchanged. More formally, if u is an eigenvector of W, Wul ¼ ul ll ,

ð75Þ

where ll is a scalar called the eigenvalue associated with the eigenvector ul. Traditionally, the set of eigenvectors of a given matrix is represented by a matrix U in which the first column represents the vector with the largest eigenvalue, the second column the eigenvector with the second largest eigenvalue, and so on. The corresponding eigenvalues are represented by a diagonal matrix L. Using this notation, matrix W can be expressed as a function of its eigenvalues and eigenvectors, W ¼ ULU1 :

ð76Þ

In the particular case where W is symmetric and can be obtained as the cross-product of a matrix by its transpose (W is positive semidefinite), its eigenvalues are all positive or zeros and its eigenvectors are orthogonal. In this particular case, U1 ¼ UT

ð77Þ

W ¼ ULUT :

ð78Þ

and

192

YANG AND PAINDAVOINE

If the eigenvectors are normalized, then UUT = I. The notions of eigenvectors and eigenvalues of a positive semidefinite matrix can be used to define the singular value decomposition (SVD) of a rectangular matrix. Specifically, any rectangular matrix A can be expressed as A ¼ UDVT

ð79Þ

Wðtþ1Þ ¼ WðtÞ þ ðA  WðtÞ AÞAT

ð80Þ

where U is the matrix of eigenvectors of AAT, AAT = ULUT, with UUT = I; V is the matrix of eigenvectors of AT A, AT A = VLVT, with VVT = I; and 1 D is the diagonal matrix of singular values of A, D = L2 with L matrix of T T eigenvalues of AA and A A. Abdi et al. (1998) pointed that the Widrow–Hoff equation [see also Eqs. (21) and (74)]

can be expressed in terms of the eigen decomposition of W (or the SVD of A) as Wðtþ1Þ ¼ UFðtþ1Þ UT with Fðtþ1Þ ¼ ½I  ðI  LÞtþ1 :

ð81Þ

limt!1 ðI  LÞtþ1 ¼ 0:

ð82Þ

When  is smaller than 2l1 max ;

Therefore, at convergence, W reduces to Wð1Þ ¼ UUT

ð83Þ

which is the value used in this section. The matrix U is an L  N matrix with N being the number of nonzero eigenvalues. Typically, N is significantly smaller than L (i.e., N  L). As a consequence, using U directly instead of W will lead to an important gain in processing speed and storage. For example, when dealing with a face recognition application, matrix W was a 33,975  33,975 matrix, whereas eigenvector matrix U was only a 33,975  400 matrix. In terms of the eigenvector matrix U, the response ^ to a pattern x [Eq. (73)] is obtained as of the model x ^ ¼ UUT x: x

ð84Þ

2. Preprocessing Using Multiscale Edges The goal of learning is to find values for the connections between cells such that the response of the model approximates the input as well as possible. To assess the performance of the model, degraded versions of the previously

193

PREFILTERING FOR PATTERN RECOGNITION

learned stimuli are presented to the model as a test. If learning has been successful, then the response pattern will be more similar to the original pattern than the degraded stimulus was (for an illustration, see Kohonen, 1977). In other words, autoassociators can act as pattern completion devices. Here, we explore different approaches for improving the performance of a linear autoassociator storing face images. The general strategy is to store, in addition to the original images, several filtered versions of the images (see Figure 34). We refer to this technique as preprocessing. Then the model is evaluated by its reconstruction performance when presented with probes that are versions of the original faces degraded by the addition of Gaussian random noise. Because we are interested in image patterns, we choose filtering techniques meaningful in this context. Because it is generally agreed that edges are essential for recognition (Jia and Nixon, 1995), we decided to increase their importance in the image. Quite a large number of algorithms have been proposed in the literature for performing edge extraction. We decided to implement three algorithms. 1. The Sobel operator (a differential operator) is considered a standard procedure well suited for noiseless images. The Sobel operator was implemented with a convolution and a 3  3 mask. 2. The GCD operator because it is known to be optimal for edge extraction in noisy images (Deriche, 1987; Bourennane et al., 1993). 3. The multiscale GCD edge detector (Bourennane et al., 1993), which is equivalent to finding local maxima of a wavelet transform as suggested in Mallat and Zhong (1992). The GCD filter is a separable filter when applied to two-dimensional images. Its impulse response for a one-dimensional signal (because of its separability, the filter can be seen as two one-dimensional filters) is given by

s ðxÞ ¼



Cs ðk0 sxemsx þ emsx  esx Þ Cs ðk0 sxemsx  emsx þ esx Þ

x0 x0

ð85Þ

with k0 = 0.564, m = 0.215, and where s = 2 j is the scale factor (with, in our case, j 2 {0, 1, 2, 3}) and x being the pixel position. Figure 34 displays the impulse response of the generalized CannyDeriche filter for different scales. This method is implemented as a wavelet transform using a convolution between the image and the edge detection filter for different scales (s = 2 j ). As a result, this filter detects edges occurring at different scale resolutions in the image (Mallat and Zhong, 1992).

194

YANG AND PAINDAVOINE

1 s=8 s=4 s=2 s=1

0.5

0

−0.5

−1

−1.5 50

55

60

65

70

75

80

85

90

95

100

Figure 34. Impulse response of the generalized Canny-Deriche filter for different scales.

In order to compare these different techniques, we implemented these three models along with two control conditions corresponding to a standard associator, and for denoising used a Wiener filter (because it is a standard algorithm applied one image at a time). All the models were tested using the same procedure. The patterns stored were a set of 80 face images of size 225  151 (with 16 gray levels per pixel). The performance of each model was assessed as follows. Each face was degraded by adding to each pixel a random number [chosen such that the noise values belong to the interval (0 to 45)]. The degraded face was then recalled by the model (or filtered in the case of the Wiener filter model). The correlation between the original face and the model-reconstructed face reflects the performance of the model for this face: the higher the correlation, the better the performance. Specifically, the models were a Wiener filter applied directly to the noise-degraded stimulus and four autoassociators (see Figure 35). 1. A standard autoassociator storing the original 80 face images. 2. An autoassociator storing the original 80 face images plus, for each face, a Sobel-filtered image of the face (hence a total of 160 face images). 3. An autoassociator storing the original 80 face images plus, for each face, a GCD-filtered image of the face (hence a total of 160 face images).

PREFILTERING FOR PATTERN RECOGNITION

195

4. An autoassociator storing the original 80 face images plus, for each face, four wavelet-transformed (by a multiscale GCD filter) face images (one face image per scale resolution, hence a total of 400 images). For the last three models, the complete set of patterns to be learned (matrix A) is composed of original and filtered images. The eigenvector matrix U and the synaptic connection matrix W have to be obtained using Eqs. (79) and (83). Figure 36 displays an example of the responses of the models to the test face. The top panels present a face learned previously by the system and a stimulus with additive random noise added used as a probe to evaluate the

Figure 35. Patterns to be learned by four autoassociators: Filtered images have been obtained, respectively, with the Sorel operator, the optimized Canny-Deriche operator, and the wavelet transform.

196

YANG AND PAINDAVOINE

Figure 36. Response of the models. (Top) A target face (learned previously by the autoassociator) and the stimulus used to test the model (the stimulus is a noisy version of the target). (Bottom) Responses of the different models. The wavelet model performed best.

1 Mean correlation

0.938 0.9

0.852 0.796

0.871

0.828

0.8

0.7

Standard

Wiener Sobel Deriche Type of models used

Wavelet

Figure 37. Mean correlation between the model responses and their targets (the higher the correlation, the better the performance). The wavelet model performed best.

performance of the models. The bottom panels show the estimation of the original face by a standard autoassociator, a Wiener filter, an autoassociator plus Sobel preprocessing, on autoassociator plus a GCD filter, and an autoassociator plus a wavelet transform. The quality of recognition (estimation) can be measured by computing the ^ (i.e., the response of model) and xk (i.e., the original cosine between vector x stimulus, which is also the desired response or target). Figure 37 gives the

PREFILTERING FOR PATTERN RECOGNITION

197

average correlation between response and target for the five models used. Clearly, the standard method is the worst of all. Preprocessing the images improves the performance of the autoassociator, and the wavelet transform gives the best result. In conclusion, the multiscale resolution (i.e., wavelet preprocessing) approach leads to the best performance for the autoassociator. Therefore, we decided, in what follows, to consider only this approach. 3. Pattern Completion of Noisy Patterns We have applied the multiscale edge preprocessing to store a set of 80 Caucasian faces (40 males and 40 females). In order to evaluate the effect due to preprocessing, we tested the models with different levels of Gaussian random noise added to the test stimulus. Learning was implemented as described previously. For simplicity, we decided to keep only two models: the standard autoassociator and the wavelet-enhanced autoassociator. Testing was implemented as described previously except that faces were tested under four different levels of noise. The noise intensity was chosen such that its range was, respectively, [0..15], [0..30], [0..45], and [0..60]. Figure 38 displays an example of the noisy test stimuli used along with the response of each model (standard and wavelet).

Figure 38. (Top) Four stimuli obtained by adding to a target stimulus a noise component of magnitude equal to one, two, three, and four times the magnitude of the signal of the original stimulus. (Middle) Responses produced by the standard autoassociator. (Bottom) The response of the autoassociator when learning is ‘‘enhanced’’ with wavelet-filtered stimuli.

198

YANG AND PAINDAVOINE

Figure 39. Stimuli and responses. (Top) Performance for a new Caucasian face and (bottom) for a new Japanese face. (Left to right) A noise-degraded stimulus (the magnitude of the noise is equal to twice the magnitude of the original signal); the response of the standard autoassociator; and the response of the wavelet-enhanced autoassociator.

We also decided to explore the performance of the model with three different types of face stimuli: (1) previously learned faces, (2) new faces similar to the learned faces, and (3) new faces coming from another race than the learned faces. This was done in order to evaluate the robustness of the models in terms of response generalization to new stimuli. Figure 39 displays, as an example, the responses of both models for two new faces (from top to bottom): (1) a new face similar to the set of learned faces (Caucasian face) and (2) a new face different from the set of learned faces (Japanese face). The autoassociator trained with the standard learning is not capable of producing distinguishable responses. As can be seen in Figure 39, better results are obtained with wavelet preprocessing. Figure 40 displays the mean correlation between noiseless face images and the output for each model (1) for 80 previously learned Caucasian faces, (2) for 80 new Caucasian faces, and (3) for 80 new Japanese faces. In all cases, preprocessing the image improves the performance of the autoassociator with the improvement being more important when the noise added is larger. This section explored the effects of storing, in a linear autoassociator, filtered versions of face images in addition to the original images. Compared to the Sobel operator and the simple GCD operator, the multiscale GCD operator (i.e., a wavelet filter) gives the best performance for a pattern completion task involving degraded face images. The multiscale generalized Canny-Deriche operator produces better generalization performance than the control with or without noise added to the image. The larger the amount

199

PREFILTERING FOR PATTERN RECOGNITION

Mean correlation

1

Learned Caucasian faces

0.9

0.8

New Caucasian faces

0.7

New Japanese faces

0.6

With wavelet Without wavelet 0

15 30 45 Maximum noise magnitude

60

Figure 40. Mean correlation between model response as a function of the magnitude of the noise for Caucasian faces learned previously, new Caucasian faces, and new Japanese faces. Filled lines correspond to the wavelet-enhanced model, and dotted lines correspond to the standard autoassociator. The wavelet-enhanced autoassociator always performed best.

of noise added, the larger the improvement in performance. This research confirms better performances obtained using wavelet transform representation for face tracking and face recognition (Rangannath and Arun, 1997; Kru¨ger et al., 1999; Kru¨ger and Sommer, 2000; Huntsberger et al., 1998). Appendix Wavelet Filter The wavelet filter

2j

2 j ðzÞ

with þ1 X 0

is defined as follows: ¼

þ1 X 0

n  2 j ðnÞZ

n  2 j ðnÞZ

¼

þ

þ1 X 0

þ 2 j ðn

þ 1ÞZ ðnþ1Þ

a1 Z 1 þ a2 Z2

ð1  em2 j Z1 Þ2 ð1  e2 j Z 1 Þ

200

YANG AND PAINDAVOINE þ1 X 0

n  2 j ðnÞZ

a1 Z 1 þ a2 Z 2

¼

ð1  em2 j Z 1 Þ2 ð1  e2 j Z 1 Þ j

,

j

where a1 = C2 j [(1  k02j) em2  e2 ] and a2 = C2 j [(1 + k02 j) em2 j+1  em2 +1)] C2 j is a normalization constant given by ð1  em2 Þ2 ð1  e2 Þ : þ ð1 þ k0 2 j Þem2 j 2 j  em2 jþ1  e2 j j

C2 j ¼

ð1  k0 2j Þem2 j

j2 j

j

The impulse response y(n) corresponding to the filter þ

is given by



yðnÞ ¼ y ðnÞ þ y ðnÞ yþ ðnÞ ¼ a1 xðn  1Þ  a2 xðn  2Þ þ b1 yþ ðn  1Þ  b2 yþ ðn  2Þ þ b3 yþ ðn  3Þ y ðnÞ ¼ a1 xðn þ 1Þ þ a2 xðn þ 2Þ þ b1 y ðn þ 1Þ  b2 y ðn þ 2Þ þ b3 y ðn þ 3Þ j

j

with b1 ¼ 2em2 þ e2 , b2 ¼ em2

jþ1

j

j

þ 2em2 2 , and b3 ¼ em2

jþ1 2 j

.

Smoothing Filter The smoothing filter 2j is defined as follows: 2j ðzÞ ¼ with

þ1 X 0

þ1 X 0

þ1 X 0

n  2j ðnÞZ þ

n  2j ðnÞZ ¼

ðnþ1Þ þ ¼ 2j ðn þ 1ÞZ

þ1 X 0

ðnþ1Þ þ 2j ðn þ 1ÞZ

c0 þ c1 Z þ c2 Z 2

ð1  em2j ZÞ2 ð1  e2j ZÞ

c3 Z 1 þ c4 Z 2 þ c5 Z3

ð1  em2j Z 1 Þ2 ð1  e2j Z 1 Þ

where j c0 ¼ A2 j ðm2  k0 2 j  m2 2 j Þ h i j j c1 ¼ A2 j ðk0 m22j þ k0 2 j  m2 j þ m2 2 jþ1 Þem2 þ ðk0 2 j  m2 j Þe2 h i j j jþ1 c2 ¼ A2 j ðk0 m22j  k0 2 j þ m2 j Þem2 2  m2 2 j em2 h i j j c3 ¼ A2 j ðk0 m22j  k0 2 j þ m2 j Þem2  m2 2 j e2 h i j j jþ1 c4 ¼ A2 j ðk0 m22j þ k0 2 j  m2 j þ m2 2 jþ1 Þem2 2 þ ðk0 2 j  m2 j Þem2 c5 ¼ A2 j ðk0 2 j þ m2 j  m2 2 j Þem2

jþ1 2 j

A2j is a given normalization constant defined as

PREFILTERING FOR PATTERN RECOGNITION

ð1  em2 Þ2 ð1  e2 Þ d 0 þ d 1 þ d2 þ d3 þ d4 þ d 5 j

A2 j ¼

201

j

with d0 d1 d2 d3 d4 d5

¼ ¼ ¼ ¼ ¼ ¼

m2 j  k0 2 j  m2 2 j j ð2k0 m22j þ m2 2 jþ1 Þem2 j j 2 j 2 j ðk0 2  m2  m 2 Þe j j ð2k0 m22j þ m2 2 jþ1 Þem2 2 j j 2 j m2 jþ1 ðk0 2  m2  m 2 Þe jþ1 j ðk0 2 j þ m2 j  m2 2 j Þem2 2 :

The impulse response y(n) corresponding to the filter  is given by yðnÞ ¼ yþ ðnÞ þ y ðnÞ yþ ðnÞ ¼ c3 xðn  1Þ þ c4 xðn  2Þ þ c5 xðn  3Þ þ b1 yþ ðn  1Þ b2 yþ ðn  2Þ þ b3 yþ ðn  3Þ þ y ðnÞ ¼ c0 þ c1 xðn þ 1Þ þ c2 xðn þ 2Þ þ b1 y ðn þ 1Þ  b2 y ðn þ 2Þ þb3 y ðn þ 3Þ:

References Abdi, H. (1994). Les Re´seaux de neurones. Presses Universitaires de Grenoble, Grenoble. Abdi, H., Valentin, D., and Edelman, B. (1998). ‘‘Neural Networks,’’ Sage, Thousand Oaks, CA. Abdi, H., Valentin, D., and Edelman, B. (1999). ‘‘Neural Networks.’’ Sage, Thousand Oaks, CA. Amico, M. D., and Ferrigno, G. (1992). Comparison between the more recent techniques for smoothing and derivative assessment in biomechanics. Med. Biol. Engin. Comput. 30, 193–204. Anderson, J. A., Silverstein, J. W., et al. (1977). Distinctive features, categorical perception, and probability learning: Some applications of a neural model. Psychol. Rev. 84, 413–451. Arslan, L. M., and Hansen, J. L. (1999). Selective training for hidden markov models with application to speech classification. IEEE Trans. Speech Audio Process 7(1), 46–54. Ayrulu, B., and Barshan, B. (2001). Neural networks for improved target differentiation and localization with sonar. Neural Networks 14, 355–373. Bachman, G., Narici, L., and Beckentein, E. (2000). ‘‘Fourier and Wavelet Analysis.’’ SpringerVerlag, New York. Bartlett, M. S. (1998). ‘‘Face Image Analysis by Unsupervised Learning and Redundancy Reduction.’’ Ph.D. thesis of Cognitive science and psychology, University of California. Benedetto, J. J., and Frazier, M. (1994). ‘‘Wavelets: Mathematics and Applications.’’ CRC Press, Boca Raton, FL. Bishop, C. M. (2000). ‘‘Neural Networks for Pattern Recognition,’’ Oxford Univ. Press, London.

202

YANG AND PAINDAVOINE

Bontempi, B., and Marcelli, A. (1996). On-line segmentation of cursive script using an arclength representation. In ‘‘Handwriting and Drawing Research: Basic and Applied Issues’’ (M. L. Simmer, C. G. Leedham, and A. J. W. Thomassen, eds.), pp. 315–327. IOS Press, Amsterdam. Bourennane, E., Paindavoine, M., and Truchetet, F. (1993). Ame´lioration du filtre de CannyDeriche pour la de´tection des contours sous forme de rampe. Traitement du signal, Recherche, Vol. 10, pp. 297–310. Broomhead, D. S., and Lowe, D. (1988). Multivariable functional interpolation and adaptive networks. Complex Syst. 2, 321–355. Chen, H., and Hewit, J. (2000). Application of wavelet transforms and neural networks to the recognition and classification of blemishes. Mechatronics 10, 699–711. Chen, C. Y., Yang, Z. L., et al. (2001). A programmable parallel VLSI architecture for 2-D discrete wavelet transform. J. VLSI Signal Process. 28, 151–163. Chen, C.H., Lee, J.D., and Lin, M.C. (1998). Classification of underwater signals using wavelet transform and neural networks, Math. Comput. Model 27(2), 47–60. Chui, C. (1992). ‘‘An Introduction to Wavelets.’’ Academic Press, San Diego. Coifman, R., Meyer, Y., and Wickerhauser, V. (1992). Wavelet analysis and signal processing. In ‘‘Wavelet and Their Applications.’’ Jones and Barlett, Boston. Coifman, R., and Wickerhauser, V. (1992). Entropy-based algorithms for best basis selection. IEEE Trans. Inform. Theory 38(2), 713–718. Das, M., Paulik, M. J., and Loh, N. K. (1990). A bivariate autoregressive modelling technique for analysis and classification of planar shapes. IEEE Trans. Pattern Anal. Mach. Intell. 12(1), 97–103. Daubechies, I. (1989). Orthonormal bases of compactly supported wavelets. Comm. Pure Appl. Math. 41(7), 909–996. Daubechies, I. (1994). Ten lectures on wavelets. CBMS, SIAM Vol. 61. Deriche, R. (1987). Using canny’s criteria to derive a recursively implemented optimal edge detector. Int. J. Comput. Vis. 1, 167–187. Desbiez, D., Vinter, A., and Meulenbroek, R. G. J. (1996). Biomechanical and perceptual determinants of drawing angles. Acta Psychol. 94, 253–271. Drolon, H., Druaux, F., and Faure, A. (2000). Particles shape analysis and classification using the wavelet transform. Pattern Recogn. Lett. 21, 473–482. Eide, A., and Lindblad, T. (1992). Artificial neural networks as measuring devices. Nuclear Instruments Methods physics Res. A317, 607–608. Englehart, K., Hudgins, B., et al. (1999). Classification of the myoelectric signal using timefrequency based representation. Med. Engineer. Phys. 21, 431–438. Farge, M. (1992). Wavelet transform and their applications to turbulance. Annu. Rev. Fluid Mech. 24, 395–457. Fausett, L. (1994). ‘‘Fundamentals of Neural Networks.’’ Prentice-Hall, New York. Fukunage, K. (1990). ‘‘Introduction to Statistical Pattern Recognition,’’ ed 2., Academic Press, San Diego. Gabor, D. (1946). Theory of communication. J. IEE 93, 429–457. Grosman, D., and Morlet, J. (1984) Decomposition of hardy functions into square integrable wavelets of constant shape. SIAM J. Math. 15, 723–736. Gurney, K. (1997). ‘‘An Introduction to Neural Networks.’’ UCL Press. Haykin, S. (1999). ‘‘Neural Networks,’’ ed 2., Prentice-Hall, New York. Hazarika, N., Zhu, J., et al. (1997). Classification of EEG signals using the wavelet transform. Signal Process. 59, 61–72. Hebb, D. O. (1949). ‘‘The Organization of Behaviour.’’ Wiley, New York.

PREFILTERING FOR PATTERN RECOGNITION

203

Hogan, N., and Flash, T. (1987). Moving gracefully: Quantitative theories of motor coordination. TINS 4, 170–174. Hudgins, B., Parker, P. A., and Scott, R. N. (1993). A new strategy for multifunctions myoelectric control. IEEE Trans. Biomed. Engineer. 40(1), 82–94. Huntsberger, T. L., Rose, J., and Ramaka, A. (1998). Fuzzy-Face: A hybrid wavelet/fuzzy selforganizing feature map system for face processing. J. Biol. Syst. 6(3), 281–298. Jia, X., and Nixon, S. (1995). Extending the feature vector for automatic face recognition. IEEE-Trans. Patterns Anal. Mach. Intelligence 17, No. 12, 1167–1176. Jobert, M., Timer, C., et al. (1994). Wavelet: A new tool in sleep biosignal analysis. J. Sleep Res. 3, 223–232. Kermit, M., Bodal, K. A., et al. (1999). Steady state vowel recognition of speech sound using O-algorithm. In ‘‘Artificial Intelligence and Soft Computing’’ (M. H. Hamza, ed.), pp. 362–367. Kermit, M., and Eide, A. J. (2000). Audio signal identification via pattern capture and template matching. Pattern Recogn. Lett. 21, 269–275. Koc, C. K., Chen, G., and Chui, C. K. (1994). Complexity analysis of wavelet signal decomposition and reconstruction. IEEE Trans. Aerospace Electr. syst. 30(3), 910–918. Kohonen, T. (1977). ‘‘Associative Memory: A System Theoretic Approach.’’ Springer-Verlag, Berlin. Kohonen, T. (1984). ‘‘Self Organization and Associative Memory.’’ Springer-Verlag, Berlin. Kru¨ger, V., Herpers, R. , et al. (1999). Teleconferencing using an attentive camera system. In ‘‘Proc. Int. Conf. on Audio- and Video-Based Biometric Person Authentification,’’ pp. 142–147. Barcelona, Spain. Kru¨ger, V., and Sommer, G. (2000). Affine real-time face tracking using gabor wavelet networks. In ‘‘Proc. Int. Conf. on Pattern Recognition.’’ Barcelona, Spain. Lee, S. W., Kim, C. H., et al. (1996). Multiresolution recognition of unconstrained handwritten numerals with wavelet transform and multilayer cluster neural network. Pattern Recognition 29(12), 1953–1961. Lee, J. S., Sun, Y. N., and Chen, C. H. (1995). Multiscale corner detection by using wavelet transform. IEEE Transacti. Image Process 4, 100–104. Lee, J. S., Sun, Y. N. , et al. (1993). Wavelet based corner detection. Pattern Recogn. 26(6), 853–865. Lemarie, P. G., and Meyer, Y. (1986). Ondelettes et bases Hibertiennes. Revista Mathematica Ibero Americana, Vol. 2, 1–18. Lin, W. H., Lee, J. S., et al. (1998). A new multiscale-based shape recognition method. Signal Process 65, 103–113. Lindbald, T., Lindsey, C. S., and Eide, A. (1997). Radial basis function neural networks. In ‘‘The Industrial Electronics Handbook’’ (J. D. Irvin, ed.), pp. 1014–1018. CRC Press, Boca Raton, FL. Lingvall, F., and Stepinski, T. (1999). Neural network classifier for eddy current signals. http://www.signal.uu.se/Research/CANDIA. Lingvall, F., and Stepinski, T. (2000). Automatic detecting and classifying defects during eddy current inspection of riveted lap-joints. NDT & E Int. 33, 47–55. Liu, Z. Q., Lu, M. D., and Wei, M. (1997). Structure noise reduction of ultrasonic signals using artificial neural network adaptive filtering. Ultrasonics 35, 325–328. Lu, J. (1993). Parallelizing Mallat algorithm for 2-D wavelet transforms. Inform. Process. Lett. 45(5), 255–259. Mallat, S. (1989). A theory for multiresolution signal decomposition: The wavelet representation. IEEE Transact. Pattern Anal. Machine Intell. 11(7), 674–693.

204

YANG AND PAINDAVOINE

Mallat, S. (1991). Zero-crossing of a wavelet transform. IEEE Transact. Inform. Theory 37, 1019–1033. Mallet, Y., Coomans, D. , et al. (1997). Classification using adaptive wavelets for feature extraction. IEEE Trans. Pattern Analy. Mach. Intell. 19(10), 1058–1066. Mallat, S., and Zhong, Z. (1992). Characterization of signal from multiscale edges. IEEE-PAMI 14, 710–732. Marquardt, C., and Mai, N. (1994). A computational procedure for movement analysis in hand-writing. J. Neurosci. Methods 52, 39–45. Meyer, Y. (1993). Wavelet: Algorithms and applications. SIAM. Minsky, M., and Papert, S. (1969). Perceptron: An Introduction to Computational Geometry. MIT Press, Cambridge, MA. Misra, M., and Nichols, T. (1994). Computation of the 2-D wavelet transforms on the connection machine-2. In ‘‘Proc. the IFIP WG10.3 Working Conf. on Applications in Parallel and Distributed Computing,’’ pp. 3–12. Caracas. Moody, J., and Darken, C. J. (1989). Fast learning in networks of locally-tuned processing units. Neural Comput. 1(2), 281–294. Musavi, M. T., Ahmed, W., et al. (1992). On the training of radial basis function classifiers. Neural Networks 5, 595–603. Nunez, P. L. (1981). ‘‘Electrical Fields of the Brain.’’ Oxford Univ. Press, New York. Oropesa, E., Cycon, H. L., and Jobert, M. (1999). ‘‘Sleep Stage Classification Using Wavelet Transform and Neural Network.’’ International Computer Science Institute, TR-99-008. Parker, D. B. (1985). ‘‘Learning-Logic.’’ Tech. Rep. TR-47, Massachusetts Institute of Technology, Center for Computational Research in Economics and Management Science, Cambridge, MA. Polikar, R., Greer, M. H., et al. (1997). Multiresolution wavelet analysis of ERPs for the detection of Alzheimer’s disease. In ‘‘Proc. 19th Int. Conf. IEEE/EMBS,’’ Chicago, IL. Powell, M. J. D. (1987). Radial basis functions for multivariable interpolation: A review. In ‘‘Algorithms for Approximation.’’ (J. C. Mason and M. G. Gox, eds.), pp. 143–167. Clarendon Press, Oxford. Rangannath, S., and Arun, K. (1997). Face recognition using transform features and neural networks. Pattern Recogn. 30(10), 1615–1622. Rechtschaffen, A., and Kales, A. (1968). ‘‘A Manual of Standardized Terminology, Technique and Scoring System for Sleep Stages of Human Subjects.’’ Public Health service, U.S. Government Printing Office. Reid, M. B., Spirlovsla, L., and Ochoa, E. (1989). Simultaneous position, scale and rotation invariant pattern classification using third-order neural networks. Int. J. Neural Networks 154 –159. Renals, S., and Rohwer, R. (1989). Phoneme classification experiments using radial basis functions. Int. Joint Conf. Neural Network 1, 461–467. Ripley, B. D. (1996). ‘‘Pattern Recognition and Neural Networks.’’ Cambridge Univ. Press, Cambridge, UK. Robert, C., Guilpin, C., and Limoge, A. (1998). Review of neural network application in sleep research. J. Neurosci. Methods 79, 187–193. Rojas, P. (1996). ‘‘Neural Networks: A Systematic Introduction.’’ Springer-Verlag, New York. Rosenblatt, F. (1959). ‘‘Principles of Neurodynamics.’’ Spartan Books, New York. Rumlhart, D. E., Hinton, G. E., and Williams, R. J. (1986). Learning representation by backpropagation errors. Nature 323, 533–536. Sasi, S., Schwiebert, L., and Bedi, J. S. (1997) Wavelet packet transform and fuzzy logic approach to handwritten character recognition. In ‘‘IEEE Workshop on Nonlinear Signal and Image Processing, NSIP-97.’’ Mackinac Island.

PREFILTERING FOR PATTERN RECOGNITION

205

Schomaker, L. R. B., and Thomassen, A. J. W. (1986). On the use and limitations of averaging handwriting signals. In ‘‘Graphonomics: Contemporary Research in Handwriting,’’ (H. S. R. Kao, G. P. Van Galen, and R. Hoosain, eds.), pp. 225–238. Elsevier, Amsterdam. Sekita, I., Kurita, T., and Otsu, N. (1992). Complex autoregressive model for shape recognition. IEEE Trans. Pattern Anal. Mach. Intell. 14(4), 489–496. Simoncelli, E. P., Freeman, W. T., et al. (1992). Shiftable multiscale transforms. IEEE Transact. Inform. Theory 38, 587–607. Strang, G., and Nguyen, T. (1997). ‘‘Wavelet and Filter banks.’’ Wellesley-Cambridge Press. Szu, H., Le Moigne, J., et al. (1997). Wavelet index of texture for artificial neural network classification of Landsat images. In ‘‘Proc. SPIE 25th AIPR Workshop: Emerging Applications of Computer Vision,’’ Vol. 2962, 36–44. Szu, H., Telfer, B., and Garcia, J. (1996). Wavelet transforms and neural networks for compression and recognition. Neural Networks 9(4), 695–708. Teulings, H.-L. H. M., and Maarse, F. J. (1984). Digital recording and processing of handwriting movements. Hum. Movement Sci. 3, 193–217. Teulings, H.-L. H. M., and Stelmach, G. E. (1991). Control of stroke size, peak acceleration, and stroke duration in parkinsonian handwriting. Hum. Movement Sci. 10, 315–333. Teulings, H.-L. H. M., Thomassen, A. J. W., and Van Galen, G. P. (1986). Invariants in handwriting: The information contained in a motor program. In ‘‘Graphonomics: Contemporary Research in Handwriting,’’ (H. S. R. Kao, G. P. Van Galen, and R. Hoosain, eds.), pp. 305–315. Elsevier, Amsterdam. Tsang, K. M., and Billings, S. A. (1992). Orthogonal estimation algorithm for complex number system. Int. J. Syst. Sci. 23(6), 1011–1018. Tungkanawanich, A., Kawasaki, Z. I., et al. (2000). Ground fault discrimination based on wavelet transform using artificial neural networks. T.IEE Japon 120-B(10), 1263–1270. Tzanetakis, G., Essl, G., and Cook, P. (2001). Audio analysis using the discrete wavelet transform. In ‘‘Proc. WSES Int. Conf. Acoustics and Music: Theory and Applications,’’ Skiathos, Greece. Unser, M. (1995). Texture classification and segmentation using Wavelet frames. IEEE Trans. Image Process 4(11), 269–281. Urick, R.J. (1983) ‘‘Principles of Underwater Sound.’’ McGraw-Hill, New York. Valentin, D., Abdi, H., O’Toole, A. J., and Cottrell, G. W. (1994). Connectionist models of face processing: A survey. Pattern Recogn. 27, 1209–1230. Van Galen, G. P. (1991). Handwriting: Issues for a psychomotor theory. Hum. Movement Sci. 10, 165–191. Van Galen, G. P., Van Doorn, R. R., and Schomaker, L. R. (1990). Effects of motor programming on the power spectral density function of finger and wirst movements. Hum. Percept. Perform 16, 755–765. Vinter, A., Desbiez, D., et al. (1996). The drawing of angular figures in Parkinson’s disease patients: Preliminary report. In ‘‘Handwriting and Drawing Research: Basic and Applied Issues’’ (M. L. Simmer, C. G. Leedham, and A. J. W. Thomassen, eds.), pp. 251–264. IOS Press, Amsterdam. Viviani, P., and Terzuolo, C. A. (1980). Space-time invariance in learned motor skills. In ‘‘Tutorials in Motor Behavior,’’ (G. E. Stelmach and J. Requin, eds.). North-Holland, Amsterdam. pp. 523–533. Ward, J. R., and Phillips, M. J. (1987). Digitizer technology: Performance, characteristics and the effects on the user surface. IEEE Comput. Graph. Appl. 4, 31–44. Wermter, S., and Sun, R. (2000). ‘‘Hybrid Neural Systems.’’ Springer, Heidelberg.

206

YANG AND PAINDAVOINE

Wesolkowski, S. (1996). Cursive script recognition: A survey. In ‘‘Handwriting and Drawing Research: Basic and Applied Issues,’’ (M. L. Simmer, C. G. Leedham, and A. J. W. Thomassen, eds.), pp. 267–284. IOS Press, Amsterdam. Widrow, B., and Hoff, M. E. (1960). ‘‘Adaptive Switching Circuits.’’ IRE WESCON Convention Record, pp. 96–104. Woltring, H. J. (1985). On optimal smoothing and derivative estimation from noisy displacement data in biomechanics. Hum. Movement Sci. 4, 229–245. Wu, J. S., Amaratunga, K., and Lui, T. M. (2000). Design of an online GIS viewer by wavelet technology. In ‘‘Proc. the 8th Int. Conf. On Computing in Civil and Building Engineering,’’ Stanford, CA. Yao, S. J., Song, Y. H., et al. (2000). Wavelet transform and neural networks for short-term electrical load forecasting. Energy Conver. Manage. 41, 1975–1988. Zhu, J., Hazarika, N., et al. (1994). Classification of EEG signals using Wavelet and an ANN. In ‘‘Workshop on Brain Electric and Magnetic Topography,’’ Sydney, Australia.

ADVANCES IN IMAGING AND ELECTRON PHYSICS, VOL. 127

Electron Optics and Electron Microscopy: Conference Proceedings and Abstracts as Source Material P. W. HAWKES CEMES–CNRS, F–31055 Toulouse Cedex, France

I. Introduction . . . . . . . . . . . . . . . . . . . . . II. International Congresses (ICEM) . . . . . . . . . . . . . III. Regional Congresses . . . . . . . . . . . . . . . . . . A. Europe (EUREM, EMC) . . . . . . . . . . . . . . . B. The Asia–Pacific region (APEM) . . . . . . . . . . . . C. The Balkans . . . . . . . . . . . . . . . . . . . . D. South America . . . . . . . . . . . . . . . . . . . IV. National Conferences . . . . . . . . . . . . . . . . . A. Europe . . . . . . . . . . . . . . . . . . . . . . 1. Armenia . . . . . . . . . . . . . . . . . . . 2. Austria . . . . . . . . . . . . . . . . . . . 3. Belgium . . . . . . . . . . . . . . . . . . . 4. Bulgaria . . . . . . . . . . . . . . . . . . . 5. Croatia . . . . . . . . . . . . . . . . . . . 6. Czechoslovakia, the Czech and Slovak Republics . . . 7. France . . . . . . . . . . . . . . . . . . . 8. Germany . . . . . . . . . . . . . . . . . . 9. Greece . . . . . . . . . . . . . . . . . . . 10. Hungary . . . . . . . . . . . . . . . . . . 11. Ireland . . . . . . . . . . . . . . . . . . . 12. Israel . . . . . . . . . . . . . . . . . . . . 13. Italy . . . . . . . . . . . . . . . . . . . . 14. Latvia . . . . . . . . . . . . . . . . . . . 15. The Netherlands . . . . . . . . . . . . . . . . 16. Poland . . . . . . . . . . . . . . . . . . . 17. Portugal . . . . . . . . . . . . . . . . . . . 18. Rumania . . . . . . . . . . . . . . . . . . 19. Scandinavia . . . . . . . . . . . . . . . . . 20. Slovenia . . . . . . . . . . . . . . . . . . . 21. Spain . . . . . . . . . . . . . . . . . . . . 22. Switzerland . . . . . . . . . . . . . . . . . 23. Turkey . . . . . . . . . . . . . . . . . . . 24. The United Kingdom . . . . . . . . . . . . . . 25. The USSR, now Russia . . . . . . . . . . . . . 26. Yugoslavia together with Serbia, Bosnia, and Herzegovina B. North America . . . . . . . . . . . . . . . . . . . 1. USA . . . . . . . . . . . . . . . . . . . . 2. Canada . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

209 214 217 217 219 220 221 222 222 223 224 226 228 228 230 232 235 244 244 246 248 250 252 252 256 260 262 262 265 266 268 270 271 280 284 286 286 293

207 Copyright 2003, Elsevier (USA). All rights reserved. ISSN 1076-5670/03

208

HAWKES

C. South America . . . . . . . . . . . . . . . . . . . . . . . . 1. Argentina . . . . . . . . . . . . . . . . . . . . . . . . 2. Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Chile . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Columbia . . . . . . . . . . . . . . . . . . . . . . . . 5. Cuba . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Ecuador . . . . . . . . . . . . . . . . . . . . . . . . . 7. Mexico . . . . . . . . . . . . . . . . . . . . . . . . . 8. Peru . . . . . . . . . . . . . . . . . . . . . . . . . . 9. Uruguay . . . . . . . . . . . . . . . . . . . . . . . . . 10. Venezuela . . . . . . . . . . . . . . . . . . . . . . . . D. Asia and Oceania . . . . . . . . . . . . . . . . . . . . . . . 1. Burma. . . . . . . . . . . . . . . . . . . . . . . . . . 2. China . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Hong Kong. . . . . . . . . . . . . . . . . . . . . . . . 4. India . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Indonesia . . . . . . . . . . . . . . . . . . . . . . . . 6. Japan . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Korea . . . . . . . . . . . . . . . . . . . . . . . . . . 8. Malaysia . . . . . . . . . . . . . . . . . . . . . . . . . 9. Pakistan . . . . . . . . . . . . . . . . . . . . . . . . . 10. The Philippines . . . . . . . . . . . . . . . . . . . . . . 11. Singapore . . . . . . . . . . . . . . . . . . . . . . . . 12. Taiwan, China . . . . . . . . . . . . . . . . . . . . . . 13. Thailand . . . . . . . . . . . . . . . . . . . . . . . . . E. Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. South Africa . . . . . . . . . . . . . . . . . . . . . . . 2. Egypt . . . . . . . . . . . . . . . . . . . . . . . . . . F. Australia and New Zealand . . . . . . . . . . . . . . . . . . . 1. Australia . . . . . . . . . . . . . . . . . . . . . . . . . 2. New Zealand . . . . . . . . . . . . . . . . . . . . . . . V. Thematic meetings . . . . . . . . . . . . . . . . . . . . . . . . A. High-Voltage Electron Microscopy (HVEM) . . . . . . . . . . . . B. International Conferences on X-Ray Optics and Microanalysis (ICXOM) . C. Charged-Particle Optics . . . . . . . . . . . . . . . . . . . . 1. The Charged-Particle Optics Conferences (CPO). . . . . . . . . . 2. Society of Photo-Optical Instrumentation Engineers (SPIE) . . . . . D. Scanning . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Scanning Electron Microscopy and the Pfefferkorn Conferences . . . . 2. Scanning . . . . . . . . . . . . . . . . . . . . . . . . E. Electron, Ion, and Photon Beam Conferences . . . . . . . . . . . . F. Microcircuit Engineering, later Micro- and Nanoengineering . . . . . . G. European Conferences on Electron & Optical Testing of Integrated Circuits (later, of Electronic Devices) and European Symposia on Reliability of Electron Devices, Failure Physics, and Analysis (ESREF) . . H. Microprocess, later Microprocesses and Nanotechnology . . . . . . . . I. Microanalysis . . . . . . . . . . . . . . . . . . . . . . . . 1. The Electron Probe Analysis Society of America (EPASA), subsequently the Microbeam Analysis Society (MAS) . . . . . . . . . . . . . 2. International Union of Microbeam Analysis Societies. . . . . . . .

295 295 296 298 298 298 298 299 299 300 300 302 302 302 304 304 306 306 315 317 317 317 318 318 318 320 320 323 323 323 325 326 326 328 329 329 330 331 331 333 334 339

342 344 346 346 348

209

ELECTRON OPTICS AND ELECTRON MICROSCOPY 3. Analytical Electron Microscopy . . . . . . . . . . . . 4. European Microbeam Analysis Society . . . . . . . . . 5. European Workshops on Electron Spectroscopic Imaging and Analysis Techniques . . . . . . . . . . . . . . . . 6. Workshops on Electron Energy-Loss Spectroscopy and Imaging (EELSI) . . . . . . . . . . . . . . . . . . . . . J. Frontiers of Electron Microscopy in Materials Science. . . . . . K. Compumag . . . . . . . . . . . . . . . . . . . . . . VI. Acknowledgments . . . . . . . . . . . . . . . . . . . . . VII. Acronyms . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

349 349

. . . .

351

. . . . . .

351 352 353 353 358 361

. . . . . .

. . . . . .

. . . . . .

I. Introduction The transmission electron microscope is now 70 years old and the scanning microscope is not much younger, even if the first commercial model appeared only in the mid-1960s, just a few years before the arrival of the scanning transmission electron microscope. As these instruments proliferated, microscopists began to organize themselves into user groups, many of which formed the nuclei of national societies of electron microscopy. These in turn convened their own meetings and participated in the regional and international meetings that came to be held at regular intervals, often under the overall tutelage of what we now call the International Federation of Societies for Microscopy (IFSM). Today, international congresses are held every four years (in even years, the date of which is not divisible by 4), as they have been ever since the second international congress was held in Paris in 1950, one year after the first such congress in Delft. European regional meetings have been held at four-yearly intervals, midway between the international congresses (and hence in even years divisible by 4), since 1956. Asia-Pacific regional meetings are now also held at four-yearly intervals, to coincide with the European regional conferences; earlier Asia-Pacific meetings were less regular. Conferences that bring together the electron microscopists of South America have been held since 1972. Three Balkan meetings have been held. In addition to all these conferences with an international character, a great many countries hold national meetings, typically annually or biennially; not infrequently, a small number of countries join together on a more or less regular basis to hold ‘‘local’’ multi-national conferences. Thus the German, Austrian and Swiss societies regularly convene a ‘‘Dreila¨ndertagung.’’ Four of the Nordic countries (Denmark, Sweden, Norway, and Finland) have long held their meeting in common, rotating between the four nations. The French society intermittently holds its annual meeting

210

HAWKES

in collaboration with that of a neighbouring society: in Barcelona with the Spanish and Portuguese societies in 1991 and 2001, for example, in Lausanne (1995) and Strasbourg (1998) with the Swiss and Belgian societies and in Lille (2002) with the Belgian, Dutch, and Swiss societies. Since 1993, societies from several of the countries of eastern and southern Europe have met biennially at Multi-National Conferences on Electron Microscopy. A series of joint Chinese–Japanese symposia has been organized since 1981. What remains of all this activity and what purpose do any records serve? The nature and accessibility of the printed material are highly diverse. Some meetings have generated regular proceedings volumes and, even if not all of these are easy to find, they can usually be tracked down. Thus in the case of the international and European regional congresses, formal proceedings have always been published and various major libraries hold copies of these; the Patent Office Library in London, for example, has a near-complete run. Individual volumes can in principle always be found in the copyright libraries of the host countries. The situation is not so clear for the other regional conferences. All the Asia–Pacific conferences have generated proceedings volumes, but I doubt whether many libraries have a complete set. The proceedings of the first two Balkan meetings are even more difficult to come by since they were distributed to participants but not, so far as I know, made more generally available. In South America, the creation of a Latin American Society with regular meetings was proposed at the Primer Curso Internacional de Microscopı´a Electro´nica para Cientı´ficos Latinoamericanos held in Mexico in November 1970; the Sociedad Latinoamericana de Microscopı´a Electro´nica (SLAME) was officially launched at the first congress, in 1972. The proceedings of some of the earlier meetings appeared in the Revista de Microscopı´a Electro´nica (subsequently renamed Revista de Microscopı´a Electro´nica y Biologia Celular, then Microscopı´a Electro´nica y Biologia Celular and now Biocell). Since 1992, however, these have been supplanted by a new series of meetings and the electron microscope community of South America has formed the Committee of Interamerican Societies for Electron Microscopy (CIASEM). At the next level, the situation is again very different in the various countries concerned. Some national societies regard the preservation of a printed record of their principal meetings as important and have taken care to ensure the existence of such a record. Thus the Canadian society has always published proceedings in the form of a regular (annual) serial, with its own ISSN; this means that it emerges from any methodical search of the indexes to scientific serials, Ulrich in particular (Ulrich, published annually), and makes it likely that a number of libraries will subscribe. The same is true of the Electron Microscopy Society of Southern Africa, which began

ELECTRON OPTICS AND ELECTRON MICROSCOPY

211

publication of an annual serial containing the proceedings of its annual meeting with the 10th conference in 1971. The Electron Microscope Society of America (EMSA) originally published abstracts of its meetings in the Journal of Applied Physics; in 1967, however, to mark the 25th anniversary of the founding of the society, a separate bound proceedings volume was published, the first of a series that continued essentially unaltered despite a change of publisher up to 1994. In that year, EMSA changed its name to the Microscopy Society of America and in 1995, a house journal was launched, the Journal of the Microscopy Society of America; in 1997, this was renamed Microscopy and Microanalysis. The proceedings of the annual meeting formed a supplement to that serial in 1995 but the San Francisco Press resumed publication in 1996; in 1997, however, publication of the proceedings as a supplement recommenced. Many other societies likewise publish proceedings volumes: the Electron Microscopy and Analysis Group (EMAG) of the (British) Institute of Physics has produced such a volume for each of its biennial meetings (odd years) since 1971; all but one of these have appeared in the Conference Series of Institute of Physics Publishing. The Italian Society published a biennial proceedings volume from 1987 to 1995; with the participation of the SIME in the series of Multi-National Conferences on Electron Microscopy (MCEM), publication has been less regular. All but the first of the biennial Chinese–Japanese Symposia are published by the Japanese Society of Electron Microscopy via the Business Center for Academic Societies Japan. Other societies publish short or extended abstracts in a regular journal, either in a special issue or as part of a normal issue. The Japanese society, for example, included abstracts of its national and regional meetings in its society journal, the Journal of Electron Microscopy until 1996. In 1997, publication of this serial was transferred to Oxford University Press and the abstracts of the national meeting were subsequently published as a Supplement to Denshikenbikyo¯ [Electron Microscopy]; a second Supplement was later created to cover the annual Symposia. The Korean Journal of Electron Microscopy includes short abstracts of the Korean society meetings although, for a few years after 1993, a separate abstracts book was issued. The Journal of the Chinese Electron Microscopy Society publishes extended abstracts of Chinese conference papers. Supplements to the Journal of the Electron Microscopy Society of Thailand contain the abstracts of papers delivered at the annual meetings of the Thai society. The German society has published the material presented at its meetings in a variety of journals over the years, notably as supplements to Optik until 1999. The proceedings of the Soviet All-Union meetings, now Russian meetings, have always appeared as all or part of issues of Izvestiya Akademiya Nauk SSSR (Seriya Fizika) and those of all but the first meeting are available in English translation in the Bulletin of the Academy of Sciences

212

HAWKES

of the USSR (Physical Series); the titles of both Izvestiya and the Bulletin have been modified to indicate the change from ‘‘USSR’’ to ‘‘Russia.’’ Various commercial serials have also welcomed the conference abstracts of national societies. Ultramicroscopy in particular is associated with numerous national societies and has published the abstracts of some of their meetings. Others have been published in Micron (or Micron and Microscopica Acta). For many years, the abstracts of papers in the life sciences delivered at the Scandinavian meetings were published in the Journal of Ultrastructure Research (now Journal of Structural Biology). For many countries, however, there is no easily accessible or central record of the papers delivered at national meetings: abstracts booklets are issued to participants and these are often difficult to trace. With no ISSN or ISBN and sometimes not even copyrighted, they may survive only on the shelves of those who were at the meetings and, all too often, not even there. Worse, they may never be cited: the abstracts booklets distributed to participants at Institute of Physics meetings used to carry the stern warning: ‘‘These outlines of conference talks . . . are not for publication; they should not be quoted in the literature nor may they be reproduced in abstracting journals or similar publications since they do not necessarily relate to papers intended for publication.’’ The material that does survive, especially in the form of books or special issues of journals or as short abstracts included in regular serials, is of course often published more fully elsewhere, duly refereed, and then indexed by the abstracting services–inspec, biosis, medline, and the like. Some work is, however, not republished elsewhere and, even when a full paper is produced, months or sometimes years later, the original and perhaps obscure abstract may not be mentioned, despite the fact that it fixes the date of the ‘‘discovery.’’ It is perhaps worth recalling that it is not necessarily the ‘‘minor’’ or ‘‘unimportant’’ abstracts that are not republished in fuller form; authors may consider that a conference abstract, especially if it occupies a page or more, is a sufficient publication of their results and that another paper on the same subject would simply duplicate the material and encumber the literature. Clearly, conference proceedings and abstracts have an important role to play in recording and charting the progress of our subject; sometimes, they contain the earliest statement of some new finding and occasionally, the only statement–if the reader needs to be convinced, I might mention the heavily cited 1968 paper of Schiske (republished in English in 2002). It therefore seems useful to chronicle as much of this elusive material as possible and to list the many national meetings, however inaccessible the proceedings or abstracts books may be; ideally, of course, the authors and topics would be indexed but such a task is beyond my resources. A beginning

ELECTRON OPTICS AND ELECTRON MICROSCOPY

213

has already been made, for many major proceedings volumes were catalogued in Hawkes & Kasper (1989 and especially 1994, supplemented in 1996) with full publishing details; several others are included in the list that forms the appendix (Hawkes, 1996b) to the IFSEM-sponsored volume entitled ‘‘The Growth of Electron Microscopy.’’ The latter list was, however, seriously deficient in a number of ways and it is these shortcomings that have led me to prepare the present much more complete account. In that list, there was no mention of the South American conferences; only one of the two series of Japanese meetings was included; and there was no serious attempt to chronicle many of the national meetings. In the present lists, much more thorough coverage has been attempted, though some gaps remain. The information gathered in Hawkes (1996b) is repeated here but in a different presentation; there, all the material was organized into a single chronological list. Here, I give a separate list for each of the numerous series of conferences, beginning with the international and regional meetings and continuing with the national and ‘‘supra-national’’ meetings. I have not, however, gone below the national level though I recognise that the North American regional societies (some of whose abstracts are to be found in Micron, Journal of Ultrastructure Research, Ultramicroscopy, Microscopy Research and Technique and doubtless elsewhere) may have more members and bigger meetings than some of the societies that are included; the same may be true of the Japanese regional meetings, the abstracts of which used to be published in the Journal of Electron Microscopy. I have also included details of a number of related meetings: the International Congresses on X-ray Optics and Microanalysis (ICXOM), the International Conferences on Charged-particle Optics (CPO) and those organized by SPIE, the HighVoltage Electron Microscopy Conferences (HVEM), the Pfefferkorn Conferences, Scanning, the Microcircuit Engineering (now Micro- and Nanoengineering, MNE) Conferences, the Electron and Optical Beam Testing of Integrated Circuits (recently, ‘‘of Electronic Devices’’) Conferences, the ‘‘Three-Beams’’ and related meetings, the Japanese ‘‘MicroProcess’’ conferences (now ‘‘Microprocesses and Nanotechnology’’), the Analytical Electron Microscopy and European and American Microbeam Analysis Society meetings, the European Workshops on Electron Spectroscopic Imaging and Analysis Techniques, the workshops on Electron Energy-Loss Spectroscopy and Imaging, the Frontiers of Electron Microscopy in Materials Science meetings and compumag. For some conference series, those on Scanning Electron Microscopy, for example, only summary bibliographic information is provided. This is a personal selection, no doubt there are other candidates equally deserving of inclusion, especially in the life sciences.

214

HAWKES

In the introductory paragraphs, I have frequently included bibliographical information about meetings not appropriate in the main list; readers should be aware that I have made no real attempt at completeness, the material reflects my own interests and is certainly invidious but it may save future students some trouble. I have of course tried to be as systematic as possible, but, nonetheless, some decisions remain arbitrary. Where national societies occasionally meet together, I have included these joint meetings in the each of the national lists, with a note to indicate their multi-national character. The Scandinavian meetings are treated as ‘‘national’’ meetings, though one could very well argue that they should be treated as ‘‘regional.’’ Where national boundaries have changed, I have occasionally grouped together society meetings that are now separate. There are still gaps in this account: I shall be most grateful to any readers who care to send missing information, accompanied wherever possible by references or proceedings volumes or photocopies of supporting material. I shall particularly appreciate specimens of society publications not mentioned here, notably newsletters and abstracts books. If a substantial amount of additional material emerges, I hope to publish it as a complement to the present text.

II. International Congresses (ICEM) In 1955, nine national societies of electron microscopy agreed to form an ‘‘International Federation of Electron Microscope Societies’’ (IFEMS): the founder members were Belgium, France, Germany, Great Britain, Japan, the Netherlands, Scandinavia, Switzerland and the USA. In 1958, these were joined by the Czechoslovak, Hungarian, Italian and Spanish Societies and the federation became the ‘‘International Federation of Societies of Electron Microscopy,’’ IFSEM; in 2002, the word ‘electron’ was dropped, to give ‘‘International Federation of Societies for Microscopy’’ (IFSM). Since its foundation, the international and most regional meetings have been held under its aegis and the number of member societies has grown steadily. The countries that are currently (2002) members of IFSM or have applied for membership are listed in Table 1. The history of the creation of the Federations and their many activities has been recapitulated in detail by Cosslett (1996) and by Maunsbach and Thomas (1996), and we say no more about it here. For personal selections of the highlights of these and some of the regional conferences of electron microscopy, see Hutchison (1996), Afzelius (1996) and Hawkes (1996a). The acronym ICEM was first used for

ELECTRON OPTICS AND ELECTRON MICROSCOPY TABLE 1 Countries that are Members or Associate Members of IFSEM or have Applied for Membership Armenia Australia Austria Belgium Brazil Bulgaria Canada China Colombia Croatia Cuba Czech Republic and Slovakia Egypt France Germany Hungary India Ireland Israel Italy Japan Korea Latvia Mexico Moldova The Netherlands New Zealand Poland Portugal Romania Russia Scandinavia (Denmark, Finland, Iceland, Norway, and Sweden) Singapore Slovenia Southern Africa Spain Switzerland Taiwan, China Thailand Turkey United Kingdom USA Venezuela Yugoslavia

215

216

HAWKES

the eleventh congress, held in Kyoto in 1986, but for convenience we employ it elsewhere in this article to refer to any of them. Since 1990, an International Workshop on Electron Energy-Loss Spectroscopy and Imaging (EELSI) has been held shortly after the International Congress; these are listed separately in Section V.I.5. Delft, 1949: Proceedings of the Conference on Electron Microscopy, Delft, 4–8 July, 1949 (Houwink, A. L., Le Poole, J. B., and Le Ru¨tte, W. A., eds.; Hoogland, Delft, 1950). Paris, 1950: Comptes Rendus du Premier Congre`s International de Microscopie Electronique, Paris, 14–22 September, 1950 (Editions de la Revue d’Optique The´orique et Instrumentale, Paris, 1953) 2 Vols. London, 1954: The Proceedings of the Third International Conference on Electron Microscopy, London School of Hygiene and Tropical Medicine, 15–21 July 1954 (Ross, R., ed.; Royal Microscopical Society, London, 1956). Berlin, 1958: Vierter Internationaler Kongress fu¨r Elektronenmikroskopie, Berlin, 10–17 September 1958, Verhandlungen (Bargmann, W., Mo¨llenstedt, G., Niehrs, H., Peters, D., Ruska, E., and Wolpers, C., eds.; Springer, Berlin, 1960) 2 Vols. Philadelphia, 1962: Electron Microscopy. Fifth International Congress for Electron Microscopy, Philadelphia, Pennsylvania, 29 August–5 September, 1962 (Breese, S. S., ed.; Academic Press, New York, 1962) 2 Vols. Kyoto, 1966: Electron Microscopy 1966. Sixth International Congress for Electron Microscopy, Kyoto, 28 August–4 September 1966 (Uyeda, R., ed.; Maruzen, Tokyo, 1966) 2 Vols. Grenoble, 1970: Microscopie Electronique 1970. Re´sume´s des Communications Pre´sente´es au Septie`me Congre`s International, Grenoble, 30 August–5 September 1970 (Favard, P., ed.; Socie´te´ Franc˛aise de Microscopie Electronique, Paris, 1970) 3 Vols. Canberra, 1974: Electron Microscopy 1974. Abstracts of Papers Presented to the Eighth International Congress on Electron Microscopy, Canberra, 25–31 August 1974 (Sanders, J. V. and Goodchild, D. J., eds.; Australian Academy of Science, Canberra, 1974) 2 Vols. Toronto, 1978: Electron Microscopy 1978. Papers Presented at the Ninth International Congress on Electron Microscopy, Toronto, 1–9 August 1978 (Sturgess, J. M., ed.; Microscopical Society of Canada, Toronto, 1978) 3 Vols. Hamburg, 1982: Electron Microscopy, 1982. Papers Presented at the Tenth International Congress on Electron Microscopy, Hamburg, 17–24 August 1982 (Deutsche Gesellschaft fu¨r Elektronenmikroskopie, Frankfurt, 1982) 3 Vols.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

217

Kyoto, 1986: Electron Microscopy 1986. Proceedings of the XIth International Congress on Electron Microscopy, Kyoto, 31 August–7 September 1986 (Imura, T., Maruse, S., and Suzuki, T., eds.; Japanese Society of Electron Microscopy, Tokyo) 4 Vols.; published as a supplement to Journal of Electron Microscopy 35 (1986). Seattle, 1990: Electron Microscopy 1990. Proceedings of the XIIth International Congress for Electron Microscopy, Seattle WA, 12–18 August 1990 (Peachey, L. D., and Williams, D. B., eds.; San Francisco Press, San Francisco) 4 Vols. See also Ultramicroscopy 36 (1991) Nos. 1–3, 1–274. Paris, 1994: Electron Microscopy 1994. Proceedings of the 13th International Congress on Electron Microscopy Paris, 17–22 July 1994 [Jouffrey, B., Colliex, C., Chevalier, J. P., Glas, F., Hawkes, P. W., Hernandez–Verdun, D., Schrevel, J. and Thomas, D. (Vol. 1), Jouffrey, B., Colliex, C., Chevalier, J. P., Glas, F., and Hawkes, P. W. (Vols. 2A and 2B) and Jouffrey, B., Colliex, C., Hernandez–Verdun, D., Schrevel, J. and Thomas, D. (Vols. 3A and 3B), eds.; Editions de Physique, Les Ulis, 1994]. Cancu´n, 1998: Electron Microscopy 1998. Proceedings of the 14th International Congress on Electron Microscopy, Cancu´n, 31 August–4 September 1998 [Memorias del 14to Congreso Internacional de Microscopı´a Electro´nica celebrado en Cancu´n (Me´xico) del 31 de Agosto al 4 de Septiembre de 1998] (Caldero´n Benavides, H. A., and Yacama´n, M. J., eds.; Institute of Physics Publishing, Bristol and Philadelphia 1998) 4 Vols. See also Micron 31 (2000), No. 5. Durban, 2002: Electron Microscopy 2002. Proceedings of the 15th International Congress on Electron Microscopy, International Convention Centre, Durban, 1–6 September 2002 [Cross, R., Richards, P., Witcomb, M., and Engelbrecht, J (Vol. 1, Physical, Materials and Earth Sciences), Cross, R., Richards, P., Witcomb, M., and Sewell, T. (Vol. 2, Life Sciences) and Cross, R., Richards, P., Witcomb, M., Engelbrecht, J., and Sewell, T. (Vol. 3, Interdisciplinary), eds.; Microscopy Society of Southern Africa, 2002]. Sapporo, 2006: 4–8 September 2006 III. Regional Congresses A. Europe (EUREM, EMC) The European regional meetings are today held under the aegis of the European Microscopy Society, which was created in 1998. Until then, they were under the responsibility of the Committee of European Societies of Electron Microscopy (CESEM, renamed Committee of European Societies of Microscopy, CESM, in 1994), which was founded at the sixth European

218

HAWKES

conference, in Jerusalem (see Section IV.A). The acronym EUREM appears to have been first used for the seventh conference, held in The Hague in 1980, but here we use it for any of this series of conferences. Some earlier meetings should also not be forgotten: the Deutscher Physiker -und Mathematikertagung of 1936, see Section IV.A.8; the Re´union d’Etudes on electron optics held in 1946, presided over by Louis de Broglie, see Section IV.A.7; and the European Congress of Applied Electron Microscopy held in Ghent, 7–10 April 1954 (Vandermeersche, 1954), see Section IV.A.3. The absence of information at this Ghent meeting about developments in Japan was regretted by the organizers and a meeting was held a few months later (28 January 1955) to remedy matters (see Bull. Microsc. Appl. 5, 1955, 28). An extremely well-documented survey of European commercial microscope development has been prepared by Agar (1996). Stockholm, 1956: Electron Microscopy. Proceedings of the Stockholm Conference, 17–20 September 1956 (Sjo¨strand, F. J. and Rhodin, J., eds.; Almqvist and Wiksells, Stockholm; 1957). Delft, 1960: The Proceedings of the European Regional Conference on Electron Microscopy, Delft, 29 August–3 September 1960 (Houwink, A. L. and Spit, B. J., eds.; Nederlandse Vereniging voor Elektronenmicroscopie, Delft n.d.) 2 Vols. Prague, 1964: Electron Microscopy 1964. Proceedings of the Third European Regional Conference, Prague, 26 August–3 September 1964 (Titlbach, M., ed.; Publishing House of the Czechoslovak Academy of Sciences, Prague, 1964) 2 Vols. Rome, 1968: Electron Microscopy 1968. Pre-Congress Abstracts of Papers Presented at the Fourth Regional Conference, Rome, 1–7 September 1968 (Bocciarelli, D. S., ed.; Tipographia Poliglotta Vaticana, Rome, 1968) 2 Vols. Manchester, 1972: Electron Microscopy 1972. Proceedings of the Fifth European Congress on Electron Microscopy, Manchester, 5–12 September 1972 (Institute of Physics, London 1972). Full versions of the contributions to the symposia on image processing and on computer-aided design in electron optics are available in Hawkes (1973) and those on high-voltage electron microscopy (sponsored by the Royal Microscopical Society) in Swann (1973). Jerusalem, 1976: Electron Microscopy 1976. Proceedings of the Sixth European Congress on Electron Microscopy, Jerusalem, 14–20 September 1976 [Brandon, D. G. (Vol. I) and Ben-Shaul, Y. (Vol. II), eds.; Tal International, Jerusalem, 1976] 2 Vols. The Hague, 1980: Electron Microscopy 1980. Proceedings of the Seventh European Congress on Electron Microscopy, The Hague, 24–29 August 1980 [Brederoo, P. and Boom, G. (Vol. I), Brederoo, P. and Priester, W. de (Vol. II), Brederoo, P. and Cosslett, V. E. (Vol. III), and Brederoo, P. and

ELECTRON OPTICS AND ELECTRON MICROSCOPY

219

Landuyt, J. van (Vol. IV), eds.]. Volumes I and II contain the proceedings of the Seventh European Congress on Electron Microscopy, Vol. III those of the Ninth International Conference on X-Ray Optics and Microanalysis, and Vol. IV those of the Sixth International Conference on High Voltage Electron Microscopy, which was held in Antwerp, 1–3 September 1980 (Seventh European Congress on Electron Microscopy Foundation, Leiden, 1980). Budapest, 1984: Electron Microscopy 1984. Proceedings of the Eighth European Congress on Electron Microscopy, Budapest 13–18 August 1984 ´ ., Ro¨hlich, P., and Szabo´, D., eds.; Programme Committee of the (Csana´dy, A Eighth European Congress on Electron Microscopy, Budapest, 1984) 3 Vols. York, 1988: Proceedings of the Ninth European Congress on Electron Microscopy, York, 4–9 September 1988 (Goodhew, P. J. and Dickinson, H. G., eds.; Institute of Physics, Bristol and Philadelphia, 1988) Conference Series 93, 3 Vols. Granada, 1992: Electron Microscopy 92. Proceedings of the 10th European Congress on Electron Microscopy, Granada, 7–11 September 1992 [Rı´os, A., Arias, J. M., Megı´as-Megı´as, L., and Lo´pez-Galindo, A. (Vol. I), Lo´pezGalindo, A. and Rodrı´guez-Garcı´a, M. I. (Vol. II) and Megı´as-Megı´as, L., Rodrı´guez-Garcı´a, M. I., Rı´os, A., and Arias, J. M. (Vol. III), eds.; Secretariado de Publicaciones de la Universidad de Granada, Granada] 3 Vols. Dublin, 1996: Electron Microscopy 1996. Proceedings of the 11th European Conference on Electron Microscopy, Dublin, 26–30 August 1996, distributed on CD-ROM. Subsequently published in book form by CESM, the Committee of European Societies of Microscopy (Brussels 1998), 3 Vols. Brno, 2000: Electron Microscopy 2000. Proceedings of the 12th European Conference on Electron Microscopy, Brno, 9–14 July 2000. (Frank, L. and ˇ iampor, F., general eds.; Vol. I, Biological Sciences, ed. by C ˇ ech S. and C Janisch R.; Vol. II, Physical Sciences, ed. by Gemperlova´ J. and Va´vra I.; Vol. III, Instrumentation and Methodology, ed. by Toma´nek P. and ˇ iampor F.; Vols Kolarˇ´ık R.; Vol. IV, Supplement, ed. by Frank L. and C I–III also distributed on CD-ROM; Czechoslovak Society of Electron Microscopy, Brno 2000). Antwerp, 2004: Proceedings European Microscopy Congress, Antwerp, 23–27 August 2004. (Schryvers, D., Timmermans, J.-P. and Pirard, E., general eds.; Biological Sciences, ed. by J.-P. Verbelen and E. Wisse; Materials Sciences, ed. by G. van Tendeloo and C. van Haesendonck; Instrumentation and Methodology, ed. by D. van Dyck and P. van Oostveldt.) B. The Asia-Pacific region (APEM) In 1956, the first Asia-Pacific Conference on Electron Microscopy was held in Tokyo; some information about the origins of this series of meetings is

220

HAWKES

given by Yasumasa Tani (1956a) in the Proceedings of that first meeting. The Asia–Pacific regional conferences are now held under the general supervision of the Committee of Asia–Pacific Societies of Electron Microscopy (CAPSEM). Tokyo, 1956: Electron Microscopy. Proceedings of the First Regional Conference in Asia and Oceania, Tokyo, 23–27 October 1956 (Electrotechnical Laboratory, Tokyo, 1957). Calcutta, 1965: Proceedings of the Second Regional Conference on Electron Microscopy in Far East and Oceania, Calcutta 2–6 February 1965 (Electron Microscopy Society of India, Calcutta). Singapore, 1984: Conference Proceedings 3rd Asia Pacific Conference on Electron Microscopy, Singapore, 29 August—3 September 1984 (Chung Mui Fatt, ed.; Applied Research Corporation, Singapore). Bangkok, 1988: Electron Microscopy 1988. Proceedings of the IVth Asia-Pacific Conference and Workshop on Electron Microscopy, Bangkok, 26 July–4 August 1988 (Mangclaviraj, V., Banchorndhevakul, W., and Ingkaninun, P., eds.; Electron Microscopy Society of Thailand, Bangkok). Beijing, 1992: Electron Microscopy I and II. 5th Asia-Pacific Electron Microscopy Conference, Beijing, 2–6 August 1992 (Kuo, K. H. and Zhai, Z. H., eds.; World Scientific, Singapore, River Edge NJ, London and Hong Kong) 2 Vols. See also Ultramicroscopy 48 (1993) No. 4, 367–490. Hong Kong, 1996: Proceedings of the 6th Asia–Pacific Conference on Electron Microscopy, Hong Kong, 1–5 July, 1996 (Barber, D., Chan, P. Y., Chew, E. C., Dixon, J. S., and Lai, J. K. L., eds.; Chinetek Promotion, Kowloon, Hong Kong). Singapore, 2000: Proceedings of the 7th Asia–Pacific Conference on Electron Microscopy, Singapore International Convention & Exhibition Centre, Suntec City, Singapore, 26–30 June 2000 (two volumes and CDROM, Yong, Y. T., Tang, C., Leong, M., Ng, C., and Netto, P., eds.; 7th APEM Committee, Singapore 2000). Kanazawa, 2004: Kanazawa, Ishikawa Prefecture, 7–11 June 2004. C. The Balkans Three Balkan congresses on electron microscopy have been held. These have been in some sense superseded by the multi-national congresses on electron microscopy (MCEM), which bring together microscopists from several several countries of south-eastern Europe, though at the present date (2002) none of the three countries that organized the Balkan meetings is among the MCEM group.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

221

Sarajevo; 1974: Electron Microscopy 1974. Pre-Congress Abstracts of Papers Presented at the First Balkan Congress on Electron Microscopy, Sarajevo, 22–26 May 1974 (Devide´, Z., Dobardzˇic, R., Jerkovic´, L., Marinkovic´, V., Pantic´, V., Pejowski, S., and Pipan, N., eds.). Istanbul, 1977: Abstracts of Communications, Second Balkan Congress on Electron Microscopy, Istanbul, 25–30 September 1977 (Erbengi, T., Chairman of the Scientific Programme Committee; Istanbul Faculty of Medecine and Turkish Society of Electron Microscopy, Istanbul). Athens, 1989: Proceedings III Balkan Congress on Electron Microscopy, Athens, 18–22 September, 1989 (Margaritis, L. H., ed.). D. South America The first society grouping the electron microscopists of South America was the Sociedad Latinoamericana de Microscopı´a Electro´nica (SLAME), founded in 1972, the founder members of which were Argentina, Brazil, Chile, Colombia and Venezuela. The Revista de Microscopı´a Electro´nica, which has changed names several times over the years (Revista de Microscopı´a Electro´nica y Biologı´a Celular, Microscopı´a Electro´nica y Biologı´a Celular, Biocell ), was the official organ of the society and later, of other learned societies as well. Today, however, a new association has been formed, the Committee of Inter-American Societies of Electron Microscopy (CIASEM), the members of which are listed in Section IV.C. The various regional congresses organized by SLAME and CIASEM are listed here. 1. First Latin-American Congress of Electron Microscopy, Maracaibo (Venezuela), 26–30 May 1972. Rev. Microsc. Electro´n. 1 (1972) Nos. 1 (Resumenes) and 2 (Simposios). 2. Second Latin-American Congress for Electron Microscopy and IV Colo´quio Brazileiro de Microscopia Electroˆnica, Faculdade de Medicina, Ribeira˜o Preto SP (Brazil), 1–5 December 1974 [Proceedings not published in a serial]. 3. Third Latin-American Congress for Electron Microscopy, Santiago (Chile), 22–26 November 1976. Rev. Microsc. Electro´n. 3 (1976) No. 1 (Resumenes) and 4 (1977) No. 1 (Simposios). Cf. Ipohorski (1978). 4. Fourth Latin-American Congress for Electron Microscopy, Mendoza (Argentina) 12–18 October, 1978. Rev. Microsc. Electro´n. 5 (1978) No. 1 and Rev. Microsc. Electro´n. y Biol. Cel. 6 (1979) Nos. 1–2. 5. Fifth Latin-American Congress for Electron Microscopy, Bogota´ (Columbia), 17–20 November 1981 [Proceedings not published in a serial].

222

HAWKES

6. Sixth Latin-American Congress for Electron Microscopy, Maracaibo (Venezuela), 2–8 December 1984. Microsc. Electro´n. y Biol. Cel. 8 (1984) No. 2 and Supplement. 7. Seventh Latin-American Congress for Electron Microscopy, Barcelona (Spain), 16–18 March 1987 [Proceedings not published in a serial]. 8. Eighth Latin-American Congress for Electron Microscopy, La Habana (Cuba), 2–5 May 1989 [Proceedings not published in a serial]. New Series 1. Memorias del 1er. Congreso Atla´ntico de Microscopı´a Electro´nica/ The Proceedings of the First Atlantic Congress of Electron Microscopy, Facultad de Ciencias, Universidad de los Andes, Me´rida (Venezuela), 25–29 May 1992, edited by E. Valiente (Editorial Venezolana, Me´rida, 1992). Held jointly with the V Jornadas Venezolanas de Microscopı´a Electro´nica. 2. Second Interamerican Conference on Electron Microscopy, Cancu´n (Me´xico), 26 September–1 October, 1993. 3. Third Interamerican Conference on Electron Microscopy and XVth Meeting of the Brazilian Society of Electron Microscopy, Caxambu´ MG (Brazil), 2–6 September 1995. Acta Microsco´pica 4 (1995), Supplements A (Biological Science) and B (Materials Science). 4. Fourth Interamerican Congress on Electron Microscopy and II Ecuadorian Congress on Electron Microscopy related to Medical, Biological and Materials Sciences, Guayaquil (Ecuador), 23–26 September 1997. Acta Microsco´pica 6 (1997) 39–57 [posters]. 5. Fifth Interamerican Congress on Electron Microscopy, Margarita Island (Venezuela), 24–28 October 1999. Full proceedings were produced as a CD-ROM, edited by E. Carrasquero, G. Castan˜eda, and E. Ramos with a preface by A. Castellano. 6. Sixth Interamerican Congress on Electron Microscopy, Veracruz (Mexico), 7–11 October 2001. Acta Microsco´pica, October 2001 [no volume number], 614 pp. 7. Seventh Interamerican Congress on Electron Microscopy, San Antonio TX (USA), 3–7 August 2003. Held jointly with Microscopy and Microanalysis, 2003 (Section IV.B.1). IV. National Conferences A. Europe The electron microscopy societies of Europe, several of which have dropped the word ‘‘electron,’’ used to belong to a Committee created at the sixth

ELECTRON OPTICS AND ELECTRON MICROSCOPY

223

EUREM congress in Jerusalem, in 1976, the Committee of European Societies of Electron Microscopy (CESEM, since 1994, CESM). In 1998, at the CESM meeting held at ICEM-14 in Cancu´n, the decision to disband CESM was taken and the European Microscopy Society was created in its stead. The following countries have national societies and many have adopted en-bloc membership of the EMS, whereby all members of the national society are automatically members of the EMS (such countries are indicated by an asterisk, correct at May 2003): Austria, Belgium,* Bosnia & Herzegovina, Bulgaria, Croatia,* the Czech Republic* and Slovakia,* France,* Germany, Greece,* Hungary,* Ireland,* Israel,* Italy,* Latvia, the Netherlands,* Poland,* Portugal, Rumania, Russia, ‘‘Scandinavia’’ (Denmark, Finland, Norway and Sweden), Slovenia,* Spain,* Switzerland,* Turkey,* the United Kingdom (EMAG*) and Yugoslavia.* Some of these societies are very young, the fruit of changing national boundaries, and there is as yet little to record about them. The following alphabetical list contains as much information as I have been able to collect about the meetings of the various societies. Note that this list includes some countries that are not strictly part of Europe but do not fit in conveniently anywhere else. 1. Armenia The Armenian Electron Microscopy Society was founded on 8 October 1991 and its byelaws were accepted on 2 July 1992. Meetings have been held regularly and in 1996, a serial entitled World of Microstructure was created in which the proceedings of these meeting first appeared. 1. Buniatyan Institute of Biochemistry of the National Academy of Science of Armenia, Yerevan, 27–28 October 1992. 2. National Academy of Science of Armenia, Yerevan and the Composers’ House in Dilijan, 11–13 September 1993. 3. National Academy of Science of Armenia, Yerevan and at the Cosmicray Physics House in Amberd, 7–10 October 1994. 4. Writers’ House on the Sevan Peninsula, 26–28 September 1995. 5. National Academy of Science of Armenia, Yerevan and at Sevan Blue House, 19–23 September 1996. World of Microstructure, No. 1 (1996). 6. Yerevan Mkhitar Heratsy State Medical University, 23–26 September 1997. World of Microstructure, No. 2 (1998). 7. 1998. National Academy of Science of Armenia, Yerevan, 5–6 October 1998. 8. 1999. National Academy of Science of Armenia, Yerevan, 27–28 December 1999. 9. ‘‘Electron Microscopy—2000.’’ International Conference Centre of the National Academy of Sciences of the Armenian Republic,

224

HAWKES

Yerevan, 17–20 October 2000. Proceedings edited by K. A. Hovnanyan, 94 pp. 10. 2001. Yerevan, 23–26 October 2001. Abstracts book (Hovnanyan, K. O., ed.), 89 pp. 2. Austria (www.univie.ac.at/asem/) ¨ sterreichische Gesellschaft fu¨r The creation and activities of the O ¨ Elektronenmikroskopie (OGE) are recounted by Ho¨rl (1996). We merely ¨ sterreichische note here that the Austrian society began in 1965 as the O ¨ Arbeitsgemeinschaft fu¨r Ultrastrukturforschung (OAU) and adopted its present name in Spring 1974 (see notes by H. Adam in Mikroskopie 30, 1974, 169–172 and especially 356). The meeting at which the Arbeitsgemeinschaft was founded was held on 26 June 1965. Today, the Austrian society meets jointly with the German and Swiss societies at the Dreila¨ndertagungen and with several other countries of eastern and southern Europe at the biennial Multi-national congresses on Electron Microscopy. Until 1999, the former were reported in Supplements published jointly by Optik and the European Journal of Cell Biology. See too memories of the great electron optician Walter Glaser by Gru¨mm and Schiske (1996). ¨ GE1: Wien, 22–25 September 1969 (joint meeting with the DGEM); Optik O 31 (1970) 111–112; Mikroskopie 26 (1970) 81–144. ¨ GE2: Graz, 5–7 April 1972 (Kolloquium with the DGEM); BEDO 5 (1972). O Forschungszentrum fu¨r Elektronenmikroskopie der Steirischen Hochschulen, Graz, 28 June 1974. Mikroskopie 30 (1974) 356. Symposium u¨ber Peroxisomen, Schule fu¨r Medizinisch-technische Assistentinnen, Vienna, 8 March 1975; Mikroskopie 31 (1975) 294–299. ¨ GE: Institute of Pathological Anatomy of the University of Vienna, 9 O ¨ sterreichische Gesellschaft fu¨r Pathologie); December 1975 (with the O Mikroskopie 32 (1976) Nos 3/4 and 5/6, 318–323 [short versions of the papers of G. Granditsch, H. Pamperl, G. Lassmann and L. Stockinger, and P. Bo¨ck, edited by M. Pavelka]. ¨ GE3: Institute of Physiology, Medical Faculty of the University of O Innsbruck, 12–14 April 1976, ‘‘Morphometrie biologischer Objekte mit Hilfe der Elektronenmikroskopie’’; Mikroskopie 33 (1977) 73 [report by W. Pfaller]. Seminar fu¨r Elektronenmikroskopie im Forschungszentrum Seibersdorf, 25 June 1976. Mikroskopie 33 (1977) 232–237 and 34 (1978) 242 (M. Pavelka). ¨ GE4: Innsbruck, 25–28 September 1978, Jahrestagung with the O ¨ sterreiO ¨ chische Physikalische Gesellschaft and the Osterreichische Gesellschaft fu¨r Vakuumtechnik.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

225

¨ GE5: Innsbruck, 23–27 August 1981 (joint meeting with the DGEM); O Optik 62 (1982) 329–331; BEDO 14 (1981). Dreila¨ndertagung fu¨r Elektronenmikroskopie: Konstanz, 15–21 September 1985 (joint meeting with the German and Swiss societies); Optik (1985) Supplement 1 or Eur. J. Cell Biol. (1985) Supplement 10, BEDO 18 (1985). ¨ GE7: Balatonaliga, 25–27 April 1985 (Joint meeting with the Hungarian O Society). ¨ GE8: Seggau–Leibnitz (Styria), 21–23 May 1987 (joint meeting with the O Group for Electron Microscopy of the Scientific Society for Measurements and Automation, Hungary); Optik 76 (1987) Supplement 2 or Eur. J. Cell Biol. 43 (1987) Supplement 18. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Salzburg, 10–16 September 1989 (joint meeting with the German and Swiss societies); Optik 83 (1989) Supplement 4 or Eur. J. Cell Biol. 49 (1989) Supplement 27. ¨ GE10: Balatonalma´di, 19–21 September 1991 (joint conference with the O Hungarian Society); Optik 88 (1991) Supplement 5 or Eur. J. Cell Biol. 55 (1991) Supplement 34. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Zu¨rich, 5–11 September 1993 (joint meeting with the German and Swiss societies); Optik 94 (1993) Supplement 5 or Eur. J. Cell Biol. 61 (1993) Supplement 39. MCEM-95: Proceedings Multinational Conference on Electron Microscopy, Stara´ Lesna´ (High Tatra Mountains), 16–20 October 1995, together with the Czechoslovak, Hungarian, Slovenian, Croatian and Italian Societies for Electron Microscopy (Slovak Academic Press, Bratislava 1995). Dreila¨ndertagung fu¨r Elektronenmikroskopie: Regensburg 7–12 September 1997 (joint meeting with the German and Swiss societies); Optik 106 (1997) Supplement 7 or Eur. J. Cell Biol. 74 (1997) Supplement 45. MCEM-97: Proceedings Multinational Congress on Electron Microscopy, Portorozˇ (Slovenia), 5–8 October 1997, together with the Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary. Part I, Microscopy Applications in the Life Sciences; Part II, Microscopy Applications in the Material Sciences; Part III, Microscopy Methods and Instrumentation. J. Computer-assisted Microsc. 8 (1996) No. 4 and 9 (1997) Nos. 1 and 2. MCEM-99: Proceedings 4th Multinational Congress on Electron Microscopy, Veszpre´m (Hungary), 5–8 September 1999, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Kova´cs, K., ed.; University of Veszpre´m 1999).

226

HAWKES

MCEM-5: Proceedings of the 5th Multinational Congress on Electron Microscopy, Department of Biology, University of Lecce (Italy), 20–25 September 2001, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Dini, L. and Catalano, M., eds.; Rinton Press, Princeton NJ 2001). Dreila¨ndertagung fu¨r Elektronenmikroskopie: Innsbruck, 9–14 September 2001 (joint meeting with the German and Swiss societies); Abstracts book (168 pp) not published as a Supplement to Optik or Eur. J. Cell Biol. MCM-6: Pula (Croatia), 1–5 June 2003, uniting the Austrian, Croatian, Italian and Slovenian Societies for Electron Microscopy, the Czechoslovak Microscopy Society and the Microscopy Society of Hungary. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Davos, 28 August–2 September, 2005 (joint meeting with the German and Swiss societies). 3. Belgium (wwc.ruca.ac.be) In 1957, a Comite´ Belge de Microscopie Electronique, Belgische Comiteit voor Elektronenmicroscopie was formed; the forerunner of the present society was founded in 1966 as the Socie´te´ Belge de Microscopie Electronique, Belgische Vereniging voor Elektronenmicroscopie. For more details, see van Dyck (1996). In 1994, the name was changed to Socie´te´ Belge de Microscopie, Belgische Vereniging voor Microscopie. A European Congress on Applied Electron Microscopy was held in Ghent, 7–10 April 1954 (see Vandermeerssche, 1954), the Proceedings of which contain an article on a planned Belgian electron microscope (Bruaux, 1954). 1957: Universitaire Stichting Brussel, 7 November 1957. 1958: Fondation Universitaire de Bruxelles, 9 December 1958. 1959: Fondation Universitaire de Bruxelles, 26 May 1959. 1960: University of Ghent, 6 April 1960. 1961: Fondation Universitaire de Bruxelles, 16 May 1961. 1962: Universitaire Stichting Brussel, 20 December 1962. 1963: and 1964 [not traced]. 1965: Universite´ de Lie`ge, 15 May 1965. 1966: Universitaire Stichting Brussel, 25 March 1966. 1967: Joint Meeting with the SFME, Brussels, 22–24 May 1967. J. Microscopie 6 (1967) No. 4, la–93a. 1968: Universite´ Catholique de Louvain, Louvain-le-Neuve, 9 February 1968. 1969: University of Ghent, 29 November 1969. 1970: FUNDP, Namur, 9 May 1970. 1971: Universite´ de Lie`ge, 13 February 1971. 1972: Universite´ Catholique de Louvain, Louvain-le-Neuve, 22 March 1972.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

227

1973: Meeting with the NVEM and DGEM in Lie`ge/Lu¨ttich, 3–6 September 1973. Optik 40 (1974) 233–283 and BEDO 6 (1974). Reported in Microsc. Acta 75 (1974) 364. 1974: Joint Meeting with the NVEM, Rotterdam, 16–17 May 1974. 1975: Universitaire Stichting Brussel, 13 March 1975. 1976: Rijksuniversitair Centrum, Antwerpen, 26 March 1976. 1977: FUNDP, Namur, 1 April 1977. 1978: University of Ghent, 14 April 1978. 1979: Universite´ de Lie`ge, 30 March 1979. 1980: Janssen Pharmaceutica, Beerse, 21 March 1980. 1981: Universite´ Libre de Bruxelles, 8 May 1981. 1982: Universite´ de Mons, 24 April 1982. 1983: Joint Meeting with the SFME, Lie`ge, 16–19 May 1983. J. Microsc. Spectrosc. Electron. 8 (1983) No. 2, la–48a and 111–278 and 341–488. 1983: Joint Meeting with the DGEM, Antwerp 11–16 September 1983. BEDO 16 (1983). 1984: Joint Meeting with the NVEM, Vrije Universiteit Brussel, 3–4 May 1984 1985: KULAK, Kortrijk, 26 April 1985 1985: Joint meeting with the Deutsche Anatomengesellschaft, Antwerp, September 1985 1986: Universite´ Catholique de Louvain, Louvain-le-Neuve, 30 May 1986. 1987: Abstracts of Papers Communicated at the Annual Meeting of the Belgian Society of Electron Microscopy, University of Antwerp (Universitaire Instelling Antwerpen), 27 March 1987. Micron Microsc. Acta 18 (1987) No. 3, 193–250. 1988: Joint Meeting with the SFME, Villeneuve d’Ascq–Lille, 17–20 May, 1988. J. Microsc. Spectrosc. Electron. 13 (1988) No. 3, 1a–64a, 65a–72a and 257–312 and 451–510. 1989: N. Goormaghtigh Institute for Pathological Anatomy, State University of Ghent, 20 October 1989. Micron Microsc. Acta 20 (1989) No. 2 111–158. 1990: Extended Abstracts of the Joint Meeting of the Belgian and Dutch Societies for Electron Microscopy, Wageningen, 6–7 December 1990. Micron Microsc. Acta 21 (1990) No. 4, 207–292. 1991: Abstracts of the Belgian Society for Electron Microscopy, Namur, 19 April 1991. Micron Microsc. Acta 22 (1991), No. 3, 247–285. 1992: AntwerpEM. Joint Meeting with the NVEM, t’Elzenveld, Antwerp, 10–11 December 1992. Eur. J. Morphol. 31 (1993) No. 1/2, 1–110 (Biomedical communications edited by F. Roels and F. van Meir); Microsc. Res. Tech. 25 (1992) 169–186. 1993: Annual meeting of the Belgian Society for Electron Microscopy, Catholic University of Louvain (KUL), Leuven 15 December 1993.

228

HAWKES

1994: Annual meeting of the Belgian Society for Electron Microscopy, Free University of Brussels, 20 May 1994. 1994: Joint Meeting with the NVEM, Papendal, Arnhem, 1–2 December 1994. Abstracts published in Jaarboek NVEM, 1994, pp. 37–123 (Berendsen, W., Drost, N., and Koerten, H. K., eds.). 1995: Joint Meeting with the French and Swiss Societies, Lausanne, 26–30 June 1995. For biological abstracts, see Biol. Cell 84 (1995) No. 3, 219–234. 1996: Joint Meeting with the NVvM, Het Pand, Gent, 12–13 December 1996. Abstracts published in Jaarboek NVvM, 1996, pp. 9–190 (Koerten, H. K., Drost, N., and Berendsen, W., eds.). 1997: Annual meeting of the Belgian Microscopy Society, Janssens Pharmaceutica, Beerse, 29 November 1997. 1998: Strasbourg, 29 June–3 July 1998. Joint Meeting with the French and Swiss Societies. For biological abstracts, see Biol. Cell 90 (1998) 247–292. 1999: Annual meeting of the Belgian Microscopy Society, Research Site Cockerill Sambre, Universite´ de Lie`ge, 20 November 1999. 2000: ‘‘Microscopy beyond Traditional Limits.’’ Joint Meeting with the NVvM and German microscopists of the border region, Papendal, 7–8 December 2000. Abstracts published in Jaarboek NVvM, 2000, pp. 13–194 (Koerten, H. K., ed.). 2001: Annual meeting of the Belgian Microscopy Society, Rixensart, 1 December 2001. 2002: Lille, 25–28 June 2002. Joint meeting with the French, Dutch and Swiss Societies. Abstracts issued by the Socie´te´ Franc¸aise des Microscopies. 2003: Annual meeting of the Belgian Microscopy Society, University of Antwerp, RUCA campus, 23 May 2003. Severin Amelinckx and Jozef van Landuyt memorial meeting. 2004: EMC Antwerp, 23–27 August 2004. 4. Bulgaria No information available. 5. Croatia (jagor.srce.hr/csem/) The Fifth Yugoslav Symposium on Electron Microscopy was held in Croatia (Plitvice Lakes, 27–30 May 1986), see Section IV.A.26 below. Since 1992, the Croatian Society of Electron Microscopy [CSEM, Hrvatsko Drustvo za Elektronsku Mikroskopiju] has held annual meetings, some of which have generated proceedings, as detailed below.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

229

1992: First Croatian Symposium on Electron Microscopy, Ruðer Bosˇkovic´ Institute, Zagreb, 18 December 1992. Periodicum Biologorum 95 (1993) 259–294. 1993: Second Annual Scientific Meeting of the CSEM, Ruðer Bosˇkovic´ Institute, Zagreb, 18 December 1993. Programme leaflet only. 1994: Third Annual Scientific Meeting of the CSEM, Department of Histology and Embryology of the Medical Faculty, Zagreb, 19 December 1994. Programme leaflet only. 1995: Fourth Annual Scientific Meeting of the CSEM, Institute of Physics, Zagreb, 20 December 1995. Programme leaflet only. 1995: Proceedings Multinational Conference on Electron Microscopy, Stara´ Lesna´ (High Tatra Mountains), 16–20 October 1995: Austrian, Croatian, Czechoslovak, Hungarian, Italian and Slovenian Societies for Electron Microscopy (Slovak Academic Press, Bratislava 1995). 1996: Second Croatian Symposium on Applications of Electron Microscopy in Life Sciences and Materials Science, Zagreb, 4 October 1996. Proceedings booklet edited by O. Milat and D. Jezek. 1997: MCEM-97. Proceedings Multinational Congress on Electron Microscopy, Portorozˇ (Slovenia), 5–8 October 1997, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary. Part I, Microscopy Applications in the Life Sciences; Part II, Microscopy Applications in the Material Sciences; Part III, Microscopy Methods and Instrumentation. J. Computer-assisted Microsc. 8 (1996) No. 4 and 9 (1997) Nos. 1 and 2. 1997: Fifth Annual Scientific Meeting of the Croatian Society for Electron Microscopy, Faculty of Veterinary Medicine, Zagreb, 19 December 1997. Programme leaflet only. 1998: Sixth Annual Scientific Meeting of the Croatian Society for Electron Microscopy, Institut Ruðer Bosˇkovic´, Zagreb, 18 December 1998. Programme leaflet only. 1999: First Congress of the Croatian Society for Electron Microscopy, Zagreb, 13–16 May 1999. MCEM-99: Proceedings 4th Multinational Congress on Electron Microscopy, Veszpre´m (Hungary), 5–8 September 1999, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Kova´cs, K. ed.; University of Veszpre´m 1999). MCEM-5: Proceedings of the 5th Multinational Congress on Electron Microscopy, Department of Biology, University of Lecce (Italy), 20–25 September 2001, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy

230

HAWKES

Society of Hungary (Dini, L. and Catalano, M., eds.; Rinton Press, Princeton NJ 2001). MCM-6: Pula (Croatia), 1–5 June 2003, uniting the Austrian, Croatian, Italian and Slovenian Societies for Electron Microscopy, the Czechoslovak Microscopy Society and the Microscopy Society of Hungary. 6. Czechoslovakia, the Czech and Slovak Republics (www.microscopy.cz) Conferences on electron microscopy have been organized since 1952 by the ˇ eskoslovenska´ Spolecˇnost pro Elektronovou Mikroskopii (Czechoslovak C Society for Electron Microscopy). For further information, see Delong (1985, 2000). Since 1993, a Bulletin has been distributed to members and from No. 5 (1995) is accessible at the CSEM/CSMS website. In 2002, the name of the society was changed to the Czechoslovak Microscopy Society ˇ eskoslovenska´ Mikroskopicka´ Spolecˇnost, CSMS). (C Czechoslovak Conferences on Electron Microscopy. CSEM 1952: Brno, autumn, 1952. CSEM 1953: Prague, 26–27 November 1953 (16 contributions). CSEM 1954: Smolenice, 11–13 November 1954 (15 contributions). CSEM 1956: Brno, 18–19 June 1956 (20 contributions). CSEM 1957: Prague, 25–26 June 1957 (25 contributions). CSEM 1959: Smolenice, 8–11 September 1959. CSEM 1961: Liblice, 1961 [unconfirmed]. CSEM 1963: Smolenice, 1963 [unconfirmed]. EUREM–3, Prague, 26 August–3 September 1964. CSEM 1965: Olomouc, 6–9 September 1965. CSEM 1967: Brno, 13–15 June 1967 (82 contributions). CSEM 1969: Prague–Suchdol, 8–11 September 1969. CSEM 1971: Stary´ Smokovec, 12–15 October 1971 (75 contributions), Proceedings, 81 pp. CSEM 1973: Olomouc, 18–21 September 1973. CSEM 1975: Bratislava, 8–12 September 1975 (86 contributions), Proceedings, 79 pp. CSEM 1977: Prague, 22–26 August 1977; with international participation. Proceedings (Ludvı´k, J. and Viklicky´, V., eds.), vol. A: Biological Science (425 pp.), vol. B: Nonbiological Sciences (240 pp.), vol. C: Appendix (43 pp.). CSEM 1979: Brno, 11–13 June 1979 (96 contributions). CSEM 1981: Kosice, 23–26 June 1981 (95 contributions), Proceedings, 95 pp. CSEM 1983: Nitra, 6–8 September 1983. CSEM 1985: Olomouc, 9–12 September 1985 (115 contributions), Proceedings, 139 pp.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

231

CSEM 1989: Prague, 26–29 June 1989 (100 contributions), Proceedings, 118 pp. CSEM 1991: Nitra, 3–5 September 1991. Abstracts volume, 120 pp. 1993: Multinational Congress on Electron Microscopy (Italian, Hungarian, Czechoslovak and Slovenian Societies), Parma, 13–17 September 1993; Proceedings issued as Supplement to 14 (2) of Microscopia Elettronica. 1995: First Annual Assembly CSEM, Laboratory of Electron Microscopy, Institute of Parasitology AVCR, Ceskych Budejovicich, 9–10 July 1998. 1995: Proceedings Multinational Conference on Electron Microscopy, Stara´ Lesna´ (High Tatra Mountains), 16–20 October 1995, together with the Austrian, Hungarian, Slovenian, Croatian and Italian Societies for Electron Microscopy (Slovak Academic Press, Bratislava 1995). MCEM-97: Proceedings Multinational Congress on Electron Microscopy, Portorozˇ (slovenia), 5–8 October 1997, together with the Austrian, Croatian, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary. Part I, Microscopy Applications in the Life Sciences; Part II, Microscopy Applications in the Material Sciences; Part III, Microscopy Methods and Instrumentation. J. Computer-assisted Microsc. 8 (1996) No. 4 and 9 (1997) Nos. 1 and 2. 1998: First Annual Assembly CSEM, Laboratory of Electron Microscopy, ˇ R, C ˇ esky´ch Budeˇjovicˇich, 9–10 July 1998. Parasitology Institute AVC MCEM-99: Proceedings 4th Multinational Congress on Electron Microscopy, Veszpre´m (Hungary), 5–8 September 1999, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Kova´cs, K., ed.; University of Veszpre´m 1999). EUREM-12: Brno, 9–14 July 2000. MCEM-5: Proceedings of the 5th Multinational Congress on Electron Microscopy, Department of Biology, University of Lecce (Italy), 20–25 September 2001, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Dini, L. and Catalano, M., eds.; Rinton Press, Princeton NJ 2001). ˇ SMS, Hoˆtel Club, Vranovska´ Ves, 8–9 2002: Second Annual meeting C February 2002. MCM-6: Pula (Croatia), 1–5 June 2003, uniting the Austrian, Croatian, Italian and Slovenian Societies for Electron Microscopy, the Czechoslovak Microscopy Society and the Microscopy Society of Hungary 2003, Dresden, 7–12 July 2003. Joint meeting with the German Electron Microscopy Society.

232

HAWKES

International Seminars on Recent Trends in Charged Particle Optics and Surface Physics Instrumentation. 1989: First Seminar, Brno, 4–6 September 1989 (no proceedings). 1990: Second Seminar, Brno, 27–29 September 1990 (no proceedings). 1992: Third Seminar, Skalsky´ Dvu˚r (near Brno), 15–19 June 1992 (no proceedings). 1994: Fourth Seminar, Skalsky´ Dvu˚r, 5–9 September 1994 (no proceedings). 1996: Fifth Seminar, Skalsky´ Dvu˚r, 24–28 June 1996. Proceedings edited by I. Mu¨llerova´ and L. Frank (92 pp). 1998: Sixth Seminar, Skalsky´ Dvu˚r, 29 June–3 July 1998. Proceedings edited by I. Mu¨llerova´ and L. Frank (84 pp). Published by the CSEM (Brno 1998). 2000: 7th Seminar, Skalsky´ Dvu˚r, 15–19 July 2000. No proceedings book. 2002: 8th Seminar, Skalsky´ Dvu˚r, 8–12 July 2002. Proceedings edited by L. Frank (96 pp þ Supplement, 6 pp). Published by the CSMS (Brno 2002). 7. France (sfmu.snv.jussieu.fr) For information about the genesis and formation of the Socie´te´ Franc¸aise de Microscopie Electronique (SFME), which changed its name to Socie´te´ Franc¸aise des Microscopies (SF) in 1995, see Couteaux (1989, 1994), Haguenau (1984, 1996) and Jouffrey (1996). For many details of the development of electron microscopy in France, see Grivet (1985) and Dupouy (1968, 1985). Before the creation of the SFME, there existed a Section de Microscopie Electronique (which soon became the Section de Microscopie et Diffraction Electroniques), created in 1953, within the Socie´te´ Franc¸aise de Microscopie The´orique et Applique´e. The first meeting was held in the Centre de Recherches des Charbonnages de France (Verneuil, Oise) on 30 April 1954 (Bull. Microsc. Appl. 4, 1954, 98–102) and several further meetings were held, notably in the Centre de Recherches sur le Cancer at Villejuif (7 March 1955, see Bull. Microsc. Appl. 5, 1955, 1–27) and the Institut Pasteur (22 March 1956, see Bull. Microsc. Appl. 6, 1956, 142). For details of the transformation of the Section into the SFME, see the Bull. Microsc. Appl. 9 (1959) 74–75. The pre-history of the SFME is briefly recapitulated in Hawkes (1997a). When the SFME was founded, the Journal de Microscopie, launched in 1962, was at once a scientific journal and the house magazine (the Bulletin de Microscopie Applique´e ceased publication in 1963). The latter role was occupied from 1987 to 1999 by the Bulletin d’Information of the Socie´te´ Franc¸aise de Microscopie Electronique, now, des Microscopies, distributed to members of the society twice a year. From 2000, society news was

ELECTRON OPTICS AND ELECTRON MICROSCOPY

233

disseminated only on the Web Site (sfmu.snv.jussieu.fr). The SF also publishes a series of monographs (Morniroli, 1998, 2002). Three major conferences were held in France before the creation of the SFME. These are included at the head of the list. The courses given at some Summer Schools and Workshops have also been recorded in permanent form; see Jouffrey (1972), Jouffrey et al. (1983), Maurice et al. (1979, 1980), Colliex and Isaacson (1989) and Morniroli (2001). L’Optique Electronique, Re´unions d’Etudes et de Mises au Point tenues en mai–juin 1945 sous la pre´sidence de Louis de Broglie, see de Broglie (1946). ICEM-2, Paris, 14–22 September, 1950. Les Techniques Re´centes en Microscopie Electronique et Corpusculaire, Toulouse, 4–8 April 1955. In the series, ‘‘Colloques Internationaux du Centre National de la Recherche Scientifique’’ (Fert, C., Introduction; Centre National de la Recherche Scientifique, Paris 1956). SFME 1959: Colle`ge de France, Paris, 17–18 December; presided over by Raimond Castaing and Emmanuel Faure´–Fremiet, with a speech by Gaston Dupouy in conclusion; Bull. Microsc. Appl. 9 (1959) 108–109. SFME 1961: Lyon–Villeurbanne, 16–17 February. SFME 1962: Toulouse, 15–16 February. SFME 1963: Orsay, 14–16 February; J. Microscopie 2 (1963) 1–43. SFME 1964: Strasbourg, 10–12 February; J. Microscopie 3 (1964) 1–60. SFME 1965: Marseille, 1–3 March; J. Microscopie 4 (1965) 99–175. SFME 1966: Bordeaux, 23–25 May; J. Microscopie 5 (1966), No. 3, 1a–91a. SFME 1967: Brussels (with the Belgian Society), 22–24 May; J. Microscopie 6 (1967) No. 4, 1a–93a. SFME 1968: Lille, 17–20 May; J. Microscopie 7 (1968), No. 4, 1a–65a. SFME 1969: Lausanne (with the Swiss Society), 19–21 May; J. Microscopie 8 (1969) No. 4, 1a–99a. ICEM–7, Grenoble, 30 August–5 September 1970. SFME 1971: Caen, 7–10 May; J. Microscopie 11 (1971) 1–106. SFME 1972: Nantes, 29–31 May; J. Microscopie 14 (1972) No. 2, 1a–108a. SFME 1973: Dijon, 4–7 June; J. Microscopie 17 (1973) No. 3, 1a–116a. SFME 1974: Rennes, 27–30 May; J. Microscopie 20 (1974) No. 1, 1a–104a. SFME 1975: Montreal (with the Canadian Society) 26–30 May; J. Microsc. Biol. Cell. 23 (1975) No. 2, 1a–88a. SFME 1976: Clermont-Ferrand, 9–11 June; J. Microsc. Biol. Cell. 26 (1976) Nos. 2–3, 1a–35a. SFME 1977: Nice, 31 May–2 June; J. Microsc. Spectrosc. Electron. 2 (1977) No. 3, 1a–18a; Biol. Cell. 29 (1977) No. 1, 1a–45a.

234

HAWKES

SFME 1978: Nancy, 23–25 May; J. Microsc. Spectrosc. Electron. 3 (1978) No. 3, 1a–16a and 271–386 and 551–631;; Biol. Cell 32 (1978) Nos 2–3, 1a–31a. SFME 1979: Lyon–Villeurbanne, 21–23 May; J. Microsc. Spectrosc. Electron. 4 (1979) No. 3, 1a–26a and 269–519 and 581–612; Biol. Cell. 35 (1979) No. 2, 1a–40a. SFME 1980: Poitiers 4–6 June; J. Microsc. Spectrosc. Electron. 5 (1980) No. 3, 1a–18a and 451–527 and 539–727; Biol. Cell 38 (1980) No. 3, 1a–31a. SFME 1981: Besanc¸on, 25–27 May; J. Microsc. Spectrosc. Electron. 6 (1981) No. 3, 1a–10a and 345–462; Biol. Cell. 41 (1981) No. 1, 1a–31a. SFME 1982: Reims, 25–27 May; J. Microsc. Spectrosc. Electron. 7 (1982) No. 2, 1a–28a and 315–423 and 487–553; Biol. Cell 44 (1982) No. 1, 1a–43a. SFME 1983: Lie`ge (with the Belgian Society), 16–19 May; J. Microsc. Spectrosc. Electron. 8 (1983) No. 2, 1a–48a and 111–278 and 341–488; Biol. Cell. 48 (1983) No. 1, 1a–54a. SFME 1984: Montpellier, 21–24 May; J. Microsc. Spectrosc. Electron. 9 (1984) No. 1, 1a–51a and 147–340; Biol. Cell. 51 (1984) No. 1, 1a–44a. SFME 1985: Strasbourg, 28–31 May; J. Microsc. Spectrosc. Electron. 10 (1985) No. 2, 1a–55a and 149–249 and 311–514; Biol. Cell 53 (1985) No. 3, 1a–51a. SFME 1986: Nantes, 9–11 June; J. Microsc. Spectrosc. Electron. 11 (1986) No. 2, 1a–55a and 129–214 and 359–396: Biol. Cell 57 (1986) No. 2, 1a–18a. SFME 1987: Talence, 20–22 May; J. Microsc. Spectrosc. Electron. 12 (1987) No. 3, 1a–39a and 353–421; Biol. Cell. 60 (1987) No. 1, 1a–40a. SFME 1988: Villeneuve d’Ascq–Lille (with the Belgian Society), 17–20 May; J. Microsc. Spectrosc. Electron. 13 (1988) No. 3 1a–64a, 65a–72a and 257–312 and 451–510; Biol. Cell. 63 (1988) No. 1, Supplement. SFME 1989: Grenoble–Saint Martin d’He`res, 8–12 July; J. Microsc. Spectrosc. Electron. 14 (1989) No. 2, 1a–80a and 315–384 and 387–414; Biol. Cell. 67 (1989) Supplement to No. 3, 1a–20a. SFME 1990: Toulouse, 19–22 June; Biol. Cell. 69 (1990) No. 2, 19a–42a. SFME 1991: Barcelona (with the Spanish and Portuguese Societies), 2–5 July. Abstracts CFIME [Coloquio Franco–Ibe´rico de Microscopı´a Electro´nica, Colloque Franco–Iberique de Microscopie Electronique, Colo´quio Franco–Ibe´rico de Microscopia Electroˆnica], Universitat de Barcelona Publicacions, 429 pp. SFME 1992: Rouen–Mont Saint-Aignan, 30 June–3 July; Biol. Cell. 75 (1992) No. 3, 257–268. SFME 1993: Villeurbanne, 29 June–3 July; Biol. Cell. 79 (1993) No. 1, 81–95.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

235

ICEM-13, Paris, 17–22 July 1994; see also Biol. Cell 80 (1994) No. 2/3. Trinoculaire 1995: Lausanne (with the Swiss and Belgian Societies), 26–30 June. For biological abstracts, see Biol. Cell 84 (1995) No. 3, 219–234. SF 1: Rennes, 26–28 June; Biol. Cell 86 (1996) Nos. 2/3, 185–195. SF 2: Nancy, 30 June–4 July 1997; Biol. Cell 89 (1997) No. 2, 153–163. Trinoculaire 1998: Strasbourg—Illkirch–Grafenstaden (with the Swiss and Belgian Societies), 29 June–3 July 1998. For biological abstracts, see Biol. Cell 90 (1998) 247–292. SF 3: Ecole Polytechnique, Palaiseau, 29 June–2 July 1999. For biological abstracts and papers presented at the Castaing Symposium, see Biol. Cell 91 (1999) 221–284. SF 4: Institut National des Sciences Applique´es, Toulouse, 4–8 September 2000. Abstracts included in the Annuaire of the Society; for biological abstracts, see Biol. Cell 92 (2000) 363–384. SF 2001: Barcelona (with the Spanish and Portuguese Societies), 3–7 September 2001. Abstracts Microscopy Barcelona 2001 (Universitat de Barcelona 2001) 614 pp; for biological abstracts, see Biol. Cell 93 (2001) 325–439. SF 2002: Lille (with the Belgian, Dutch and Swiss Societies), 25–28 June 2002. Abstracts volume issued by the Socie´te´ Franc¸aise des Microscopies. SF 2003: University of Toulon, La Garde, 23–26 June 2003. Abstracts volume issued by the Socie´te´ Franc¸aise des Microscopies. 8. Germany (www.dge-homepage.de) The founding of the Deutsche Gesellschaft fu¨r Elektronenmikroskopie (DGE or DGEM) is described fully by Schimmel (1996) and the ‘‘Gru¨ndungsprotokoll’’ is reproduced in Elektronenmikroskopie (1990). From 1959 until 1988, separate meetings were held in East Germany, organized by the Electron Microscopy Section of the Physical Society of the German Democratic Republic and, for biological material, by the Society of Topochemistry and Electron Microscopy (Heydenreich et al., 1996). From 1991 to 1995, meetings on specific topics in electron microscopy were held in Halle, in what is now the Max-Planck-Institut fu¨r Mikrostrukturphysik. In 1968, the Arbeitskreis ‘‘Elektronenmikroskopische Direktabbildung von Oberfla¨chen’’ of the DGEM held the first of a series of annual colloquia; the proceedings of these were published in Beitra¨ge zur elektronenmikroskopischen Direktabbildung von Oberfla¨chen (BEDO), originally published by Verlag R. A. Remy in Mu¨nster; from the 26th volume onwards, the title changed to Beitra¨ge zur elektronenmikroskopischen Direktabbildung und Analyse von Oberfla¨chen (BEDAO) and publication was taken over by

236

HAWKES

Christel Ehrenwerth (Mu¨nster). Publication of BEDAO was discontinued after the 29th volume (1996). We give separate lists for the various series. For BED(A)O, only brief information is given; note that some volumes include left-over papers from the previous meeting or papers presented at other meetings but of direct interest to the readership; such minutiae are not signalled below. The DGEM has seven other Arbeitskreise at present (2003): ‘‘Analytische Elektronenmikroskopie in Biologie und Medizin’’ (AEBM), ‘‘Rastersondenmikroskopie’’ (SPM), ‘‘Energiefilterung und ElektronenEnergieverlustspektroskopie’’ (EF & EES), ‘‘Pra¨paration und Abbildung nativer Systeme’’ (PANS), ‘‘Digitale Informationserfassung, -verarbeitung und - archivierung’’ (DIVA), ‘‘Elektronenmikroskopische Erregerdiagnostik’’ (EMED), and ‘‘Hochauflo¨sende Transmissions-Elektronenmikroskopie’’ (HREM). In 1990, the DGE began publication of a house journal, distributed to its members, entitled Elektronenmikroskopie (published by Hirzel, Stuttgart). A number of historical articles have appeared there; see Kinder (1990), Lenz (1995), Rang (1992), Schulze (1992a,b, 1994, 1996), and Schimmel (1998). Number 18 includes papers arising from the celebrations at the Dortmund meeting (1999) of the fiftieth anniversary of the DGE and the sixtieth anniversary of the first commercial electron microscope (Lenz, 1999; Schulze, 1999; Sitte, 1999). There is also a collection of photographs illustrating ‘‘60 Jahre ‘kommerzielle’ Elektronen Mikroskopie’’ (pp. 23–27). See also www.dge-homepage.de. One major meeting that was held long before the creation of the DGEM is of particular importance. From 13–19 September 1936, the twelfth Deutscher Physiker- und Mathematikertagung was held in Bad Salzbrunn; a large section was devoted to the new subject of electron microscopy and full records are to found in the Zeitschrift fu¨r technische Physik 17 (1936), No. 12, 584–666 and in the book edited by Busch and Bru¨che (1937). General lectures were given by H. Busch, E. Bru¨che, O. Scherzer, W. Schaffernicht, M. Knoll, and W. Glaser; individual contributions were made by G. Weiss, W. Heimann, H. Rothe, and W. Kleen, A. Recknagel, W. Heinz and S. Wagener, H. Mahl, A. Gehrts, and M. von Ardenne. Numerous other meetings have given rise to proceedings volumes, see Hoppe et al. (1970), Beer et al. (1975) to which Johansen and Høglund (1975) is directly relevant, Hoppe and Mason (1979), Baumeister and Vogell (1980), Messerschmidt et al. (1990), Kontron (1971, 1972, 1973, 1975a,b) and Baumeister and Zeitler (1992). The issue of Beitra¨ge zur Forschungstechnologie dedicated to Heinz Bethe on his 65th birthday (Hillebrand et al., 1984) is also relevant here as are the tributes to Gottfried Mo¨llenstedt published in the Journal of Electron Microscopy (Lenz, 1998; Lichte, 1998; Mo¨llenstedt, 1998).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

237

The history of electron microscopy in Germany now has a considerable literature (AEG, 1940; von Ardenne, 1972, 1978, 1985, 1996, 1997, cf. Charles, 1996; Borasky, 1982, cf. Ruska, 1985; von Borries, 1991; von Borries and Ruska, 1944; Bru¨che, 1940, 1941, 1943a,b, 1957, 1966/7; Deutsch, 1960; ECBO, 1980; Freundlich, 1963, 1985, 1994; Gabor, 1957, 1966/7, 1974; Giesbrecht, 1981; Gloede, 1986; Gru¨newald, 1976, cf. Ruska, 1977; Knoll, 1968; Kruger et al., 2000; Ku¨pfmu¨ller, 1944; Lambert, 1986; Lambert and Mulvey, 1996; Lenz, 1986, 1995, 1996; Matthias, 1942; Mo¨llenstedt, 1996, 1997, 1999; Mulvey, 1962, 1967, 1971, 1973, 1987; Niedrig, 1987, 1996; Ramsauer, 1941, 1942, 1943; Ru¨denberg, 1943; Ru¨denberg, 1992; Ru¨denberg and Ru¨denberg, 1992, 1994; Ruska, 1956a, b, 1957a, 1974, 1979, 1980, 1984, 1986, 1987; Schulze, 1992a,b, 1994, 1996; Weichan, 1985; Wolff, 1985; Wolpers, 1991; Ximen and Lenz, 1995, cf. Elektronenmikroskopie Nr. 12, 1995, 35–38). For guidance through all this material, the articles by Niedrig (1987) and Freundlich (1994) are particularly useful, while the book by Qing (1995) contains material from newspapers and obituaries as well as the better known sources. The house journals of the electron microscope manufacturers (notably Siemens and Carl Zeiss) likewise contain valuable information and the Jenaer Jahrbuch (published by Carl Zeiss Jena) is a rich mine. a. Deutsche Gesellschaft fu¨r Elektronenmikroskopie DGEM 1: Mosbach, 23–24 April 1949; Optik 5 (1949) 457–575 and Phys. Bla¨tt. 5 (1949) 237 (E. Bru¨che). DGEM 2: Bad Soden, 14–16 April 1950; Optik 7 (1950) 185–335 and Phys. Bla¨tt. 6 (1950) 327 (B. von Borries). DGEM 3: Hamburg, 18–20 May, 1951; Phys. Verhandl. 2 (1951) 63–92. DGEM 4: Tu¨bingen, 6–9 June 1952; Optik 9 (1952) 189–191 and 10 (1953) 1–205 and Phys. Verhandl. 3 (1952) 97–124. DGEM 5: Innsbruck, 16–19 September 1953; Optik 11 (1954) 97ff and Phys. Verhandl. 4 (1953) 83–124. DGEM 6: Mu¨nster, 28–31 March, 1955; Phys. Verhandl. 6 (1955) 9–42. DGEM 7: Darmstadt, 23–25 September, 1957; Phys. Verhandl. 8 (1957) 211–240. ICEM–4, Berlin, 10–17 September 1958. Coincides with DGEM 8. DGEM 9: Freiburg i. Br., 18–21 October 1959; Phys. Verhandl. 10 (1959) 189–207. DGEM 10: Kiel, 24–27 September, 1961; Mikroskopie 17 (1962) 25–56 and Phys. Verhandl. 12 (1961) 133–154. DGEM 11: Zu¨rich, 22–25 September, 1963 (with the Swiss Society); Mikroskopie 19 (1964) 1–73.

238

HAWKES

DGEM 12: Aachen, 26–30 September, 1965 (with the Dutch Society); Mikroskopie 21 (1966) 105–174. DGEM 13: Marburg, 17–21 September 1967; Optik 27 (1968) 61 and Mikroskopie 23 (1968) 61–129. DGEM 14: Wien, 22–25 September 1969 (with the Austrian Society); Optik 31 (1970) 111–112 and Mikroskopie 26 (1970) 81–144. DGEM 15: Karlsruhe, 19–23 September 1971; Optik 35 (1972) 257–345 and Mikroskopie 28 (1972) No. 11/12, 321–367. DGEM 16: Lie`ge, 3–6 September 1973 (with the Belgian and Dutch Societies); Optik 40 (1974) 233–283 and Beitra¨ge zur Elektronenmikroskopische Direktabbildung von Oberfla¨chen 6 (1974). DGEM 17: Berlin, 21–26 September 1975; Optik 45 (1976) 105–109 and Mikroskopie 32 (1976) No. 5/6, 145–190 and No. 7/8, 204–255. DGEM 18: Mu¨nster-in-Westfa¨len, 4–9 September 1977 (with the Royal Microscopical Society, ‘‘International Conference on Microprobe Analysis in Biology and Medicine’’); Optik 52 (1978/9) 261–162; Mikroskopie 34 (1978) No. 3/4, 79–124 and No. 5/6, 134–193. DGEM 19: Tu¨bingen, 9–11 September 1979; Optik 55 (1980) 217–218; Mikroskopie 36 (1980) Teil 1: 274–282 and 282–318; Teil 2: 336–355. DGEM 20: Innsbruck, 23–27 August 1981 (with the Austrian Society); Optik 62 (1982) 329–331 and Beitra¨ge zur Elektronenmikroskopische Direktabbildung von Oberfla¨chen 14 (1981). ICEM–10, Hamburg, 17–24 August 1982. DGEM 21: Antwerp, 11–16 September 1983 (with the Belgian Society); Beitra¨ge zur Elektronenmikroskopische Direktabbildung von Oberfla¨chen 16 (1983). Dreila¨ndertagung fu¨r Elektronenmikroskopie: Konstanz, 15–21 September 1985 (with the Swiss and Austrian Societies); Optik (1985) Supplement 1 or Eur. J. Cell Biol. (1985) Supplement 10. See also Beitra¨ge zur Elektronenmikroskopische Direktabbildung von Oberfla¨chen 18 (1985). DGEM 23: Bremen, 13–19 September 1987; Optik 77 (1987) Supplement 3 or Eur. J. Cell Biol. 44 (1987) Supplement 19. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Salzburg, 10–16 September 1989 (with the Swiss and Austrian Societies); Optik 83 (1989) Supplement 4 or Eur. J. Cell Biol. 49 (1989) Supplement 27. Abstracts of papers presented at the 23rd Scientific Meeting of the Israel Society of Electron Microscopy and the First Joint Israeli–German Electron Microscopy Symposium, Weizmann Institute of Science, Rehovot, 15–16 May, 1989. Ultramicroscopy 32 (1990) 185–203. Second Joint German–Israeli Electron Microscopy Symposium, MaxPlanck-Institut fu¨r Biochemie, Martinsried bei Mu¨nchen, 23–24 October

ELECTRON OPTICS AND ELECTRON MICROSCOPY

239

1990; the programme is reproduced in Elektronenmikroskopie Nr. 2 (1990) 15–16. DGEM 25: Darmstadt 1–7 September 1991; Optik 88 (1991) Supplement 4 [sic] or Eur. J. Cell Biol. 55 (1991) Supplement 33. 26th Annual Meeting of the Israel Society for Microscopy and Third Israeli–German Symposium, Hebrew University, Jerusalem, 18–19 May 1992. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Zu¨rich, 5–11 September 1993 (with the Swiss and Austrian Societies); Optik 94 (1993) Supplement 5 or Eur. J. Cell Biol. 61 (1993) Supplement 39. DGEM 27: Leipzig 10–15 September 1995; Optik 100 (1995) Supplement 6 or Eur. J. Cell Biol. 67 (1995) Supplement 41. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Regensburg, 7–12 September 1997 (with the Swiss and Austrian Societies); Optik 106 (1997) Supplement 7 or Eur. J. Cell Biol. 74 (1997) Supplement 45. DGEM 29: Dortmund, 5–10 September 1999. Optik 110 (1999) Supplement 8 or Eur. J. Cell Biol. 78 (1999) Supplement 50. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Innsbruck, 9–14 September 2001 (joint meeting with the Austrian and Swiss societies); Abstracts book (168 pp) not published as a Supplement to Optik or Eur. J. Cell Biol. DGEM 31: Dresden, 7–12 September 2003 (with the Czechoslovak Microscopy Society). Dreila¨ndertagung fu¨r Elektronenmikroskopie: Davos, 28 August–2 September, 2005 (joint meeting with the Austrian and Swiss societies).

b. DDR The first six conferences were organized by the Physikalische Gesellschaft der DDR (PG), the remainder jointly by this society and the Gesellschaft fu¨r Topochemie und Elektronenmikroskopie (GTE). After the reunification of Germany, the electron microscopy group of the Physikalische Gesellschaft joined the DGEM and the GTE was disbanded; the Arbeitsgruppe ‘‘Mikrosonde’’ is now active in the Fachverband of the Deutsche Physikalische Gesellschaft and organizes sessions in the meetings of this society. 1. Arbeitstagung ‘‘Elektronenmikroskopie,’’ Halle 1959 (PG), no proceedings. 2. Arbeitstagung ‘‘Elektronenmikroskopie,’’ Dresden 1962 (PG), no proceedings. 3. Arbeitstagung ‘‘Elektronenmikroskopie,’’ Jena 1964 (PG), no proceedings.

240

HAWKES

4. Arbeitstagung ‘‘Elektronenmikroskopie,’’ Erfurt 1966 (PG), no proceedings. 5. Arbeitstagung ‘‘Elektronenmikroskopie,’’ with 1. Arbeitstagung ‘‘Mikrosonde,’’ Dresden 1969 (PG), no proceedings. 6. Arbeitstagung ‘‘Elektronenmikroskopie,’’ Berlin 1971 (PG), no proceedings. 7. Vero¨ffentlichungen zur VII Arbeitstagung ‘‘Elektronenmikroskopie,’’ 2–5 April 1973 und II Arbeitstagung ‘‘Mikrosonde,’’ 4–7 April 1973, Berlin (Vortra¨ge and Bildanhang, 2 vols). Edited by T. Heyn, A. Ro¨der, S. Da¨britz, and L. Ku¨chler (PG and GTE). 8. Vortra¨ge der III. Arbeitstagung ‘‘Mikrosonde,’’ 29 January–1 February 1975 und der VIII. Arbeitstagung ‘‘Elektronenmikroskopie,’’ 27–30 January 1975, Berlin (Vortra¨ge and Bildanhang, 2 vols). edited by A. Ro¨der, S. Da¨britz, L. Ku¨chler, and T. Heyn (PG and GTE). 9. Vero¨ffentlichungen zur 9. Tagung ‘‘Elektronenmikroskopie,’’ 23–25 January 1978, Dresden (Vero¨ffentlichungen and Bildanhang, 2 vols). Edited by J. Heydenreich and H. Luppa (PG und GTE der DDR); Beitra¨ge zur 4. Tagung ‘‘Mikrosonde,’’ 26–28 January 1978, Dresden (PG). 10. Vero¨ffentlichungen zur 10. Tagung ‘‘Elektronenmikroskopie,’’ 19–21 January 1981, Leipzig (Vero¨ffentlichungen, 2 vols, and Bildanhang, 1 vol.), edited by H. Luppa and J. Heydenreich (GTE und PG der DDR); Beitra¨ge zur 5. Tagung ‘‘Mikrosonde,’’ 22–24 January 1981, Leipzig (Beitra¨ge and Bildanhang, 2 Vols), edited by A. Ro¨der, L. Ku¨chler and S. Da¨britz (PG). 11. Vero¨ffentlichungen zur 11. Tagung ‘‘Elektronenmikroskopie,’’ 16–18 January 1984, Dresden (Vero¨ffentlichungen, 2 vols, and Bildanhang, 1 vol.), edited by J. Heydenreich and H. Luppa (PG and GTE), report by J. Heydenreich and H. Luppa in Ultramicroscopy 13 (1984) 415–416; Beitra¨ge zur 6. Tagung ‘‘Mikrosonde,’’ 18–20 January 1984, Dresden (Beitra¨ge and Bildanhang, 2 Vols), edited by A. Ro¨der, S. Da¨britz, and L. Ku¨chler (PG). 12. Vero¨ffentlichungen zur 12. Tagung ‘‘Elektronenmikroskopie,’’ 18–20 January 1988, Dresden (Vero¨ffentlichungen, 2 vols, and Bildanhang, 1 vol.), edited by J. Heydenreich, H. Luppa and D. Stiller (PG and GTE); Beitra¨ge zur 7. Tagung ‘‘Mikrosonde,’’ 20–22 January 1988, Dresden (Beitra¨ge and Bildanhang, 2 Vols), edited by A. Ro¨der, L. Ku¨chler, and S. Da¨britz (PG). c. Halle Proceedings ‘‘High-Resolution Electron Microscopy–Fundamentals and Applications,’’ Proceedings of the Autumn School of the International Centre of Electron Microscopy at the Institute of Solid-state Physics and Electron

ELECTRON OPTICS AND ELECTRON MICROSCOPY

241

microscopy, Halle/Saale, 12–17 October 1991, J. Heydenreich and W. Neumann, eds. ‘‘Image Interpretation and Image Processing in Electron Microscopy,’’ Proceedings of the Autumn School of the International Centre of Electron Microscopy at the Max Planck Institute of Microstructure Physics, Halle/Saale, 4–9 October 1992, J. Heydenreich and W. Neumann, eds. ‘‘Analytical Transmission Electron Microscopy in Materials Science— Fundamentals and Techniques,’’ Proceedings of the Autumn School of the International Centre of Electron Microscopy at the Max Planck Institute of Microstructure Physics, Halle/Saale, 26 September–1 October 1993, J. Heydenreich and W. Neumann, eds. ‘‘Electron Microscopy of Boundaries and Interfaces in Materials Science,’’ Proceedings of the Autumn School of the International Centre of Electron Microscopy at the Max Planck Institute of Microstructure Physics, Halle/ Saale, 25 September–1 October 1994, J. Heydenreich and W. Neumann, eds. ‘‘In Situ Electron Microscopy in Materials Research—Present State and Future Prospects,’’ Autumn School of the International Centre of Electron Microscopy at the Max Planck Institute of Microstructure Physics, Halle/Saale, 24–30 September 1995; Abstracts Booklet only. There was no school in 1996; from 1997 onward, the autumn schools were held alternately in Halle and Berlin. Proceedings of the 1997 and 1999 schools were not published. ‘‘Advanced Semiconductors: Formation, Properties and Characterization of Nano-Scale Structures,’’ Autumn School organized in Halle by the Centre of Electron Microscopy and Materials Science, MPI for Microstructure Physics, Halle/Saale and the Department of Inorganic Chemistry, FritzHaber-Institute of the MPG, Berlin, 20–25 September 1997. ‘‘Metal Clusters,’’ Autumn School organized in Berlin by the Centre of Materials Science and Electron Microscopy, MPI for Microstructure Physics, Halle/ Saale and the Department of Inorganic Chemistry, Fritz-Haber-Institute of the MPG, Berlin, 1–6 November 1998. Crystal Res. Technol. 33 (1998) No. 7/8, 973–1186. ‘‘New Techniques in Electron Microscopy for Materials Sciences,’’ Autumn School organized in Halle by the Centre of Electron Microscopy and Materials Science, MPI for Microstructure Physics, Halle/Saale and the Department of Inorganic Chemistry, Fritz-Haber-Institute of the MPG, Berlin, 25–30 September 1999. ‘‘Electron Microscopy of Catalysts and Nanostructured Materials,’’ Autumn School organized in Berlin by the Centre of Electron Microscopy and Materials Science, MPI for Microstructure Physics, Halle/Saale and the Department of Inorganic Chemistry, Fritz-Haber-Institute of the MPG,

242

HAWKES

Berlin, 10–15 October 2000. Poster contributions placed on the website w3.rz-berlin.mpg.de/ac/ac.html (Su, D. S. and Wrabetz, S., eds.).

d. Arbeitskreis fu¨r Elektronenmikroskopischen Direkabbildung von Oberfla¨chen (Arbeitskreis EDO, AEDO) 1. Kolloquium u¨ber ‘‘Pra¨parationsmethoden und Aufnahmetechnik in der Raster Elektronenmikroskopie,’’ Mu¨nster, 8–9 April 1968. BEDO vol. 1, ed. by G. Pfefferkorn (Remy, Mu¨nster 1969). 2. Kolloquium AEDO, Karlsruhe, 20–21 March 1969. BEDO vol. 2, ed. by G. Pfefferkorn (Remy, Mu¨nster 1969). 3. Kolloquium ‘‘Mikroanalyse sowie mikromorphologische Abbildung und Vermessung von Oberfla¨chen,’’ Arbeitskreis ‘‘Mikrosonde’’ and Arbeitskreis EDO, Berlin, 8–11 April, 1970. BEDO vol. 3, ed. by G. Pfefferkorn (Remy, Mu¨nster 1970). 4. Kolloquium ‘‘Mikroanalyse und mikromorphologische Abbildung von Oberfla¨chen,’’ Arbeitskreis ‘‘Mikrosonde’’ and Arbeitskreis EDO, Bremen, 14–16 April 1971. BEDO vol. 4/2, ed. by G. Pfefferkorn (Remy, Mu¨nster 1971). 5. Kolloquium EDO, Graz, 1972. BEDO vol. 5, ed. by G. Pfefferkorn (Remy, Mu¨nster 1975). 6. Kolloquium EDO, Lu¨ttich/Lie`ge, 3–6 September 1973. BEDO vol. 6, ed. by G. Pfefferkorn (Remy, Mu¨nster, 1974). Together with the joint Belgian, Dutch and German Electron Microscopy Society meeting. 7. Kolloquium EDO, Frankfurt-am-Main, 4–7 June 1974. BEDO vol. 7, ed. by G. Pfefferkorn (Remy, Mu¨nster 1975). 8. Kolloquium des Arbeitskreises ‘‘Elektronenmikroskopische Direktabbildung und Analyse der Oberfla¨chen,’’ Berlin, 21–26 September 1975. BEDO vol. 8, ed. by G. Pfefferkorn (Remy, Mu¨nster 1977). 9. Kolloquium des Arbeitskreises ‘‘Elektronenmikroskopische Direktabbildung und Analyse der Oberfla¨chen,’’ Mainz, 9–11 June 1976. BEDO vol. 9, ed. by G. Pfefferkorn (Remy, Mu¨nster 1978). 10. Kolloquium des AEDO, Mu¨nster, 4–9 September 1977. BEDO vol. 10, ed. by G. Pfefferkorn (Remy, Mu¨nster 1979). This volume of BEDO also contains a laudatio for G. Pfefferkorn on his 65th birthday, essays in his honour and three papers delivered at ‘‘25 Jahre Elektronenmikroskopie in Mu¨nster’’ (20 June 1975). 11. Kolloquium AEDO, Duisburg, 4–6 September 1978. BEDO vol. 11, ed. by G. Pfefferkorn (Remy, Mu¨nster 1978). 12. Kolloquium AEDO, Tu¨bingen, 9–14 September 1979. BEDO vol. 12/1, ed. by G. Pfefferkorn (Remy, Mu¨nster 1979). BEDO vol. 12/2 (edited by G. Pfefferkorn and K. Schur) contains the proceedings of the First

ELECTRON OPTICS AND ELECTRON MICROSCOPY

243

International Conference on Emission Electron Microscopy, held in Tu¨bingen, 13–14 September 1979. 13. Kolloquium AEDO, Clausthal–Zellerfeld, 15–17 September 1980. BEDO vol. 13, ed. by G. Pfefferkorn (Remy, Mu¨nster 1980). 14. Kolloquium AEDO, Innsbruck, 23–27 August 1981. BEDO vol. 14, ed. by G. Pfefferkorn (Remy, Mu¨nster 1981). Joint with the DGEM and the ¨ GE. O 15. Kolloquium AEDO, Bremen–Vegesack, 12–15 September 1982. BEDO vol. 15/1, ed. by G. Pfefferkorn (Remy, Mu¨nster 1982). 16. Kolloquium AEDO, Antwerp, 11–16 September 1983. BEDO vol. 16, ed. by G. Pfefferkorn (Remy, Mu¨nster 1983). Joint with the DGEM and the Belgian Society of Electron Microscopy. 17. Kolloquium AEDO, Homburg/Saar, 16–19 September 1984. BEDO vol. 17, ed. by G. Pfefferkorn (Remy, Mu¨nster 1984). 18. Kolloquium AEDO, Konstanz, 15–21 September 1985. BEDO vol. 18, ed. by G. Pfefferkorn (Remy, Mu¨nster 1985). Joint with the DGEM and the Swiss and Austrian societies. 19. Kolloquium AEDO, Aachen, 16–19 September 1986. BEDO vol. 19, ed. by G. Pfefferkorn (Remy, Mu¨nster 1986). 20. Kolloquium AEDO, Bremen, 13–19 September 1987. BEDO vol. 20, ed. by G. Pfefferkorn (Remy, Mu¨nster 1987). 21. Kolloquium AEDO, Kassel, 3–6 October 1988. BEDO vol. 21, ed. by G. Pfefferkorn (Remy, Mu¨nster 1988). 22. Kolloquium AEDO, Salzburg, 10–16 September 1989. BEDO vol. 22, ed. by U. Ehrenwerth (Remy, Mu¨nster 1989). 23. Kolloquium AEDO, Berlin, 10–13 September 1990. BEDO vol. 23, ed. by U. Ehrenwerth (Remy, Mu¨nster 1990). 24. Kolloquium AEDO, Darmstadt, 1–7 September 1991. BEDO vol. 24, ed. by U. Ehrenwerth (Remy, Mu¨nster 1993). 25. Kolloquium AEDO, Congress-Hotel Koningshof, Veldhoven (Netherlands), 5–8 October 1992. BEDO vol. 25, ed. by U. Ehrenwerth (Remy, Mu¨nster 1993). 26. Kolloquium AEDO, Zu¨rich, 5–11 September 1993. BEDAO vol. 26, ed. by U. Ehrenwerth (Christel Ehrenwerth, Mu¨nster 1993). 27. Kolloquium AEDO, Saarbru¨cken, 4–7 October 1994. BEDAO vol. 27, ed. by U. Ehrenwerth (Christel Ehrenwerth, Mu¨nster 1994). 28. Kolloquium AEDO, Leipzig 10–15 September 1995. BEDAO vol. 28, ed. by U. Ehrenwerth (Christel Ehrenwerth, Mu¨nster 1995). 29. Kolloquium AEDO, Mu¨nster, 9–11 October 1996. BEDAO vol. 29, ed. by U. Ehrenwerth (Christel Ehrenwerth, Mu¨nster 1996); this was the last published volume. For details of subsequent meetings, see the DGE website.

244

HAWKES

9. Greece The Ell E Hlo Moo (Hellenic Society of Electron Microscopy) was founded in 1983 in Thessaloniki. Scientific meetings have been held regularly since 1992. Two earlier meetings should also be noted, as follows. These were organized by the Electron Microscopy Group of the Hellenic Society for Biological Sciences. Athens, 1989: Proceedings III Balkan Congress on Electron Microscopy, Athens, 18–22 September, 1989 (Margaritis, L. H., ed.). Abstracts of papers presented at the 24th Scientific Meeting of the Israel Society for Electron Microscopy and the First Joint Israeli–Greek Symposium, Bar Ilan University, Ramat Gan, 8–9 May 1990. Micron Microsc. Acta 22 (1991) No. 3, 287–308. First Scientific Meeting: Thessaloniki, 16 April 1992. Second Scientific Meeting: Thessaloniki, 8 April 1993. Third Scientific Meeting: Thessaloniki, 21 April 1994. Fourth Scientific Meeting: Karditsa, 13–14 April 1995. Fifth Scientific Meeting: Xanthi, May 1997. Sixth Scientific Meeting: Kavala, 2–4 April 1999. Seventh Scientific Meeting, Thessaloniki, 23 November 2001. 10. Hungary (picasso.elte.nu/microscopy/) For information about the organization of electron microscopy in Hungary, see Vira´gh and Csana´dy (1996). A biography of L. L. Marton, who was born in Budapest on 15 April 1901, has been written by Su¨sskind (1985). 1959: I Magyar Elektronmikroszko´pos Konferencia. 1961: II Magyar Elektronmikroszko´pos Konferencia, Tihany, 6–9 September 1961. 1963: III Magyar Elektronmikroszko´pos Konferencia, Pe´cs, 10–12 October 1963. 1965: IV Magyar Elektronmikroszko´pos Konferencia, Balatonsze´plak, 27–29 September 1965. 1967: V Magyar Elektronmikroszko´pos Konferencia, Balatonsze´plak, 5–7 September 1967. Mikroskopie 22 (1967), Nos 11–12, 344–355 (Adam, H., Pibermann, V., and Guba, F., guest eds.).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

245

1969: VI Magyar Elektronmikroszko´pos Konferencia, 6th Annual Conference of the Hungarian Electron Microscopy Society, Balatonsze´plak, 4–6 September 1969. 1971: VII Magyar Elektronmikroszko´pos Konferencia, Szeged, 26–28 August 1971. Mikroskopie 29 (1973), Nos. 1–2, 37–61 (Guba, F. and Adam, H., guest eds.); Proceedings were also issued by the Me´re´stechnikai e´s Automatiza´la´si Tudma´nyos Egyesu¨let: VII Magyar Elektronmikroszko´pos Konferencia. 1973: VIII Magyar Elektronmikroszko´pos Konferencia, Balatonfu¨red, 21–24 September 1973. 1975: IX Magyar Regiona´lis Elektronmikroszko´pos Konferencia, Veszpre´m, 20–24 August 1975 (Balaton Conference on Electron Microscopy). Short report in Micron 6 (1975) 177–180. Proceedings were published by the Scientific Society of Measurement and Automation (MATE), Budapest. 1977: X Magyar Regiona´lis Elektronmikroszko´pos Konferencia, Esztergom, 5–7 September 1977. 1979: XI Magyar Elektronmikroszko´pos e´s Mikroanalı´zis Konferencia, Szeged, 29 August –1 September 1979. Mikroskopie 38 (1981) 27–54 and 103–113 (D. Szabo´). 5th Multinational Philips Conference on ‘‘The Application of Electron Microscopes for Research in Biology, Medicine and Technology,’’ Visegrad, 1–2 April 1982. Abstracts booklet edited by the organizers, Csana´dy, A., Ro¨hlich, P. and Szabo´, D. 1982: XII Magyar Elektronmikroszko´pos e´s Mikroanalı´zis Konferencia, Eger, 29–31 March 1982. Mikroskopie 40 (1983) 103–126 and 140–154. EUREM-8, Budapest 13–18 August 1984. 1985: Joint Hungarian–Austrain Meeting, Balatonaliga 25–27 April 1985. 1987: Austrian–Hungarian Joint Conference on Electron Microscopy, Seggau-Leibnitz (Styria) 21–23 May 1987; Optik 76 (1987) Supplement 2; Eur. J. Cell Biol. 43 (1987) Supplement 18. 1991: Third Hungarian–Austrian Joint Conference on Electron Microscopy, Balatonalma´di, 19–21 September 1991. Optik 88 (1991) Supplement 5; Eur. J. Cell. Biol. 55 (1991) Supplement 34. Multinational Congress on Electron Microscopy (Italian, Hungarian, Czechoslovak and Slovenian Societies), Parma, 13–17 September 1993; Proceedings issued as Supplement to 14 (2) of Microscopia Elettronica. Proceedings Multinational Conference on Electron Microscopy, Stara´ Lesna´ (High Tatra Mountains), 16–20 October 1995, together with the Austrian, Czechoslovak, Slovenian, Croatian and Italian Societies for Electron Microscopy (Slovak Academic Press, Bratislava 1995).

246

HAWKES

MCEM-97: Proceedings Multinational Congress on Electron Microscopy, Portorozˇ (Slovenia), 5–8 October 1997, together with the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy. Part I, Microscopy Applications in the Life Sciences; Part II, Microscopy Applications in the Material Sciences; Part III, Microscopy Methods and Instrumentation. J. Computer-assisted Microsc. 8 (1996) No. 4 and 9 (1997) Nos. 1 and 2. MCEM-99: Proceedings 4th Multinational Congress on Electron Microscopy, Veszpre´m (Hungary), 5–8 September 1999, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Kova´cs, K., ed.; University of Veszpre´m 1999). MCEM-5: Proceedings of the 5th Multinational Congress on Electron Microscopy, Department of Biology, University of Lecce (Italy), 20–25 September 2001, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Dini, L. and Catalano, M., eds.; Rinton Press, Princeton NJ 2001). MCM-6: Pula (Croatia), 1–5 June 2003, uniting the Austrian, Croatian, Italian and Slovenian Societies for Electron Microscopy, the Czechoslovak Microscopy Society and the Microscopy Society of Hungary. 11. Ireland (www.nuigalway.ie/msi/) The Irish Society began life as the Irish Electron Microscope Users’ Group; in 1979, the Users’ Group became the Irish Society of Electron Microscopists, later ‘‘Microscopy,’’ Cumann Leictreon Mhiocrasco´paithe na hE´ireann. At the September 1990 meeting, the Irish Society for Electron Microscopy became the Microscopical Society of Ireland. IEMUG-1: Dublin, 19–20 May, 1977. IEMUG-2: Not traced. ISEM-3: National Institute for Higher Education, Plassey Campus, Limerick, 26–27 April 1979. ISEM-4: University College, Cork, 11–12 September 1980. Communication to the 1979 and 1980 Meetings published as a Supplement to Proc. Roy. Microsc. Soc. 15 (1980) No. 5. ISEM-5: New University of Ulster, Coleraine, 10–11 September 1981. Proc. Roy. Microsc. Soc. 16 (1981), No. 6, Supplement. ISEM-6: Trinity College Dublin, 20–22 September 1982. Proc. Roy. Microsc. Soc. 17 (1982), No. 6, Supplement. ISEM-7: University College, Galway, 21–22 September, 1983. Proc. Roy. Microsc. Soc. 18 (1983), No. 6, Supplement.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

247

ISEM-8: University College, Cork, 19–20 September, 1984. Proc. Roy. Microsc. Soc. 19 (1984), No. 6, Supplement. ISEM-9: University College, Earlsfort Terrace, Dublin, 26–27 September 1985. Proc. Roy. Microsc. Soc. 20 (1985), No. 6, Supplement. ISEM-10: University of Ulster, Jordanstown (Northern Ireland), 17–18 September 1986. Proc. Roy. Microsc. Soc. 21 (1986), No. 6, Supplement. ISEM-11: University College, Galway, 2–3 September 1987. Proc. Roy. Microsc. Soc. 23 (1988), No. 1, Supplement. ISEM-12: [Site and date not given in proceedings] 1988. Proc. Roy. Microsc. Soc. 24 (1989), No. 1, Supplement. ISEM-13: Queen’s University, Belfast [exact date not given in Proceedings] 1989. Proc. Roy. Microsc. Soc. 25 (1990), No. 3, Supplement. ISEM-14: University College, Cork, 5–6 September 1990. Proc. Roy. Microsc. Soc. 25 (1990), No. 6, Supplement. MSI-15: University of Ulster, Coleraine (Northern Ireland), [exact date not given in Proceedings] 1991. Proc. Roy. Microsc. Soc. 27 (1992), No. 1, Supplement. MSI-16: Trinity College Dublin, 2–3 September 1992. Proc. Roy. Microsc. Soc. 28 (1993), No. 1, Supplement. MSI-17: Hodson Bay Hotel, Athlone, County Roscommon, 7–9 September, 1993. Proc. Roy. Microsc. Soc. 29 (1994), No. 1, Supplement. MSI-18: University of Ulster, Coleraine (Northern Ireland), 6–8 September 1994. Proc. Roy. Microsc. Soc. 30 (1995), No. 1, Supplement. MSI-19: University College, Dublin, 5–7 September 1995. Proc. Roy. Microsc. Soc. 30 (1995), No. 4, Supplement. EUREM-11, Dublin, 26–30 August 1996, held in conjunction with the 20th Annual MSI Symposium; for communications to the latter, see Proc. Roy. Microsc. Soc. 32 (1997), No. 1, Supplement. MSI-21: The Queen’s University of Belfast, 9–11 September 1997. Proc. Roy. Microsc. Soc. 33 (1998), No. 1, Supplement. MSI-22: National University of Ireland, Galway, 1–3 September 1998. Proc. Roy. Microsc. Soc. 34 (1999) No. 2, Supplement. MSI-23: Northern Ireland Bioengineering Centre, University of Ulster at Jordanstown, 31 August–2 September 1999. Proc. Roy. Microsc. Soc. 35 (2000) No. 1, Supplement. MSI-24: University College, Dublin, 10–12 September 2000. Proc. Roy. Microsc. Soc. 35 (2000), No. 4, Supplement. MSI-25: Queen’s University Belfast, 28–30 August 2001. Proc. Roy. Microsc. Soc. 37 (2002), No. 1, Supplement. MSI-26: National University of Ireland, Galway, 28–30 August 2002. Proc. Roy. Microsc. Soc. 37 (2002), No. 4 Supplement.

248

HAWKES

MSI-27: University College Cork, 11–12 September (2003). 12. Israel (www.technion.ac.il/technion/materials/ism) The Israel Society of Electron Microscopy was created in 1964. In 1991, the name was changed to Israel Society for Microscopy. Foundation meeting: Weizmann Institute of Science, Rehovot, 1964. First Annual Scientific Meeting, Weizmann Institute of Science, Rehovot, 1966. Second Annual Scientific Meeting, 1967. Third Annual Scientific Meeting, 1968. Fourth Annual Scientific Meeting, Van Lir [Leer] Centre, Jerusalem, 4 May 1969. Fifth Annual Scientific Meeting, Weizmann Institute of Science, Rehovot, 23 December 1970. Sixth Annual Scientific Meeting, University of Haifa, Haifa, 6 June 1971. Seventh Annual Scientific Meeting, Hasharon Hospital, Petah Tikva, 27 April 1972. Eighth Annual Scientific Meeting, Bar-Ilan University, Ramat Gan, 13 December 1972. Israel J. Med. Sci. 9 (1973) 1120–1124. [No meeting in 1973 owing to the Yom Kippur war] Ninth annual Scientific Meeting, Weizmann Institute of Science, 5 June 1974. Israel J. Med. Sci. 10 (1974) 1574–1580. Tenth Annual Scientific Meeting, Ben Gurion University of the Negev, Beer-Sheva, 9 April 1975. Israel J. Med. Sci. 11 (1975) 398–408. EUREM-6, Jerusalem, 14–20 September 1976. Eleventh Annual Scientific Meeting, Bar–Ilan University, Ramat Gan, 30 March 1977. Israel J. Med. Sci. 13 (1977) 335–342 (Hammerman, I., ed.). Twelfth Annual Scientific Meeting, Faculty of Agriculture, Hebrew University, Rehovot, 29 March 1978. Thirteenth Annual Scientific Meeting, Kaplan Hospital, Rehovot, 20 March 1979. Fourteenth Annual Scientific Meeting, Silver Institute, Technion, Haifa, 15 May 1980. Israel J. Med. Sci. 17 (1981) 304–308 (Nir, E., ed.). Fifteenth Annual Scientific Meeting, Tel Aviv University, Tel Aviv, 21 May 1981. Abstracts of papers presented at the 16th Annual Scientific Meeting of the Israel Society of Electron Microscopy, Jerusalem, 5 April 1982. Ultramicroscopy 8 (1982) 442–453.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

249

Abstracts of papers presented at the 17th Annual Scientific Meeting of the Israel Society of Electron Microscopy, Weizmann Institute of Science, Rehovot, 7 April 1983. Ultramicroscopy 11 (1983) 327–339. Abstracts of papers presented at the Festive 18th Scientific Meeting of the Israel Society of Electron Microscopy, Israel Institute for Biological Research, Nes-Ziona, 26 April 1984. Ultramicroscopy 14 (1984) 375–389. Abstracts of papers presented at the 19th Scientific Meeting of the Israel Society of Electron Microscopy, Hebrew University, Jerusalem, 7 May 1985. Ultramicroscopy 17 (1985) 157–168. Abstracts of papers presented at the 20th Scientific Meeting of the Israel Society of Electron Microscopy, Kaplan Hospital, Rehovot, 29 May, 1986. Ultramicroscopy 19 (1986) 385–398. Abstracts of papers presented at the 21st Scientific Meeting of the Israel Society of Electron Microscopy, Technion–Israel Institute of Technology, Haifa, 11–12 May 1987. Ultramicroscopy 23 (1987) 229–253. Abstracts of papers presented at the 22nd Scientific Meeting of the Israel Society of Electron Microscopy, University of Tel Aviv, 7 June 1988. Ultramicroscopy 26 (1988) 423–436. Abstracts of papers presented at the 23rd Scientific Meeting of the Israel Society of Electron Microscopy and the First Joint Israeli–German Electron Microscopy Symposium, Weizmann Institute of Science, Rehovot, 15–16 May, 1989. Ultramicroscopy 32 (1990) 185–203 (Gru¨nbaum, E., guest ed.). Abstracts of papers presented at the 24th Scientific Meeting of the Israel Society for Electron Microscopy and the First Joint Israeli–Greek Symposium, Bar-Ilan University, Ramat-Gan, 8–9 May 1990. Micron Microsc. Acta 22 (1991) No. 3, 287–308. Second Joint German–Israeli Electron Microscopy Symposium, Max-PlanckInstitut fu¨r Biochemie, Martinsried bei Mu¨nchen, 23–24 October 1990; the programme is reproduced in Elektronenmikroskopie Nr. 2 (1990) 15–16. 25th Annual Meeting, Shefayim, 3–4 June 1991. 26th Annual Meeting and Third Israeli–German Symposium, Hebrew University, Jerusalem, 18–19 May 1992. 27th Annual Meeting (with the 12th Annual Meeting of the Israel Society for Histochemistry and Cytochemistry), Weizmannn Institute of Science, Rehovot, 23–24 May 1993. 28th Annual Meeting, Tel Aviv University, Tel Aviv, 7 June 1994. 29th Annual Meeting, Centre for Technological Education, Holon, 12 June 1995. 30th Annual Meeting, Technion–Israel Institute of Technology, Haifa, 30 May 1996.

250

HAWKES

31st Annual Meeting, Life Science Building, Givat Ram, Hebrew University of Jerusalem, 15 May 1997. 32nd Annual Meeting, Weizmann Institute of Science, Rehovot, 11 June 1998. 33rd Annual Meeting, Bar-Ilan University, Ramat-Gan, 2 June 1999. 34th Annual Meeting, Ben-Gurion University, Beer-Sheva, 18 May 2000. 35th Annual Meeting, Technion—Israel Institute of Technology, Haifa, 15 May 2001. 36th Annual Meeting, Inter-University Institute for Marine Sciences, Eilat, 30 April–2 May 2002. 37th Annual Meeting, Marine Science Program, Michmoret, 22 May 2003. 13. Italy (www.sime.unile.it) The Societa` Italiana di Microscopia Elettronica (SIME) was founded in 1956; for details of the development of electron microscopy in Italy, see Valdre` (1996) and Donelli et al. (1982). Since 1980, members of SIME have received Microscopia Elettronica, the Bolletino della Societa` Italiana di Microscopia Elettronica, twice a year. This contains scientific articles as well as society news. The lectures delivered at various international schools are recorded in Valdre` (1971), Valdre` and Ruedl (1976), Merli and Antisari (1992), Rickerby et al. (1999) and Calestani et al. (2002) and at an Italian Seminar in Merli and Antisari (1985). SIME, 1957: Rome, 7 May 1957. SIME, 1959: Atti del II Congresso Italiano di Microscopia Elettronica, Fondazione Carlo Erba, Palazzo Visconti, Milan, May 1959 (Societa` Italiana di Microscopia Elettronica). SIME 1961: Atti del III Congresso Italiano di Microscopia Elettronica, Fondazione Carlo Erba, Milan, 3–4 March 1961 (Societa` Italiana di Microscopia Elettronica). SIME 1963: Atti del IV Congresso Italiano di Microscopia Elettronica, Padova, 25–26 November 1963 (Societa` Italiana di Microscopia Elettronica). SIME 1965: Atti del V Congresso Italiano di Microscopia Elettronica, Bologna, 5–7 October, 1965 (Societa` Italiana di Microscopia Elettronica). SIME 1967: Atti del VI Congresso Italiano di Microscopia Elettronica, Siena, 29–31 October, 1967 (Societa` Italiana di Microscopia Elettronica). EUREM–4, Rome, 1–7 September 1968. SIME 1969: Atti del VII Congresso Italiano di Microscopia Elettronica, Modena, 28–30 September, 1969 (Societa` Italiana di Microscopia Elettronica).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

251

SIME 1971: Abstracts of the VIII Italian Congress for Electron Microscopy, Milano, 23–25 September, 1971; J. Submicrosc. Cytol. 4 (1972) 101–134. SIME 1973: Abstracts of the IX Italian Congress for Electron Microscopy, Saint-Vincent, 8–10 October, 1973; J. Submicrosc. Cytol. 6 (1974) 103–141. SIME 1974: Atti del Convegno del Gruppo Strumentazione e Tecniche non Biologiche della Societa` Italiana di Microscopia Elettronica, Laboratorio di Chimica e Tecnologia dei materiali e dei Componenti per l’Elettronica [LAMEL], Bologna, 28 June 1974 (Cooperativa Libreria Universitaria, Bologna, 1975). SIME 1975: Abstracts of the X Italian Congress for Electron Microscopy, Rosa Marina di Ostuni (Brindisi), 2–4 October, 1975; J. Submicrosc. Cytol. 8 (1976) 243–268. SIME 1977: Cosenza, 10–12 October, 1977; J. Microsc. Spectrosc. Electron. 3 (1978), No. 1, 1ab–14ab. SIME 1979: Abstracts of the Papers Presented at the 12th Congress of the Italian Society of Electron Microscopy, Ancona, 20–22 September, 1979; Ultramicroscopy 5 (1980) 363–428. SIME 1981: Titles of General Lectures and Abstracts of papers presented by the Instrumentation and Material Group, Florence, 30 September–3 October, 1981; J. Microsc. Spectrosc. Electron. 7 (1982), No. 3, 29a–78a. SIME 1983: Abstracts of the Papers Presented at the 14th Congress of the Italian Society of Electron Microscopy, Ferrara, 21–24 September, 1983; Ultramicroscopy 12 (1983/4) 87–166. SIME 1985: Abstracts of the Papers Presented at the XV Congress of the Italian Society of Electron Microscopy, Rome, 28–31 May, 1985; Ultramicroscopy 17 (1985) 399–409. SIME, 1987: Atti del XVI Congresso di Microscopia Elettronica, Bologna, 14–17 October 1987; Supplement to 8(2) of Microscopia Elettronica. SIME, 1989: Atti del XVII Congresso di Microscopia Elettronica, Lecce, 4–7 October 1989; Supplement to 10(2) of Microscopia Elettronica. SIME, 1991: Atti del XVIII Congresso di Microscopia Elettronica, Padova, 24–28 September 1991; Supplement to 12(2) of Microscopia Elettronica. MCEM-1993: Multinational Congress on Electron Microscopy (Italian, Hungarian, Czechoslovak and Slovenian Societies), Parma, 13–17 September 1993; Proceedings issued as Supplement to 14(2) of Microscopia Elettronica. SIME 1995: Atti del XX Congresso di Microscopia Elettronica, Rimini, 11–14 September 1995; Supplement to 16(2) of Microscopia Elettronica. MCEM-1995: Proceedings Multinational Conference on Electron Microscopy, Stara´ Lesna´ (High Tatra Mountains), 16–20 October 1995, together

252

HAWKES

with the Austrian, Czechoslovak, Hungarian, Slovenian and Croatian Societies for Electron Microscopy (Slovak Academic Press, Bratislava 1995). SIME 1997: XXI Congresso di Microscopia Elettronica, Taormina, 19–23 October 1997 (Proceedings not published). MCEM-97: Proceedings Multinational Congress on Electron Microscopy, Portorozˇ (Slovenia), 5–8 October 1997, together with the Austrian, Croatian, Czechoslovak and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary. Part I, Microscopy Applications in the Life Sciences; Part II, Microscopy Applications in the Material Sciences; Part III, Microscopy Methods and Instrumentation. J. Computer-assisted Microsc. 8 (1996) No. 4 and 9 (1997) Nos. 1 and 2. MCEM-99: Proceedings 4th Multinational Congress on Electron Microscopy, Veszpre´m (Hungary), 5–8 September 1999, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Kova´cs, K., ed.; University of Veszpre´m 1999). MCEM-5: Proceedings of the 5th Multinational Congress on Electron Microscopy, Department of Biology, University of Lecce (Italy), 20–25 September 2001, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Dini, L. and Catalano, M., eds.; Rinton Press, Princeton NJ 2001). MCM-6: Pula (Croatia), 1–5 June 2003, uniting the Austrian, Croatian, Italian and Slovenian Societies for Electron Microscopy, the Czechoslovak Microscopy Society and the Microscopy Society of Hungary. 14. Latvia No information available. 15. The Netherlands (www.microscopie.nl) Much has been published about early activity in electron microscopy in the Netherlands. See Le Poole (1966/7, 1978 and especially, 1985), Kruit et al. (1996), van Iterson (1996) and the book produced to accompany the 25th Anniversary meeting of the Society in Leiden in 1977 (Brederoo, 1977); the biographical memoir of J. B. Le Poole (Mulvey and van de Laak-Tijssen, 2000) is likewise very informative. See also Brederoo (1978), Spit (1978), Le Poole (1978), van Dorsten (1978), Elbers (1978), Stadhouders and van Haelst (1978), and van Bruggen (1978). The Nederlandse Vereniging voor

ELECTRON OPTICS AND ELECTRON MICROSCOPY

253

Electronenmicroscopie was founded in the Hortus Botanicus in Amsterdam as Nederlandse Kring voor Electronenmicroscopie on 4 July, 1952. Informal meetings were held annually until 1955, after which two meetings were held per year for several years. It was in 1959 that the Kring [Circle] became a Vereniging [Society], the NVEM, and the name remained unaltered until 1995, when ‘‘Electron’’ was dropped and the society became the Nederlandse Vereniging voor Microscopie (NVvM). Publication of a Jaarboek began in 1994. Bakker (1977), Henstra (1986), Thompson (1987), and Boekestein (1991) also contain much historical information. Instrumental developments may be traced through the serials and bulletins issued by Philips, notably Philips Research Reports, Philips Research Journal, Philips Electron Optics Bulletin and Norelco Reporter. Some of the meetings of the Royal Microscopical Society, listed in full in the UK section (IV.A.24), are also relevant. 1949. ICEM-1: First International Conference on Electron Microscopy, Delft, 4–8 July. 1960: EUREM-2, Delft, 29 August–3 September. Proceedings of the Symposium on Cytochemical Progress in Electron Microscopy, Oxford 2–4 July 1962. J. Roy. Microsc. Soc. 81 (1963) 106– 244. [First of a series of meetings held alternately in Holland and the UK.] NVEM 1965: Joint meeting with the DGEM, Aachen, 26–30 September, 1965; Mikroskopie 21 (1966) 105–174. Proceedings of the International Symposium on Electron Microscopy and Cytochemistry, Leiden, 31 May–4 June 1966. J. Histochem. Cytochem. 14 (1966) No. 10, 739–771. [Second of a series of meetings held alternately in Holland and the UK. The RMS is not, however, mentioned in the introductory material by R. J. Barrnett and A. M. Seligman.] NVEM 1973: Joint meeting with the DGEM and BVEM/SBME, Lie`ge, 3–6 September 1973; Optik 40 (1974) 233–283 and Beitra¨ge zur Elektronenmikroskopische Direktabbildung von Oberfla¨chen 6 (1974). NVEM 1974: Abstracts of papers presented at the Joint Meeting of the Belgian and Dutch Societies for Electron Microscopy, Rotterdam, 16–17 May 1974. J. Microscopie 19 (1974) 1a–18a. NVEM 1975: Abstracts of Papers presented at the EM Conference organized by the Netherlands Society of Electron Microscopy and the Medical Faculty of Nijmegen, Nijmegen, 27–28 November 1975. J. Microsc. Biol. Cell. 24 (1975) 175–202. NVEM 1975: Abstracts of papers presented at the EM Conference organized by the Netherlands Society of Electron Microscopy, Twente

254

HAWKES

University of Technology, Enschede, 25–26 November 1975. Ultramicroscopy 2 (1976) 113–140 (Wisse, E., ed.). NVEM 1977: Abstracts of papers presented at the Annual Conference of the Netherlands Society of Electron Microscopy, Collegegebouw, Gorlaeus Laboratoria, University of Leiden, 10–11 November 1977. Ultramicroscopy 3 (1978) 111–148 (Brederoo, P., ed.). For this 25th Anniversary meeting, the NVEM published EM Conferentie Leiden (Brederoo, P., ed.; New Rhine Publishers, Leiden 1977, ISBN: 90-6277-510-9). NVEM 1978: Abstracts of papers presented at the Annual Conference of the Dutch Society of Electron Microscopy, Amsterdam, 23–24 November 1978. Ultramicroscopy 4 (1979) 113–152 (Woldringh, C. L., ed.). NVEM 1979: Abstracts of papers presented at the Annual Conference of the Dutch Society of Electron Microscopy, Utrecht, 8–9 November 1979. Ultramicroscopy 5 (1980) 87–130 (Leene, W., ed.). 1980: EUREM-7, The Hague, 24–29 August. NVEM 1981: Abstracts of papers presented at the Annual Conference of the Dutch Society of Electron Microscopy, Lunteren, 17–18 December 1981. Ultramicroscopy 9 (1982) 401–427 (Boom, G., ed.). NVEM 1982: Annual Conference of the Netherlands Society for Electron Microscopy, Wageningen, 16–17 December 1982 [no publication]. NVEM 1983: Abstracts of papers presented at the Annual Conference of the Netherlands Society of Electron Microscopy, Wageningen, 8–9 December 1983. Ultramicroscopy 14 (1984) 391–419 (Leene, W., ed.). NVEM 1984: Joint meeting with the Belgian Society, Vrije Universiteit Brussel, 3–4 May 1984 [no publication]. NVEM 1984: Abstracts of papers presented at the Netherlands Society for Electron Microscopy and Royal Microscopical Society Joint Meeting on Image Analysis and Interpretation, Wageningen, 28–30 November 1984. Ultramicroscopy 15 (1984) 375–408 (Mast, K. van der, ed.). NVEM 1985: Abstracts of papers presented at the Annual Conference of the Netherlands Society of Electron Microscopy, Wageningen, 28–29 November 1985. Ultramicroscopy 19 (1986) 77–117 (Brederoo, P. and Priester, W. de, eds.). NVEM 1986: Abstracts of papers presented at the Annual Conference of the Netherlands Society of Electron Microscopy, International Agricultural Centre, Wageningen, 3–4 December 1986. Ultramicroscopy 21 (1987) 181–208. NVEM 1987: Abstracts of papers presented at the Annual Conference of the Netherlands Society of Electron Microscopy, International Agricultural Centre, Wageningen, 3–4 December 1987. Ultramicroscopy 24 (1988) 421–453.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

255

NVEM 1988: Abstracts of papers presented at the Annual Conference of the Netherlands Society of Electron Microscopy, Kerkrade, 30 November–2 December 1988. Ultramicroscopy 27 (1989) 193–221. NVEM 1989: Abstracts of papers presented at the Annual Conference of the Netherlands Society of Electron Microscopy, International Agricultural Centre, Wageningen, 29–30 November 1989. Ultramicroscopy 31 (1989) 455–487. NVEM 1990: Extended Abstracts of the Joint meeting of the Belgian and Dutch Societies for Electron Microscopy, Wageningen, 6–7 December 1990. Micron Microsc. Acta 21 (1990) No. 4, 207–292. NVEM 1991: Annual Conference of the Netherlands Society for Electron Microscopy, Noordwijkerhout, 11–12 December 1991 [no publication]. NVEM 1992: AntwerpEM, joint meeting with the Belgian Society, Antwerp 10–11 December 1992. Eur. J. Morphol. 31 (1993) No. 1/2, 1–110 (Biomedical communications, Roels, F. and Meir, F. van, eds.); Microsc. Res. Tech. 25 (1993) 169–186. NVEM 1993: Annual Conference of the Netherlands Society for Electron Microscopy, Arnhem, 2–3 December 1993 [no publication]. NVEM 1994: Joint Meeting with the Belgian Society, Arnhem 1–2 December 1994. Jaarboek ’94, Nederlandse Vereniging voor Electronenmicroscopie (Berendsen, W., Drost, N., and Koerten, H. K., eds.), pp. 11–123 (Karsten Drukkers, Leiden 1994). [In 1995, the name of the society changed to Nederlandse Vereniging voor Microscopie.] NVvM 1995: Meeting of the Dutch Society for Microscopy, Papendal, Arnhem 30 November–1 December 1995. Jaarboek ’95, Nederlandse Vereniging voor Microscopie (Berendsen, W., Drost, N., and Koerten, H. K., eds.) pp. 11–132. NVvM 1996: Joint Meeting with the Belgian Society, Het Pand, Gent, 12–13 December 1996. Abstracts published in Jaarboek NVvM, 1996 (Koerten, H. K., Drost, N., and Berendsen, W., eds.) pp. 9–190. NVvM 1997: Meeting of the Dutch Society for Microscopy, National Sports Centre ‘Papendal,’ Arnhem 27–28 November, 1997. Abstracts published in Jaarboek NVvM, 1997 (Koerten, H. K., Drost, N., and Berendsen, W., eds.) pp. 9–121. NVvM 1998: Meeting of the Dutch Society for Microscopy, National Sports Centre ‘Papendal,’ Arnhem 10–11 December 1998. Abstracts published in Jaarboek NVvM, 1998 (Koerten, H. K. and Berendsen, W., eds.) pp. 9–110. NVvM 1999: ‘‘Light and Matter.’’ Meeting of the Dutch Society for Microscopy, National Sports Centre ‘Papendal,’ Arnhem 16–17 December

256

HAWKES

1998. Abstracts published in Jaarboek NVvM, 1999 (Koerten, H. K. and Berendsen, W., eds.) pp. 14–99. NVvM 2000: ‘‘Microscopy beyond Traditional Limits.’’ Joint Meeting with the Belgian Society and German microscopists of the border region. Papendal, 7–8 December 2000. Abstracts published in Jaarboek NVvM, 2000 (Koerten, H. K., ed) pp. 13–193. NVvM 2001: Papendal, 13–14 December 2001. Abstracts published in Jaarboek NVvM, 2001 (Koerten, H. K., ed) pp. 13–151. 2002. Lille, 25–28 June 2002. Joint meeting with the French, Belgian and Swiss societies. Abstracts issued by the Socie´te´ Franc˛aise des Microscopies. NVvM 2003: Papendal, 11–12 December 2003. 16. Poland The earliest meeting on electron microscopy in Poland of which I am aware was held in Poznan´ in 1959 during the First Conference of the Polish Society of Anatamo-pathologists; the theme was ‘‘The role of electron microscopy in medicine and biology.’’ In 1960, the Polish Commission for Electron Microscopy was founded and from 1960–1968 the Chairman was Janusz Groniowski; he was succeeded by Andrzej Vorbrodt (1968–1975), Wincenty Kilarski (1975–1984), Wenecjusz Domaga¢a (1984–1987), Leszek Cieciura (1987–1996), and Wies¢awa Biczysko (1996–). The conferences organized by this Commission, from 1963 onwards, are listed below. In 2001, a Polish Society for Microscopy, Polskie Towarzystwo Mikroskopii, was created, with Aleksandra Czyrska–Filemonowicz as founder-president. A second series of conferences, on the electron microscopy of solids, was launched in 1969. These are listed separately. In addition to these two series of meetings, numerous symposia and workshops on more specialized topics have been held. A final list provides a record of those which have been indicated to me. Between 1974 and 1980, some international conferences of Philips electron microscope users were held in Zakopane. Conferences of the Polish Commission for Electron Microscopy 1. First Conference of the Polish Commission for Electron Microscopy, Poznan´, June 1963 (Groniowski, J., Chairman of the Organizing Committee). 2. Second Conference of the Polish Commission for Electron Microscopy, Warszawa, 1966 (Groniowski, J., Chairman of the Organizing Committee).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

257

3. Third Conference of the Polish Commission for Electron Microscopy, Krako´w, 1968 (Groniowski, J., Chairman of the Organizing Committee). 4. Fourth Conference of the Polish Commission for Electron Microscopy, Lublin, 1969 (Groniowski, J., Chairman of the Organizing Committee). 5. Fifth Conference of the Polish Commission for Electron Microscopy, Szczecin, 27–28 November 1970. 6. Sixth Conference of the Polish Commission for Electron Microscopy, Gliwice, November 1971. 7. Proceedings of the 7th Annual Conference of the Commission for Electron Microscopy (Vorbrodt, A., ed.), Uniejo´w, 26–28 November 1972. Folia Histochem. Cytochem. 11 (1973), Nos. 3/4, 293–364 (Vorbrodt, A., Chairman of the Commission; Cieciura, L. and Bartel, H., Organizing Committee). 8. Eighth Annual Conference of the Polish Commission for Electron Microscopy, Poznan´, 8–10 November 1073 (Garyel, P., Chairman of the Organizing Committee). 9. Ninth Conference of the Polish Commission for Electron Microscopy, Gdan´sk, 24–26 October 1974. 10. Tenth Conference of the Polish Commission for Electron Microscopy, Uniejo´w, 6–8 November 1975. 11. Eleventh Conference of the Polish Commission for Electron Microscopy, Kazimierz nad Wis¢a, 18–20 October 1976. 12. Twelfth Conference of the Polish Commission for Electron Microscopy, Miedzyzdroje, 27–29 October 1977 (Domaga¢a, W., Chairman of the Organizing Committee). 13. Thirteenth Conference of the Polish Commission for Electron Microscopy, Oles´nica, 5–7 October 1978. 14. 14th Polish Conference on Electron Microscopy (Polish Academy of Sciences), Jablonna, 25–27 October 1979. Acta Med. Polon. 21 (1980) No. 4. 15. Proceedings of the XVth Conference on Electron Microscopy, Rydzyna, 9–11 October 1980. Folia Histochem. Cytochem. 19 (1981) No. 3, 105 pp. (following p. 178). (Kilarski, W., Commission Chairman; Gabyrel, P., Chairman of the Organizing Committee). 16. Proceedings of the XVth [sic] Conference on Electron Microscopy, Szczecin, 15–16 September 1983. Folia Histochem. Cytobiol. 22 (1984) No. 2, 121–164 (Domaga¢a, W., Chairman of the Organizing Committee). 17. Seventeenth Conference of the Polish Commission for Electron Microscopy, Krako´w, 12–14 September 1985.

258

HAWKES

18. Eighteenth Conference of the Polish Commission for Electron Microscopy, Bydgoszcz, 23–24 September 1987 (Cieciura, L., Commission Chairman; Domaniewski, J., Chairman of the Organizing Committee). Proceedings printed in Polish and English. 19. Nineteenth Conference of the Polish Commission for Electron Microscopy, Porabka-Kozubnik, 11–12 September 1989 (Cieciura, L., Commission Chairman; Gruca, S., Chairman of the Organizing Committee). Proceedings printed in Polish and English. 20. Twentieth Conference of the Polish Commission for Electron Microscopy, Kiekrz, 28–29 November 1991 (Cieciura, L., Commission Chairman; Biczysko, W., Chairman of the Organizing Committee). Proceedings [Streszczenia z Sympozjum Mikroskopii Elektronowej, Kiekrz 1991] printed in Polish. 21. Twenty-first Conference of the Polish Commission for Electron Microscopy, Kiekrz, 16–17 December 1993 (Cieciura, L., Commission Chairman; Biczysko, W., Chairman of the Organizing Committee). 22. Twenty-second Conference of the Polish Commission for Electron Microscopy, Wroc¢aw, 7–9 September 1995 (Cieciura, L., Commission Chairman). 23. Twenty-third Conference of the Polish Commission for Electron Microscopy, Bialystok, 11–13 September 1997 (Musiatowicz, B., Chairman of the Organizing Committee; Biczysko, W., Conference Chairman). Electron microscopy of solids I. Institute of Ferrous Metallurgy, Gliwice, 1969, organized by S. Gorczyca, the proceedings (in Polish) contain 10 papers. II. University of Technology, Warsaw, 11–12 June 1971, organized by S. Gorczyca, the proceedings (in Polish) contain 36 papers. III. University of Mining and Metallurgy, Krako´w–Bartkowa, 4–7 June 1973, organized by S. Gorczyca, the proceedings (in Polish) contain 36 papers. IV. Silesian Technical University, Gliwice–Wis¢a, 19–21 May 1975, organized by J. Adamczyk and S. Gorczyca, the proceedings (in Polish) contain 90 papers. V. University of Technology, Warszawa–Jadwisin, 30 September 1979, organized by M. W. Grabski and S. Gorczyca, the proceedings (in Polish), edited by J. A. Kozubowski, contain 77 papers. VI. University of Mining and Metallurgy and Steel Plant, Krako´w– Krynica, September 1981, organized by S. Gorczyca, the proceedings (in Polish) contain 93 papers.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

259

VII. University of Mining and Metallurgy, Krako´w–Krynica, 17–20 April 1989, organized by S. Gorczyca, the proceedings (in Polish) contain 55 papers. VIII. Technical University and University of Mining and Metallurgy, Wroc¢aw–Szklarska Poreba, 20–23 April 1993, organized by H. Szyman´ski and S. Gorczyca, the proceedings (in Polish) contain 58 papers. IX. University of Mining and Metallurgy, Krako´w–Zakopane, 6–9 May 1996, organized by A. Czyrska-Filemonowicz. Proceedings of the IX Conference on Electron Microscopy of Solids, EM ’96 (CzyrskaFilemonowicz, A., Garbarz, B., Adrian, H., Wojtas, J., and ZielinskaLipiec, eds.; Fotobit, Krako´w 1996). X. Warsaw–Serock, 20–23 September 1999, organized by J. A. Kozubowski. Proceedings of the X Conference on Electron Microscopy of Solids, EM 99 (Jezierska, E. and Kozubowski, J. A., eds.; Fotobit, Krako´w 1999). XI. Hotel Motyl, Krynica, 19–23 May 2002. Abstracts volume, 150 pp. Proceedings published in Mater. Chem. Phys. (2003). Occasional Symposia Symposium on Electron Microscopy, Gdan´sk, June 1970 (Groniowski, J.). Conference of the Polish Commission for Electron Microscopy, Warsaw 1970 (Groniowski, J.). Symposium on Electron Microscopy, Poznan´ 1972 (Groniowski, J.). Symposium on Scanning Electron Microscopy, Bia¢owieza, 24–25 May 1974 (Nowak, H. F.). Proceedings were printed (in Polish): Mikroskopia Elektronowa Skaningowa. Symposium on Electron Microscopy: ‘‘Electron microscopy in oncology,’’ Poznan´, 26 June 1975. Conference on Electron Microscopy: ‘‘Collagen, its structure and function,’’ Warsaw, 16 May 1977. Conference and Workshops on Electron Microscopy: ‘‘Elastic fibres,’’ Gdan´sk, 13 May 1978 (Zawrocka-Wrzolkowa, T.). Symposium and Workshops on Electron Microscopy: ‘‘Freeze-etching and freeze-fracturing,’’ Gdansk, 9–10 September 1980 (ZawrockaWrzo¢kowa, T.). Symposium and Workshops on Electron Microscopy: ‘‘Lung tissue ultrastructure,’’ Bialystok, 29 June 1981 (Groniowski, J. and Nowak, H. F.). Conference and Workshops on Electron Microscopy: ‘‘Ultrastructure in pathology,’’ Poznan´, 22 June 1983 (Biczysko, W.). Symposium on Electron Microscopy, Warsaw, November 1983.

260

HAWKES

Symposium on Electron Microscopy: ‘‘Morphometric methods in biology and medicine,’’ Warsaw, September 1984. Conference and Workshops on Electron Microscopy: ‘‘Electron microscopy in pathology,’’ Poznan´, 21–24 April 1986 (Biczysko, W.). Polish Conference on Electron Microscopy, Bia¢ystok 1988. Conference on Cell Biology, Ło´dz´, 4–5 July 1988 (Cieciura, L.); Proceedings were printed in English. Symposium on Electron Microscopy in Memory of Professor Janusz Groniowski, Poznan´, 14 December 1990 (Biczysko, W.). Conference on Electron Microscopy, Lodz, 16–17 September 1994 (Cieciura, L.). Symposium on Electron Microscopy, Poznan, 6 December 1996 (Biczysko, W.). ‘‘The ultrastructure of prion-related diseases,’’ Poznan, December 1997. Symposium of the Commission of Genetics and Electron Microscopy, Poznan´, 10–11 December 1998. Symposium of the the Commission of Microscopy, Poznan´, 7–8 December 2000. 17. Portugal (www.spme-bc.pt) The Sociedade Portuguesa de Microscopia Electroˆnica was founded in 1966 and has held meetings in the last quarter of every year since. The abstracts distributed to participants are now all in both Portuguese and English. In 1990, the Society became the Sociedade Portuguesa de Microscopia Eletroˆnica e Biologia Celular. A much fuller account of the history of the society, prepared by A. Coimbra for the 1986 meeting of the SPME, is to be found on the society website (www.spme-bc.pt/20anos.htm) I. Porto, 1966 II. Porto, 1967 III. Coimbra, 1968 IV. Lisboa, 1969 V. Porto, 1970 VI. Coimbra, 1971 VII. Oeiras, 1972 VIII. Luanda, 1973 IX. Porto, 1974 X. Coimbra, 1975 XI. Lisboa, 1976 XII. Porto, 16–17 December 1977. Cieˆncia Biologı´ca, Mol & Cell. Biol. 3 (1978) No. 1, 1A–26A. XIII. Instituto Bota´nico, Universidade de Coimbra, December 1978. Cieˆncia Biologı´ca, Mol & Cell. Biol. 5 (1980) No. 1, 1a–21a.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

261

XIV. Oeiras, 1979. Cieˆncia Biologı´ca, Mol & Cell. Biol. 5 (1980) No. 4, 25a–41a. XV. Porto, 1980. Abstracts of papers presented at the I. Meeting of Morfologistas Portugueses (29 Reunia˜o da Sociedade Anato´mica Portuguesa e 15a Reunia˜o Anual da SPME). Cieˆncia Biologı´ca, Mol & Cell. Biol. 7 (1982) No. 1/2, 1a–30a. XVI. Coimbra, 1981. Cieˆncia Biologı´ca, Mol & Cell. Biol. 7 (1982) No. 3, 31a–52a. XVII. Oeiras, 1982. Cieˆncia Biologı´ca, Mol & Cell. Biol. 7 (1982) No. 4, 53a–79a. XVIII. Porto,1983.CieˆnciaBiologı´ca,Mol&Cell.Biol.9(1984)No.1,1–155. XIX. I. Congresso Ibe´rico de Microscopı´a Electro´nica, joint meeting with the SEME in Ronda (Ma´laga), 23–25 October 1984. Proceedings edited by F. Sa´nchez Garrido, published by Imprenta de la Universidad de Ma´laga 1984. Abstracts published in Cieˆncia Biologı´ca, Mol & Cell. Biol. 9 (1984) No. 4, 245–262. XX. Faculdade de Cieˆncias Me´dicas, Lisboa, 1985. Cieˆncia Biologı´ca, Mol & Cell. Biol. 11 (1986) No. 1/2, 1a–41a. XXI. Porto, 23–25 October 1986. Cieˆncia Biologı´ca, Mol & Cell. Biol. 12 (1987) No. 1/2, 69–98. XXII. E´vora, December 1987. Cieˆncia Biologı´ca, Mol & Cell. Biol. 13 (1988) No. 3/4, 99–127. XXIII. Fundac¸a˜o Calouste Gulbenkian, Lisboa, 14–16 December 1988, held jointly with the Spanish Society (IBEREM 88). Cieˆncia Biologı´ca, Mol & Cell. Biol. 14 (1989) No. 3/4, 83–302. XXIV. Coimbra, December 1989. Cieˆncia Biologı´ca, Mol & Cell Biol. 15 (1990) No. 1–4, 1–83. XXV. Porto, 8–9 November 1990. Cieˆncia Biologı´ca, Mol & Cell Biol. 16 (1991) No. 1–4, 1–163. XXVI. Joint meeting with the Spanish and French Societies, Barcelona, 2–5 July. Abstracts CFIME [Coloquio Franco–Ibe´rico de Microscopı´a Electro´nica, Colloque Franco–Iberique de Microscopie Electronique, Colo´quio Franco–Ibe´rico de Microscopia Electroˆnica], Universitat de Barcelona Publicacions, 429 pp. XXVII. Oeiras, December 1992. Cieˆncia Biologı´ca, Mol & Cell. Biol. 17 (1992) Nos. 1–4, 1–78. XXVIII. Oeiras, 1993 [abstracts not traced, not published in Cieˆncia Biologı´ca, Mol & Cell. Biol.]. XXIX. Coimbra, December 1994. Cieˆncia Biologı´ca, Mol & Cell Biol. 19 (1994) Nos. 1–4, 15–89. XXX. Porto, December 1995. Cieˆncia Biologı´ca, Mol & Cell Biol. 20 (1995) Nos. 1–4, 1–93.

262

HAWKES

XXXI. Oeiras, 7–8 October 1996. XXXII. Coimbra, 1997. XXXIII. Engo. Antonio de Almeida Foundation, Oporto, 9–11 December 1998. XXXIV. Rector’s Buildings, University of Lisbon, 10–11 December 1999. XXXV. Primeiro Congreso Luso-Brasileiro de Morfologia Funcional, Goiaˆnia, Goia´s (Brazil), 26–31 August 2000. Joint meeting with the Sociedade Brasileira de Anatomia, the Sociedade Brasileira de Microscopia e Microana´lise, the Sociedade Brasileira de Biologia Celular, the Associac¸a˜o Paranaense para o Desenvolvimento do Ensino da Cieˆncia and the Sociedade Anatoˆmica Portuguesa. Brazil. J. Morphol. Sci. 17 (2000) Supplement. XXXVI. Barcelona, 3–7 September 2001. Joint meeting with the Spanish and French Societies. Abstracts Microscopy Barcelona 2001 (Universitat de Barcelona 2001) 614 pp; for biological abstracts, see Biol. Cell 93 (2001) 325–439. XXXVII. University of Madeira, Funchal, 13–14 December 2002. Abstracts available as a CD-ROM. XXXVIII. La Garde, Toulon, July 2003. Joint meeting with the Spanish and French Societies. 18. Rumania No information available. 19. Scandinavia (www.scandem.org) The Skandinaviska Fo¨reningen fo¨r Elektronmikroskopi was founded in Stockholm on 16 October 1948 and its subsequent history is described by Maunsbach and Afzelius (1996). The generic title ‘‘Scandinavia’’ has been retained above, though Maunsbach tells us that this should really be ‘‘The Nordic Countries,’’ since the latter includes Finland and Iceland, whereas Scandinavia is Denmark, Norway and Sweden. In 2002, the expanded title, ‘‘Scandinavian Society for Electron Microscopy,’’ was replaced by ‘‘Nordic Microscopy Society.’’ An extremely full account of the life and work of Alvar Wilska is available (in Finnish), see Wilska (1991) and his own earlier historical article (Wilska, 1964). See too Afzelius (1981). The earlier publications do not mention the meeting number; the first to which a number is attributed is the 29th (1976). Research Institute of Experimental Physics, Stockholm, 16 October 1948. [Meeting at which the Scandinavian Society was created.]

ELECTRON OPTICS AND ELECTRON MICROSCOPY

263

Karolinska Institutet, Stockholm, 27 May 1949. Karolinska Institutet, Stockholm, 17 June 1950. Danmarks Tekniska Ho¨gskola (Danish Technical University), Copenhagen, 22 May 1951. Chemical Institute, Uppsala, 7 June 1952. Karolinska Institutet, Stockholm, 13 February 1954. Karolinska Institutet, Stockholm, 13 May 1955. EUREM–1, Stockholm, 17–20 September 1956. Norges Teknisk-Naturvitenskaplige Forskningsra¨ds Sentralinstitutt for Industriell Forskning, Blindern, Oslo, 27 May 1958. Anatomiska Institutionen, Go¨teborg (Department of Anatomy, University of Gothenburg), 12 June 1959. J. Ultrastruct. Res. 3 (1959) No. 2, 234–240. Tandlika¨reho¨gskola (School of Dentistry), Copenhagen, 1 September 1961. J. Ultrastruct. Res. 6 (1962) No. 2, 135–140. Anatomiska Institutionen (Department of Anatomy), University of Uppsala, 1 June 1962. J. Ultrastruct. Res. 8 (1963) No. 1/2, 189–196. Voksena˚sen, Oslo, 30–31 May 1963. J. Ultrastruct. Res. 9 (1963) No. 3/4, 393–401. Lund, 14–15 May 1964. J. Ultrastruct. Res. 12 (1965) No. 1/2, 232–245. ˚ rhus, 20–21 May 1965. J. Ultrastruct. Res. 14 (1966) No. 3/4, 411–426. A Anatomiska Institutionen, Gothenburg, 2–3 June 1966. J. Ultrastruct. Res. 18 (1967) No. 1/2, 224–236. ˚ bo, Turku, 1–2 June 1967. J. Ultrastruct. Res. 20 (1967) No. 3/4, 290–305. A [From now on, J. Ultrastruct. Res. covers biological subjects only; a note states that non-biological abstracts will be published ‘‘elsewhere’’ but I have not managed to trace these.] Stockholm, 4–5 June 1968. J. Ultrastruct. Res. 25 (1968) No. 1/2, 156–172. Oslo, 5–6 June 1969. J. Ultrastruct. Res. 29 (1969) No. 5/6, 563–580. Copenhagen, 4–5 June 1970. J. Ultrastruct. Res. 36 (1971) No. 3/4, 504–563. Gothenburg, 10–12 June 1971. J. Ultrastruct. Res. 38 (1972) No. 1/2, 188–212. Aarhus, 8–9 June 1972. J. Ultrastruct. Res. 42 (1973) No. 3/4, 394–414. Umea˚, 7–8 June 1973. J. Ultrastruct. Res. 44 (1973) No. 5/6, 430–450. SCANDEM-74; Helsinki, 10–12 June 1974. J. Ultrastruct. Res. 50 (1975) No. 3, 362–395. SCANDEM-75; Bergen, 2–3 June 1975. J. Ultrastruct. Res. 54 (1976) No. 3, 476–493. 29. SCANDEM-76; Lyngby, 8–9 June 1976. J. Ultrastruct. Res. 57 (1976) No. 2, 211–235. 30. SCANDEM-77; Uppsala, 6–7 June 1977. J. Ultrastruct. Res. 63 (1978) No. 1, 86–109.

264

HAWKES

31. SCANDEM-78; Tampere, 5–6 June 1978. J. Ultrastruct. Res. 66 (1979) No. 1, 78–96. 32. SCANDEM-79; Oslo, 11–13 June 1979. J. Ultrastruct. Res. 69 (1979) No. 1, 134–156. 33. SCANDEM-80; Aarhus, 9–11 June 1980. J. Ultrastruct. Res. 73 (1980) No. 1, 91–133 and erratum, 77 (1981) No. 2, 232. 34. SCANDEM-81; Department of Pathology, Huddinge Hospital, Stockholm, 3–5 June 1981. J. Ultrastruct. Res. 76 (1981) No. 3, 312–351; Ultramicroscopy 6 (1981) 409–417. 35. SCANDEM-82; Department of Cell Biology, University of Jyva¨skyla¨ (Finland), 7–9 June 1982. J. Ultrastruct. Res. 81 (1982) No. 3, 375–409; Ultramicroscopy 9 (1982) 393–399. 36. SCANDEM-83; Norwegian Institute of Technology, Trondheim, 5–8 June 1983. J. Ultrastruct. Res. 85 (1983) No. 1, 95–125; Ultramicroscopy 12 (1983/4) 265–277. 37. SCANDEM-84; Panum Institute, Faculty of Medicine, University of Copenhagen, 6–8 June 1984. J. Ultrastruct. Res. 88 (1984) No. 3, 287–311; Ultramicroscopy 13 (1984) 417–428. 38. SCANDEM-85; Vardyrkesskolan, Faculties of Medicine and Technical Sciences, University of Linko¨ ping, 10–13 June 1985. J. Ultrastruct. Res. 91 (1985) No. 3, 243–275; Ultramicroscopy 17 (1985) 169–184. 39. SCANDEM-86; University of Oulu, Linnanmaa, Oulu, 4–6 June 1986. J. Ultrastruct. Mol. Struct. Res. 94 (1986) No. 3, 268–296; Ultramicroscopy 19 (1986) 399–411. 40. SCANDEM-87; University of Bergen, 1–3 June 1987. J. Ultrastruct. Mol. Struct. Res. 98 (1988) No. 3, 325–342; Ultramicroscopy 24 (1988) 65–79. 40. [sic: 40th Anniversary meeting]. SCANDEM-88; Institute of Anatomy, Aarhus, 6–8 June 1988. J. Ultrastruct. Mol. Struct. Res. 100 (1988) No. 3, 278–307. Ultramicroscopy 26 (1988) 411–422. 41. SCANDEM-89; Uppsala. J. Ultrastruct. Mol. Struct. Res. 102 (1989) No. 3, 279–307. ˚ bo Akademi, Turku, 42. SCANDEM-90; University of Turku and A 10–13 June 1990, joint meeting with the 15th Annual Meeting of the Finnish Electron Microscopists. Micron Microsc. Acta 21 (1990) 145–191. 43. SCANDEM-91; University of Oslo, 12–14 June 1991. Micron Microsc. Acta 22 (1991) No. 1/2, 21–192. 44. SCANDEM-92; Technical University of Denmark, Lyngby, 1–3 June 1992. Micron Microsc. Acta 23 (1992) No. 1/2, 65–234. 45. SCANDEM-93; Lund, 9–11 June 1993. Proceedings Forty-fifth Annual Meeting, the Scandinavian Society for Electron Microscopy (Karlsson, G., ed.). Published by SCANDEM-93, printed in Lund.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

265

46. SCANDEM-94; Kuopio, 13–15 June 1994. Proceedings Forty-sixth Annual Meeting, the Scandinavian Society for Electron Microscopy (Tammi, R. and Sorvari, R., eds.). Published by SCANDEM-94, printed in Kuopio. 47. SCANDEM-95; Trondheim, 12–14 June 1995. Proceedings Fortyseventh Annual Meeting, the Scandinavian Society for Electron Microscopy (T. Beisva˚g, T., Iversen, T.-H., and Solberg, J. K., eds.). Published by SCANDEM-95, printed in Trondheim. 48. SCANDEM-96; Aarhus, 2–5 June 1996. Proceedings Forty-eighth Annual Meeting, the Scandinavian Society for Electron Microscopy (Maunsbach, A. B., ed.). Published by SCANDEM-96, printed in Aarhus. 49. SCANDEM-97; Chalmers University of Technology, Go¨teborg, 10–13 June 1997. Proceedings Forty-ninth Annual Meeting, the Scandinavian Society for Electron Microscopy (Tho¨le´n, A. R., ed.). Published by SCANDEM-97, printed in Go¨teborg. 50. SCANDEM-98; Helsinki University of Technology in Otaniemi, Espoo and the University of Helsinki, Viiki Biocenter, Helsinki, 7–10 June 1998. Extended Abstracts of the Fiftieth Annual Meeting of the Scandinavian Society for Electron Microscopy (Punnonen, E.-L. and Heikinheimo, E., eds.). Published by SCANDEM-98, printed in Espoo. 51. SCANDEM-99; Realfagsbygget, UiB, Bergen, 2–5 June 1999. Proceedings, the Fifty-first Annual Meeting, the Scandinavian Society for Electron Microscopy (Bjørkelund, O. A., ed.). Published by SCANDEM99, printed in Bergen. [No meeting in 2000] 52. SCANDEM-2001; Huddinge (Stockholm), 12–15 June 2001. Final Programme and Proceedings SCANDEM 2001 (Hebert, H. and Sundberg, M., eds.). Published by SCANDEM. 53. SCANDEM-2002; Tampere Hall, Tampere, 12–15 June 2002. Proceedings SCANDEM 2002 (Kera¨nen, J. and Sillanpa¨a¨, K., eds.). Published by SCANDEM. 54. SCANDEM-2003; University Library, University of Oslo, Blindern, 10–13 June 2003. 20. Slovenia 1993: Multinational Congress on Electron Microscopy (Italian, Hungarian, Czechoslovak and Slovenian Societies), Parma, 13–17 September 1993; Proceedings issued as Supplement to 14(2) of Microscopia Elettronica.

266

HAWKES

1995: Proceedings Multinational Conference on Electron Microscopy, Stara´ Lesna´ (High Tatra Mountains), 16–20 October 1995: Austrian, Croatian, Czechoslovak, Hungarian, Italian and Slovenian Societies for Electron Microscopy (Slovak Academic Press, Bratislava 1995). 1997: MCEM-97. Proceedings Multinational Congress on Electron Microscopy, Portorozˇ (Slovenia), 5–8 October 1997, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary. Part I, Microscopy Applications in the Life Sciences; Part II, Microscopy Applications in the Material Sciences; Part III, Microscopy Methods and Instrumentation. J. Computer-assisted Microsc. 8 (1996) No. 4 and 9 (1997) Nos. 1 and 2. MCEM-99. Proceedings 4th Multinational Congress on Electron Microscopy, Veszpre´m (Hungary), 5–8 September 1999, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Kova´cs, K., ed.; University of Veszpre´m 1999). MCEM-5. Proceedings of the 5th Multinational Congress on Electron Microscopy, Department of Biology, University of Lecce (Italy), 20–25 September 2001, uniting the Austrian, Croatian, Czechoslovak, Italian and Slovenian Societies for Electron Microscopy and the Microscopy Society of Hungary (Dini, L. and Catalano, M., eds.; Rinton Press, Princeton NJ 2001). MCM-6. Pula (Croatia), 1–5 June 2003, uniting the Austrian, Croatian, Italian and Slovenian Societies for Electron Microscopy, the Czechoslovak Microscopy Society and the Microscopy Society of Hungary. 21. Spain (sme.cnb.uam.es) The Sociedad Espan˜ola de Microscopı´a Electro´nica (SEME) was founded in 1956. A few details of the development of electron microscopy in Spain are to be found in Bru´ (1992); a longer account is in preparation for Advances in Imaging & Electron Physics. A serial Microscopı´a, Boletin de la Sociedad Espan˜ola de Microscopı´a Electro´nica, has been distributed to members since January 1998 and is also available on the society website. In 1999, the name of the society was changed to Sociedad de Microscopı´a de Espan˜a. Reunio´n de la SEME, Madrid, 1960. Reunio´n de la SEME, Madrid, 1962. I. Reunio´n de la SEME, Madrid, 1964. II. Reunio´n de la SEME, Sevilla, 1966. III. Reunio´n de la SEME, Madrid, 1968.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

267

IV. Reunio´n de la SEME, Lloret de Mar (Barcelona), 1970. V. Reunio´n de la SEME, Cullera (Valencia), 2–4 November 1972. Proceedings published by Universidad Polite´cnica de Valencia, Valencia 1973. VI. Reunio´n de la SEME, Salamanca, 2–4 October 1974. Proceedings published by Graficesa, Salamanca 1974. VII. Reunio´n de la SEME, Cordo´ba, 1976. VIII. Reunio´n Bienal de la SEME, Bilbao, 4–7 July, 1978. IX. Reunio´n de la SEME, Las Palmas (Gran Canaria), 2–4 October 1980. X. Reunio´n de la SEME, La Corun˜a, 21–23 October 1982. XI. CIME: Congreso Ibe´rico de Microscopı´a Electro´nica (XI Reunio´n de la SEME held jointly with the Sociedade Portuguesa de Microscopia Electro´nica), Ronda (Ma´laga), 23–25 October 1984. Proceedings edited by F. Sa´nchez Garrido, published by Imprenta de la Universidad de Ma´laga 1984. Abstracts published in Cieˆncia Biologı´ca, Mol & Cell. Biol. 9 (1984) No. 4, 245–262. XII. Congreso Nacional de Microscopı´a Electro´nica, Reunio´n Bienal de la SEME, Avila, 29–31 October 1986. XIII. ‘‘IBEREM 88.’’ Reunio´n conjunta de la SEME y de la Sociedade Portuguesa de Microscopia Electro´nica, Lisboa, 14–16 December 1988. XIV. Congreso Nacional de Microscopı´a Electro´nica, Reunio´n Bienal de la SEME, Ca´diz, 10–13 December 1990. Proceedings edited by J. Vilches and A. Lo´pez, published by Jime´nez-Mena Artes Gra´ficas, Ca´diz 1992. XV. CFIME: Coloquio Franco-Ibe´rico de Microscopı´a Electro´nica, Barcelona 2-5.7.91. (with the SFME and the SPME). Abstracts published by Universitat de Barcelona Publicacions, 1991, 429 pp. XVI. EUREM-10, Granada, 7–11 September 1992. XVII. Reunio´n Bienal de la SEME, Oviedo, 5–8 April 1995. Proceedings published by Gra´ficas Oviedo 1995. XVIII. Reunio´n Bienal de la SEME, Toledo, 15–18 April 1997. Abstracts volume, 292 pp. XIX. Reunio´n Bienal de la SEME, Centro de Congresos, Murcia, 28–30 April 1999. Abstracts volume (Sa´nchez-Pina, M. A., Garcı´a, M., Ferrer, C., and Zuasti, A., eds.), 443 pp. XX. Reunio´n Bienal de la SME (with the French and Portuguese Societies), Barcelona, 3–7 September 2001. Abstracts Microscopy Barcelona 2001 (Universitat de Barcelona 2001) 614 pp; for biological abstracts, see Biol. Cell 93 (2001) 325–439. XXI. Reunio´n Bienal de la SME, Cadiz, 28 September–1 October 2003.

268

HAWKES

22. Switzerland (www.ssom.ch) The development of electron microscopy in Switzerland is presented in detail in a book edited by Gu¨nther (1990), from which the details of earlier meetings listed below have been extracted. The Swiss Society for Optics and Electron Microscopy (Schweizerische Gesellschaft fu¨r Optik und Elektronenmikroskopie, Socie´te´ Suisse d’Optique et de Microscopie Electronique) distributes a Mitteilungsblatt, Bulletin d’Information to its members, which contains the programmes of the annual meetings. In 1997 the name of the society was changed to Swiss Society for Optics and Microscopy, Schweizerische Gesellschaft fu¨r Optik und Mikroskopie, Socie´te´ Suisse pour l’Optique et la Microscopie. Schweizerisches Komitee fu¨r Optik, Comite´ Suisse d’Optique 1. Zu¨rich, 23 June 1954 Schweizerisches Komitee fu¨r Licht- und Elektronenoptik, Comite´ Suisse d’Optique Photonique et Electronique 1. Lausanne, 29 June 1955 2. Neuchaˆtel, 26 May 1959 3. Fribourg, 9 November 1962 4. Zu¨rich, 22–25 May 1963, joint meeting with the DGE; Mikroskopie 19 (1964) 1–73. 5. Berne, 20 October 1965 6. Basel, 4 November 1966 7. Zu¨rich, 20 October 1967 Schweizerische Gesellschaft fu¨r Optik und Elektronenmikroskopie, Socie´te´ Suisse d’Optique et de microscopie Electronique 1. Lausanne, 19 May 1969, joint meeting with the SFME; J. Microscopie 8 (1969) No. 4, 1a–99a. 2. Fribourg, 8 October 1971 3. Lugano, 19–20 October 1973 4. Zu¨rich, 30 October 1974 5. Aarau, 3–4 October 1975 6. Zu¨rich, 14–15 October 1976 7. Berne, 7 October 1977 8. Basel, 12–13 October 1978, see report by J. R. Gu¨nther in Ultramicroscopy 4 (1979) 107. 9. Lausanne, 5–6 October 1979

ELECTRON OPTICS AND ELECTRON MICROSCOPY

269

10. Lausanne, 13–14 October 1980 11. Davos, 26 September 1981 12. Vaduz, 30 September 1982 13. Berne, 22–23 September 1983 14. Zu¨rich, 5 October 1984 15. Dreila¨ndertagung fu¨r Elektronenmikroskopie, Konstanz, 15–21 September 1985, joint meeting with the German and Austrian societies; Optik (1985) Supplement 1 or Eur. J. Cell Biol. (1985) Supplement 10. See also Beitra¨ge zur Elektronenmikroskopische Direktabbildung von Oberfla¨chen 18 (1985). 16. Berne, 10 October 1986 17. Fribourg, 28 January 1987, joint meeting with the Schweizerische Arbeitsgemeinschaft fu¨r Oberfla¨chen und Grenzfla¨chen. 18. Neuchaˆtel, 24 September 1987 19. Lausanne, 29–30 September 1988 20. Grenoble, 10–12 July 1989, joint meeting with the SFME; J. Microsc. Spectrosc. Electron. 14 (1989) No. 2, 1a–80a and 315–384 and 387–414. 21. Dreila¨ndertagung fu¨r Elektronenmikroskopie, Salzburg, 10–16 September 1989, joint meeting with the German and Austrian societies; Optik 83 (1989) Supplement 4 or Eur. J. Cell Biol. 49 (1989) Supplement 27. 22. Zu¨rich, 27 October 1989 23. Berne, 26 October 1990 24. Basel, 25 October 1991. ‘‘Rund um die Mikroskopie.’’ 25. Berne, 23 October 1992. ‘‘Analytical Transmission Electron Microscopy–Applications.’’ 26. Dreila¨ndertagung fu¨r Elektronenmikroskopie, Zu¨rich, 5–11 September 1993, joint meeting with the German and Austrian societies; Optik 94 (1993) Supplement 5 or Eur. J. Cell Biol. 61 (1993) Supplement 39. 27. Basel, 4 November 1994. ‘‘Microscopy in Basel.’’ 28. Lausanne, 26–30 June 1995, joint meeting with the SFME and the SBME. For biological abstracts, see Biol. Cell 84 (1995) No. 3, 219–234. 29. Zurich, 18 October 1996. ‘‘Elektronenmikroskopie in Zu¨rich.’’ 30. Dreila¨ndertagung fu¨r Elektronenmikroskopie, Regensburg, 7–12 September 1997, joint meeting with the German and Austrian societies; Optik 106 (1997) Supplement 7 or Eur. J. Cell Biol. 74 (1997) Supplement 45. 31. Strasbourg, 29 June–3 July 1998, joint meeting with the French and Belgian Societies. For biological abstracts, see Biol. Cell 90 (1998) 247–292. 32. Neuchaˆtel, 15–17 September 1999. ‘‘50 Jahre SSOM.’’ 33. Basel, 28–29 September 2000. Herbsttagung der Sektion Mikroskopie der SSOM.

270

HAWKES

34. Dreila¨ndertagung fu¨r Elektronenmikroskopie, Innsbruck, 9–14 September 2001 (joint meeting with the Austrian and German societies). Abstracts book (168 pp) not published as a Supplement to Optik or Eur. J. Cell Biol. 35. Lille, 25–28 June 2002. Joint meeting with the Belgian, French and Dutch Societies. Abstracts volume issued by the Socie´te´ Franc˛aise des Microscopies. 36. EPFL Lausanne, 4 October 2002. Herbsttagung der Sektion Mikroskopie der SSOM: ‘‘Mikroskopie in Lausanne und Genf/Microscopie a` Lausanne et Gene`ve.’’ 37. Dreila¨ndertagung fu¨r Elektronenmikroskopie: Davos, 28 August–2 September, 2005 (joint meeting with the Austrian and German societies). 23. Turkey (www.temd.org) The Tu¨rk Elektron Mikroskopi Dernegˇi (Turkish Society of Electron Microscopy), formally founded in 1971, has met regularly since 1970. Abstracts booklets are distributed at the meetings but not otherwise published. First National Symposium on Electron Microscopy, Ankara, 11–12 May 1970. Second Symposium on Electron Microscopy, Istanbul, 2–5 May 1972. Third Symposium on Electron Microscopy, Izmir, 23–25 September 1974. Fourth Symposium on Electron Microscopy, Istanbul, 2–6 February 1976. Abstracts of Communications, Second Balkan Congress on Electron Microscopy, Istanbul, 25–30 September 1977 (Erbengi, T., Chairman Sci. Prog. Comm.; Istanbul Faculty of Medecine and Turkish Society of Electron Microscopy, Istanbul). Sixth National Conference on Electron Microscopy, Istanbul, 20–25 September 1981 [with international participation]. Seventh Congress on Electron Microscopy, Istanbul, 16–20 September 1985 [with international participation]. Eighth National Conference on Electron Microscopy, Sivas, 17–20 June 1987. Ninth National Conference on Electron Microscopy, Istanbul 29–30 May 1989. Tenth National Conference on Electron Microscopy, Istanbul, 18–20 September 1991 [with international participation]. Eleventh National Conference on Electron Microscopy, Edirne, 8–10 September 1993.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

271

Twelfth National Conference on Electron Microscopy, Antalya, 11–15 September 1995 [with international participation]. Thirteenth National Congress of Electron Microscopy, Middle East Technical University, Ankara, 1–4 September 1997 [with international participation]. Proceedings/Bildiriler Kitabı´ edited by E. Tekin and Y. Canberk, 887 pp. Fourteenth National Electron Microscopy Congress, Kervansaray Termal Hotel, Bursa, 29 September–2 October 1999 [with international partici¨ zet Kitabı´/Abstract Book, 155 pp. Organized on behalf of the pation]. O Turkish Electron Microscopy Society by the Departments of Histology and Embryology, Faculties of Medicine and Veterinary Studies and of Engineering and Architecture, Uludagˇ University. Fifteenth National Electron Microscopy Congress, Pine Bay Holiday Resort, Kusˇadası´ 18–21 September 2001 [with international participation]. Abstracts book, 144 pp. Organized on behalf of the Turkish Electron Microscopy Association by Gazi University. Sixteenth National Electron Microscopy Congress, Izmir, 2–5 September 2003. 24. The United Kingdom (www.rms.org.uk, groups.iop.org/em/) Electron microscopy was represented in Great Britain for many years by the British Joint Committee for Electron Microscopy (see Nature 188, 1960, 104), which originally represented twelve learned societies and finally consisted of representatives of the Institute of Physics, the Royal Microscopical Society and the Institute of Materials. The BJCEM was formally disbanded in 1999. The Electron Microscopy Group of the Institute of Physics was formed in 1946 and a first meeting was held in Oxford in September of that year; a detailed account is given by Cosslett (1971). A Newsletter is now distributed to members of the Electron Microscopy and Analysis Group twice a year though in the past it appeared irregularly: the Newsletter dated December 1975 begins ‘‘It is some considerable time since the group produced a Newsletter . . . ’’ The Royal Microscopical Society has long been active in electron microscopy; the inaugural meeting of the Electron Microscopy Section of the RMS was held at Baden Powell House, in London, 30 September–1 October 1965; abstracts of papers presented on the second day are to be found in the Proceedings of the Royal Microscopical Society. Details of this and some other meetings are listed separately below; many of these have been co-sponsored by the RMS and another organization, such as the Institute of Physics or the Institute of Metals—the references should therefore be consulted if exact information about the roles and participation of the various players is required. Most of the earlier meetings were

272

HAWKES

organized around a specific theme, and were not general meetings, unlike the majority of those listed in this section on national meetings. Since 1970, however, the Royal Microscopical Society has organized an annual meeting, MICRO, in London, usually in July, and this is of a general character; electron microscopy, which originally played a very modest part in these conferences, now occupies a substantial role. MICRO meetings are not listed individually here, the full programme is easily found in Proc. Roy. Microsc. Soc. each year beginning at volume 5 (1970), Part 5; since 1992, a special issue of Microscopy and Analysis has frequently been devoted to it. The 1989 MICRO was held in conjunction with the EMAG meeting and the proceedings are listed below; full proceedings of the 1990 MICRO were also published (Elder, 1990). In 2002, the name changed to MicroScience (9–11 July 2002). Although it is not strictly the record of a conference, we also mention a special issue of the Journal of the Royal Microscopical Society (Drummond, 1950), which contains an ‘‘extensive treatise on the practice of electron microscopy,’’ prepared by the Electron Microscopy Group of the Institute of Physics, in cooperation with the Royal Microscopical Society (see EMG 1945 below). A series of conferences on the Microscopy of Semiconducting Materials (MSM) is held biennially, organized alternately by the RMS and EMAG; these are somewhat arbitrarily listed with the RMS meetings. One-day meetings are held on instrumentation and applications in both the life and physical sciences related to the transmission electron microscope with a field-emission gun (FEGTEM). The dates are listed below. The decision to construct ‘‘a suite of ultra-high resolution imaging and analytical scanning transmission electron microscopes (STEMs) at Daresbury Laboratories in Cheshire’’ has given rise to to two SuperSTEM Workshops, at the EMAG 2001 conference in Dundee and at the RMS MicroScience meeting in 2002. In 1970, the Society of Electron Microscope Technology was founded ‘‘by microscopists who needed a forum for the exchange of information on techniques and applications.’’ An annual one-day meeting (occasionally extended to two days) and a series of specialized afternoon meetings are organized by the committee of the Society. Brief details of these meetings are listed after the MSM section. A Summer School held in Cambridge in July 1963 inspired a book that has become a classic (Hirsch et al., 1965). A ‘‘Discussion on new developments in electron microscopy with special emphasis on their application in biology’’ was held at the Royal Society of London on 12–13 March 1970 (Huxley and Klug, 1971).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

273

The four Users Groups in Scotland (Edinburgh, Tayside, Aberdeen and West of Scotland) have held annual symposia on microscopy since 1972, rotating between the different regions; prior to that, two annual meetings were organized by the Edinburgh Microscopical Society (Elder, 1996). The creation of a Welsh Microscopical Society was announced in 1990 (Proceedings of the Royal Microscopical Society 25, 1990, 413). Two of the Scottish Summer Schools in Physics have been devoted to electron microscopy and microanalysis, see Chapman and Craven (1984) and Fitzgerald et al. (1993). Many articles on the development of electron microscopy in Britain have been published: see Agar (1986), Brown et al. (1996); Cosslett (1971, 1979, 1981, 1982, 1985, 1986), Everhart (1996), Hawkes (1979), Jervis (1971/2), Manton (1978), McMullan (1985, 1989, 1990, 1993, 1995), Mulvey (1985, 1986, 1989), Oatley (1982), Oatley et al. (1965, 1985), Preston (1983), Reed (1985), Smith (1997), and Stewart (1985). Breton et al. (2004) is devoted to Oatley’s contribution to the development of the scanning electron microscope. Institute of Physics [EMG] 1943: Royal Society, London, 29 November 1943. [EMG] 1944: National Physical Laboratory, June 1944. [EMG] 1945: National Institute of Medical Research, Mill Hill, June 1945; it was at this meeting that ‘‘it was agreed that a compilation should be made of the methods used in the practice of electron microscopy, particularly in regard to the techniques of specimen preparation.’’ This subsequently appeared in print as Drummond (1950), after being issued as a typescript edition by the Institute of Physics in 1948. [EMG] 1946/1: Midland Hotel, Manchester, 16–17 January 1946. EMG 1946/2: Clarendon Laboratory, Oxford, 17–18 September 1946; J. Sci. Instrum. 24 (1947) 113–119 (V. E. Cosslett). EMG 1947/1: British Medical Association, Tavistock Square, London, 20–21 March 1947; J. Sci. Instrum. 25 (1948) 23–27 (J. Sayer, E. F. Brown, and J. B. Todd). EMG 1947/2: Physics Department, University of Leeds, 16–17 September 1947; J. Sci. Instrum. 25 (1948) 167–170 (V. E. Cosslett). EMG 1948/1: Anatomy Theatre and Physics Department, King’s College London, 7–8 April 1948; J. Sci. Instrum. 25 (1948) 328–331 (V. E. Cosslett). EMG 1948/2: Cavendish Laboratory, Cambridge, 20–23 September 1948; J. Sci. Instrum. 26 (1949) 163–169 and Nature 163 (1949) 32–34 (V. E. Cosslett).

274

HAWKES

EMG 1949/1: The investigation of biological systems by the electron microscope and by x-ray analysis, Buxton, May 1949; Brit. J. Appl. Phys. 1 (1950) 57–59 (I. M. Dawson and M. F. Perutz). EMG 1949/2: Metallurgical Applications of the Electron Microscope, Royal Institution, London 16 November 1949 (organized by the Institute of Metals and 7 other Learned Societies); IoM Monograph and Report Series, No. 8 (Institute of Metals, London 1950); Nature 165 (1950) 390–393 (C. J. B. Clews). EMG 1950: Joint Convention on Modern Microscopy (with RMS), Newcastle, 18–21 April. EMG 1951: St Andrews, 19–20 June 1951, Brit. J. Appl. Phys. 3 (1952) 25–29 and Nature 168 (1951) 819–821 (D. G. Drummond and G. Liebmann). EMG 1952: H. H. Wills Physical Laboratory, University of Bristol, 16–18 September 1952; Brit. J. Appl. Phys. 4 (1953) 1–5 (V. E. Cosslett, J. B. Nutting, and R. Reed) and Nature 170 (1952) 861–863 (V. E. Cosslett). EMG 1953/1: Recent Research in Electron Optics, Imperial College, London, 15–16 May 1953; Nature 172 (1953) 61–62 (O. Klemperer). EMG 1953/2: Birkbeck College, London, 10–11 November; Brit. J. Appl. Phys. 5 (1954) 165–170 and Nature 173 (1954) 340–341 (C. E. Challice). ICEM–3, London, 15–21 July 1954. EMG 1955: Department of Chemistry, University of Glasgow, 5–7 July 1955; Brit. J. Appl. Phys. 7 (1956) 89–93 (C. E. Challice). EMG 1956/1: Electron Microscopy of Fibres, Department of Textile Industries, University of Leeds, 3–4 January 1956; Brit. J. Appl. Phys 8 (1957) 1–8 and erratum 218 (C. E. Challice and J. Sikorski). EMG 1956/2: Departments of Chemistry and Physics, University of Reading, 24–26 July 1956; Brit. J. Appl. Phys. 8 (1957) 259–269 (C. E. Challice). EMAG 1957: Department of Botany, University College of North Wales, Bangor, 10–12 September 1957; Brit. J. Appl. Phys. 9 (1958) 306–312 (H. W. Emerton). EMG 1958: On Precipitation in Alloys and on Metal Structures (with X-ray Analysis Group), London, 28 November 1958; Brit. J. Appl. Phys. 10 (1959) 438–444 (A. Franks and R. S. M. Revell). EMG 1959: Washington Singer Laboratory, University of Exeter, 7–10 July 1959; Brit. J. Appl. Phys. 11 (1960) 22–32 (J. A. Chapman and M. J. Whelan). EMG 1960: Co-ordination of Light and Electron Microscopy (with RMS), Leeds, 31 March–1 April 1960; J. Roy. Microsc. Soc. 79 (1959–1960), Part 3, 179–274; Nature 186 (1960) 672–673 (V. E. Cosslett). EMG 1961: Department of Chemistry, University of Nottingham, 10–14 July; Brit J. Appl. Phys. 12 (1961) 585–591 (P. M. Kelly and R. Reed).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

275

EMAG 1963/1: Cavendish Laboratory, Cambridge, 2–6 July 1963; Brit. J. Appl. Phys. 14 (1963) 733–740 (R. B. Nicholson, W. C. Nixon, and D. H. Warrington). EMAG 1963/2: Electron Probe Microanalysis, Department of Physics, University of Reading, 26–27 September 1963; Brit. J. Appl. Phys. 15 (1964) 113–120 and J. Sci. Instrum. 41 (1964) 61–65 (P. Duncumb, J. V. P. Long, and D. A. Melford). EMAG 1964: Use of Electron Microscopy, Diffraction and Probe Analysis in the Identification of Precipitates in Solids, National Physical Laboratory, Teddington, 16–17 April 1964; Brit J. Appl. Phys. 15 (1964) 867–870 (K. F. Hale and R. F. Braybrook). EMAG 1965: Non-conventional Electron Microscopy, Engineering Department, Cambridge, 31 March-2 April 1965. EMAG 1967: Electron Optics, Instrumentation and Quantitative Electron Microscopy, Buchanan Arts Theatre, University of St, Andrews, 19–21 September. EMAG 1968: Scanning Electron Microscopy, Cambridge, 8–10 July; Phys. Bull. 19 (1968) 423–424 (K. C. A. Smith). Organized jointly with the Electron Microscopy section of the RMS. EMAG 1969/1: Dynamic Experimentation in Electron Optical Instruments, Imperial College, London, 14–15 April. EMAG 1969/2: Non-conventional Electron Microscopy, St, Catherine’s College, Oxford, 14–16 July. EMAG 1970/1: Application of High-voltage Electron Microscopy, Harwell, 2–3 April. Organized jointly with the Electron Microscopy section of the RMS. EMAG 1970/2: Scanning Electron Microscopy in Materials Science, Newcastle-upon-Tyne, 7–9 July. Organized jointly with the Electron Microscopy section of the RMS. EMAG, 1971: Electron Microscopy and Analysis. Proceedings of the 25th Anniversary Meeting of the Electron Microscopy and Analysis Group of the Institute of Physics, Cambridge, 29 June-1 July 1971 (Nixon, W. C., ed.; Institute of Physics, London 1971) Conference Series 10. EUREM-5, Manchester, 5–12 September 1972. EMAG, 1973: Scanning Electron Microscopy: Systems and Applications, Newcastle-upon-Tyne, 3–5 July 1973 (Nixon, W. C., ed.; Institute of Physics, London, 1973) Conference Series 18. EMAG, 1975: Developments in Electron Microscopy and Analysis. Proceedings of EMAG 75, Bristol 8–11 September 1975 (Venables, J. A., ed.; Academic Press, London and New York, 1976).

276

HAWKES

EMAG 1977: Developments in Electron Microscopy and Analysis. Proceedings of EMAG 77, Glasgow, 12–14 September 1977 (Misell, D. L., ed.; Institute of Physics, Bristol, 1977) Conference Series 36. EMAG, 1979: Electron Microscopy and Analysis, 1979. Proceedings of EMAG 79, Brighton, 3–6 September 1979 (Mulvey, T., ed.; Institute of Physics, Bristol, 1980) Conference Series 52. EMAG, 1981: Electron Microscopy and Analysis, 1981. Proceedings of EMAG 81, Cambridge, 7–10 September 1981 (Goringe, M. J., ed.; Institute of Physics, Bristol, 1982) Conference Series 61. EMAG, 1983: Electron Microscopy and Analysis, 1983. Proceedings of EMAG 83, Guildford, 30 August-2 September 1983 (Doig, P., ed.; Institute of Physics, Bristol, 1984) Conference Series 68. EMAG, 1985: Electron Microscopy and Analysis, 1985. Proceedings of EMAG 85. Newcastle-upon-Tyne, 2–5 September 1985 (Tatlock, G. J., ed.; Institute of Physics, Bristol, 1986) Conference Series 78. EMAG, 1987: Electron Microscopy and Analysis, 1987. Proceedings of EMAG 87, Manchester, 8–9 September 1987 (Brown, L. M., ed.; Institute of Physics, Bristol and Philadelphia, 1987) Conference Series 90. EUREM-9, York, 4–9 September 1988. EMAG, 1989: EMAG-MICRO 89. Proceedings of the Institute of Physics Electron Microscopy and Analysis Group and Royal Microscopical Society Conference, London, 13–15 September 1989 (Goodhew, P. J. and Elder, H. Y., eds.; Institute of Physics, Bristol and New York, 1990) Conference Series 98, 2 Vols. EMAG, 1991: Electron Microscopy and Analysis 1991. Proceedings of EMAG 91, Bristol, 10–13 September 1991 (Humphreys, F. J., ed.; Institute of Physics, Bristol, Philadelphia and New York, 1991) Conference Series 119. EMAG, 1993: Electron Microscopy and Analysis 1993. Proceedings of EMAG 93, Liverpool, 15–17 September 1993 (Craven, A. J., ed.; Institute of Physics, Bristol, Philadelphia and New York, 1994) Conference Series 138. EMAG 1995: Electron Microscopy and Analysis 1995. Proceedings of EMAG 95. Birmingham, 12–15 September 1995 (Cherns, D., ed.; Institute of Physics, Bristol, Philadelphia and New York, 1995) Conference Series 147. EMAG 1997: Electron Microscopy and Analysis 1997. Proceedings of EMAG 97, Cavendish Laboratory, Cambridge, 2–5 September 1997 (Rodenburg, J. M., ed.; Institute of Physics, Bristol and Philadelphia, 1997) Conference Series 153. EMAG 1999: Electron Microscopy and Analysis 1999. Proceedings of EMAG 99, University of Sheffield, 25–27 August 1999 (Kiely, C. J., ed.; Institute of Physics, Bristol and Philadelphia, 1999) Conference Series 161.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

277

EMAG 2001: Electron Microscopy and Analysis 2001. Proceedings of the Institute of Physics Electron Microscopy and Analysis Group Conference, University of Dundee, 5–7 September 2001 (Aindow, M. and Kiely, C. J., eds.; Institute of Physics, Bristol and Philadelphia 2001). Conference Series 168. EMAG 2003: Electron Microscopy and Analysis 2003. Proceedings of the Institute of Physics Electron Microscopy and Analysis Group Conference, Examination Schools, University of Oxford, 3–5 September 2003 (McVitie, S., ed.; Institute of Physics, Bristol and Philadelphia, 2003). Royal Microscopical Society Proceedings of the Symposium on Cytochemical Progress in Electron Microscopy, Oxford 2–4 July 1962. J. Roy. Microsc. Soc. 81 (1963) 106–244. [First of a series of meetings held alternately in Holland and the UK.] Abstracts of Papers presented to the Second Day of the Inaugural Meeting of the Electron Microscopy Section of the Royal Microscopical Society, Baden Powell House, London, 30 September-1 October 1965. Proc. Roy. Microsc. Soc. 1 (1966) No. 1, 25–31. ‘‘Electron Microscope Studies on the Biosynthesis and Assembly of Fibrous Proteins.’’ Second Meeting of the Electron Microscopy Section of the Royal Microscopical Society, University of Birmingham, 31 March–1 April 1966. Proc. Roy. Microsc. Soc. 1 (1966) No. 2, 65–70. Proceedings of the International Symposium on Electron Microscopy and Cytochemistry, Leiden, 31 May–4 June 1966. J. Histochem. Cytochem. 14 (1966) No. 10, 739–771. (Second of a series of meetings held alternately in Holland and the UK. The RMS is not, however, mentioned in the introductory material by R. J. Barrnett and A. M. Seligman.) Third Meeting of the Electron Microscopy Section of the Royal Microscopical Society [not traced]. ‘‘Plant Fine Structure.’’ Fourth Meeting of the Electron Microscopy Section of the Royal Microscopical Society, University of Leeds, 29 March–2 April 1967. Proc. Roy. Microsc. Soc. 2 (1967) No. 3, 366–384. ‘‘Applications of High-voltage Electron Microscopy.’’ Abstracts of papers presented at the meeting of EMAS (IPPS) and the Electron Microscopy Section (RMS), UKAEA, Harwell, 2–3 April, 1970. Proc. Roy. Microsc. Soc. 5 (1970) No. 3, 96–124. ‘‘Interpretation of Electron Scattering in Various Modes of Instrumental Operations.’’ University of York, 14–16 December, 1971. Proc. Roy. Microsc. Soc. 6 (1971) No. 6, 267–278 and 7 (1972) No. 1, 41–42. Electron Microscopy and Cytochemistry, Proceedings of the Second [in fact the fourth] International Symposium on Electron Microscopy and

278

HAWKES

Cytochemistry, Driernlo, 25–29 June 1973 (Wisse, E., Daems, W. T., Molenaar, I., and Duijn, P. van, eds.; North Holland, Amsterdam and London and American Elsevier, New York 1975). See also Histochem. J. 6 (1974) 55–114. ‘‘Current developments and Future Prospects in Electron Microscopy.’’ Gibraltar, 3–7 January, 1974. Proc. Roy. Microsc. Soc. 9 (1974) No. 2, 89–100. International Conference on Microprobe Analysis in Biology and Medicine, Mu¨nster, 4–8 September 1977 (coincides with the 18th meeting of the DGE). ‘‘The Applications of High Resolution Electron Microscopy in Materials Science (microscopy of amorphous, disordered and partially ordered solids),’’ Cambridge, March 1979 RMS in collaboration with the IoP). J. Microscopy 119 (1980) Part 1, guest editor W. M. Stobbs. Abstracts of papers presented at the Netherlands Society for Electron Microscopy and RMS Joint Meeting on Image Analysis and Interpretation, Wageningen 28–30 November 1984. Ultramicroscopy 15 (1984) 375–408. Third International Meeting on Low-temperature Biological Microscopy and Analysis, Cambridge, 1–4 April 1985, joint meeting with the NVEM. J. Microscopy 141 (1986), No. 3, 243–391. Fourth International meeting on Low-temperature Biological Microscopy and Analysis, Cambridge, April 1990, sponsored by the RMS and the DGE, organized by P. Echlin and K. Zierold. J. Microscopy 161 (1991) Nos. 1 and 2, 1–385. Conferences on the Microscopy of Semiconducting Materials have been held biennially since 1979 organized alternately by the Royal Microscopical Society and EMAG. The following list gives the dates and venues of these meetings and details of the corresponding publication, all of which except the first have appeared in the Conference Series of what is now Institute of Physics Publishing. MSM I, St Catherine’s College, Oxford, 9–11 April 1979. J. Microscopy 118 (1980) 3–126 and 255–388. MSM II, St Catherine’s College, Oxford, 6–10 April 1981. Proceedings (Institute of Physics, Bristol and London 1981; Cullis, A. G. and Joy, D. C., eds.) Conference Series 60. MSM III, St Catherine’s College, Oxford, 21–23 March 1983. Proceedings (Institute of Physics, Bristol and London 1981; Cullis, A. G., Davidson, S. M., and Booker, G. R., eds.) Conference Series 67.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

279

MSM IV, St Catherine’s College, Oxford, 25–27 March 1985. Proceedings (Adam Hilger, Bristol and Boston 1985; Cullis, A. G. and Holt, D. B., eds.) Conference Series 76. MSM V, Oxford University, 6–8 April 1987. Proceedings (Institute of Physics, Bristol and Philadelphia 1987; Cullis, A. G. and Augustus, P. D., eds.). Conference Series 87. MSM VI, Oxford University, 10–13 April 1989. Proceedings (Institute of Physics, Bristol and New York 1989; Cullis, A. G. and Hutchison, J. L., eds.). Conference Series 100. MSM VII, Oxford University, 25–28 March 1991. Proceedings (Institute of Physics, Bristol, Philadelphia and New York 1991; Cullis, A. G. and Long, N. J., eds.). Conference Series 117. MSM VIII, Oxford University, 5–8 April 1993. Proceedings (Institute of Physics, Bristol and Philadelphia 1993; Cullis, A. G., Staton-Bevan, A. E., and Hutchison, J. L., eds.). Conference Series 134. MSM IX, Oxford University, 20–23 March 1995. Proceedings (Institute of Physics, Bristol and Philadelphia 1995; Cullis, A. G. and Staton-Bevan, A. E., eds.). Conference Series 146. MSM X, Oxford University, 7–10 April 1997. Proceedings (Institute of Physics, Bristol and Philadelphia 1997; Cullis, A. G. and Hutchison, J. L., eds.). Conference Series 157. MSM XI, Oxford University, 22–25 March 1999. Proceedings (Institute of Physics, Bristol and Philadelphia 1999; Cullis, A. G. and Beanland, R., eds.). Conference Series 164. MSM XII, Oxford University, 25–29 March 2001. Proceedings (Institute of Physics, Bristol and Philadelphia 2001; Cullis, A. G. and Hutchison. J. L., eds.). Conference Series 169. MSM XIII, Churchill College, Cambridge, 31 March–3 April 2003. Developments in FEGTEM (RMS and EMAG), Department of Materials, Oxford FEGTEM I: Department of Materials, Oxford, 12 July 1999 FEGTEM II: Department of Materials, Oxford, 3 July 2000 FEGTEM III: Department of Materials, Oxford, 3 July 2001, followed on 4 July by ‘‘Developments in Energy-filtered Electron Microscopy’’ FEGTEM IV: Department of Materials, Oxford, 1 July 2002. FEGTEM V: Institute of Materials Research, University of Leeds, 14 July 2003. Society of Electron Microscope Technology (SEMT). www.semt.org.uk 1980. Bedford College, London, 23 April.

280

HAWKES

1981. Guy’s Hospital, London, 15 April. 1982. Guy’s Hospital, London, 14 April. 1983. British Museum (Natural Hitory), London, 20 April. 1985. St Bartholomew’s Hospital, London, 9 October. 1986, not traced 1987. St. Bartholomew’s Hospital, London, 26 October. 1988. Eastman Dental Hospital, London, 28 October. 1989. Eastman Dental Hospital, London, 27 October. 1990. Charing Cross Hospital, London, 24 October. 1991. Charing Cross and Westminster Medical School, London, 23 October. 1992. Eastman Dental Hospital, London, 9 October. 1993. Eastman Dental Hospital, London, 22 October. 1994. Eastman Dental Hospital, London, 9 October. 1995. Royal Veterinary College, London, 18 October. 1996. Royal Veterinary College, London, 13 March. 1997. Royal Veterinary College, London, 19 March. 1998. School of Pharmacy, London, 25 March. 1999. School of Pharmacy, London, 24 March. 2000. Millenium Microscopy Meeting, Open University, Milton Keynes, 3–4 February. 2001. School of Pharmacy, London, 28 March. 2001. Diagnostic EM: investigations in materials and life sciences. School of Pharmacy, London, 12 December. 2002. Immuno-Workshop, joint with Aurion, Guy’s Hospital, London, 11–13 September; half-day meeting, 6 November, School of Pharmacy, London. 25. The USSR, now Russia The Electron Microscopy Council of the Academy of Sciences of the USSR was created in 1963; the founder-Chairman was B. K. Vainshtein, who remained in office until 1992 when N. A. Kiselev became Chairman. This council plays the role of an electron microscopy society in Russia, and has been responsible for organizaing all the All-Union/Russian conferences on electron microscopy and scanning electron microscopy as well as a number of more specialized meetings. I know of only one publication describing the development of electron microscopy in the Soviet Union, namely Nyrikov and Kuschnier (1956), and this gives few details. The most useful source is the survey by Agar (1996). 1. Proceedings of the 1st Soviet All-Union Conference on Electron Microscopy, Moscow, 15–19 Dec. 1950, Izv. Akad. Nauk SSSR (Ser. Fiz.) 15 (1951) Nos. 3 and 4 (no English translation).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

281

2. Proceedings of the 2nd Soviet All-Union Conference on Electron Microscopy, Moscow 9–13 May 1958; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 23 (1959) Nos. 4 and 6. 3. Proceedings of the 3rd Soviet All-Union Conference on Electron Microscopy, Leningrad, 24–29 October 1960; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 25 (1961) No. 6. 4. Proceedings of the 4th Soviet All-Union Conference on Electron Microscopy, Sumy, 12–14 March 1963; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 27 (1963) No. 9. 5. Proceedings of the 5th Soviet All-Union Conference on Electron Microscopy, Sumy, 6–8 July 1965; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 30 (1966) No. 5. 6. Proceedings of the 6th Soviet All-Union Conference on Electron Microscopy, Novosibirsk, 11–16 July 1967; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 32 (1968) Nos 6 and 7. 7. Proceedings of the 7th Soviet All-Union Conference on Electron Microscopy, Kiev, 14–21 July 1969; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 34 (1970) No. 7. 8. Proceedings of the 8th Soviet All-Union Conference on Electron Microscopy and 1st All-Union Symposium on Scanning Electron Microscopy, Moscow 15–20 November 1971; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 36 (1972) Nos 6 and 9. 9. Proceedings of the 9th Soviet All-Union Conference on Electron Microscopy and 2nd All-Union Symposium on Scanning Electron Microscopy, Tbilisi, 28 October-2 November 1973; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 38 (1974) Nos 7 and 11. 10. Proceedings of the 10th Soviet All-Union Conference on Electron Microscopy, Tashkent 5–8 Oct. 1976; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 41 (1977) Nos. 5 and 11. 11. Proceedings of the 11th Soviet All-Union Conference on Electron Microscopy, Tallin, October 1979; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 44 (1980) Nos. 6 and 10. 12. Proceedings of the 12th Soviet All-Union Conference on Electron Microscopy, Sumy 1982; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 48 (1984) No. 2. 13. Proceedings of the 13th Soviet All-Union Conference on Electron Microscopy, Sumy October 1987; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 52 (1988) No. 7 and 53 (1989) No.2. 14. Proceedings of the 14th Soviet All-Union Conference on Electron Microscopy, Suzdal, October and November 1990; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 55 (1991) No. 8.

282

HAWKES

15. Proceedings of the 15th Conference on Electron Microscopy, Chernogolovka, May 1994; Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 59 (1995) No. 2. 16. Proceedings of the 16th Conference on Electron Microscopy, Chernogolovka, December 1996; Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 61 (1997) No. 10. 17. Proceedings of the 17th Russian Conference on Electron Microscopy, Chernogolovka, June 1998; Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 63 (1999) No. 7 (Petrov, V. I. and Luk’yanov, A. E., guest eds.). 18. Proceedings of the 18th Russian Conference on Electron Microscopy, Chernogolovka, 5–8 June 2000; Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 65 (2001) No. 9. 19. Proceedings of the 19th Russian Conference on Electron Microscopy, Chernogolovka, 27–31 May 2002; Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 67 (2003). Scanning Microscopy 1. Proceedings of the 8th Soviet All-Union Conference on Electron Microscopy and 1st All-Union Symposium on Scanning Electron Microscopy, Moscow 15–20 November 1971; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 36 (1972) Nos. 6 and 9. 2. Proceedings of the 9th Soviet All-Union Conference on Electron Microscopy and 2nd All-Union Symposium on Scanning Electron Microscopy, Tbilisi, 28 October-2 November 1973; Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 38 (1974) Nos. 7 and 11. 3. Proceedings of the 3rd All-Union Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the Solid State, REM81. Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 46 (1982) No. 12 and 47 (1983) No. 6. 4. Proceedings of the Fourth All-Union Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the Solid State, Zvenigorod, April 1984. Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 48 (1984) No. 12. 5. Proceedings of the Fifth All-Union Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the Solid State, Zvenigorod, May 1986. Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 51 (1987) No. 3, 417–506. 6. Proceedings of the Fifth [sic] All-Union Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the

ELECTRON OPTICS AND ELECTRON MICROSCOPY

283

Solid State, Zvenigorod, April 1989. Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 54 (1990) No. 2, 193–365. 7. Proceedings of the Seventh All-Union Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the Solid State, Zvenigorod, March 1991. Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 56 (1992) No. 3. 8. Proceedings of the Eighth Russian Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the Solid State, Chernogolovka, May 1993. Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 57 (1993) No. 8 and 58 (1994) No. 1. 9. Proceedings of the Ninth Russian Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the Solid State, SEM–95, Chernogolovka, May 1995. Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 60 (1996) No. 2. 10. Proceedings of the Tenth Russian Symposium on Scanning Electron Microscopy and Analytical Methods of investigating the Solid State, SEM–97, Chernogolovka, 8–11 June 1997. Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 62 (1998) No. 3. 11. Proceedings of the Eleventh National Symposium on Scanning Electron Microscopy and Analytical Methods in the Study of Solids, SEM–99, Chernogolovka, May 1999. Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 64 (2000) No. 8. 12. Proceedings of the Twelfth National Symposium on Scanning Electron Microscopy and Analytical Methods in the Study of Solids, SEM–2001, Chernogolovka, 4–6 June 2001. Izv. Ross. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 66 (2002) No. 9. Diffraction methods for structural studies Proceedings of the All-Union Symposium on Electron Microscopy and Electron Diffraction in the Study of the Organization, Structure and Properties of the Solid State, Zvenigorod, May 1983. Izv. Akad. Nauk SSSR (Ser. Fiz.) or Bull. Acad. Sci. USSR (Phys. Ser.) 48 (1984) No. 9. Proceedings First All-Union Symposium on Diffraction methods for Structural Studies, Zvenigorod 3–6 November 1991. Izv. Akad. Nauk (Ser. Fiz.) or Bull. Russ. Acad. Sci. (Phys.) 57 (1993) Nos. 2 and 8. Problems of theoretical and applied electron optics [Problemyi Teoeticheskoi i Prikladnoi Elektronnoi Optiki] Proceedings of the First All-Russia Seminar, Scientific Research Institute for Electron and Ion Optics, Moscow 1996. Prikladnaya Fizika (1996) No. 3.

284

HAWKES

Proceedings of the Second All-Russia Seminar, Scientific Research Institute for Electron and Ion Optics, Moscow, 25 April 1997. Prikladnaya Fizika (1997) No. 2–3. Proceedings of the Third All-Russia Seminar, Scientific Research Institute for Electron and Ion Optics, Moscow, 31 March–2 April 1998. Prikladnaya Fizika (1998) Nos. 2 and 3/4. Proceedings of the Fourth All-Russia Seminar, Scientific Research Institute for Electron and Ion Optics, Moscow, 21–22 October 1999. Prikladnaya Fizika (2000) Nos. 2 and 3; Proc SPIE 4187 (2000), edited by A. M. Filachev and I. S. Gaidoukova. Proceedings of the Fifth All-Russia Seminar, Scientific Research Institute for Electron and Ion Optics, Moscow, 14–15 November 2001. Prikladnaya Fizika (2002) No. 3; Proc SPIE 5025 (2002), edited by A. M. Filachev and I. S. Gaidoukova. Proceedings of the Sixth All-Russia Seminar, Scientific Research Institute for Electron and Ion Optics, Moscow, 28–30 May 2003. Prikladnaya Fizika and Proc SPIE. 26. Yugoslavia together with Serbia, Bosnia, and Herzegovina Yugoslavia. An informal meeting of electron microscope users was held in Ljubljana in 1959 and resulted in the production of a Basic Electron Microscopy Handbook; a multi-author textbook followed soon after (Pantic, 1962), with contributions by A. Strojnik, V. Marinkovic´, J. Vukovic´, B. Navinsek, V. Pantic´, and N. Pipan. The Yugoslav Society may be said to have been founded at the first Yugoslav Symposium on Electron Microscopy, in 1969. The development of the subject is recounted by Vukovic´ (1996). 1. Jozef Stefan Institute of Nuclear Science, Ljubljana, 20–21 November 1969. Proceedings of the First Yugoslav Symposium on Electron Microscopy, edited by V. Marinkovic´. 2. Veterinary Department, Beograd, 1970 [no Proceedings], organized by V. Pantic´. 1974: First Balkan Conference on Electron Microscopy in Sarajevo (see Section III.C). 3. Medical School, Beograd, 16–18 May 1980. Proceedings of the Third Yugoslav Symposium on Electron Microscopy, edited by J. B. Vukovic´ and T. M. Nenadovic´. 4. Kranjska Gora 26–28 May 1983. Zbornik, YUSEM-83, 4 Jugoslovanski Simpozij o Elektronski Mikroskopiji [Proceedings of the Fourth Yugoslav Symposium on Electron Microscopy], edited by M. Psˇenicˇnik and B. Drinovec´.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

285

5. Plitvicka jezera, 27–30 May 1986. Zbornik Radova 5. Jugoslavenski Simpozij iz Elektronske Mikroskopije [Proceedings of the Fifth Yugoslav Symposium on Electron Microscopy], edited by M. Wrischler, A. Hlousˇek–Radojcˇic´ and N. Ljubesˇic´. 6. Sarajevo—Igman, 29 May–1 June 1989. Zbornik Radova YUSEM–’89, 6. Jugoslovenski Simpozijum za Elektronsku Mikroskopiju [Proceedings of the Sixth Yugoslav Symposium on Electron Microscopy], edited by B. Plavsˇic´. 7. University of Nis, 3–4 December 1992. Zbornik Abstracta SEM ’92, Treci Simpozijum za Elektronsku Mikroskopiju Srbije [This is at once the seventh YUSEM conference and the third Serbian Symposium on Electron Microscopy.] Proceedings edited by D. Dojcinov. Serbia The Serbian Society of Electron Microscopy (Drusˇtvo za Elektronsku Mikroskopiju Srbije) was founded in 1979. 1. Beograd, 23–24 April 1982. Zbornik Radova, Sovremena Elektronska Mikroskopija i Mikroanaliza u Biomeditsinskim Istazhivanjima [Modern Electron Microscopy and Microanalysis in Biomedical Research]. Proceedings edited by J. Vukovic´. 2. Beograd, 1986. ‘‘Thirty Years of Electron Microscopy in Serbia,’’ Proceedings edited by J. B. Vukovic´. 3. University of Nisˇ, 3–4 December 1992. See YUSEM VII above. New series 1994: Novi Sad, 2–3 June 1994. I Electron Microscopy Congress. Proceedings edited by J. Milin. 1996: Hotel Jugoslavija, Belgrade, 2–5 October 1996. Drugi Kongres za Elektronsku Mikroskopiju [II National Congress on Electron Microscopy]. Elektronska Mikroskopija u Biomedicini i Nauci o Materijalima, 40 godina elektronske mikroskopije u Srbiji. (Bumbasˇirevic´, V., ed.; Excelsior, Novi Beograd 1997); Invited papers published in Bull. Acad. Serbe Sci. Arts, Cl. Sci. Math. Nat. 115 (1998) No. 37. Bosnia and Herzegovina 1. Sarajevo—Zenica, 30 September–2 October 1987. Zbornik Apstrakta Kolokvijum iz Elektronske Mikroskopije (Drusˇtvo za Elektronsku Mikroskopiju Bosne i Hercegovine) [Proceedings of the First Symposium of the Bosnia and Herzegovina Society of Electron Microscopy], edited by B. Plavsˇic´.

286

HAWKES

B. North America 1. USA (www.microscopy.msa.com) Extensive historical material concerning the creation and expansion of the Electron Microscopy Society of America is available; see in particular the book edited by Newberry (1992) and the articles by Hall (1985), Reisner (1989a) and Fisher (1996). With the Spring 1981 issue of the EMSA Bulletin (vol. 11, No. 1), a series of ‘‘Reflections’’ was launched by J. H. Reisner; these are all listed separately in the bibliography although it should be noted that some are concerned with the history of electron microscopy, not necessarily in the USA, and not specifically with the EMSA. See the entries under Calbick (1988), Cohen (1987), Cosslett (1982), Freundlich (1994), Fullam (1986), Hansl (1987), Hillier (1986), Meryman (1987), Newberry (1985a,b, 1986, 1993), Porter (1993), Reimann (1989), Reisner (1981a,b, 1982, 1983, 1984, 1989b, 1990, 1991, 1992), Rempfer (1993), Rochow (1983), Rudenberg and Rudenberg (1994), Smith (1984), Watson (1992, 1993). Many of the publications of Marton are also directly relevant, though his earlier work in Europe is usually covered as well; see Marton (1960, 1965, 1968, 1994) and Su¨sskind (1985). See too Calbick (1944), Wyckoff (1954), the article by Yoshii (1970) and that by Cohen and Steever (1971) on the Washington State University microscope and the scholarly studies by Rasmussen (1996, 1998a,b) of microscope development at RCA. A few historical articles have appeared in the EMSA Proceedings, notably by Rudenberg (1992) and Rudenberg and Rudenberg (1992). In 1971, the EMSA began issuing a Bulletin to its members; in 1993, this changed title from EMSA Bulletin to MSA Bulletin but the latter was discontinued with the creation of the Journal of the Microscopy Society of America in 1995. This in turn gave way to Microscopy and Microanalysis in 1997; the volume numbering was continued without interruption so that Microscopy and Microanalysis commences with volume 3. In 1997, the MSA Bulletin was resuscitated with no break in volume numbering; the 1997 issues form volume 23. Many local societies are affiliated to the national society: Alabama Imaging and Microscopy Society, Appalachian Regional Electron Microscopy Society, Arizona Imaging and Microanalysis Society, Central States Microscopy Society, Chesapeake Society for Microscopy, Connecticut Microscopy Society, Microscopy Society of Northeastern Ohio, Microscopy Society of the Ohio River Valley, Florida Society for Microscopy, Iowa Microscopy Society, Louisiana Society for Microscopy, Metropolitan Microscopy Society, Michigan Microscopy Society, Midwest Society of Electron Microscopists, Minnesota Microscopy Society, Mountain

ELECTRON OPTICS AND ELECTRON MICROSCOPY

287

States Society for Electron Microscopy, New England Society for Electron Microscopy, New York Society of Experimental Microscopists, Northern California Society for Microscopy, North Carolina Society for Microscopy and Microbeam Analysis, Northwestern Ohio Microscopy Society, Oklahoma Society for Microscopy, Pacific Northwest Microscopy Society, Philadelphia Microscopy Society, San Diego Society for Electron Microscopy, Southern California Society for Electron Microscopy Technologists, Southern California Society for Microscopy, South Carolina Society for Electron Microscopy, Southeastern Microscopy Society, and Texas Society for Electron Microscopy. Many of these distribute Newsletters or journals to their members (the Texas Society for Electron Microscopy Journal and The Beam, for example) and abstracts of many of their meetings have been published. These are not listed here. Some other major conferences should also be mentioned. The fiftieth anniversary of the National Bureau of Standards fell in 1951 and to mark the occasion, a Symposium on Electron Physics was held from 5–7 November at the NBS in Washington DC. The proceedings were published as National Bureau of Standards Circular No. 527 (1954). From 5 to 8 July, 1961, a conference on ‘‘The Impact of Electron Microscopy on Theories of the Strength of Crystals’’ was held in Berkeley (Thomas and Washburn, 1963). A Symposium on Quantitative Electron Microscopy was held in the Armed Forces Institute of Pathology in Washington DC from 30 March–3 April 1964 and full proceedings were published in book form and in a serial (Bahr and Zeitler, 1965). In 1966 (13 June–15 July), a Summer Workshop on High-voltage Electron Microscopy was organized at Argonne National Laboratory by R. K. Hart under the overall direction of A. V. Crewe; the Proceedings are available as an ANL Report (Gilroy et al., 1966) and eight AMU–ANL High-voltage Electron Microscope Newsletters were also produced. From 13–17 September 1971, a Symposium on ‘‘The Structure and Properties of Materials, Techniques and Applications of Electron Microscopy’’ was held in Berkeley (Thomas et al., 1972). The series of US–Japan High-Voltage Electron Microscopy Conferences are listed in the section on HVEM (5.1). Materials Research Society (MRS) Symposia on electron microscopical themes are recorded in Krakow et al. (1984), Hobbs et al. (1986), Bravman et al. (1988), Sinclair et al. (1990), Anderson (1990), Biegelsen et al. (1993), Sharma et al. (1996), Smith (1997), Anderson and Walck (1997), and Bentley et al. (2001) and relevant Minerals, Metals and Materials Society (TMS) Symposia in Krakow and O’Keefe (1989) and Disko et al. (1992). Also see Parsons (1978) and Buseck (1992). For the Proceedings of the International Workshop on Electron Holography, held at the Holiday Inn World’s Fair in Knoxville TN from 29–31 August 1994, see Tonomura et al. (1995).

288

HAWKES

EMSA 1: [First National Conference on the Electron Microscope] Hotel Sherman, Chicago IL, 27–28 November 1942; a transcript of the proceedings is reproduced in EMSA and its People, the First Fifty Years by S. P. Newberry (Electron Microscopy Society of America, 1992). EMSA 2: La Salle Hotel, Chicago IL, 16–18 November 1944; J. Appl. Phys. 16 (1945) 263–266. EMSA 3: Frick Chemical Laboratory, Princeton University, Princeton NJ, 30 November–1 December 1945; J. Appl. Phys. 17 (1946) 66–68. EMSA 4: Mellon Institute, Pittsburgh PA, 5–7 December 1946; J. Appl. Phys. 18 (1947) 269–273. EMSA 5: Franklin Institute, Philadelphia PA, 11–13 December 1947; J. Appl. Phys. 19 (1948) 118–126. See also Anal. Chem. 20 (1948) 90–92. EMSA 6: E. F. Burton Memorial Meeting, University of Toronto, 9–11 September 1948; J. Appl. Phys 19 (1948) 1186–1192. See also Anal. Chem. 20 (1948) 990–993. EMSA 7: National Bureau of Standards, Washington DC, 6–8 October 1949; J. Appl. Phys. 21 (1950) 66–72. See also Anal. Chem. 21 (1949) 1434–1437. EMSA 8: Statler Hotel, Detroit MI, 14–16 September 1950; J. Appl. Phys. 22 (1951) 110–116. Washington, 1951: Electron Physics. Proceedings of the NBS Semicentennial Symposium on Electron Physics, Washington, 5–7 November, 1951. Issued as National Bureau of Standards Circular 527 (1954). See also Anal. Chem. 23 (1951) 1885–1887. EMSA 9: Franklin Institute, Philadelphia PA, 8–10 October 1951; J. Appl. Phys. 23 (1952) 156–164. See also Anal. Chem. 23 (1951) 1887–1890. EMSA 10: Hotel Statler, Cleveland OH, 6–8 November 1952; J. Appl. Phys. 24 (1953) 111–118. See also Anal. Chem. 24 (1952) 1865–1868. EMSA 11: Pocono Manor Inn, Pocono Manor PA, 5–7 November 1953; J. Appl. Phys. 24 (1953) 1414–1426. See also Anal. Chem. 26 (1954) 437–438. EMSA 12: Moraine-on-the-Lake Hotel, Highland Park IL, 14–16 October 1954; J. Appl. Phys. 25 (1954) 1453–1468. EMSA 13: Pennsylvania State University, University Park PA, 27–29 October 1955; J. Appl. Phys. 26 (1955) 1391–1398. EMSA 14: University of Wisconsin, Madison WI, 10–12 September 1956; J. Appl. Phys. 27 (1956) 1389–1398. EMSA 15: Massachusetts Institute of Technology, Cambridge MA, 9–11 September 1957; J. Appl. Phys. 28 (1957) 1368–1386. EMSA 16: Santa Monica Civic Auditorium, Santa Monica CA, 7–9 August 1958; J. Appl. Phys. 29 (1958) 1615–1626.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

289

EMSA 17: Ohio State University, Columbus OH, 9–12 September 1959; J. Appl. Phys. 30 (1959) 2024–2042. EMSA 18: Marquette University, Milwaukee WI, 29–31 August 1960; J. Appl. Phys 31 (1960) 1831–1848. EMSA 19: Pittsburgh Hilton Hotel, Pittsburgh PA, 23–26 August 1961; J. Appl. Phys 32 (1961) 1626–1646. ICEM–5, Philadelphia PA, 29 August to 5 September, 1962. EMSA 21: Denver CO, 28–31 August 1963; J. Appl. Phys 34 (1963) 2502–2534. EMSA 22: Detroit MI, 13–16 October 1964; J. Appl. Phys 35 (1964) 3074–3102. EMSA 23: New York NY, 25–28 August 1965; J. Appl. Phys 36 (1965) 2603–2632. EMSA 24: San Francisco Hilton, San Francisco CA, 22–25 August 1966; J. Appl. Phys 37 (1966) 3919–3952. EMSA 25: Proceedings of the 25th Anniversary Meeting Electron Microscopy Society of America, Chicago IL, 29 August-1 September 1967 (Arceneaux, C. J., ed.; Claitor, Baton Rouge 1967). EMSA 26: Proceedings of the 26th Annual Meeting Electron Microscopy Society of America, New Orleans LA, 16–19 September 1968, (Arceneaux, C. J., ed.; Claitor, Baton Rouge 1968). EMSA 27: Proceedings of the 27th Annual Meeting Electron Microscopy Society of America, St Paul MN, 26–29 August 1969 (Arceneaux, C. J., ed.; Claitor, Baton Rouge 1969). EMSA 28: Proceedings of the 28th Annual Meeting Electron Microscopy Society of America, Houston TX, 5–8 October 1970 (Arceneaux, C. J., ed.; Claitor, Baton Rouge 1970). EMSA 29: Proceedings of the 29th Annual Meeting Electron Microscopy Society of America, Boston MA, 9–13 August 1971 (Arceneaux, C. J., ed.; Claitor, Baton Rouge 1971). EMSA 30: Proceedings of the 30th Annual Meeting Electron Microscopy Society of America and First Pacific Regional Conference on Electron Microscopy, Los Angeles CA, 14–17 August 1972 (Arceneaux, C. J., ed.; Claitor, Baton Rouge 1972). EMSA 31: Proceedings of the 31st Annual Meeting Electron Microscopy Society of America, New Orleans LA, 14–17 August 1973 (Arceneaux, C. J., ed.; Claitor, Baton Rouge 1973). EMSA 32: Proceedings of the 32nd Annual Meeting Electron Microscopy Society of America, St. Louis MO, 13–15 August 1974 (Arceneaux, C. J. and G. W. Bailey, eds.; Claitor, Baton Rouge 1974). EMSA 33: Proceedings of the 33rd Annual Meeting Electron Microscopy Society of America, Las Vegas NV, 11–15 August 1975 (Bailey G. W. and Arceneaux, C. J., eds.; Claitor, Baton Rouge 1975).

290

HAWKES

EMSA 34: Proceedings of the 34th Annual Meeting Electron Microscopy Society of America, Miami Beach FL, 9–13 August 1976 (Bailey G. W., ed.; Claitor, Baton Rouge 1976). EMSA 35: Proceedings of the 35th Annual Meeting Electron Microscopy Society of America, Boston MA, 22–26 August 1977 (Bailey G. W., ed.; Claitor, Baton Rouge 1977). ICEM-9, Toronto, 1–9 August 1978 (incorporates EMSA 36). EMSA 37: Proceedings of the 37th Annual Meeting Electron Microscopy Society of America, San Antonio TX, 13–17 August 1979 (Bailey G. W., ed.; Claitor, Baton Rouge 1979). EMSA 38: Proceedings of the 38th Annual Meeting Electron Microscopy Society of America, San Francisco CA, 4–8 August 1980 (Bailey G. W., ed.; Claitor, Baton Rouge 1980). EMSA 39: Proceedings of the 39th Annual Meeting Electron Microscopy Society of America, Atlanta GA, 10–14 August 1981 (Bailey G. W., ed.; Claitor, Baton Rouge 1981). EMSA 40: Proceedings of the 40th Annual Meeting Electron Microscopy Society of America, Washington DC, 9–13 August 1982 (Bailey G. W., ed.; Claitor, Baton Rouge 1982). EMSA 41: Proceedings of the 41st Annual Meeting Electron Microscopy Society of America, Phoenix AZ, 8–12 August 1983 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1983). See also Ultramicroscopy 13 (1984) 1–183. EMSA 42: Proceedings 42nd Annual Meeting Electron Microscopy Society of America jointly with Microscopical Society of Canada, Eleventh Annual Meeting, Detroit MI, 13–17 August 1984 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1984). EMSA 43: Proceedings 43rd Annual Meeting Electron Microscopy Society of America, Louisville KY, 5–9 August 1985 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1985). See also Basu and Millette (1986). EMSA 44: Proceedings 44th Annual Meeting Electron Microscopy Society of America, Albuquerque NM, 10–15 August 1986 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1986). EMSA 45: Proceedings 45th Annual Meeting Electron Microscopy Society of America, Baltimore MD, 2–7 August 1987 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1987). EMSA 46: Proceedings 46th Annual Meeting Electron Microscopy Society of America jointly with Microscopical Society of Canada, Fifteenth Annual Meeting, Milwaukee WI, 7–12 August 1988 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1988).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

291

EMSA 47: Proceedings 47th Annual Meeting Electron Microscopy Society of America, San Antonio TX, 6–11 August 1989 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1989). ICEM-XII, Seattle, 12–18 August 1990 (incorporates EMSA 48). EMSA 49: Proceedings 49th Annual Meeting Electron Microscopy Society of America, San Jose CA, 4–9 August 1991 (Bailey, G. W., and Hall, E. L., eds.; San Francisco Press, San Francisco 1991). See also Ultramicroscopy 47 (1992) Nos. 1–3, 1–306. EMSA 50: Proceedings 50th Annual Meeting Electron Microscopy Society of America, 27th Annual Meeting Microbeam Analysis Society, Nineteenth Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, Boston MA, 16–21 August 1992 (Bailey, G. W., Bentley, J., and Small, J. A., eds.; San Francisco Press, San Francisco 1992) 2 Vols. MSA 51: Proceedings 51st Annual Meeting Microscopy Society of America, Cincinnati OH, 1–6 August 1993 (Bailey, G. W., and Rieder, C. L., eds.; San Francisco Press, San Francisco 1993). MSA 52: Proceedings 52nd Annual Meeting Microscopy Society of America, 29th Annual Meeting Microbeam Analysis Society, New Orleans LA, 31 July–5 August 1994 (Bailey, G. W. and Garratt-Reed, A. J., eds. San Francisco Press, San Francisco 1994). MSA 53: Microscopy and Microanalysis 1995, 53rd Annual Meeting Microscopy Society of America, Kansas City KS, 13–16 August 1995; J. Microsc. Soc. Am. Proceedings (Bailey, G. W., Ellisman, M. H., Henniger, R. A., and Zaluzec, N. J., eds.; Jones & Begell, New York and Boston 1995). MSA 54: Proceedings Microscopy and Microanalysis 1996. 54th Annual Meeting Microscopy Society of America, Twenty-third Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, 30th Annual Meeting Microbeam Analysis Society, Minneapolis MN, 11–15 August, 1996 (Bailey, G. W., Corbett, J. M., Dimlich, R. V. W., Michael, J. R., and Zaluzec, N. J., eds.; San Francisco Press, San Francisco 1996). MSA 55: Proceedings Microscopy and Microanalysis 1997. 55th Annual Meeting Microscopy Society of America, 31st Annual Meeting Microbeam Analysis Society, 48th Annual Meeting Histochemical Society, Cleveland OH, 10–14 August, 1997; Microsc. Microanal. 3 (1997) Supplement 2 (Bailey, G. W., Dimlich, R. V. W., Alexander, K. B., McCarthy, J. J., and Pretlow, T. P., eds.; Springer, New York 1997). MSA 56: Proceedings Microscopy and Microanalysis 1998. 56th Annual Meeting Microscopy Society of America, 32nd Annual Meeting Microbeam Analysis Society, Atlanta GA, 12–16 July, 1998; Microsc. Microanal.

292

HAWKES

4 (1998) Supplement 2 (Bailey, G. W., Alexander, K. B., Jerome, W. G., Bond, M. G., and McCarthy, J. J., eds.; Springer, New York 1998). MSA 57: Proceedings Microscopy and Microanalysis 1999. 57th Annual Meeting Microscopy Society of America, 33rd Annual Meeting Microbeam Analysis Society, Portland OR, 1–5 August, 1999; Microsc. Microanal. 5 (1999) Supplement 2 (Bailey, G. W., Jerome, W. G., McKernan, S., Mansfield, J. F., and Price, R. L., eds.; Springer, New York 1999). MSA 58: Proceedings Microscopy and Microanalysis 2000. 58th Annual Meeting Microscopy Society of America, 27th Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, 34th Annual Meeting Microbeam Analysis Society, Philadelphia PA, 13–17 August, 2000; Microsc. Microanal. 6 (2000) Supplement 2 (Bailey, G. W., McKernan, S., Price, R. L., Walck, S. D., Charest, P.-M., and Gauvin, R., eds.; Springer, New York 2000). MSA 59: Proceedings Microscopy and Microanalysis 2001. 59th Annual Meeting Microscopy Society of America, 35th Annual Meeting Microbeam Analysis Society, Long Beach CA, 5–9 August, 2001; Microsc. Microanal. 7 (2001) Supplement 2 (Bailey, G. W., Price, R. L., Voelkl, E., and Musselman, I. H., eds.; Springer, New York 2001). MSA 60: Proceedings Microscopy and Microanalysis 2002. 60th Annual Meeting Microscopy Society of America, 29th Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, 36th Annual Meeting Microbeam Analysis Society, 35th Meeting International Metallographic Society, Quebec City PQ, 4–8 August, 2002. Microsc. Microanal. 8 (2002) Supplement 2 (Voelkl, E., Piston, D., Gauvin, R., Lockley, A. J., Bailey, G. W., and McKernan, S., eds.; Cambridge University Press, New York 2002). The printed proceeedings contain only the invited papers, all others are distributed as a CD-ROM. MSA 61: 61st Annual Meeting Microscopy Society of America, 7th Interamerican Congress on Electron Microscopy, 37th Annual Meeting Microbeam Analysis Society, 36th Meeting International Metallographic Society, San Antonio TX, 3–7 August 2003. MSA 62: 62nd Annual Meeting Microscopy Society of America, 38th Annual Meeting Microbeam Analysis Society, Savannah GA, 1–5 August 2004. MSA 63: 63rd Annual Meeting Microscopy Society of America, 39th Annual Meeting Microbeam Analysis Society, Honolulu HI, 31 July–4 August 2005. MSA 64: 64th Annual Meeting Microscopy Society of America, 40th Annual Meeting Microbeam Analysis Society, Chicago IL, 6–10 August 2006.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

293

2. Canada (msc.rsvs.ulaval.ca) The forerunner of the present Canadian Society was the Ontario Group of Electron Microscopists, which first met in October 1958. This gave way to the Burton Society of Electron Microscopists in October 1960, but this survived only until 1965. In October 1965, the present society, the Microscopical Society of Canada or la Socie´te´ de Microscopie du Canada, came into being and began publication of a quarterly Bulletin in 1973. A proceedings volume has been published each year since 1974, with the exception of those years in which the society met jointly with EMSA in the USA; the EMSA Proceedings then include the Canadian contributions. For accounts of the early days of Canadian electron microscopy and related matters, see Watson (1964/5, 1974, 1992, 1993, 1995), Franklin et al. (1978), Kohl et al. (1978), Howatson (1982), Hillier (1986, 1995), Ottensmeyer (1995), Prebus (1998), and especially Doane et al. (1993), and Simon and Doane (1996). Doane (1992) examines an aspect of electron microscopy that is rarely treated. For comments on the articles by Watson (1992, 1993) and the book by Doane et al. (1993), see Agar (1994). MSC 1: Ontario Science Centre, Toronto, 24–25 June, 1974 (Simon, G. T. and Sturgess, J. M., eds.). MSC 2: Pavillon Comtois, Universite´ Laval, Quebec City, 15–17 June 1975 (de Estable-Puig, R. F. and Sturgess, J. M., eds.). MSC 3: University of Ottawa, Ottawa, 20–23 June, 1976 (Sturgess, J. M., ed.). MSC 4: University of Western Ontario, London, 13–15 June, 1977 (Sturgess, J. M. and Beveridge, T. J., eds.). ICEM-9, Toronto, 1–9 August 1978. Coincides with MSC 5. MSC 6: University of British Columbia, Vancouver, 15–17 June, 1979 (Sturgess, J. M., Omar, S. A., and Culling, C. F. A., eds.). MSC 7: Memorial University, St John’s, Newfoundland, 5–7 June 1980 (Omar, S. A., Barber, V. C., and Northwood, D. O., eds.). MSC 8: McGill University, Montreal, 13–15 June, 1981 (Sturgess, J. M., ed.). MSC 9: University of Alberta, Edmonton, 11–13 June, 1982 (Kuster, J. E., Batz, H. W., and Craig, D. A., eds.). MSC 10: Chalk River Nuclear Laboratories, Chalk River, Ontario, 17–19 May, 1983 (Parsons, J. R., ed.). MSC 11: Proceedings 42nd Annual Meeting Electron Microscopy Society of America jointly with Microscopical Society of Canada, Eleventh Annual Meeting, Detroit MI, 13–17 August 1984 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1984). MSC 12: University of New Brunswick, Fredericton, 18–20 May, 1985 (Bance, G. N., ed.).

294

HAWKES

MSC 13: McMaster University, Hamilton, Ontario, 13–15 June, 1986 (Simon, G. T., ed.). MSC 14: University of Manitoba, Winnipeg, 17–19 June, 1987 (Cann, C. D., ed.). MSC 15: Proceedings 46th Annual Meeting Electron Microscopy Society of America jointly with Microscopical Society of Canada, Fifteenth Annual Meeting, Milwaukee WI, 7–12 August 1988 (Bailey, G. W., ed.; San Francisco Press, San Francisco 1988). MSC 16: University of Guelph, Guelph, Ontario, 30 May–2 June, 1989 (Beveridge, T. J., ed.). MSC 17: Dalhousie University, Halifax, Nova Scotia, 10–12 June, 1990 (Rowden, G and Murphy, J. G., eds.). MSC 18: Health Sciences Centre, University of Calgary, Calgary, Alberta, 23–26 June, 1991 (Craig, D. A., ed.). MSC 19: Proceedings 50th Annual Meeting Electron Microscopy Society of America, 27th Annual Meeting Microbeam Analysis Society, Nineteenth Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, Boston MA, 16–21 August 1992 (Bailey, G. W., Bentley, J., and Small, J. A., eds.; San Francisco Press, San Francisco 1992) 2 vols. MSC 20: University of Toronto, 3–5 June, 1993 (Doane, F. W., ed.). MSC 21: University of Montreal, 12–15 June, 1994 (Doane, F. W., ed.). MSC 22: University of Ottawa, 4–7 June, 1995 (Doane, F. W., ed.). MSC 23: Proceedings Microscopy and Microanalysis 1996. 54th Annual Meeting Microscopy Society of America, Twenty-third Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, 30th Annual Meeting Microbeam Analysis Society, Minneapolis MN, 11–15 August, 1996 (Bailey, G. W., Corbett, J. M., Dimlich, R. V. W., Michael, J. R., and Zaluzec, N. J., eds.; San Francisco Press, San Francisco 1996). MSC 24: Medical Sciences Building, University of Alberta, Edmonton, 4–7 June 1997 (Egerton, R. F., ed.). MSC 25: Ecole de Technologie Supe´rieure, Montre´al PQ, 27–29 May 1998 (Montpetit, D., ed.). MSC 26: University of Guelph, Guelph ON, 26–28 May 1999 (Corbett, J. M., ed.). MSC 27: Proceedings Microscopy and Microanalysis 2000. 58th Annual Meeting Microscopy Society of America, 27th Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, 34th Annual Meeting Microbeam Analysis Society, Philadelphia PA, 13–17 August, 2000; Microsc. Microanal. 6 (2000) Supplement 2 (Bailey, G. W., McKernan, S., Price, R. L., Walck, S. D., Charest, P.-M., and Gauvin, R., eds; Springer, New York 2000).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

295

MSC 28: University of New Brunswick, Fredericton NB, 6–8 June 2001 (Hall, D. C., ed.). MSC 29: Proceedings Microscopy and Microanalysis 2002. 60th Annual Meeting Microscopy Society of America, 29th Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, 36th Annual Meeting Microbeam Analysis Society, 35th Meeting International Metallographic Society, Quebec City PQ, 4–8 August, 2002. Microsc. Microanal. 8 (2002) Supplement 2 (Voelkl, E., Piston, D., Gauvin, R., Lockley, A. J., Bailey, G. W., and McKeman, S., eds; Cambridge University Press, New York, 2002). The printed proceedings contain only the invited papers, all others are distributed as a CD-ROM. MSC 30: Vancouver BC, 4–6 June 2003. MSC 31: Wolfville NS, 2004.

C. South America A society to bring together the electron microscopists of South America was first created in 1972, a year after the foundation of the Venezuelan society on which it was to some extent based; this was the Sociedad Latinoamericana de Microscopı´a Electro´nica (SLAME) under whose auspices eight congresses were held, see Section III.D (Regional Conferences). Now, however, the electron microscopy societies of several South American countries belong to the Comite´ de Sociedades Interamericanas para Microscopı´a Electro´nica (Committee of Interamerican Societies of Electron Microscopy, CIASEM), the objects of which are to hold an Interamerican Congress every two years, to sponsor regional meetings and international publications in the field of microscopy and more generally, to promote research activity in this area. Details of these Interamerican congresses are also to be found in Section III.D. The countries listed below are those currently (2003) represented within CIASEM. 1. Argentina The present Sociedad Argentina de Microscopı´a (SAMIC) was formally launched in 1995. It is not, however, very active in electron microscopy and there are only two modern instruments in the country (in 2002), as well as about ten older TEMs and approximately 30 SEMs. Work in the materials sciences tends to be presented at the meetings of the Asociacio´n Fı´sica Argentina and the Sociedad Argentina de Materiales. In 1978, the Fourth Latin-American Congress on Electron Microscopy was held in Mendoza (see Section III.D).

296

HAWKES

2. Brazil (www.dema.ufscar.br/sbmm/) From 1971 until 1983, the Sociedade Brasileira de Microscopia Eletroˆnica (Brazilian Society for Electron Microscopy) held a series of general colloquia; from 1986 onwards, these have been supplemented by various more specialized meetings. These are listed here in overall chronological order. In September 1997, the name of the society was changed to Sociedade Brasileira de Microscopia e Microana´lise. I. II. III. IV.

V. VI. VII. VIII. IX. X.

XI.

XII.

XIII.

XIV.

XV.

Coloquium, Sa˜o Paolo (SP), 1971. Coloquium, Sa˜o Paolo (SP), 1972. Coloquium, Rio de Janeiro (RJ), 1973. Coloquium, Ribeira˜o Preto (SP), 1–5 December, 1974, together with the Second Latin-American Congress for Electron Microscopy (see Section III.D). Coloquium, Piracicaba (SP), 1976. Coloquium, Sa˜o Paolo (SP), 1978. Coloquium, Rio de Janeiro (RJ), 1980. Coloquium, Rio de Janeiro (RJ), 1981. Coloquium, Ribeira˜o Preto (SP), 1983. Coloquium, Sa˜o Paolo (SP), 1985. First Symposium on Special Techniques for Electron Microscopy, Caxambu´ (MG), 10–11 September 1986. Coloquium, Caxambu´ (MG), 1–3 September 1987. First Symposium on Ultrastructural Cytochemistry, Curitiba (PR), November 1988. MICROMAT I: Brazilian Conference on Microscopy of Materials, Sa˜o Paolo (SP), 24–26 October 1988. Coloquium, Caxambu´ (MG), 1989. MICROMAT II: Brazilian Conference on Microscopy of Materials, Sa˜o Paolo (SP), 1990. Coloquium, Caxambu´ (MG), 1991. First Symposium on Biological Membranes, Nova Friburgo (RJ), 1992. MICROMAT III: Brazilian Conference on Microscopy of Materials, Rio de Janeiro (RJ), 1992. Coloquium, Caxambu´ (MG), 1993. MICROMAT IV: Brazilian Conference on Microscopy of Materials, Sa˜o Carlos (SP), 1994. Coloquium, Caxambu´ (MG), 2–6 September 1995. Held jointly with the III Interamerican Congress on Electron Microscopy (see Section III.D), Proceedings published as Acta Microsco´pica 4 (1995), Supplements A (Biological Science) and B (Materials Science), see too Acta Microsco´pica 6 (1997) 59–91.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

297

Second Symposium on Ultrastructural Cytochemistry, Rio de Janeiro (RJ), 6–9 October, 1996. Proceedings published as Acta Microsco´pica 5 (1996), Supplement A. MICROMAT V: Brazilian Conference on Microscopy of Materials, Rio de Janeiro (RJ), 13–16 October, 1996. Proceedings published as Acta Microsco´pica 5 (1996), Supplement B. XVI. Coloquium, Caxambu´ (MG), 1–5 September 1997. Proceedings published as Acta Microsco´pica 6 (1997), Supplements A [Physical Sciences] and B [Life Sciences]; see also Acta Microsco´pica 6 (1997) 59–91. MICROMAT 98: Proceedings of the 6th Brazilian Conference on Microscopy of Materials, Aguas de Lindo´ia, Sa˜o Paolo, 25–27 October 1998. Proceedings published as Acta Microsco´pica 7 (1998), Supplement A. XVII. Congress of the Brazilian Society for Microscopy and Microanalysis and X Congress of the Brazilian Society for Cell Biology, Santos (SP), 13–16 October 1999. Proceedings published as Acta Microsco´pica 8 (1999), Supplements A [General Contributions and Conferences], B [Invited papers] and C [Posters]; selected papers were also published in a Brazilian Anniversary Issue of Tissue & Cell 31 (1999) No. 3 which gives the dates of the meeting as December 1998 and the venue as Sa˜o Paolo. Primeiro Congreso Luso-Brasileiro de Morfologia Functional, Goiaˆnia, Goia´s (Brazil), 26–31 August 2000. Joint meeting with the Sociedade Brasileira de Anatomia, the Sociedade Brasileira de Microscopia e Microana´lise, the Sociedade Brasileira de Biologia Celular, the Associac¸a˜o Paranaense para o Desenvolvimento do Ensino da Cieˆncia and The Sociedade Anatoˆmica Portuguesa. Brazil. J. Morphol. Sci. 17 (2000) Supplement.  MICROMAT VII: Proceedings of the 7 Congreso Brasileiro de Micro scopia de Materiais together with CBECiMat-14, the 14 Congreso Brasileiro de Engenharia e Cieˆncias dos Materiais, Hotel Fazenda Fonte Colina Verde, Sao Pedro (SP), 3–6 December 2000. XVIII. Congreso de Sociedade Brasileira de Microscopia e Microana´lise, Aguas de Lindo´ia (SP), 28–31 October 2001. SBMM-2002, this congress unites MICROMAT and a Simpo´sio de Metodologias Integradas no Estudo de Biologia, Centro Integrado dos Empresa´rias e Trabalhadores das Indu´strias do Parana (CIETEP), Curitiba (PR), 20–22 November 2002.

298

HAWKES

3. Chile There is at present no formal Chilean Society of Electron Microscopy although I believe that such a society was in the process of creation in 1997. The Third Latin-American Congress on Electron Microscopy was held in Santiago de Chile in 1976 (see Section III.D). Abstracts of publications in this field are published in Noticiero de Biologı´a and electron microscopy in the life sciences is well represented in Biological Research (formerly Archivos de Biologı´a y Medecina Experimentales). 4. Colombia Columbia has an active Society of Electron Microscopy. Several meetings have been held, on themes drawn from both the life and the physical sciences. Thus in 1995, a meeting was held on ‘‘Electron microscopy in research and diagnosis,’’ which resulted in a publication covering both biology and materials science. In 1996, a Symposium on ‘‘Electron microscopy applied to research in the geosciences’’ was held. A further volume of abstracts recording work in biology is planned. Courses on topics in electron microscopy are also organized; in particular, courses on ‘‘Quantitative analysis in geoscience’’ and ‘‘Methodology and foundations of transmission electron microscopy’’ are planned. The Fifth Latin-American Congress on Electron Microscopy was held in Bogota´ in 1981 (see Section III.D). Memorias, II Simposio Colombiano de Microscopı´a Electro´nica aplicada a la Investigacio´n en Geociencias, Santafe´ de Bogota´, 21–23 August 1996. Special Publication No. 2 of the Columbian Society of Electron Microscopy. 5. Cuba Electron microscopy in Cuba is represented, so far as the life sciences are concerned, by the Sociedad de Morfologı´a, which holds regular meetings. The Eighth Latin-American Congress on Electron Microscopy was held in La Habana in 1989 (see Section III.D). 6. Ecuador The Sociedad Ecuatoriana de Microscopı´a Electro´nica was founded on the initiative of Amado Freire Potes, who proposed the creation of such a society on 27 March 1979. This suggestion was accepted unanimously and the statutes were published in the Registro Oficial on 6 January 1980.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

299

1. Memorias del I. Congreso Ecuatoriano de Microscopı´a Electro´nica en Microbiologı´a, Patologı´a, Biologı´a y Geologı´a, Guayaquil, 26–28 September 1995. 2. Fourth Interamerican Congress on Electron Microscopy and II Ecuadorian Congress on Electron Microscopy related to Medical, Biological and Materials Sciences, Guayaquil, 23–26 September 1997. Acta Microsco´pica 6 (1997) 39–57 [posters]. 7. Mexico Microscopy in Mexico is represented by the Asociacio´n Mexicana de Microscopı´a. 1993. Second Interamerican Conference on Electron Microscopy and Primero Congreso Mexicano de Microscopı´a Electro´nica, Cancu´n, 26 September–1 October 1993 (see Section III.D). 1994. Secundo Congreso Mexicano de Microscopı´a Electro´nica, Cancu´n, 26–29 September 1994. Organized by Asociacio´n Mexicana de Microscopı´a. 1996. III Congreso de la Asociacio´n Mexicana de Microscopı´a Cancu´n, 1–5 September 1996. Final programme and Abstracts edited by E. M. Rivera Mun˜oz. ICEM-14, Cancu´n, 31 August–4 September 1998. 2001. Sixth Interamerican Congress on Electron Microscopy, Veracruz (Mexico), 7–11 October 2001. 8. Peru On 6–7 December 1994, a Seminar–Workshop on ‘‘Ultraestructura Celular: Tres De´cadas en el Peru´’’ was held in the Centro Internacional de la Papa and it was at this meeting that the Asociacio´n Peruana de Microscopı´a Electro´nica (APEMEL) was formally founded; a microscope had, however, been installed in Peru many years earlier, in 1959, in the National University of San Augustı´n, in Arequipa. A very full account of activity in electron microscopy in Peru has been prepared (Castilla de Maruenda et al., 1996), which contains the papers delivered at the Seminar–Workshop (including a survey of microscopy in Peru by Takano Moro´n, 1996), the Statutes of the new Asociacio´n, abstracts of theses by Peruvian microscopists, a list of publications on this subject and information about the CIASEM. This volume was launched at a meeting of the APEMEL (23 July 1996), held at the Colegio Me´dico del Peru´. An idea of the range of activities can be obtained from the titles of the other 1996 meetings: ‘‘La microscopı´a electro´nica y la criminalistica’’ (P. Ruiz Chunga, F. Yupanqui Garcia and Andre´s Chavieri Salazar; 25 April); Microscopı´a electro´nica del effecto tunel’’ (F. Camino; 30 May); ‘‘Microscopı´a electro´nica en patologia’’

300

HAWKES

(S. Antunez de Mayolo; 25 June); ‘‘La microscopı´a electro´nica en la geologı´a’’ (G. Ocharan de Vilca; 26 September); and ‘‘20 an˜os de microscopı´a electro´nica en el CIP’’ (E. Velit Tamanaja; 31 October). 9. Uruguay There is at present no formal Uruguayan Society of Electron Microscopy but such a society was in the process of creation in 1997. 10. Venezuela (electra.ciens.ucv.ve/svme/) The Sociedad Venezolana de Microscopı´a Electro´nica was founded in Me´rida in 1971. The first scientific meeting of the society was held in 1984, under the name ‘‘Jornadas Venezolanas de Microscopı´a Electro´nica’’ and meetings have been held every two years since then. In 1994, it was decided to change the name from ‘‘Jornadas’’ to ‘‘Congreso’’ with no interruption in the numbering. Electron microscopy began in Venezuela long before 1971, however, when Humberto Ferna´ndez–Mora´n created the Instituto Venezolano de Neurologia e Investigaciones Cerebrales (IVNIC) in Altos de Pipe (Miranda) in 1950; the electron microscopic studies of Ferna´ndez–Mora´n go back to 1948, in the University of Zulia. In 1964, Orlando J. Castejo´n founded an electron microscopy section in the Instituto de Investigacio´n Clı´nica of this same university and further microscopes were installed over the years. A Centro de Microscopı´a Electro´nica was also created during the 1960s in the Universidad de los Andes, in Me´rida. It was formally inaugurated on 21 September 1968, with Julio M. Sosa as Director and Jose´ A. Serrano as Assistant Director. The main themes of research were concerned with ultrastructural studies in the life sciences and a manual of electron microscopy and a book on various aspects of biological ultrastructure were published, as well as an Atlas of Biological Ultrastructure. In 1971, Serrano founded the Sociedad Venezolana de Microscopı´a Electro´nica and accepted the editorship of the Revista de Microscopı´a Electro´nica. In 1975, Palacios Pru¨ became Director of the Centre, to be succeeded in 1981 by Rosa Virginia Mendoza. In 1979, a second pole of ultrastructural research was created, within the Department of Morphology of the Faculty of Medicine and University Hospital of the Andes. A third major centre was created in 1976 with the installation of a laboratory of electron microscopy in the School of Biology of the Universidad Central de Venezuela. The Centro de Microscopı´a Electro´nica was inaugurated in September 1977 and both physical and biological themes were developed there under the leadership of Mitsuo Ogura (cell biology), Tatiana Me´rida (botany), He´ctor Finol (zoology), Carlos E. Rojas (physics) and Carmelo Bolı´var (chemistry).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

301

For accounts of electron microscopy in Venezuela, see Ferna´ndez-Mora´n (1985) and the brief abstract by Ogura (1992). Three of the Latin-American Congresses on Electron Microscopy have been held in Maracaibo, the first in 1972 and the sixth in 1984 and the fifth Interamerican Congress in 1999 (see Section III.D). I. Jornadas Venezolanas de Microscopı´a Electro´nica, Edif. Postgrado de Facultad de Ciencias Veterinarias, Nu´cleo Maracay, Maracay (Aragua), 23–28 July 1984. Organized by the Faculties of Agriculture and Veterinary Sciences, Universidad Central de Venezuela. Abstracts volume, 260 pp. II. Jornadas Venezolanas de Microscopı´a Electro´nica, Caracas, 14–17 July 1986. Organized by the Centro de Microscopı´a Electro´nica, Facultad de Ciencias, Universidad Central de Venezuela. Abstracts volume, 112 pp. III. Jornadas Venezolanas de Microscopı´a Electro´nica, Chemistry Building, IVIC, Altos de Pipe (Caracas), 18–22 July 1988. Organized by the Centro de Microscopı´a Electro´nica, Facultad de Ciencias, Universidad Central de Venezuela. Abstracts volume, 91 pp. IV. Jornadas de Microscopı´a Electro´nica, Hotel Bordo´, Cumana´ (Sucre), 16–18 July 1990. Organized by the Instituto de Investigaciones Biome´dicas y de Ciencias Aplicadas, Universidad de Oriente. Memorias edited by Comite Organizador IV JME, 260 pp. V. Jornadas Venezolanas de Microscopı´a Electro´nica, Me´rida (Me´rida), 25–29 May 1992, joint with the First Atlantic Congress of Electron Microscopy (see Section III.D). VI. Jornadas Venezolanas de Microscopı´a Electro´nica, Hotel Maruma, Maracaibo (Zulia), 3–6 July 1994. Organized by the Capitulo Zuliano de la SVME under the auspices of the Universidad de Zulia. Abstracts volume, 230 pp. VII. Congreso Venezolano de Microscopı´a Electro´nica, Hotel Suite Ucaima, Urb. La Vin˜a, Valencia (Carabobo), 25–27 September 1996. Organized by the Centro de Investigaciones Me´dicas y Biotecnolo´gicas de la Universidad de Carabobo (CIMBUC). Resumen, 300 pp. V. Interamericano Congreso de Microscopı´a Electro´nica and VIII Congreso Venezolano de Microscopı´a Electro´nica, Isla de Margarita, 24–28 October 1999 (see Section III.D). IX. Congreso Venezolano de Microscopı´a Electro´nica, Instituto de Investigaciones en Biomedicina y Ciencias Aplicadas, Universdad de Oriente, Cumana´, 5–8 November, 2000. Acta Microsco´pica (2000), Suplemento No. 1.

302

HAWKES

X. Congreso Venezolano de Microscopı´a Electro´nica, Maracaibo, 14–16 October 2002. Proceedings distributed on CD-ROM.

D. Asia and Oceania The Committee of Asia–Pacific Societies of Electron Microscopy has representatives from Australia (Section IV.F.1), Burma (Section IV.D.1), China (Section IV.D.2), Hong Kong (Section IV.D.3), India (Section IV.D.4) Indonesia (Section IV.D.5), Japan (Section IV.D.6), Korea (Section IV.D.7), Malaysia (Section IV.D.8), New Zealand (Section IV.F.2), Pakistan (Section IV.D.9), the Philippines (Section IV.D.10), Singapore (Section IV.D.11), Taiwan-China (Section IV.D.12), and Thailand (Section IV.D.13). These range from long-established societies to informal user groups. 1. Burma (Myanmar) No information obtained. 2. China The Chinese Electron Microscopy Society. Extended abstracts of the papers delivered at the biennial meetings of the Chinese Electron Microscopy Society are published in the Journal of the Chinese Electron Microscopy Society (Dianzi Xianwei Xuebao). The Society was founded on the occasion of the First National Conference on Electron Microscopy, held in Chengdu in 1980. Details of the Chinese–Japanese Seminars are to be found below. Electron microscopy is also a major feature of the Beijing Conferences and Exhibitions on Instrumental Analysis (BCEIA), which have been held biennially since 1985; for bibliographic details of the corresponding publications, see Yao (1985, 1987, 1989, 1991, 1993) and Lin (1995, 1997, 1999). For accounts of the development of electron microscopy in China, see Qian and Kuo (1996), Yao (1996) and the very vivid personal record written by Huang (1996a), who has also contributed a historical paper to the Journal of the Chinese Electron Microscopy Society (Huang, 1996b). 1. Chengdu, 5 November 1980. Three Abstracts Volumes were distributed to participants: Vol. 1, Instrumentation and Techniques; Vol. 2, Biology; Vol. 3, Materials Sciences. 2. Guangzhou, 9–14 November 1982 (372 abstracts). 3. Beijing, 9–11 December 1984. J. Chinese Electron Microsc. Soc. 3 (1984) Nos. 3 and 4.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

303

4. Baotou,1–7August1986.J.ChineseElectronMicrosc.Soc.5(1986)No.3. 5. Wuhan, 15–21 October 1988. J. Chinese Electron Microsc. Soc. 7 (1988) No. 3. 6. Beijing, 22–26 October 1990. J. Chinese Electron Microsc. Soc. 9 (1990) No. 3. An international symposium was held at the same date and the proceedings were published as International Symposium on Electron Microscopy (Kuo, K. and Yao, J, eds.), Beijing, 22–23 October 1990 (World Scientific, Singapore, New Jersey, London & Hong Kong 1991). APEM-5, Beijing, 2–6 August 1992. 7. Huangshan, 6–11 April 1993. J. Chinese Electron Microsc. Soc. 12 (1993) Nos. 1 and 2. 8. Xi’an, 9–14 October 1994. J. Chinese Electron Microsc. Soc. 13 (1994) Nos. 5 and 6. 9. Beijing, 9–14 October 1995. J. Chinese Electron Microsc. Soc. 15 (1996) Nos. 5 and 6. 10. Hefei,24–28June1996. J.ChineseElectronMicrosc. Soc.16(1997) No.4. 11. Dalian, 15–19 September 1998. J. Chinese Electron Microsc. Soc. 17 (1998) Nos. 4 and 5. 12. Kunmin, Yunnan, 4–9 September, 2000. J. Chinese Electron Microsc. Soc. 19 (2000) Nos. 3 and 4. 13. Taiyuan City, Shanxi Province, 19–23 August 2002. J. Chinese Electron Microsc. Soc. 21 (2002). Chinese–Japanese seminars New Trends of Electron Microscopy in Atom Resolution Materials Science and Biology, Proceedings of the First Chinese–Japanese Electron Microscopy Seminar, Dalian, 27–31 July, 1981 (Hashimoto, H., Kuo, K. H., and Ko, T., eds.; Science Press, Beijing 1982). Recent Developments of Electron Microscopy, Proceedings of the Second Chinese–Japanese Electron Microscopy Seminar, Beijing, 17–19 October 1983 (Hashimoto, H., Ko, T., Kuo, K. H., and Ogawa, K., eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Developments of Electron Microscopy, Proceedings of the Third Chinese–Japanese Electron Microscopy Seminar, Hanzhou, 4–7 November 1985 (Hashimoto, H., Kuo, K. H., Lee, K., and Ogawa, K, eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Developments of Electron Microscopy, Proceedings of the Fourth Chinese–Japanese Electron Microscopy Seminar, Kunming, 8–12 November 1987 (Hashimoto, H., Kuo, K. H., Lee, K., and Ogawa, K., eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Developments of Electron Microscopy, Proceedings of the Fifth Chinese–Japanese Electron Microscopy Seminar, Urumqi, 5–7 September

304

HAWKES

1990 (Hashimoto, H., Kuo, K. H., Lee, K., and Ogawa, K., eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Developments of Electron Microscopy, Proceedings of the Sixth Chinese–Japanese Electron Microscopy Seminar, Okayama, 5–9 November 1991 (Hashimoto, H., Kuo, K. H., Lee, K., and Ogawa, K., eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Development of Electron Microscopy, Proceedings of the Seventh Chinese–Japanese Electron Microscopy Seminar, Zhang Jia Jie, 1–4 November 1993 (Hashimoto, H., Kuo, K. H., Lee, K., and Suzuki, T., eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Development of Electron Microscopy, Proceedings of the Eighth Chinese–Japanese Electron Microscopy Seminar, Wuishan 1–3 May 1995 (Hashimoto, H., Kuo, K. H., Lee, K., and Ogawa, K., eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Development of Electron Microscopy, Proceedings of the Ninth Chinese–Japanese Electron Microscopy Seminar, Dalian 20–24 September 1998 (Hashimoto, H. and Li, F. H., eds.; Japanese Society of Electron Microscopy, Tokyo). Recent Development of Electron Microscopy, Proceedings of the Tenth Chinese–Japanese Electron Microscopy Seminar, Chendu, 3–7 November 1999 (Hashimoto, H. and Li, F. H., eds.; Japanese Society of Electron Microscopy, Tokyo). 3. Hong Kong APEM-6, Hong Kong, 1–5 July 1996. 4. India For each of the meetings of the Electron Microscope Society of India (EMSI), a proceedings book has been produced, containing one or two full lectures and abstracts of the papers presented during the conference. Some of these form special or regular issues of the original EMSI Bulletin; a new series of EMSI Bulletins was launched in December 2000 (starting with volume 1, number 1). See too Das Gupta and De (1968). An International Symposium on Electron Microscopy in Life Science was held in Calcutta in 1969, see Das Gupta (1969). The second Regional Conference on Electron Microscopy in Far East and Oceania was held in Calcutta in 1965 (see Section III.B). 1. Saha Institute of Nuclear Physics, Calcutta, 1961. 2. Tata Cancer Research Centre, Bombay, 1963.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

305

3. APEM-2, Saha Institute of Nuclear Physics, Calcutta, 2–6 February 1965. 4. National Physical Laboratory, New Delhi, 1966. 5. Indian Institute of Technology, Kharagpur, 1967. 6. Tata Institute of Fundamental Research, Bombay, 1969. 7. Vigyan Bhawan, Defence Science Laboratory, Delhi, 1972. Abstracts of the Seventh Annual Conference of the Electron Microscope Society of India. 8. Department of Physics, Faculty of Science, Banaras Hindu University, Varanasi, 18–20 October 1973. Proceedings of the Eighth Annual Conference of the Electron Microscope Society of India (Bhattacharya, D. L., ed.). 9. Central Drug Research Institute, Lucknow, 6–8 December, 1976. Ninth Annual Conference of the Electron Microscope Society of India, Abstracts. 10. Bhaba Atomic Research Institute, Bombay, 19–21 December, 1977. Tenth Annual Conference Programme and Abstracts, Bull. Electron Microsc. Soc. India 1 (1977) No. 4. 11. Madras Medical College, Madras, 8–10 January, 1979. Eleventh Annual Conference of the Electron Microscope Society of India, Abstracts. 12. CSIO Chandigarh, 17–19 December 1979. Proceedings XII Annual Conference of the Electron Microscope Society of India. 13. Indian Institute of Science, Bangalore, 1981. 14. Indian Association for the Cultivation of Science, Calcutta, 18–20 January 1982. 15. Banaras Hindu University, Varanasi, 16–18 December 1982. 16. Kamraj University, Madurai, 24–26 November 1983. 17. Department of Biophysics, Punjab University, Chandigarh, 31 January–3 February 1986. 18. RSIC, North Eastern Hill University, Shillong, Meghalaya, 4–6 October 1988. 19. National Physical Laboratory, New Delhi, 14–16 December 1994. Electron Microsc. Soc. India Bull., Special Issue 1994. 20. Indian Association for the Cultivation of Science, Jadavpur, Calcutta, 5–7 December 1996; Souvenir (Chaudhuri, S., Pal, R., and Banerjee, S., eds.) 102 pp. 21. Regional Research Laboratory, Trivendrum, Kerala, 17–19 December, 1997. 22. Centre for Cellular and Molecular Biology, Hyderabad (Andhra Pradesh), 9–11 November 1998 (Gupta, P. D., convener). 23. DefenceMaterialandStoresResearchandDevelopmentEstablishment, Kanpur (Uttar Pradesh), 1–3 December 1999 (Dwivedi, C. D., convener).

306

HAWKES

24. ‘‘Electron Microscopy and Allied Fields.’’ Regional Sophisticated Instrumentation Centre, Central Instrumentation Laboratory, Panjab University, Chandigarh, 9–11 February 2001 (Abstract Book, 172 pp.; Sahni, A, president, Johal, M. S., Hon. General Secretary, Sharma, M. L., treasurer). 5. Indonesia The Indonesian Society of Microscopy and Microanalysis was created at the first National Seminar on Microscopy and Microanalysis in 1997. Annual meetings are held and the Microscopy and Microanalysis Journal is published semestrially. 1. 2. 3. 4.

First National Seminar, 14–15 April 1997. August 3–4 1998. December 14 1998. Graha Widya Bakti (DRN) Hall, Kawasan Puspiptek, Serpong, near Jakarta, 8 September 1999. 5. PUSPIPTEK (Centre for Research, Science and Technology), Serpong, near Jakarta. 6. Bandung, 5 February 2002 and Yogyakarta, 8 February 2002. 6. Japan (jsem.bcasj.or.jp/index_e.html) For extensive accounts of the development of electron microscopy in Japan, see the books produced to accompany the two ICEM meetings in Kyoto (Sugata, 1968; Fujita, 1986) and Chapters 2.9 and 4.5 by Yada, Komoda and many other authors in ‘‘The Growth of Electron Microscopy’’ (Yada et al., 1996; Komoda et al., 1996). Much other material is available: Hashimoto (1983, 1996), Hibi (1985), Higashi (1966, 1967), Hitachi (1959), JEOL (1959–1973), Kanaya (1985), Tadano (1953) and Tani (1956b). The organized study of electron microscopy in Japan may be said to have begun in 1939, when the 37th Co-operative Research Committee came into being, with S. Seto as Chairman. Reports were produced approximately four times a year and when the Japanese Society of Electron Microscopy was formed, in 1949, the Committee reports were printed together with the Society proceedings. The Society was launched on 13 May 1949 in the Engineering Department of the University of Tokyo, with S. Seto as president. From 1953 onwards, the Society organized an annual meeting in Spring and an annual Symposium in the Autumn. Meanwhile, the 37th Cooperative Committee had divided into two Committees: the High Resolution Electron Microscope Committee (HREM), with Y. Tani as Chairman, and the Thin Sectioning

ELECTRON OPTICS AND ELECTRON MICROSCOPY

307

Methods and Techniques Committee (TSMT), with K. Sasagawa as chairman; these two Committees met between 1952 and 1954. The Society began publication of the Journal of Electronmicroscopy (later Journal of Electron Microscopy) in 1953; the first eight volumes were annual, after which the journal appeared quarterly. A house journal, Denshikenbikyo¯, has been distributed to members since 1950. A useful index in English and Japanese to the relatively elusive first eight volumes of both serials is included in volume 9 (1960) of Journal of Electron Microscopy (pp. 162–179). In 1997, Oxford University Press took over publication of the Journal of Electron Microscopy and the frequency became bimonthly. Various electron microscopy laboratories and electron microscope manufacturers produce (or have produced) serial publications (in particular, Hitachi, JEOL and Shimazu). The Centre for Ultra-high Voltage Electron Microscopy in Osaka University (Yamadaoka, Suita, Osaka) has issued Denken, later Ultra-Denken since 1972. The High-Voltage Electron Microscopy Laboratory in Kyushu University publishes a collection of Annual Reports (No. 25 in 2001). The High-voltage Electron Microscopy Laboratory in Nagoya has published Nagoya Daigaku Denshi Kogaku Kenyu no Ayumi [Progress in Electron Optics Research at Nagoya University] since 1973; this reached No. 17 in 2001. There is also a very active Japanese Society of Electron Microscopy Technology for Medicine and Biology, which holds an annual three-day meeting, an annual one-day symposium and training sessions. Its journal and transactions are at present published only in Japanese but English versions are planned. Several technical texts have been published by the Society and English translations of some of these are contemplated. For details, see www02.so-net.ne.jp/emtech/ guide.html. Details of the Chinese–Japanese Seminars are to be found in Section IV.D.3. Finally, we cite the illustrated memoir of Hatsujiro Hashimoto (1999). Regular meetings JSEM 1: Engineering Department, University of Tokyo, 13 May, 1949 [no Proceedings]. Four lectures were delivered, by Hideo Yamashita, Yasumasa Tani, Keinosuke Kobayashi and Noboru Higashi. JSEM 2: School of Medicine, Kyoto University, 5 October 1949 [Pre-prints distributed to participants at these and subsequent JSEM meetings]. Lectures were delivered by Eijii Sugata, Bunichi Tamamushi and Kyugo Sasagawa, followed by 9 biological, 12 physical and 14 theoretical papers. JSEM 3: Engineering Department, University of Tokyo, 23 March 1950. JSEM 4: School of Medicine, Osaka University, 5 November 1950.

308

HAWKES

JSEM 5: School of Medicine, University of Tokyo, 8 July 1951. JSEM 6: Engineering Department, Kyoto University, 8–9 December 1951. JSEM 7: School of Medicine, Keio University, 24–25 May, 1952. JSEM 8: School of Medicine, University of Nagoya, 16–17 October 1952. JSEM 9: Engineering High School, Institute of Tokyo, 23–24 May 1953. JSEM 10: School of Medicine, Jikei University, Tokyo, 29–30 April 1954. JSEM 11: Yokohama Medical University, 7–8 May, 1955. JSEM 12: Kyotofuritsu School of Medicine, Kyoto, 13–14 June, 1956. APEM-1, Tokyo, 23–27 October 1956. JSEM 13: Jikei Medical University, Tokyo, 14–15 May, 1957; Denshikenbikyo¯ 6 (1957) No. 1/2. JSEM 14: Kyoto University Medical Department, 24–26 June. 1958; Denshikenbikyo¯ 7 (1958) No. 2/3. JSEM 15: Faculty of General Education, Tohoku University, Sendai, 13–15 May, 1959; Denshikenbikyo¯ 8 (1959) No. 2/3. JSEM 16: Dept of Agriculture, Tokyo University, 19–20 May, 1960; J. Electron Microsc. 9 (1960), No. 1 (titles and authors only). JSEM 17: School of Medicine, Jikei University, Tokyo, 18–19 May, 1961; J. Electron Microsc. 10 (1961), No. 2, 128–129 (titles and authors only). JSEM 18: Engineering Faculty, Nagoya University 19–20 May, 1962; J. Electron Microsc. 11 (1962), No. 1, 64–65 (titles and authors only). JSEM 19: 19th Scientific Meeting of the Society of Electron-Microscopy, Japan, Hiroshima University 18–19 May 1963; J. Electron Microsc. 12 (1963) 110–126. JSEM 20: 20th Scientific Meeting of the Japanese Society of ElectronMicroscopy, Tokushima University 16–17 May 1964; J. Electron Microsc. 13 (1964) 29–54. JSEM 21: 21st Scientific Meeting of the Japanese Society of Electron Microscopy, Kagoshima University 15–16 May 1965; J. Electron Microsc. 14 (1965) 127–161. JSEM 22: 22nd Scientific Meeting of the Japanese Society of Electron Microscopy, Tokyo University 7–9 April 1966; J. Electron Microsc. 15 (1966) 30–66. ICEM–6, Kyoto, 28 August–4 September 1966. JSEM 23: 23rd Scientific Meeting of the Japanese Society of Electron Microscopy, Kyushu University 5–6 May 1967; J. Electron Microsc. 16 (1967) 179–220. JSEM 24: 24th Scientific Meeting of the Japanese Society of Electron Microscopy, Yamanashi University 11–12 May 1968; J. Electron Microsc. 17 (1968) 237–282.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

309

JSEM 25: 25th Scientific Meeting of the Japanese Society of Electron Microscopy, Jikei University School of Medicine, 10–11 May 1969; J. Electron Microsc. 18 (1969) 198–233. JSEM 26: 26th Scientific Meeting of the Japanese Society of Electron Microscopy, Science Museum, Tokyo, 20–22 May 1970; J. Electron Microsc. 19 (1970) 284–324. JSEM 27: 27th Scientific Meeting of the Japanese Society of Electron Microscopy, Kyoto, 18–20 May 1971; J. Electron Microsc. 20 (1971) 215–265. JSEM 28: 28th Scientific Meeting of the Japanese Society of Electron Microscopy, Okayama, 23–25 May 1972; J. Electron Microsc. 21 (1972) 203–257. JSEM 29: 29th Scientific Meeting of the Japanese Society of Electron Microscopy, Tokyo, 18–20 May 1973; J. Electron Microsc. 22 (1973) 281–320. JSEM 30: 30th Scientific Meeting of the Japanese Society of Electron Microscopy, Osaka, 22–24 May 1974; J. Electron Microsc. 23 (1974) 201–245. JSEM 31: 31st Scientific Meeting of the Japanese Society of Electron Microscopy, Tokyo, 22–24 May 1975; J. Electron Microsc. 24 (1975) 179–218. JSEM 32: 32nd Scientific Meeting of the Japanese Society of Electron Microscopy, Nagoya 20–22 May 1976; J. Electron Microsc. 25 (1976) 175–224. JSEM 33: 33rd Scientific Meeting of the Japanese Society of Electron Microscopy, Fukuoka 12–14 May 1977; J. Electron Microsc. 26 (1977) 225–276. JSEM 34: 34th Scientific Meeting of the Japanese Society of Electron Microscopy, Sapporo, 20–22 June 1978; J. Electron Microsc. 27 (1978) 333–390. JSEM 35: 35th Scientific Meeting of the Japanese Society of Electron Microscopy, Takarazuka, 23–25 May 1979; J. Electron Microsc. 28 (1979) 201–262; Development of Electron Microscopy and its Future, Proceedings of 30th Anniversary of Japanese Society of Electron Microscopy, 22 May 1979; J. Electron Micros. 28 (1979) Supplement. JSEM 36: 36th Scientific Meeting of the Japanese Society of Electron Microscopy, Yokohama 27–29 May 1980; J. Electron Microsc. 29 (1980) 274–339. JSEM 37: 37th Scientific Meeting of the Japanese Society of Electron Microscopy, Kyoto 20–22 May 1981; J. Electron Microsc. 30 (1981) 213–280.

310

HAWKES

JSEM 38: 38th Scientific Meeting of the Japanese Society of Electron Microscopy, Komaba Eminence, Tokyo, 26–28 May 1982; J. Electron Microsc. 31 (1982) 285–344. JSEM 39: 39th Scientific Meeting of the Japanese Society of Electron Microscopy, Aichi Sangyo Boeki-kan, Nagoya, 31 May–2 June 1983; J. Electron Microsc. 32 (1983) 219–287. JSEM 40: 40th Scientific Meeting of the Japanese Society of Electron Microscopy, Sendai Shimin Kaikan, Sendai, 27–29 June 1984; J. Electron Microsc. 33 (1984) 261–322. JSEM 41: 41st Scientific Meeting of the Japanese Society of Electron Microscopy, Hokkaido University, Sapporo, 25–27 June 1985; J. Electron Microsc. 34 (1985) 183–247. ICEM-11, Kyoto, 31 August–7 September 1986. JSEM 43: 43rd Scientific Meeting of the Japanese Society of Electron Microscopy, Kanagawa-kenritsu Kenmin Hall, Yokohama, 27–29 May 1987; J. Electron Microsc. 36 (1987) 294–352. JSEM 44: 44th Scientific Meeting of the Japanese Society of Electron Microscopy, Sendai Shimin Kaikan, Sendai, 1–3 June 1988; J. Electron Microsc. 37 (1988) 232–286. JSEM 45: 45th Scientific Meeting of the Japanese Society of Electron Microscopy, Osaka Shoko Kaigishu, Osaka, 31 May–2 June 1989, Abstracts of the Presentation in Commemoration of the 40th Anniversary of the Japanese Society of Electron Microscopy; J. Electron Microsc. 38 (1989) 250–320. JSEM 46: 46th Scientific Meeting of the Japanese Society of Electron Microscopy, Gunma University, Maebashi, 17–19 May 1990; J. Electron Microsc. 39 (1990) 275–349. JSEM 47: 47th Scientific Meeting of the Japanese Society of Electron Microscopy, Osaka Sun Palace, Suita, Osaka, 22–24 May 1991; J. Electron Microsc. 40 (1991) 234–300. JSEM 48: 48th Scientific Meeting of the Japanese Society of Electron Microscopy, Makuhari Messe, Chiba, 2–4 June 1992; J. Electron Microsc. 41 (1992) 277–320. JSEM 49: 49th Scientific Meeting of the Japanese Society of Electron Microscopy, International Conference Center, Kobe, 26–28 May 1993; J. Electron Microsc. 42 (1993) 244–283. JSEM 50: 50th Scientific Meeting of the Japanese Society of Electron Microscopy, Hokutopia, Kita-ku, Tokyo, 25–27 May 1994; J. Electron Microsc. 43 (1994) 213–253. JSEM 51: 51st Annual Meeting of the Japanese Society of Electron Microscopy, Riihga Royal Hotel, Sakai, Osaka, 24–26 May 1995; J. Electron Microsc. 44 (1995) 231–265.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

311

JSEM 52: 52nd Annual Meeting of the Japanese Society of Electron Microscopy, Nipponseinenkan Shinjuku, Tokyo, 22–24 May 1996; J. Electron Microsc. 45 (1996) 325–358. JSEM 53: Proceedings of the 53rd Annual Meeting of the Japanese Society of Electron Microscopy, Archaic, Amagasaki, 21–23 May 1997; Denshi Kenbikyo¯ [Electron Microscopy] 32 (1997) Supplement 1. JSEM 54: Proceedings of the 54th Annual Meeting of the Japanese Society of Electron Microscopy, Sendai International Centre, 13–15 May 1998; Denshi Kenbikyo¯ [Electron Microscopy] 33 (1998) Supplement 1. JSEM 55: Proceedings of the 55th Annual Meeting of the Japanese Society of Electron Microscopy, Nagoya Congress Center, Nagoya, 18–21 May 1999; Denshi Kenbikyo¯ [Electron Microscopy] 34 (1999) Supplement 1. JSEM 56: Proceedings of the 56th Annual Meeting of the Japanese Society of Electron Microscopy, Hoku Topia, Tokyo, 17–19 May 2000; Denshi Kenbikyo¯ [Electron Microscopy] 35 (2000) Supplement 1. JSEM 57: Proceedings of the 57th Annual Meeting of the Japanese Society of Electron Microscopy, ACROS, Fukuoka, 10–12 May 2001; Denshi Kenbikyo¯ [Electron Microscopy] 36 (2001) Supplement 1. JSEM 58: Proceedings of the 58th Annual Meeting of the Japanese Society of Electron Microscopy, International House, Osaka, 14–16 May 2002; Denshi Kenbikyo¯ [Electron Microscopy] 37 (2002) Supplement 1. APEM-8, Kanazawa, Ishikawa Prefecture, 7–11 June 2004. Symposia 1. Engineering Department, Osaka University, 14–16 November 1953; Denshikenbikyo¯ 3 (1954) No. 2, 26 ff and No. 3, 34 ff. 2. School of Medicine, Tohoku University, 10–11 November 1954; Denshikenbikyo¯ 4 (1955) No. 2. 3. School of medicine, Kyushu University, 29–30 October 1955; Denshikenbikyo¯ 5 (1956) No. 1. 4. Mount Hiei, Kyoto University, 3–4 October 1957. 5. School of Medicine, Keioh University, 22–23 November 1958. 6. School of Medicine, Osaka Ichiritsu University, 30–31 October 1959. 7. Engineering Department, Nagoya University, 5–6 November 1960. 8. Kurakukaikan Hall, Hokkaido University, Sapporo, 14–16 September 1961. 9. School of Medicine, Jikei University, Tokyo, 16–17 November 1962. 10. Tokyo Metropolitan University, 3–4 November 1963; J. Electron Microsc. 12 (1963) 260–295 and 13 (1964) 172–183. 11. Kawaguchilo City Hall, Yamanashi, 21–22 November 1964; J. Electron Microsc. 13 (1964) 204–240.

312

HAWKES

12. Municipal Science Museum, Nagoya, 21–22 October 1965; J. Electron Microsc. 14 (1965) 302–359. 13. Tokyo Medical and Dental University, 25–26 November 1967; J. Electron Microsc. 17 (1968) 160–178. 14. School of Medicine, Kanazawa University, 9–10 November 1968; J. Electron Microsc. 18 (1969) 73–85. 15. Nagoya City, Aikichen, 7–8 November 1969; J. Electron Microsc. 19 (1970) 107–117. 16. Mitaka City Public Hall, Tokyo, 5–7 November 1970; J. Electron Microsc. 20 (1971) 72–78. 17. Gunma University, Maebashi, 21–23 October 1971; J. Electron Microsc. 21 (1972) 91–96. 18. Faculty of Science, Tohoku University, Sendai, 9–11 November 1972; J. Electron Microsc. 22 (1973) 105–112. 19. Faculty of Medicine, Kyushu University, Fukuoka, 9–11 November 1973; J. Electron Microsc. 23 (1974) 61–71. 20. Faculty of Science, Hokkaido University, Sapporo, 3–5 October 1974; J. Electron Microsc. 24 (1975) 53–58. 21. Kyoto-kaikan, Kyoto, 12–14 November 1975; J. Electron Microsc. 25 (1976) 51–64. 22. Chiba-ken Bunka-kaikan, Chiba, 29–30 November 1976; J. Electron Microsc. 26 (1977) 63–68. 23. Hiroshima-ken Ishi-kaikan, Hiroshima, 18–19 November 1977; J. Electron Microsc. 27 (1978) 63–69. 24. Yokohama-shi Kyoiku Bunka Senta¯, Yokohama, 15–16 November 1978; J. Electron Microsc. 28 (1979) 58–65. 25. Tottori University School of Medicine, Yonago, 26–27 October 1979; J. Electron Microsc. 29 (1980) 75–84. 26. Kagoshima-ken Sangyo Kaikan, Kagoshima, 18–19 November 1980; J. Electron Microsc. 30 (1981) 89–98. 27. Niigata Ongaku Bunka Kaikan, Niigata, 14–15 October 1981; J. Electron Microsc. 31 (1982) 101–112. 28. Hiroshima-ken Yakuji Eisei Kaikan, Hiroshima, 5–6 November 1982; J. Electron Microsc. 32 (1983) 66–75. [From now on, many symposia have a ‘‘main theme’’; whenever this is given in the abstracts, it is mentioned here.] 29. ‘‘200 kV transmission electron microscopy.’’ Matsumoto Kosei Bunka Kaikan, Matsumoto, 7–8 October 1983; J. Electron Microsc. 33 (1984) 78–92. 30. ‘‘Processing and reconstruction of electron microscope images.’’ Ehime-ken Ishi-kaikan, Matsuyama, 19–20 October 1984; J. Electron Microsc. 34 (1985) 56–65.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

313

31. ‘‘Localization of constituents.’’ Faculty of Engineering, University of Tokyo, Tokyo, 17–18 October 1985; J. Electron Microsc. 35 (1986) 81–90. 32. ‘‘The present state and future view of electron microscopy—the application of the electron microscope to the promotion of new materials science, ultramodern technology and life science.’’ Okayama Eiseikaikan, Okayama, 29–30 October 1987; J. Electron Microsc. 37 (1988) 92–105. 33. Kyushu University, Fukuoka, 13–14 October 1988; J. Electron Microsc. 38 (1989) 70–76. 34. Ishikawa-ken Bunkyo Kaikan, Kanazawa, 14–15 October 1989; J. Electron Microsc. 39 (1990) 193–199. 35. Shimbun Hoso Kaikan, Kochi, 25–26 October 1990; J. Electron Microsc. 40 (1991) 203–210. 36. ‘‘Future of microscopy science and industry.’’ Yamagata University School of Medicine, Yamagata, 10–11 October 1991; J. Electron Microsc. 41 (1992) 199–204. 37. ‘‘New microscopy–principle and application.’’ Big Roof (Naraken New Public Hall), Nara, 29–30 October 1992; J. Electron Microsc. 42 (1993) 126–129. 38. ‘‘New trend of electron microscopy.’’ Tsukuba Center for Institute, Tsukuba, 26–27 October 1993; J. Electron Microsc. 43 (1994) 51–56. 39. ‘‘Recent progress in imaging science.’’ Nagoya Congress Center, Nagoya, 27–28 October 1994; J. Electron Microsc. 44 (1995) 49–54. 40. ‘‘To observe movements and functions in the ultramicroworld– development of imaging techniques.’’ Hokkaido University School of Medicine, 12–13 October 1995; J. Electron Microsc. 45 (1996) 177–183. From now on, the proceedings of the annual symposia are not included in Journal of Electron Microscopy; those of the 1996 and 1997 symposia form a separate serial publication (ISSN 0917–0642) published by the JSEM and distributed by the Business Center for Academic Societies Japan, Tokyo). From 1998 onwards, they are published a second Supplement to Denshi Kenbikyo¯ [Electron Microscopy]. 41. Kyoto Institute of Technology, 24–25 October 1996. Proceedings of the 41st Symposium Japanese Society of Electron Microscopy. 42. ‘‘The 100th anniversary of the discovery of the electron.’’ University of Tokyo, 31 October–1 November 1997. Proceedings of the 42nd Symposium Japanese Society of Electron Microscopy. 43. Proceedings of the 43rd Symposium of the Japanese Society of Electron Microscopy, Chiba University Convention Hall, 28–30 October 1998. Denshi Kenbikyo¯ [Electron Microscopy] 33 (1998) Supplement 2. 44. Proceedings of the 44th Symposium of the Japanese Society of Electron Microscopy, Juntendo University, Hongo Campus, 17–19

314

HAWKES

November 1999. Denshi Kenbikyo¯ [Electron Microscopy] 34 (1999) Supplement 2 (Sakai, T., ed.). 45. Proceedings of the 45th Symposium of the Japanese Society of Electron Microscopy, Okazaki National Research Institutes, Okazaki Conference Center, 21–22 November 2000. Denshi Kenbikyo¯ [Electron Microscopy] 35 (2000) Supplement 2. 46. Proceedings of the 46th Symposium of the Japanese Society of Electron Microscopy, in conjunction with the 7th International Symposium on Advanced Physical Fields, National Institute for Materials Science, 14–16 November 2001. Denshi Kenbikyo¯ [Electron Microscopy] 36 (2001) Supplement 2. 47. Proceedings of the 47th Symposium of the Japanese Society of Electron Microscopy, Sendaishi City War Reconstruction Memorial Hall, 27–28 November 2002. Denshi Kenbikyo¯ [Electron Microscopy] 37 (2002) Supplement 2. Committee Meetings. The 37th Cooperative Research Committee was launched in 1939, with S. Seto as chairman, and met regularly from then on. The 44th, 45th and 46th meetings of the 37th Committee were held in the Japanese Academy, Tokyo Ueno in 1947 and the 47th, 48th and 49th meetings were held in the same venue in 1948. In 1949, the 50th meeting was held in the University of Tokyo on 11–12 May 1949; the 51st meeting was held in Kyoto University on 6–7 October 1949 and the 52nd meeting in the Japanese Academy, Tokyo Ueno, in December of the same year. Subsequent meetings are as follows: 53rd meeting, Japanese Academy, Tokyo Ueno, 21–22 March 1950. 54th meeting, Science Museum, Tokyo Ueno, 7–8 July 1950. 55th meeting, Biological Institute of Osaka University, 6–7 November 1950. 56th meeting, Kyowa Building Hall, Muromachi Tokyo, 9–10 March 1951. 57th meeting, Kyowa Building Hall, Muromachi Tokyo, 9–10 July 1951. 58th meeting, Kyoto University, 11–12 December 1951. 59th meeting, Keioh University, 26–27 May 1952. 1st HREM and TSMT meeting, School of Medicine, Nagoya University, 18 November 1952. 2nd HREM and TSMT meeting, School of Medicine, Jikei university, 11–12 February 1953. 3rd HREM and TSMT meeting, Institute of Tokyo, 25–26 May 1953. 4th HREM and TSMT meeting, Kyoto University, 29–30 October 1953. 5th HREM and TSMT meeting, 12–13 November 1953. 6th HREM and TSMT meeting, 27–28 April 1954.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

315

7. Korea (www.ksem.com) Abstracts of Meetings of the Korean Society of Electron Microscopy (KSEM) were for the most part published in the Korean Journal of Electron Microscopy, which originally appeared twice a year but now appears quarterly (March, June, September and December). Since 1993, however, the Meetings Abstracts have been issued as a separate volume. Meetings were at first held annually, with occasional gaps, but from 1995 onwards, two meetings have been held, in Spring and Autumn. These are complemented by training workshops in the biological sciences and in materials science and by users’ meetings with microscope manufacturers. Inaugural General Meeting, Sungkyunkwan University, Seoul, 6 May 1967. 1st Annual Meeting, Sungkyunkwan University, Seoul, 20 November 1967. Korean J. Electron Microsc. 1(1), 1969, 49–53. 2nd Annual Meeting, Yonsei University, Seoul, 2 December 1968. 3rd Annual Meeting, Konkuk University, Seoul, 15 November 1969. Korean J. Electron Microsc. 2(1), 1972, 47–48. 4th Annual Meeting, Yonsei University, Seoul, 26 May 1973. Korean J. Electron Microsc. 3(1), 1973, 61–65. 5th Annual Meeting, Yonsei University, Seoul, 27 April 1974. Korean J. Electron Microsc. 4(1), 1974, 39–46. 6th Annual Meeting, Yonsei University, Seoul, 17 May 1975. Korean J. Electron Microsc. 5(1), 1975, 45–49. 7th Annual Meeting, Sungkyunkwan University, Seoul, 22 May 1976. Korean J. Electron Microsc. 7(1), 1977, 51–57. 8th Annual Meeting, Seoul National University, Seoul, 21 May 1977. Korean J. Electron Microsc. 8(1), 1978, 91–96. 9th Annual Meeting, Korea University, Seoul, 27 May 1978. Korean J. Electron Microsc. 9(1), 1979, 95–98. 10th Annual Meeting, Yonsei University, Seoul, 16 June 1979. Korean J. Electron Microsc. 10(1/2), 1980, 87–91. 11th Annual Meeting, Seoul National University Seoul, 15 November 1980. Korean J. Electron Microsc. 11(1), 1981, 67–75. 12th Annual Meeting, Korea University, Seoul, 27 June 1981. Korean J. Electron Microsc. 12(1), 1982, 89–95. 13th Annual Meeting, Yonsei University, Seoul, 29 May 1982 1982. Korean J. Electron Microsc. 12(2), 1982, 81–84. 14th Annual Meeting, Yonsei University, Seoul, 21 May 1983 [no abstracts in KJEM]. 15th Annual Meeting, Kyungju Chosun Hotel, Kyungju, 1–2 June 1984 [no abstracts in KJEM].

316

HAWKES

16th Annual Meeting, Yonsei University, Seoul, 1 June 1985. Korean J. Electron Microsc. 16(1), 1986, 117–124. 17th Annual Meeting, Korea University, Seoul, 31 May 1986. Korean J. Electron Microsc. 17(1), 1987, 185–201. 18th Annual Meeting, Seoul National University, Seoul, 30 May 1987. Korean J. Electron Microsc. 17(2), 1987, 85–98. 19th Annual Meeting, Yusung Tourist Hotel, Daejon, 27–28 May 1988 [no abstracts in KJEM]. 20th Annual Meeting, Seoul National University, Seoul, 27 May 1989. Korean J. Electron Microsc. 19(2), 1989, 145–160. 21st Annual Meeting, Sokrisan Tourist Hotel, Boeun, 26 May 1990. Korean J. Electron Microsc. 21(1), 1991, 121–131. 22nd Annual Meeting, Korea University, Seoul, 25 May 1991 [no abstracts in KJEM]. 23rd Annual Meeting, Yusung Tourist Hotel, Daejon, 29 May 1992. Korean J. Electron Microsc. 22(2), 1992, 141–153. 24th Annual Meeting, Korea University, Seoul, 4–5 June 1993. 25th Annual Meeting, Chonbuk National University, Jeonju, 27–28 May 1994. 26th Annual Meeting, Korea University, Seoul, 4–5 June 1995. 1995 Fall Meeting, Changwon, 2 December 1995. 27th Annual Meeting, College of Medicine, Seoul National University, 31 May–1 June 1996. 1996 Fall Meeting, College of Engineering, Chungnam National University, Daejon, 16 November 1996. 28th Annual Meeting, Research Institute of Advanced Materials, Seoul National University, 29–30 May 1997. 1997 Fall Meeting, Kyungnam University, Masan, 7–8 November 1997. 29th Annual Meeting, Yonsei University, Wonju Campus, 29–30 May 1998. 1998 Fall Meeting, Yonsei University, Seoul Campus, 13–14 November 1998. 30th Annual Meeting, Medical School, Korea University, Seoul, 28–29 May 1999. 1999 Fall Meeting, Life Sciences Building, Hallym University, Chunchon, 5–6 November 1999. 2000 Spring Meeting, Art Hall, Hanyang University, Ansan, 26–27 May 2000. 31st Annual Meeting, Catholic Medical University, Seoul, 10–11 November 2000. 2001 Spring Meeting, Korea Basic Science Institute, Daejon, 25–26 May 2001. 32nd Annual Meeting, College of Natural Science, Chosun University, Kwangju, 9–10 November 2001. 2002 Spring Meeting, Advanced Materials Building, Hanyang University, Seoul. 24–25 May 2002.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

317

8. Malaysia The Electron Microscopy Society of Malaysia organizes regular scientific conferences and has published MicroSoM (Bulletin of the Electron Microscopy Society of Malaysia) since 1998; vol. 4 appeared in 2001. 1. Subang Merlin 23–24 November 1991. 2. University of Technology of Malaysia, Johor, 14–15 November 1992. 3. Penang, 1993. 4. Kuala Lumpur, 1994. 5. Kuala Lumpur, 1995 or 1996. 6. First ASEAN Microscopy Conference, Joint meeting of the Electron Microscopy Society of Malaysia and the Microscopy Society of Singapore, 27–30 November 1997. 7. USM Perak, 27–28 November 1998. 8. Awana Genting, Pahang, 2–4 December 1999. 9. Kota Bharu, 12–14 November 2000. 10. ‘‘Advances in the applications of electron microscopy for biological, material and medical sciences.’’ Palace of the Golden Horses, Mines Resort City, Kuala Lumpur, 8–10 November 2001. Organized together with the Faculty of Science and Technology, Universiti Kebangsaan Malaysia. Proceedings of the Tenth Scientific Conference and 11th Annual General Meeting of the Electron Microscopy Society of Malaysia (Leong, Y. K., Kassim, I., Lee, H. K., and Ong, K. C., eds.; Electron Microscopy Society of Malaysia, Kuala Lumpur, 2001). 302 pp. Also available as a CD-ROM. 9. Pakistan No information obtained. 10. The Philippines The Microscopy Society of the Philippines (MicrosPhil) was created on 21 June 2000, with Meliton U. Ordillas as first president and a first Annual General Meeting was held in October 2000. On 18 August 2001, the first seminar and workshop was held. First General Assembly, University of the Philippines, Diliman, Quezon City, October 2000. Second General Assembly and Scientific Conference, National Engineering Center, University of the Philippines, Diliman, Quezon City, 16–17 November 2001. Theme: ‘‘Microscopy in the Philippines: Advancing the Frontiers of Research and Development.’’

318

HAWKES

11. Singapore (www.med.nus.sg/micsoc/about society.htm) APEM-3, Singapore, 29 August–3 September 1984 In 1992, a pro-tem committee was formed, with the creation of a Singapore society as its objective. The Microscopy Society (Singapore) was created in April 1994, with Ng Mah Lee as first president, succeeded in 1996 by Samuel Tay Sam Wah and in 1998 by Bay Boon Huat. 1996. Scientific conference at Ana Hotel, Singapore. 1997. First ASEAN Microscopy Conference, Joint meeting of the Electron Microscopy Society of Malaysia and the Microscopy Society of Singapore, Senai, Johore 27–30 November 1997. APEM-7, Singapore International Convention & Exhibition Centre, Suntec City, Singapore, 26–30 June 2000 together with the second ASEAN Microscopy Conference. 12. Taiwan, China The Chinese Society for Electron Microscopy was founded in 1981, after two years or prehistory as the Electron Microscopy Committee of the Chinese Society for Materials Science. In 1999, the name was changed to Microscopy Society of Taiwan in order to reflect the increasing diversity of the field. Annual meetings have been held since 1982. Extended abstracts have been published for every meeting including those of the Electron Microscopy Committee, so that the 2001 volume is the 22nd in the series. Information is available only about recent meetings. Annual meetings 17. National Tsing Hua University, Hsinchu, 17 September 1998; coincides with the 19th Symposium 18. National Taiwan University, Taipei, June 10 1999; coincides with the 20th Symposium 19. National Tsing Hua University, Hsinchu, June 17 2000; coincides with the 21st Symposium 20. National Tsing Hua University, Hsinchu, 18 December 2001; coincides with the 22nd Symposium 13. Thailand For some account of electron microscopy in Thailand see Mangclaviraj (1996). The Electron Microscopy Society of Thailand publishes the Journal of the Electron Microscopy Society of Thailand, which contains the abstracts of the annual meetings (volume 1 appeared in 1987). 1. The First Annual Conference of the Electron Microscopy Society of Thailand, Chulalongkorn University, Phya Thai, Bangkok, 20–22 April 1983.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

319

2. The Second Annual Conference of the Electron Microscopy Society of Thailand, Chulalongkorn University, Phaya Thai, Bangkok, 8 February 1984. 3. The Third Annual Conference of the Electron Microscopy Society of Thailand, Chulalongkorn University, Phaya Thai, Bangkok, 11 February 1985. 4. The Fourth Annual Conference of the Electron Microscopy Society of Thailand, Faculty of Science, Kasetsart University, Bang Ken, Bangkok, February 1986. 5. The Fifth Annual Conference of the Electron Microscopy Society of Thailand, Kasetsart University, Kamphaeng Saen, Nakhon Pathom, 13 February 1987. APEM-4, Bangkok, 26 July–4 August 1988. This coincided with the Sixth Annual Conference of the Electron Microscopy Society of Thailand. 7. The Seventh Annual Conference of the Electron Microscopy Society of Thailand, Thammasat University, Rangsit Centre, Khongluang, Pathum Thani, 16 December 1988. 8. The Eighth Annual Conference of the Electron Microscopy Society of Thailand, 6–8 December 1990, Chiang Mai Phu Khum Hotel, Chiang Mai, with the collaboration of the Faculty of Medicine of Chiang Mai University. 9. The Ninth Annual Conference of the Electron Microscopy Society of Thailand, Faculty of Science, Prince of Songkla University, Hat Yai, Songkla, 11–13 December 1991. 10. The Tenth Annual Conference of the Electron Microscopy Society of Thailand, Chulalongkorn University, Phaya Thai, Bangkok, 6 November 1992. 11. The Eleventh Annual Conference of the Electron Microscopy Society of Thailand, Faculty of Science, Mahidol University, Paholyothin, Bangkok, 8–9 December 1993. 12. The Twelfth Annual Conference of the Electron Microscopy Society of Thailand, Suranaree University of Technology, Nakhon Ratchasima, 14–16 December 1994. 13. The Thirteenth Annual Conference of the Electron Microscopy Society of Thailand, 6–8 December 1995, Wiang Inn Hotel, Chiang Rai, with the collaboration of the Faculty of Medicine of Chiang Mai University. 14. The Fourteenth Annual Conference of the Electron Microscopy Society of Thailand, J. B. Hotel, Hat Yai, Songkla, 11–13 December 1996. 15. The Fifteenth Annual Conference of the Electron Microscopy Society of Thailand, Chulalongkorn University, Phaya Thai, Bangkok, 19 December 1997.

320

HAWKES

16. The Sixteenth Annual Conference of the Electron Microscopy Society of Thailand, 17–18 December 1998, Kasetsart University, Bangkok. Abstracts published in J. Electron Microsc. Soc. Thailand 12(Supplement), December 1998. 17. The Seventeenth Annual Conference of the Electron Microscopy Society of Thailand, to Commemorate the 40th Anniversary of the Faculty of Medicine, Chiang Mai University, 7–9 December 1999, Suan Bua Hotel & Resort, Chiang Mai. Abstracts published in J. Electron Microsc. Soc. Thailand 13(Supplement), December 1999. 18. The Eighteenth Annual Conference of the Electron Microscopy Society of Thailand, to Commemorate the 36th Anniversary of Khon Kaen University, 17–19 January 2001, Charoen Thani Princess Hotel, Khon Kaen, Thailand. Abstracts and Full papers published in J. Electron Microsc. Soc. Thailand 15(1), January 2001. 19. Third ASEAN Conference and 19th Annual Conference of the Electron Microscopy Society of Thailand, Lotus Hotel, Pang Saun Kaew, Chiang Mai, 30 January–1 February 2002. J. Electron Microsc. Soc. Thailand 16 (2002) No. 1.

E. Africa 1. South Africa (www.uct.ac.za/depts/emu/mssa) For a most interesting and full account of the creation of the Electron Microscopy Society of Southern Africa, see Witcomb and WolfeCoote (1996). A proceedings volume has been published for each volume from the tenth meeting (1971) onwards, with the title Electron Microscopy Society of Southern Africa Proceedings or Elektronenmikroskopievereniging van Suidelike Afrika Verrigtinge; the short form EMSSA Proc. is used below. In 1996, the society became the Microscopy Society of Southern Africa/Mikroskopievereniging van Suidelike Afrika and the word ‘‘Electron/Elektron’’ disappeared from the title of the proceedings (MSSA Proc. below). EMSSA 1: University of the Witwatersrand, Johannesburg, December, 1962. EMSSA 2: University of the Witwatersrand, Johannesburg, December, 1963. EMSSA 3: Council for Scientific and Industrial Research, Pretoria, 1 December, 1964. EMSSA 4: University of the Witwatersrand, Johannesburg, 7 December, 1965. EMSSA 5: University of Pretoria, Pretoria, 6 December 1966. S. Afr. J. Sci. 63 (1967) 184–187.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

321

EMSSA 6: South African Institute for Medical Research, Johannesburg, 5 December 1967. S. Afr. J. Sci. 64 (1968) 413–419. EMSSA 7: University of Pretoria, Pretoria, 3 December 1968. S. Afr. J. Sci. 65 (1969) 296–300. EMSSA 8: University of the Witwatersrand, Johannesburg, 3 December 1969. S. Afr. J. Sci. 66 (1970) 324–326. EMSSA 9: University of Natal, Durban, 4 December 1970. EMSSA 10: Council for Scientific and Industrial Research, Pretoria, 3 December 1971. EMSSA Proc. 1 (1971). EMSSA 11: University of the Witwatersrand, Johannesburg, 30 November–1 December 1972. EMSSA Proc. 2 (1972). EMSSA 12: University of Pretoria, Pretoria, 29–30 November, 1973. EMSSA Proc. 3 (1973). EMSSA 13: University of the Witwatersrand, Johannesburg, 28–29 November 1974. EMSSA Proc. 4 (1974). EMSSA 14: University of Pretoria, Pretoria 27–28 November 1975. EMSSA Proc. 5 (1975). EMSSA 15: Rand Afrikaans University, Johannesburg 2–3 December 1976. EMSSA Proc. 6 (1976). EMSSA 16: Newlands Hotel, Cape Town, 1–2 December 1977. EMSSA Proc. 7 (1977). EMSSA 17: Council for Scientific and Industrial Research, Pretoria 4–5 December 1978. EMSSA Proc. 8 (1978). EMSSA 18: University of Port Elizabeth, Port Elizabeth 3–5 December 1979. EMSSA Proc. 9 (1979). EMSSA 19: University of the Witwatersrand, Johannesburg, 1–3 December, 1980. EMSSA Proc. 10 (1980). EMSSA 20: University of Durban Westville, Durban Westville, 2–4 December, 1981. EMSSA Proc. 11 (1981). EMSSA 21: Rhodes University, Grahamstown, 1–3 December, 1982. EMSSA Proc. 12 (1982). EMSSA 22: University of the Witwatersrand, Johannesburg, 30 November–2 December, 1983. EMSSA Proc. 13 (1983); Snyman, H. C., ed. (before this date, the name of the editor is not given in the proceedings). EMSSA 23: University of Stellenbosch, Stellenbosch, 5–7 December, 1984. EMSSA Proc. 14 (1984); Snyman, H. C., Engelbrecht, J. A. A., Fletcher, J., and Murray, P. W. le R., eds. EMSSA 24: University of Natal, Pietermaritzburg, 4–6 December, 1985. EMSSA Proc. 15 (1985); Snyman, H. C., Fletcher, J., and Timme, A. H., eds. EMSSA 25: Potchefstroom University for Christian Education, Potchefstroom 3–5 December 1986. EMSSA Proc. 16 (1986); Witcomb, M. J., Coetzee, J., and Coubrough, R. I., eds.

322

HAWKES

EMSSA 26: University of the Witwatersrand, Johannesburg 2–4 December 1987. EMSSA Proc. 17 (1987); Snyman, H. C., Coetzee, J., and Coubrough, R. I., eds. EMSSA 27: University of Natal, Durban, 6–8 December 1988 EMSSA Proc. 18 (1988); Bentley, A. P., Coetzee, J., and Rawdon, B. B., eds. EMSSA 28: University of Pretoria, Onderstepoort 5–7 December 1989. EMSSA Proc. 19 (1989); Shaw, M. P., Pienaar, R. N., and Bernard, R. T. F., eds. EMSSA 29: Rhodes University, Grahamstown 5–7 December 1990. EMSSA Proc. 20 (1990); Shaw, M. P., Piennar, R. N., and Bernard, R. T. F., eds. EMSSA 30: University of Cape Town, Cape Town 4–6 December 1991. EMSSA Proc. 21 (1991); Neethling, J. H., Baecker, A. A. W., and van der Horst, G., eds. EMSSA 31: University of Natal, Pietermaritzburg, 2–4 December, 1992. EMSSA Proc. 22 (1992); Neethling, J. H., Drennan, P. M., and Rawdon, B. B., eds. EMSSA 32: Berg-en-Dal, Kruger National Park, 1–3 December, 1993. EMSSA Proc. 23 (1993); Drennan, P. M., Hodgson, A. N., Neethling, J. H., Engelbrecht, J. A. A., von Holy, A., and Coetzee, J., eds. EMSSA 33: University of Port Elizabeth, Port Elizabeth, 29 November–2 December, 1994. EMSSA Proc. 24 (1994); Coetzer, L. A., Hodgson, A. N., Baecker, A. A. W., Neethling, J. H., and Cross, R. H. M., eds. EMSSA 34: Aventura Resort, Warmbaths, 28 November–1 December 1995. EMSSA Proc. 25 (1995); Coetzer, L. A., Hodgson, A. N., Baecker, A. A. W., Engelbrecht, J. A. A., and Cross, R. H. M., eds. MSSA 35: University of Natal, Durban, 4–6 December 1996. MSSA Proc. 26 (1996); Cloete, E., Engelbrecht, J. A. A., Hodgson, A. N., Pienaar, R. N., and Cross, R. H. M., eds. MSSA 36: University of the Western Cape, Bellville, 3–5 December 1997. MSSA Proc. 27 (1997); Cloete, E., Engelbrecht, J. A. A., Neethling, J. N., Dennison, C., Pienaar, R. N., and Cross, R. H. M., eds. MSSA 37: Rand Afrikaans University, Johannesburg, 2–4 December 1998. MSSA Proc. 28 (1998); McLean, M., Engelbrecht, J. A. A., Neethling, J. H., Dennison, C., Mycock, D., and Cross, R. H. M., eds. MSSA 38: University of the Free State, Bloemfontein, 1–3 December 1999. MSSA Proc. 29 (1999); McLean, M., Lang, C. I., Dennison, C., Mycock, D., and Cross, R. H. M., eds. MSSA 39: Rhodes University, Grahamstown, 6–8 December 2000. MSSA Proc. 30 (2000); McLean, M., Lang, C. I., Dennison, C., Mycock, D., and Cross, R. H. M., eds.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

323

MSSA 40: Oppenheimer Life Sciences Building, East Campus, University of the Witwatersrand, Johannesburg, 5–7 December 2001. MSSA Proc. 31 (2001); Bro¨zel, V. S., Cornish, L. A., Wolfe-Coote, S. A., Sym, S., and Cross, R. H. M., eds. ICEM-15, Durban, 1–6 September 2002. 2. Egypt [No information available]. F. Australia and New Zealand 1. Australia (microscopy.org.au) From 1968 until 1986, biennial conferences on electron microscopy were sponsored by the Australian Academy of Science, with the result that an Australian Society of Electron Microscopy was formed only in 1986, when the Academy discontinued its support. An Electron Microscopy Newsletter has been distributed to members of the society since its creation. For further information about electron microscopy in Australia, see Turner and Vesk (1996) and Goodchild and Dowell (1986); some additional background material may be found in Goodman (1981). The early years of biological electron microscopy in Australia have been studied by the historian of science Rasmussen (1999). The proceedings of MMEM-2000 (Microscopy and Microanalysis of Engineering Materials, held at Monash University, Clayton Campus, Melbourne from 14–16 February 2000) are published in Micron 32 (2001) No. 8, pp. 707–877. 1. Australian Conference on Electron Microscopy, Australian Academy of Science, Canberra City, 19–22 February 1968. 2. Australian Conference on Electron Microscopy, Australian Academy of Science, Canberra City, 16–19 February 1970. 3. Australian Conference on Electron Microscopy, Australian Academy of Science, Canberra City, 14–17 February 1972. ICEM-8, Canberra, 25–31 August 1974. 4. Fourth Australian Conference on Electron Microscopy, University of Sydney, 9–12 February 1976. 5. Fifth Australian Conference on Electron Microscopy, University of Adelaide, 20–24 February 1978.

324

HAWKES

6. Electron Microscopy 1980, Proceedings of the Sixth Australian Conference, Monash University, Clayton, Melbourne, 18–22 February 1980. Micron 11 (1980) Nos. 3–4, 221–521 and Supplement 1, 53 pp. 7. Seventh Australian Conference on Electron Microscopy and Cell Biology Meeting, Australian National University, Canberra, 15–19 February 1982. ‘‘Electron Microscopy and Cell Biology,’’ Proceedings of the Seventh Australian Conference, Micron 13 (1982) 241–394. 8. Eighth Australian Conference on Electron Microscopy, Griffith University Brisbane, 14–17 February 1984. 9. Ninth Australian Conference on Electron Microscopy and Eighth National Symposium of the Microscopical Society of Australia, University of New South Wales, 15–20 February 1986. 10. Tenth Australian Conference on Electron Microscopy, University of Adelaide, North Terrace Adelaide, 21–26 February 1988. 11. Eleventh Australian Conference on Electron Microscopy, University of Melbourne, 12–16 February 1990, Proceedings edited by A. E. C. Spargo. See also J. Microscopy 162 (1991) No. 3, 303–413. 12. Proceedings of the 1992 Joint Conference on Electron Microscopy and Cell Biology, ACEM-12 and ANZSCB-11, Campus of the University of Western Australia, Perth, 10–14 February 1992, edited by B. J. Griffin, A. W. S. Johnson, J. Kuo, and F. J. Lincoln. 13. Abstracts ACEM-13, Microscopy–Bridging the Sciences, 13th Biennial Conference Australian Society for Electron Microscopy Inc. and Materials-enabling Technology, sponsored by Institute of Metals and Materials Australasia Inc., Broadbeach, Gold Coast, QLD 7–11 February 1994 (Allen, D., Barry, J., Bostrom, T., Hogan, L. M., and Pennisi, M. S., eds.; Australian Society of Microscopy, 1994). Micron 25 (1994) No. 6, 499–621 (Mackinnon, I., guest ed.). 14. Microcosmopolitan, papers presented at 14th Biennial Conference on Electron Microscopy, ACEM-14, University of Sydney, 5–9 February 1996. Joint with the First Meeting of the International Union of Microbeam Analysis Societies (IUMAS–1) and the 9th Symposium of the Microscopical Society of Australia (MAS–9). (Cox, G., ed.; Australian Society for Electron Microscopy, Canberra 1996). 15. ACEM-15, 15th Australian Conference on Electron Microscopy (in association with Microscopy New Zealand Inc.), Wrest Point Convention Centre, Sandy Bay, Hobart (Tasmania), 1–6 February 1998. Abstracts printed in Conference Handbook, 130 pp. 16. micrOZcopy 2000, 16th Australian Conference on Electron Microscopy, National Convention Centre, Canberra, 6–11 February 2000. Program and Abstracts Book, 129 pp.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

325

17. ACEM-17, 17th Australian Conference on Electron Microscopy Adelaide Convention Centre, Adelaide 4–8 February 2002 (Terlet, J., Convenor). Abstracts printed in Conference Handbook, 116 pp. 18. ACEM-18. Geelong, 2–6 February 2004. 2. New Zealand A series of National Electron Microscopy Conferences has been held since the 1960s, beginning with an informal meeting in 1966; after the 18th conference, the word electron was dropped from the name of the society and the 1997 meeting is the first Microscopy of New Zealand Conference. A conference handbook containing abstracts is distributed to participants at each meeting but is not otherwise published. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

Lower Hutt (Wellington), 1966. Dunedin, 17–19 May 1967. Christchurch, August 1969. Auckland, 17–19 May 1971. Palmerston North, 23–25 May 1972. Wellington, 14–16 August 1973. Dunedin, 25–27 August 1975. School of Medicine, University of Auckland, 9–12 May 1977. Christchurch, 23–25 May 1978. University of Waikato, Hamilton, 13–16 May 1980. Massey University, Palmerston North, 19–21 August 1981. Central Institute of Technology, Upper Hutt, 17–20 May, 1983. University of Otago, Dunedin, 15–17 May 1985. Medical School, University of Auckland, 11–15 May 1987. Canterbury University, Christchurch, 15–19 May, 1989. Victoria University, Wellington, 20–24 May 1991. Massey University, Palmerston North, 17–21 May 1993. University of Otago, Dunedin, 4–8 September 1995.

Microscopy of New Zealand Conferences 19. Microscopy 97. Faculty of Medicine and Health Sciences, University of Auckland, 10–14 February 1997. The Conference Proceedings contain 26 abstracts in the Life Sciences and 26 in the Physical Sciences. 20. ACEM-15, 15th Australian Conference on Electron Microscopy (in association with Microscopy New Zealand Inc.), Wrest Point Convention Centre, Hobart (Tasmania), 1–6 February 1998. Abstracts printed in Conference Handbook, 130 pp. 21. Microscopy 2003, Victoria University, Wellington, 7–11 February 2003.

326

HAWKES

V. Thematic Meetings A. High-Voltage Electron Microscopy (HVEM) The first pictures obtained with the first of the two high-voltage microscopes constructed in the Laboratoire d’Optique Electronique du CNRS in Toulouse were published in 1960 (Dupouy et al., 1960) and full accounts of those instruments have been prepared by Gaston Dupouy (1968, 1985). With the commercial production of such microscopes by AEI, RCA, Hitachi and JEOL, a series of conferences was held and continued until 1986. The proceedings of these are listed below. Occasional conferences on particular themes in high-voltage microscopy have also been held (e.g. Fujita, 1985; Fujita et al., 1991). An International Symposium on In-situ High Voltage Electron Microscopy, Application to Plasticity and Further Topics of Materials Research was held in Halle/Saale from 2–6 April 1979, see Crystal Res. & Technol. 14 (1979) Nos. 10 and 11. An International Workshop on High-voltage and High-resolution Electron Microscopy was held in Stuttgart, 21–24 February 1994 (Ru¨hle et al., 1994). The Highvoltage Electron Microscopy Laboratory in Kyushu University publishes a collection of Annual Reports, which reached No. 25 in 2001. Each volume contains original material and articles already published. This HVEM Laboratory was founded in 1975 and its 20th anniversary is commemorated at ‘‘The Asian Science Seminar on New Directions in Transmission Electron Microscopy and Nano-characterization of Materials’’ (Fukuoka, 17–26 March 1997). The Centre for Ultra-High Voltage Electron Microscopy in Osaka University began publication of Denken, later Ultra-Denken, in 1972. The High-Voltage Electron Microscope Laboratory in Nagoya University publishes Nagoya Daigaku Denshi Kogaku Kenkyu no Ayumi [Progress in Electron Optics Research at Nagoya University]. Several US–Japan Conferences on High-Voltage Electron Microscopy have been held in Honolulu: 29 October–2 November 1967 (organized by H. Hashimoto and R. E. Ogilvie); 19–26 September 1971 (organized by H. Hashimoto and G. Thomas); and 6–10 December 1976, see Imura and Fisher (1977) and Fisher and Imura (1978). For details of the AMU–ANL Summer Workshop on High-Voltage Electron Microscopy, see Gilroy et al. (1966). Useful surveys of high-voltage electron microscopes have been prepared by Allen (1995) and Allen and Dorignac (1998). Monroeville, 1969: Current Developments in High Voltage Electron Microscopy (First National Conference), Fundamental Laboratories of the United States Steel Corporation, Monroeville (PA), 17–19

ELECTRON OPTICS AND ELECTRON MICROSCOPY

327

June 1969. Proceedings not published but Micron 1 (1969) 220–307 contains official reports of the meeting based on the session chairmen’s notes. Stockholm, 1970: The Proceedings of the Second International Conference on High-Voltage Electron Microscopy, Stockholm, 14–16 April 1971; published as Jernkontorets Annaler 155 (1971) No. 8. Oxford, 1973 : High Voltage Electron Microscopy. Proceedings of the Third International Conference, Oxford, 27–30 August, 1973 (Swann, P. R., Humphreys, C. J., and Goringe, M. J., eds.; Academic Press, London and New York, 1974). Toulouse, 1975 : Microscopie Electronique a` Haute Tension. Textes des Communications Pre´sente´es au 4e` Congre`s International, Toulouse, 1–4 Septembre 1975 (Jouffrey, B. and Favard, P., eds.; SFME Paris, 1976). Kyoto, 1977 : High Voltage Electron Microscopy 1977. Proceedings of the Fifth International Conference on High Voltage Electron Microscopy, Kyoto, 29 August–1 September 1977 (Imura, T. and Hashimoto, H., eds.; Japanese Society of Electron Microscopy, Tokyo, 1977); published as a supplement to Journal of Electron Microscopy 26 (1977). The Hague, 1980 : Electron Microscopy 1980. Proceedings of the Seventh European Congress on Electron Microscopy, The Hague [Brederoo, P. and Boom, G. (Vol. I), Brederoo, P. and Priester, W. de (Vol. II), Brederoo, P. and Cosslett, V. E. (Vol. III), and Brederoo, P. and Landuyt, J. van (Vol. IV), eds.]. Volumes I and II contain the proceedings of the Seventh European Congress on Electron Microscopy, Vol. III those of the Ninth International Conference on X-Ray Optics and Microanalysis, and Vol. IV those of the Sixth International Conference on High Voltage Electron Microscopy (Seventh European Congress on Electron Microscopy Foundation, Leiden, 1980). Berkeley, 1983 : Proceedings of the Seventh International Conference on High Voltage Electron Microscopy, Berkeley, 16–19 August 1983 (Fisher, R. M., Gronsky, R., and Westmacott, K. H., eds.). Published as a Lawrence Berkeley Laboratory Report, LBL-16031, UC-25, CONF-830819. Kyoto, 1986: The Proceedings of the Eighth International Conference on High-voltage Electron Microscopy form part of volume II of the Proceedings of ICEM XI (Kyoto, 1986), pp. 889–1192.

328

HAWKES

B. International Conferences on X-Ray Optics and Microanalysis (ICXOM) 1. X-ray Microscopy and Microradiography. Proceedings of a Symposium held at the Cavendish Laboratory, Cambridge, 16–21 August 1956 (Cosslett, V. E., Engstro¨m, A., and Pattee, H. H., eds.; Academic Press, New York 1957). 2. X-ray Microscopy and X-ray Microanalysis. Proceedings of the Second International Symposium, Stockholm 1960 (Engstro¨m, A., Cosslett, V., and Pattee, H., eds.; Elsevier, Amsterdam, London, New York & Princeton 1960). 3. X-ray Optics and X-ray Microanalysis. Third International Symposium, Stanford University, Stanford CA, 22–24 August 1962 (Pattee, H. H., Cosslett, V. E., and Engstro¨m, A. eds.; Academic Press, New York & London 1963). 4. Optique des Rayons X et Microanalyse/X-ray Optics and Microanalysis, IVe congre`s international sur l’optique des Rayons X et la Microanalyse, Orsay [7–10] septembre 1965 (Castaing, R., Deschamps, P., and Philibert, J., eds.; Hermann, Paris 1966). 5. Vth International Congress on X-ray Optics and Microanalysis/V. Internationaler KongreB fu¨r Ro¨ntgenoptik und Mikroanalyse/Ve Congre`s International sur l’Optique des Rayons X et la Microanalyse, Tu¨bingen, September 9–14, 1968 (Mo¨llenstedt, G. and Gaukler, K. H., eds.; Springer, Berlin and New York 1969). 6. Proceedings of the Sixth International Conference on X-ray Optics and Microanalysis, Osaka, 5–10 September 1971 (Shinoda, G., Kohra, K. and Ichinokawa, T., eds.; University of Tokyo Press, Tokyo 1972). 7. X-ray Optics and Microanalysis. Transactions of the VIIth International Conference on X-ray Optics and Microanalysis, Moscow–Kiev, 9–16 July 1974 (Borovsky, I. B. and Komyak, N. I., eds.; Mashinostroenie, Leningrad 1976). 8. Eighth International Conference on X-Ray Optics and Microanalysis. Papers presented at the Eighth International Conference on X-Ray Optics and Microanalysis, held in Boston, Massachusetts, August 18–24, 1977 and sponsored by the Microbeam Analysis Society (Beaman, D. R., Ogilvie, R. E. and Wittry, D. B., eds.; Pendell, Midland MI 1980). 9. Electron Microscopy 1980, Volume 3, Analysis. Proceedings of the Seventh European Congress on Electron Microscopy including the Ninth International Conference on X-ray Optics and Microanalysis, The Hague, The Netherlands, August 24–29, 1980. (Brederoo, P. and Cosslett, V. E., eds.; Seventh European Congress on Electron Microscopy Foundation, Leiden 1980). 10. ICXOM 10, Toulouse, 5–9 September 1983. 10e Congre`s International d’Optique des Rayons X et de Microanalyse/10th International

ELECTRON OPTICS AND ELECTRON MICROSCOPY

329

Congress on X-ray Optics and Microanalysis. Journal de Physique 45 (1984) Colloque C2, Supplement to No. 2. 11. ICXOM 11, London Canada 4–8 August 1986. 11th International Congress on X-ray optics and Microanalysis/11e`me Congre`s International d’Optique des Rayons X et de Microanalyse (Brown, J. D. and Packwood, R. H., eds.). 12. 12 ICXOM, 12th International Congress on X-ray Optics and Microanalysis, 28 August–1 September 1989, Cracow, organized by Faculty of Structural Analysis, Institute of Mining and Metallurgy, al. Mickiewicza 30, 30–059 Cracow, Poland (Jasienska, S. and Maksymowicz, L. J., eds.; Academy of Mining and Metallurgy Printing House, Cracow) 2 vols. 13. X-ray Optics and Microanalysis 1992. Proceedings of the Thirteenth International Congress, UMIST, Manchester, UK, 31 August–4 September 1992 (Kenway, P. B., Duke, P. J., Lorimer, G. W., Mulvey, T., Drummond, I. W., Love, G., Michette, A. G., and Stedman, M., eds.; Institute of Physics, Bristol and Philadelphia 1993) Conference Series No. 130. 14. ICXOM XIV, Proceedings of the 14th International Congress on Xray Optics and Microanalysis, Guangzhou, 29 August–2 September 1995. J. Trace & Microprobe Tech. 15 (1997) No. 4 (Janssens, K and Liu, L.-K., guest eds.; Marcel Dekker, New York, Basel and Hong Kong). 15. ICXOM XV, Proceedings of the 15th International Congress on Xray Optics and Microanalysis, Antwerp, 24–27 August 1998. J. Anal. Atom. Spectrosc. 14 (1999) No. 3 (Janssens, K., guest ed.). 16. ICXOM XVI, Proceedings of the 16th International Congress on X-ray Optics and Microanalysis, Vienna University of Technology, 2–6 July 2001. Papers submitted either to J. Anal. Atom. Spectrosc. [no special issue] or to Spectrochim. Acta B58 (2003) No. 4 (Mantler, M., Wobrauschek, P., Friedbacher, G., and Schreiner, M., guest eds.). 17. ICXOM XVII, Proceedings of the 17th International Congress on X-ray Optics and Microanalysis, Chamonix, Mont Blanc, 22–26 September 2003.

C. Charged-Particle Optics 1. The Charged-Particle Optics Conferences (CPO) This series of meetings was launched by H. Wollnik, in collaboration with K. L. Brown and P. W. Hawkes, in an attempt to bring together members of the various communities of charged-particle optics (accelerator optics, electron optics and spectrometer optics).

330

HAWKES

Giessen, 1980: Charged Particle Optics. Proceedings of the First Conference on Charged Particle Optics, Giessen, 8–11 September 1980 (Wollnik, H., ed.) Nucl. Instrum. Meth. 187 (1981) 1–314. Albuquerque, 1986: Charged Particle Optics. Proceedings of the Second International Conference on Charged Particle Optics, Albuquerque, 19–23 May 1986 (Schriber, S. O. and Taylor, L. S., eds.) Nucl. Instrum. Meth. Phys. Res. A258 (1987) 289–598. Toulouse, 1990: Charged Particle Optics. Proceedings of the Third International Conference on Charged Particle Optics, Toulouse, 24–27 April 1990 (Hawkes, P. W., ed.) Nucl. Instrum. Meth. Phys. Res. A298 (1990) 1–508. Tsukuba 1994: Charged Particle Optics. Proceedings of the Fourth International Conference on Charged Particle Optics, Tsukuba 3–6 October 1994 (Ura, K., Hibino, M., Komuro, M. Kurashige, M., Kurokawa, S., Matsuo, T., Okayama, S., Shimoyama, H., and Tsuno, K., eds.) Nucl. Instrum. Meth. Phys. Res. A363 (1995) 1–496. Delft 1998: Charged Particle Optics. Proceedings of the Fifth International Conference on Charged Particle Optics, Delft 14–17 April 1998 (Kruit, P. and Amersfoort, P. W. van, eds.) Nucl. Instrum. Meth. Phys. Res. A427 (1999) 1–422. College Park 2002: Charged Particle Optics, Sixth International Conference on Charged Particle Optics, Marriott Hotel, Greenbelt MD, 22–25 October 2002. Nucl. Instrum. Meth. Phys. Res. (2003). 2. Society of Photo-Optical Instrumentation Engineers (SPIE) Since 1993, SPIE has organized conferences on a theme related to chargedparticle optics as part of its major annual meeting. 1. Charged-particle Optics, San Diego CA, 15 July 1993 (Thompson, W. B., Sato, M. and Crewe, A. V., eds.). Proc. SPIE 2014 (1993). 2. Electron-beam Sources and Charged-particle Optics, San Diego, 10–14 July 1995 (Munro, E., and Freund, H. P., eds.). Proc. SPIE 2522 (1995). 3. Charged-particle Optics II, Denver CO, 5 August 1996 (Munro, E., ed.). Proc. SPIE 2858 (1996). 4. Charged-particle Optics III, San Diego CA, 27–28 July 1997 (Munro, E., ed.). Proc. SPIE 3155 (1997). [There was no SPIE meeting on charged-particle optics in 1998 since an international CPO conference was held in Delft in that year, see Section V.C.1.]

ELECTRON OPTICS AND ELECTRON MICROSCOPY

331

5. Charged-particle Optics IV, Denver CO, 22–23 July 1999 (Munro, E., ed.). Proc. SPIE 3777 (1999). 6. ‘‘Charged Particle Beam Optics Imaging,’’ San Diego CA, 30 July 2001. In Charged Particle Detection, Diagnostics and Imaging (Delage, O., Munro, E., and Rouse, J. A., eds.) Proc. SPIE 4510 (2001) 71–236. [There was no SPIE meeting on charged-particle optics in 2002 since an international CPO conference was held in Maryland in that year, see Section V.C.1.]

D. Scanning Two series of meetings concerned explicitly with scanning electron microscopy are of direct relevance: the ‘‘Scanning Electron Microscopy’’ meetings that began at the Illinois Institute of Technology (IIT) Research Institute in Chicago in 1968, only three years after the Cambridge Instrument Company put a scanning electron microscopy on the market for the first time (Stewart and Snelling, 1964; Stewart, 1985; Oatley et al., 1965, 1985; Jervis, 1971/2; Langrish et al., 1972; Wells, 1974; Oatley, 1982; McMullan, 1985, 1988, 1990, 1993; Brown et al., 1996; Everhart, 1996); and the ‘‘Scanning’’ meetings that have been held annually since 1989, the abstracts of which are published in Scanning. The annual ‘‘Pfefferkorn Conferences,’’ which have been loosely associated with the Scanning Electron Microscopy meetings since 1982, are also listed in this section. 1. Scanning Electron Microscopy and the Pfefferkorn Conferences A series of annual meetings on scanning electron microscopy was launched in 1968. Proceedings of the first ten meetings (1968–1977) were issued as bound volumes by the IIT Research Institute in Chicago under the general direction of O. Johari. From 1978 to 1986, bound volumes were published by Scanning Electron Microscopy Inc, edited by O. Johari. These volumes are all entitled Scanning Electron Microscopy and are identified by date only, they do not carry volume numbers. In 1987, publication of these bound volumes was discontinued, to be replaced by a quarterly serial, Scanning Microscopy, published by Scanning Microscopy International and again with O. Johari as editor-in-chief; from volume 5 (1991) onwards, the editor-in-chief has been G. M. Roomans, O. Johari remaining managing editor. The last printed volume to appear was vol. 10 (1996).

332

HAWKES

Pfefferkorn Conferences 1. Electron Beam Interactions with Solids for Microscopy, Microanalysis and Microlithography. Proceedings of the 1st Pfefferkorn Conference, Asilomar Conference Center, Monterey CA 18–23 May 1982 (Kyser, D. F., Newbury, D. E., and Shimizu, R., eds.; Scanning Electron Microscopy, AMF O’Hare 1984). 2. The Science of Biological Specimen Preparation for Microscopy and Microanalysis. Proceedings of the 2nd Pfefferkorn Conference, Sugar Loaf Mountain Resort, Traverse City MI, 23–28 April 1983 (Revel, J.-P., Barnard, T., and Haggis, G. H., eds.; Scanning Electron Microscopy, AMF O’Hare 1984). 3. Electron Optical Systems for Microscopy, Microanalysis and Microlithography. Proceedings of the 3rd Pfefferkorn Conference, Ocean City MD, 9–14 April 1984 (Hren, J. J., Lenz, F. A., Munro, E., and Sewell, P. B., eds.; Scanning Electron Microscopy, AMF O’Hare 1984). 4. The Science of Biological Specimen Preparation for Microscopy and Microanalysis. Proceedings of the 4th Pfefferkorn Conference, Grand Canyon Squire Inn, Grand Canyon AZ, 25–30 March 1985 (Mu¨ller, M., Becker, R. P., Boyde, A., and Wolosewick, J. L., eds.; Scanning Electron Microscopy, AMF O’Hare 1986). 5. Physical Aspects of Microscopic Characterization of Materials. Proceedings of the 5th Pfefferkorn Conference, Brueggen (GFR) 2–7 October 1986 (Kirschner, J., Murata, K., and Venables, J. A., eds.) Scanning Microscopy, Supplement 1 (1987). 6. Image and Signal Processing in Electron Microscopy. Proceedings of the 6th Pfefferkorn Conference, Niagara Falls, 28 April–2 May 1987 (Hawkes, P. W., Saxton, W. O., Ottensmeyer, F. P., and A. Rosenfeld, A., eds.). Scanning Microscopy, Supplement 2 (1988). 7. The Science of Biological Specimen Preparation for Microscopy and Microanalysis. Proceedings of the 7th Pfefferkorn Conference, University of Guildford, 11–16 September 1988 (Albrecht, R. M. and Ornberg, R. L., eds.). Scanning Microscopy, Supplement 3 (1989). 8. Fundamental Electron and Ion Beam Interactions with Solids for Microscopy, Microanalysis and Microlithography. Proceedings of the 8th Pfefferkorn Conference, Park City UT, 7–12 May 1989 (Schou, J., Kruit, P., and Newbury, D. E., eds.). Scanning Microscopy, Supplement 4 (1990). 9. The Science of Biological Specimen Preparation for Microscopy and Microanalysis. Proceedings of the 9th Pfefferkorn Conference, Santa Cruz CA, 6–10 August 1990 (Edelmann, L. and Roomans, G. M., eds.). Scanning Microscopy, Supplement 5, bound into 5 (1991) No. 4, pp. si–siv and s1–s118.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

333

10. Signal and Image Processing in Microscopy and Microanalysis. Proceedings of the 10th Pfefferkorn Conference, Cambridge UK, 16–19 September 1991 (Hawkes, P. W., ed.). Scanning Microscopy, Supplement 6 (1992). 11. Physics of Generation and Detection of Signals used for Microcharacterization. Proceedings of the 11th Pfefferkorn Conference, University of Massachusetts, Amherst, 9–14 August 1992 (Re´mond, G., Gijbels, R. H. H., Levenson, L. L., and Shimizu, R., eds.). Scanning Microscopy, Supplement 7 (1993). 12. The Science of Biological Microanalysis. Proceedings of the 12th Pfefferkorn Conference, University of Cambridge, 27–30 September 1993 (Roomans, G. M., Gupta, B. L., Leapman, R. D., and Zglinicki, T. von, eds.). Scanning Microscopy, Supplement 8 (1994). 13. Luminescence. Proceedings of the 13th Pfefferkorn Conference, AmeriCana Resort, Niagara Falls, 1318 May 1994 (Re´mond, G., Balk, L., and Marshall, D. J., eds.). Scanning Microscopy, Supplement 9 (1995). 14. The Science of Biological Specimen Preparation for Microscopy. Proceedings of the 14th Pfefferkorn Conference, Shrine of Our Lady of the Snows, Belleville IL, 6–11 August 1995 (Malecki, M. and Roomans, G. M., eds.). Scanning Microscopy, Supplement 10 (1996). 15. Electron Image and Signal Processing. Proceedings of the 15th Pfefferkorn Conference, Silver Bay NY, 18–22 May 1996 (Hawkes, P. W., Frank, J. Saxton, W. O., and Roomans, G. M., eds.). Scanning Microscopy 11 (1997). Issued only as a CD-ROM and placed on the website www.eurocellmat.org.uk. 16. Optimization of the Scanning Electron Microscope. Proceedings of the 16th Pfefferkorn Conference, University of Wales, Aberystwyth, 6–8 April 1998 (ap Gwynn, I. and Roomans, G. M., eds.). Scanning Microscopy 13 (1999) No. 1. Issued only as a CD-ROM and placed on the website www.eurocellmat.org.uk. 2. SCANNING (www.scanning-fams.org) Scanning 89. Hotel Queen Mary, Long Beach CA, 5–7 April 1989. No formal proceedings, full programme in Centre Section of Scanning 11 (1989) No. 2. Scanning 90. Arlington VA, 18–20 April 1990. Proceedings published as bound-in Supplement to Scanning 12 (1990) No. 5, pp. I-1–I-72. Scanning 91. Atlantic City NJ, 10–12 April 1991. Scanning 13 (1991) Supplement I (Becker, R. P., ed.). Scanning 92. Atlantic City NJ, 1–3 April (1992). Scanning 14 (1992) Supplement II.

334

HAWKES

Scanning 93. Orlando FL, 21–23 April (1993) Scanning 15 (1993) Supplement III. Scanning 94. Charleston SC, 17–20 May 1994). Scanning 16 (1994) Supplement IV. Scanning 95. Monterey CA, 29–31 March 1995. Scanning 17 (1995) Supplement V. Scanning 96. Monterey CA, 10–12 April 1996. Scanning 18 (1996) Supplement VI. Scanning 97. Monterey CA, 19–22 April 1997. Scanning 19 (1997) No. 3 (Becker, R. P., guest ed.). Scanning 98. Baltimore MD, 9–12 May 1998. Scanning 20 (1998) No. 3 (Becker, R. P., guest ed.) and No. 5. Scanning 99. Hyatt Regency O’Hare, Chicago IL, 11–14 April 1999. Scanning 21 (1999) No. 2 (Becker. R. P., guest ed.). Scanning 2000. Sheraton Four Points, Riverwalk North Hotel, San Antonio TX, 9–12 May 2000. Scanning 22 (2000) No. 2 (Becker. R. P., guest ed.). Scanning 2001. Roosevelt Hotel, New York NY, 5–7 May 2001. Scanning 23 (2001) No. 2 (Becker, R. P., guest ed.). Scanning 2002. No meeting. Scanning 2003. Double Tree Hotel, San Diego Mission Valley, CA, 3–5 May 2003. Scanning 2004. Washington DC, 1–3 May 2004. Scanning 2005. Monterey CA, 7–9 May 2005. Scanning 2006. New Orleans LA, 6–8 May 2006.

E. Electron, Ion, and Photon Beam Conferences Until 1983, two separate series of meetings were held: the ‘‘International Conferences on Electron and Ion Beam Science and Technology,’’ sponsored by the Electrochemical Society, and the ‘‘Symposia on Electron, Ion and Photon Beam Technology,’’ sponsored by the IEEE and the American Vacuum Society. These biennial meeting dovetailed until 1983, when they were united into a single series, sponsored by all three organizations and published in Part B of J. Vac. Sci. Technol. Although not formally one of the above conferences, the Fall Meeting of the Electrochemical Society, held in Washington DC, 11–15 October 1964, is of direct interest; the booklet containing the ‘‘Extended Abstracts of Electrothermic and Metallurgy Division, volume 2, Number 2’’ contains a long section on Electron Probe Technique, with papers on ‘‘Analysis of light elements,’’ ‘‘Quantitative analysis,’’ ‘‘New techniques and instrumentation,’’ and ‘‘Applications.’’

ELECTRON OPTICS AND ELECTRON MICROSCOPY

335

For an account of early work on electron-beam fabrication, see Broers (1985). Electrochemical society conferences 1964. Proceedings First International Conference on Electron and Ion Beam Science and Technology, Toronto, 28 April–2 May 1964 (Bakish, R., ed.; Wiley, New York & London 1965). 1966. Proceedings Second International Conference on Electron and Ion Beam Science and Technology, New York, April 1966 (Bakish, R., ed.; American Institute of Mining, Metallurgical and Petroleum Engineers, New York 1966). 1968. Third International Conference on Electron and Ion Beam Science and Technology, Boston MA (Bakish, R., ed.). Proceedings of the Electrochemical Society, PV 68-1 (Princeton NJ, 1968). 1970. Fourth International Conference on Electron and Ion Beam Science and Technology, Los Angeles CA (Bakish, R., ed.). Proceedings of the Electrochemical Society, PV 70-3 (Princeton NJ, 1970). 1972. Fifth International Conference on Electron and Ion Beam Science and Technology, Houston TX, May 1972 (Bakish, R., ed.). Proceedings of the Electrochemical Society, PV 72-3 (Princeton NJ, 1972). 1974. Sixth International Conference on Electron and Ion Beam Science and Technology, San Francisco CA, 12–17 May 1974 (Bakish, R., Gonzales, A. J., Amboss, K., Broers, A. N., and Smith, H. I., eds.). Proceedings of the Electrochemical Society, PV 74-1 (Princeton NJ, 1974). 1976. Seventh International Conference on Electron and Ion Beam Science and Technology, Washington DC (Bakish, R., ed.). Proceedings of the Electrochemical Society, PV 76-1 (Princeton NJ, 1976). 1978. Eighth International Conference on Electron and Ion Beam Science and Technology, Seattle (Bakish, R., ed.). Proceedings of the Electrochemical Society, PV 78-5 (Princeton NJ, 1978). 1980 Electron and Ion Beam Science and Technology, St. Louis MO (Bakish, R., ed.). Proceedings of the Electrochemical Society, PV 80-6 (Princeton NJ, 1980). 1982. Electron and Ion Beam Science and Technology (10th), Cleveland OH, May 1982 (Bakish, R., ed.). Proceedings of the Electrochemical Society, PV 83-2 (Princeton NJ, 1983). Three-beams conferences 1. 1959. Proceedings of the First Symposium on Electron Beam Melting, Hotel Somerset, Boston MA, 20 March, 1959 (Hetherington, J. S., ed.;

336

HAWKES

Alloyd Research Corporation, Waterton MA 1959). Note: these Proceedings include an account by A. Lawley of a Symposium on Electron Bombardment Floating-zone Melting and Allied Electron Bombardment Techniques, which had been held shortly before this meeting at the SERL Laboratories, Baldock (Herts), pp. 1–17. 2. 1960. Proceedings of the Second Symposium on Electron Beam Processes, Hotel Sheraton Plaza, Boston MA, 24–25 March 1960 (Bakish, R., ed.; Alloyd Corporation, Cambridge MA 1960). 3. 1961. Proceedings of the Third Symposium on Electron Beam Technology, Boston 23–24 March 1961 (R. Bakish, ed.; Alloyd Electronics Corporation, Cambridge MA 1961). 4. 1962. Proceedings of the Fourth Symposium on Electron Beam Technology, Boston MA, 29–30 March 1962 (Bakish, R., ed.; Alloyd Electronics Corporation, Cambridge MA 1962). 5. 1963. Proceedings of the Fifth Symposium on Electron Beam Technology, Boston MA, 28–29 March 1963 (Morley, J., ed.; Alloyd Electronics Corporation, Cambridge MA 1963). 6. 1964. Proceedings of the Sixth Symposium on Electron Beam Technology, Boston MA, 27–28 April 1964 (Morley, J., ed.; Alloyd Electronics Corporation, Cambridge MA 1964). 7. 1965 Proceedings of the Electron and Laser Beam Symposium, Pennsylvania State University, University Park PA, 31 March–2 April 1965 (El-Kareh, A. B., ed.; Pennsylvania State University, University Park and Alloyd General Corporation, Medford MA 1965). 8. Proceedings of the 8th Annual Electron and Laser Beam Symposium, University of Michigan, Ann Arbor MI, 6–8 April 1966 (Haddad, G. I., ed.; University of Michigan, Ann Arbor MI and IEEE, 1966). 9. Record of the IEEE Ninth Annual Symposium of Electron, Ion and Laser Beam Technology, Claremont Hotel, Berkeley CA, 9–11 April 1967 (Pease, R. F. W., ed.; San Francisco Press, San Francisco 1967). 10. Record of the 10th Symposium on Electron, Ion and Laser Beam Technology, Gaithersburg MD, 21–23 May 1969 (Marton, L., ed.; San Francisco Press, San Francisco 1969). 11. Record of the 11th Symposium on Electron, Ion and Laser Beam Technology, Boulder CO, 12–14 May 1971 (Thornley, R. F. M., ed.; San Francisco Press, San Francisco 1971). 12. Proceedings of the 12th Symposium on Electron, Ion and Laser Beam Technology, Massachusetts Institute of Technology, Cambridge MA, 21–23 May 1973 (Wolf, E. D. and Broers, A. N., eds.). J. Vacuum Sci. Technol. 10 (1973) No. 6, 909–1167. 13. Proceedings of the 13th Symposium on Electron, Ion and Photon Beam Technology, Antlers Plaza Hotel, Colorado Springs CO, 21–23 May

ELECTRON OPTICS AND ELECTRON MICROSCOPY

337

1975 (Pease, R. F. W. and Skinner, J. G., eds.). J. Vacuum Sci. Technol. 12 (1975) No. 6, 1109–1387. 14. Proceedings of the 14th Symposium on Electron, Ion and Photon Beam Technology, Rickey’s Hyatt House, Palo Alto CA, 25–27 May 1977 (Varnell, G. L. and Bartelt, J. L., eds.). J. Vacuum Sci. Technol. 15 (1978) No. 3, 835–1000. 15. Proceedings of the 15th Symposium on Electron, Ion and Photon Beam Technology, Boston MA, 29 May–1 June 1979 (Chang, T. H. P. and Hatzakis, M., eds.). J. Vacuum Sci. Technol. 16 (1979) No. 6, 1597–2036. 16. Proceedings of the 16th Symposium on Electron, Ion and Photon Beam Technology, Loews Anatole Hotel, Dallas TX, 26–29 May 1981 (Herriott, D. R. and Bruning, J. H., eds.). J. Vacuum Sci. Technol. 19 (1981) No. 4, 813–1430. [Note: the next numbered conference is the 29th, 1986] 27. Proceedings of the 1983 International Symposium on Electron Ion and Photon Beams, Westin Bonaventure Hotel, Los Angeles CA, 31 May–3 June 1983 (Hatzakis, M. and Chang, T. H. P., eds.). J. Vacuum Sci. Technol. B1 (1983) No. 4, 947–1400. 28. Proceedings of the 1984 International Symposium on Electron Ion and Photon Beams, Westchester Marriott Hotel, Tarrytown NY, 29 May–1 June 1984 (Kelly, J., ed.). J. Vacuum Sci. Technol. B3 (1985) No. 1, 29–462. 29. Proceedings of the 29th International Symposium on Electron Ion and Photon Beams, Thunderbird Motor Inn, Jantzen Beach, Portland OR, 28–31 May 1985 (Bruning, J. H., ed.). J. Vacuum Sci. Technol. B4 (1986) No. 1, 49–436. 30. Proceedings of the 30th International Symposium on Electron Ion and Photon Beams, Westin Hotel, Boston MA, 27–30 May 1986 (Neureuther, A. R., ed.). J. Vacuum Sci. Technol. B5 (1987) No. 1, 37–456. 31. Proceedings of the 31st International Symposium on Electron Ion and Photon Beams, Marriott Warner Center, Woodlands Hills CA, 26–29 May 1987 (Howard, R. E., ed.). J. Vacuum Sci. Technol. B6 (1988) No. 1, 97–504. 32. Proceedings of the 32nd International Symposium on Electron Ion and Photon Beams, Marriott Harbor Beach, Fort Lauderdale FL, 31 May–3 June 1988 (Hu, E. L., ed.). J. Vacuum Sci. Technol. B6 (1988) No. 6, 1797–2314. 33. Proceedings of the 33rd International Symposium on Electron Ion and Photon Beams, Monterey Sheraton, Monterey CA, 30 May–2 June 1989 (Wilson, A. D., ed.). J. Vacuum Sci. Technol. B7 (1989) No. 6, 1373–2068.

338

HAWKES

34. Proceedings of the 34th International Symposium on Electron Ion and Photon Beams, Hyatt Regency Hotel, San Antonio TX, 29 May–1 June 1990 (Orloff, J., ed.). J. Vacuum Sci. Technol. B8 (1990) No. 6, 1309–2059. 35. Proceedings of the 35th International Symposium on Electron Ion and Photon Beams, Seattle WA, 28–31 May 1991 (Shaw, J., ed.). J. Vacuum Sci. Technol. B9 (1991) No. 6, 2815–3619. 36. Proceedings of the 36th International Symposium on Electron Ion and Photon Beams, Peabody Hotel, Orlando FL, 26–29 May 1992 (Kubena, R., ed.). J. Vacuum Sci. Technol. B10 (1992) No. 6, 2497–3259. 37. Proceedings of the 37th International Symposium on Electron Ion and Photon Beams, Sheraton Harbor Island Hotel, San Diego CA, 1–4 June 1993 (Cerrina, F., ed.). J. Vacuum Sci. Technol. B11 (1993) No. 6, 2139–3015. 38. Proceedings of the 38th International Symposium on Electron Ion and Photon Beams, Sheraton New Orleans Hotel, New Orleans LA, 31 May–3 June 1994 (Adesida, A., ed.). J. Vacuum Sci. Technol. B12 (1994) No. 6, 3223–4057. 39. Proceedings of the 39th International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, Scottsdale Conference Resort, Scottsdale AZ, 30 May–2 June 1995 (Kern, D., ed.). J. Vacuum Sci. Technol. B13 (1995) No. 6, 2307–3123. 40. Proceedings of the 40th International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, Atlanta Marriott Marquis, Atlanta GA, 28–31 May 1996 (Pang, S. W., ed.). J. Vacuum Sci. Technol. B14 (1996) No. 6, 3607–4370. 41. Proceedings of the 41st International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, Marriott’s Laguna Cliffs Resort, Dana Point CA, 27–30 May 1997 (Owen, G., ed.). J. Vacuum Sci. Technol. B15 (1997) No. 6, 2079–2952. 42. Proceedings of the 42nd International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, The Westin Hotel, Chicago IL, 26–29 May 1998 (Marrian, C., ed.). J. Vacuum Sci. Technol. B16 (1998) No. 6, 3115–3956. 43. Proceedings of the 43rd International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, Marriott Resort, Marco Island FL, 1–4 June 1999 (Wolfe, F., ed.). J. Vacuum Sci. Technol. B17 (1999) No. 6, 2679–3455. 44. Proceedings of the 44th International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, Rancho Las Palmas, Marriott Resort, Rancho Mirage CA, 30 May–2 June 2000 (Melngailis, J., ed.). J. Vacuum Sci. Technol. B18 (2000) No. 6, 2865–3614.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

339

45. Proceedings of the 45th International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, J. W. Marriott Hotel, Washington DC, 29 May–1 June 2001 (McCord, M., ed.). J. Vacuum Sci. Technol. B19 (2001) No. 6, 2307–2937. 46. Proceedings of the 46th International Conference on Electron Ion and Photon Beam Technology and Nanofabrication, Hilton Anaheim, Anaheim CA, 28–31 May 2002 (McCord, M., ed.). J. Vacuum Sci. Technol. B20 (2002) No. 6. 47. Proceedings of the 47th International Conference on Electron Ion and Photon Beam Technology and Microfabrication, Marriot Waterside Hotel, Tampa, FL, 27–30 May 2003. J. Vac. Sci. Technol. B21 (2003) No. 6.

F. Microcircuit Engineering, later Micro- and Nanoengineering 1. Conference on Micro Electron Beam Technology for Fabrication, Recording and Dynamic Inspection, University Engineering Department, Cambridge, 18–20 March 1975. Organized by the Electron Microscopy and Analysis Group of the Institute of Physics in association with the Institution of Electrical Engineers. 2. First Workshop on Very Large Scale Integration—generation and design of microstructures, Rheinisch-Westfa¨lische Technische Hochschule, Aachen, 9–11 November 1976. Organized by the Institute of Semiconductor Electronics, sponsored by the German Ministry of Research and Technology in cooperation with the Technical University of Aachen. Chairman: H. Beneking. 3. International Conference on Microlithography, Colloque International sur le Microlithographie, Tour Olivier-de-Serres, Paris XV, 21–24 June 1977. 4. Microcircuit Engineering 1978. Proceedings of the International Conference on Microlithography, Cambridge, 11–13 April 1978 (Ahmed, H. and Nixon, W. C., eds.; Cambridge University Press, Cambridge 1980). Organized by the Electron Microscopy and Analysis Group of the Institute of Physics, co-sponsored by the Institution of Electrical Engineers. 5. Microcircuit Engineering ‘79–microstructure fabrication, RheinischWestfa¨lische Technische Hochschule, Aachen, 25–27 September 1979. Organized by the Institute of Semiconductor Electronics in cooperation with the German Section of the IEEE. Chairman: H. Beneking. Proceedings issued by the Institute of Semiconductor Electronics, Aachen, 40 contributions, 362 pp.

340

HAWKES

6. Microcircuit Engineering 1980. Proceedings of the International Conference on Microlithography, Amsterdam, 30 September and 1–2 October 1980 (Kramer, R. P., ed.; Delft University Press, Delft 1981). 7. Microcircuit Engineering 81. Ecole Fe´de´rale Suisse de Technologie, Lausanne, 28–30 September 1981. Proceedings (Oosenbrug, A., ed.) issued by the Swiss Federal Institute of Technology, Lausanne. 8. Microcircuit Engineering 82. Grenoble, 5–8 October 1982. Proceedings issued by Comite´ du Colloque International sur la Microlithographie, Grenoble. 9. Microcircuit Engineering 83. Cambridge, 25–29 September 1983 (Ahmed, H., Cleaver, J. R. A., and Jones, G. A. C., eds.; Academic Press, London and New York 1983). 10. Microcircuit Engineering 1984. Hotel Schweizerhof, Berlin, 25–29 September 1984 (Heuberger, A. and Beneking, H., eds.; Academic Press, London and Orlando 1985). 11. Microcircuit Engineering 85. Proceedings of the International Conference on Microlithography, Rotterdam, 23–25 September 1985 (Mast, K. van der and Radelaar, S., eds.). Microelectronic Engineering 3 (1985) 1–654. 12. Microcircuit Engineering 86. Proceedings of the International Conference on Microlithography, Interlaken, 23–25 September 1986 (Lehmann, H. W. and Bleiker, C., eds.). Microelectronic Engineering 5 (1986) 1–605. 13. Microcircuit Engineering 87. Proceedings of the International Conference on Microlithography, Jouy-en-Josas, 22–25 September 1987 (Castagne´, R. and Perrocheau, J., eds.). Microelectronic Engineering 6 (1987) 1–708. 14. Microcircuit Engineering 88. Proceedings of the International Conference on Microlithography, Vienna, 20–22 September 1988 (Paschke, F., Fallmann, W., and Lo¨schner, H., eds.). Microelectronic Engineering 9 (1989) 1–653. 15. Microcircuit Engineering 89. Proceedings of the International Conference on Microlithography, Cambridge, 26–28 September 1989 (Ahmed, H., Cleaver, J. R. A.,Jones, G. A. C., McMahon, R. A. M. and Broers, A. N., eds.). Microelectronic Engineering 11 (1990) 1–708. 16. Microcircuit Engineering 1990. Proceedings of the International Conference on Microlithography, Leuven, 18–20 September 1990 (Declerck, G., Hove, L. van den, and Coopmans, F., eds.). Microelectronic Engineering 13 (1991) 1–573. 17. Microcircuit Engineering 1991. Proceedings of the International Conference on Microlithography, Rome, 17–19 September 1991 (Tucciarone, A., Paoletti, A., and Paroli, P., eds.). Microelectronic Engineering 17 (1992) 1–581.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

341

18. Microcircuit Engineering 1992. Proceedings of the International Conference on Microfabrication, Erlangen, 21–24 September 1992 (Ryssel, H. and Stephani, D., eds.). Microelectronic Engineering 21 (1993) Nos. 1– 4, 1–500. 19. Microcircuit Engineering 1993. Proceedings of the International Conference on Microfabrication, Maastricht, 26–29 September 1993 (Radelaar, S., Romijn, J., and Drift, E. van der, eds.). Microelectronic Engineering 23 (1994) Nos. 1– 4, 1–495. 20. Micro- and Nano-Engineering 94. Proceedings of the International Conference on Micro- and Nanofabrication, Davos, 26–29 September 1994 (Lehmann, H. W., Stauffer, U., and Vettiger, P., eds.). Microelectronic Engineering 27 (1995) Nos. 1– 4, 1–564. 21. Micro- and Nano-Engineering 95. Proceedings of the International Conference on Micro- and Nanofabrication, Aix-en-Provence, 25–29 September 1995 (Perrocheau, J., ed.). Microelectronic Engineering 30 (1996) Nos. 1– 4, 1–625. 22. Micro- and Nano-Engineering 96. Proceedings of the International Conference on Micro- and Nanofabrication, Glasgow, 22–25 September 1996 (Beaumont, S. P. and Wilkinson, C. D. W., eds.). Microelectronic Engineering 35 (1997) Nos. 1– 4, 1–580. 23. Micro- and Nano-Engineering 97. Proceedings of the International Conference on Micro- and Nanofabrication, Athens, 15–18 September 1997 (Hatzakis, M. and Gogolides, E., eds.). Microelectronic Engineering 41/42 (1998) Nos. 1–4, 1–648. 24. Micro- and Nano-Engineering 98. Proceedings of the International Conference on Micro- and Nanofabrication, Leuven, 22–24 September 1998 (Hove, L. van den, Rossum, M. van, and Ronse, K., eds.). Microelectronic Engineering 46 (1999) Nos. 1–4, 1–513. 25. Micro- and Nano-Engineering 99. Proceedings of the International Conference on Micro- and Nanofabrication, Rome, 21–23 September 1999 (Gentile, M., di Fabrizio, E., and Meneghini, G., eds.). Microelectronic Engineering 53 (2000) Nos. 1–4, 1–712. 26. Micro- and Nano-Engineering 2000. Proceedings of the International Conference on Micro- and Nanofabrication, Jena, 18–21 September 2000 (Fortagne, O., Kern, D. P., and Behringer, U., eds.). Microelectronic Engineering 57/58 (2001). 27. Micro- and Nano-Engineering 2001. Proceedings of the International Conference on Micro- and Nanofabrication, Grenoble, 16–19 September 2001 (Joubert, O., Perrocheau, J., and Tedesco, S., eds.). Microelectronic Engineering 61/62 (2002).

342

HAWKES

28. Micro- and Nano-Engineering 2002. Proceedings of the International Conference on Micro- and Nanofabrication, Lugano, 16–19 September 2002. Microelectronic Engineering (2003). 29. Micro- and Nano-Engineering 2003. Proceedings of the International Conference on Micro- and Nanofabrication, Cavendish laboratory, Cambridge, 22–25 September 2003. Microelectronic Engineering (2004).

G. European Conferences on Electron & Optical Testing of Integrated Circuits (later, of Electronic Devices) and European Symposia on Reliability of Electron Devices, Failure Physics and Analysis (ESREF) Proceedings of the First European Conference on Electron and Optical Beam Testing of Integrated Circuits, Grenoble, 9–11 December 1987 (Wolfgang, E. and Courtois, B., eds.). Microelectronic Engineering 7 (1987) Nos. 2–4, 113–444. Proceedings of the Second European Conference on Electron and Optical Beam Testing of Integrated Circuits, Universita¨t Duisburg, 1–4 October 1989 (Kubalek, E. and Wolfgang, E., eds.). Microelectronic Engineering 12 (1990) Nos. 1–4. Proceedings of the First European Symposium on Reliability of Electron Devices, Failure Physics and Analysis, Bari, 2–5 October 1990 (Pollino, E., ed.). Quality & Reliability Engineering International 7 (1991) No. 4. Proceedings of the Third European Conference on Electron and Optical Beam Testing of Integrated Circuits, Como, 8–11 September 1991 (Melgara, M., Wolfgang, E., Courtois, B., and Fantini, F., eds.). Microelectronic Engineering 16 (1992) Nos. 1–4. Proceedings of the Second European Symposium on Reliability of Electron Devices, Failure Physics and Analysis, 1991, Bordeaux, 7–10 October 1991 (Touboul, A., ed.). Quality & Reliability Engineering International 8 (1992) No. 3. Proceedings of the Third European Symposium on Reliability of Electron Devices, Failure Physics and Analysis, Schwabisch–Gemund, 5–7 October 1992 (Berger, H. H. and Gerling, W. H., eds.). Quality & Reliability Engineering International 9 (1993) No. 4. Proceedings of the Fourth European Conference on Electron and Optical Beam Testing of Electronic Devices, Swiss Federal Institute of Technology, Zu¨rich, 1–3 September 1993 (Birolini, A., Ciappa, M. and Wolfgang, E., eds.). Microelectronic Engineering 24 (1994) Nos. 1–4, 1–441. [Note: wrongly stated to be the third conference on the cover of the journal.]

ELECTRON OPTICS AND ELECTRON MICROSCOPY

343

Proceedings of the Fourth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis, Bordeaux, 4–7 October 1993 (Labat, N. and Touboul, A., eds.). Quality & Reliability Engineering International 10 (1994) No. 4. Proceedings of the Fifth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis, Glasgow, 4–7 October 1994 (Brydon, G. M., ed.). Quality & Reliability Engineering International 11 (1995) No. 4. Proceedings of the Fifth European Conference on Electron and Optical Beam Testing of Electronic Devices, Wuppertal, 27–30 August 1995 (Wolfgang, E., Courtois, B., and Balk, L. J., eds.). Microelectronic Engineering 31 (1996) Nos. 1–4. Proceedings of the Sixth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis, Bordeaux–Arcachon, 3–6 October 1995 (Labat, N. and Touboul, A., eds.). Quality & Reliability Engineering International 12 (1996) No. 4. Proceedings of the Seventh European Symposium on Reliability of Electron Devices, Failure Physics and Analysis, Entschede, 7–11 October 1996 (Verwey, J. F., ed.). Microelectronics & Reliability 36 (1996) No. 11–12. From now on, the EOBT conferences are held in conjunction with the European Symposia on Reliability of Electron Devices, Failure Physics and Analysis (ESREF). Eighth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis and Sixth European conference on Electron and Optical Beam Testing of Electronic Devices, Arcachon, 7–10 October 1997 (Labat, N. and Touboul, A., eds.). Microelectronics & Reliability 37 (1997) No. 10–11. Ninth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis and Seventh European Conference on Electron and Optical Beam Testing of Electronic Devices, Scandic Hotel, Copenhagen, 5–9 October 1998 (Jensen, F. and Kjrgaard, eds.). Microelectronics & Reliability 38 (1998) No. 6–8. Tenth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis and Eighth European Conference on Electron and Optical Beam Testing of Electronic Devices, Arcachon, 5–8 October 1999 (Labat, N. and Touboul, A., eds.). Microelectronics Reliability 39 (1999) No. 6–7. Eleventh European Symposium on Reliability of Electron Devices, Failure Physics and Analysis and Ninth European Conference on electron and Optical Beam Testing of Electronic Devices, Dresden, 2–6 October 2000 (Balk, L. J., Wolfgang, E., and Gerling, W. H., eds.). Microelectronics Reliability 40 (2000) Nos. 8–10.

344

HAWKES

Twelfth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis and Tenth European Conference on Electron and Optical Beam Testing of Electronic Devices, Arcachon, 1–5 October 2001 (Touboul, A. and Labat, N., eds.). Microelectronics Reliability 41 (2001) Nos. 9/10. Thirteenth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis and Eleventh European Conference on Electron and Optical Beam Testing of Electronic Devices, Bellaria (near Rimini), 7–11 October 2002 (Fantini, F. and Vanzi, M., eds.). Microelectronics Reliability 42 (2002) Nos. 9–11. Fourteenth European Symposium on Reliability of Electron Devices, Failure Physics and Analysis and Twelfth European Conference on Electron and Optical Beam Testing of Electronic Devices, Palatium, Arcachon, 6–10 October 2003. Microelectronics Reliability 43 (2003).

H. MicroProcess, later Microprocesses and Nanotechnology 1. Digest of Papers, 1988, 1st MicroProcess Conference, Tokyo, 4–6 July 1988 (Japan Society of Applied Physics, Business Center for Academic Societies Japan, Tokyo 1988). 2. Kobe, 2–5 July 1989 (Gamo, K, Atoda, N., Koinuma, H., Ito, T., Shibata, T. and Yamaoka, T., eds.). Japan. J. Appl. Phys. 28 (1989) Nos. 10 and 11, 2049–2196 and 2329–2404. 3. Makuhari, Chiba, 16–19 July 1970 (Gamo, K., Harada, K., Horiike, Y., Ito, T., Okada, K., Okazaki, S. and Yamaoka, T., eds.). Japan. J. Appl. Phys. 29 (1990) Nos. 10 and 11, 2193–2325 and 2551–2683. Also MicroProcess 90 (Namba, S. and Kitayama, T., eds.) JJAP Series 4 (Publication Office of the Japanese Journal of Applied Physics, Tokyo 1991). 4. Kanazawa-shi Bunka Hall, Kanazawa, 15–18 July 1991 (Gamo, K., Horiike, Y., Ito, T., Okayama, S., Shibata, T., Tagawa, S., Todokoro, Y., and Urisu, T., eds.). Japan. J. Appl. Phys. 30 (1991) No. 11B, 2989–3317. 5. Kanagawa Science Park, Kanagawa, 13–16 July 1992 (Horiike, Y., Gamo, K., Urisu, T., Shibata, T., Okayama, S., Ito, T., Nagata, K., and Ueno, T., eds.). Japan. J. Appl. Phys. 31 (1992) No. 12B, 4101–4585. 6. International Conference Center, Hiroshima, 19–22 July 1993 (Horiike, Y., Gamo, K., Urisu, T., Shibata, T., Kanayama, T., Ito, T., Nagata, K., and Ueno, T., eds.). Japan. J. Appl. Phys. 32 (1993) No. 12B, 5821–6300. 7. Chiao Tung University, 11–14 July 1994 (Aoyagi, K., Atoda, N., Harada, M., Horiike, Y., Kanayama, T., Matsui, S., Miyashi, M., Murota,

ELECTRON OPTICS AND ELECTRON MICROSCOPY

345

J., Nakamura, M., and Takigawa, T., eds.). Japan. J. Appl. Phys. 33 (1994) No. 12B, 6745–7231. 8. Sendai International Centre, Sendai, 17–20 July 1995 (Aoyagi, K., Atoda, N., Harada, K., Horiike, Y., Itoh, J., Kanayama, T., Matsui, S., Miyoshi, M., Morimoto, H., Murota, J., Nakamura, M., Okazaki, S., Suzuki, K., Takigawa, T., and Yamaoka, T., eds.). Japan. J. Appl. Phys. 34 (1995) No. 12B, 6545–6986. 9. Kitakyushu International Conference Centre, Kitakyushu City, 8–11 July 1996 (Aoyagi, Y., Atoda, N., Fukui, T., Komuro, M., Kotera, M., Matsui, S., Matsui, Y., Okazaki, S., Samukawa, S., Shimazu, N., Suzuki, K., Takai, M., Tanaka, A., Watanabe, H., Yoneda, M., and Yoshikawa, R., eds.). Japan. J. Appl. Phys. 35 (1996) No. 12B, 6347–6695. 10. Microprocesses and Nanotechnology. Nagoya Congress Center, Nagoya City, 7–10 July 1997 (Aoyagi, Y., Atoda, N., Fukui, T., Kadomura, S., Kawai, Y., Komuro, M., Kotera, M., Matsui, S., Matsui, Y., Okazaki, S., Samukawa, S., Shimazu, N., Suzuki, K., Takai, M., Watanabe, H., and Yoshikawa, R., eds.). Japan. J. Appl. Phys. 36 (1997) No. 12B, 7473–7808. 11. Microprocesses and Nanotechnology. Kyoungju, Korea 13–16 July 1998 (Aoyagi, Y., Asano, T., Ban, H., Betsui, K., Deguchi, K., Endo, M., Fukushima, T., Hane, H., Hanyu, I., Hori, M., Itoh, J., Kadomura, S., Kanayama, T., Kawai, Y., Kikuchi, Y., Kinoshita, H., Komuro, M., Kotera, M., Makino, T., Marumoto, K., Matsui, S., Murai, F., Nakase, M., Nakayama, Y., Ohki, S., Ohta, T., Ohtsuka, H., Okazaki, S., Shimazu, N., Sugihara, K., Takakuwa, Y., Terasawa, T., Urisu, T., and Zaima, S., eds.). Japan. J. Appl. Phys. 37 (1998) No. 12B, 6667–7207. 12. Microprocesses and Nanotechnology. Yokohama, 6–8 July 1999 (Itoh, J., Kanayama, T., Kawai, Y., Komuro, M., Kotera, M., Matsui, S., Ohki, S., Ohtsuka, H., and Zaima, S., eds.). Japan. J. Appl. Phys. 38 (1999) No. 12B, 6955–7275. 13. Microprocesses and Nanotechnology. University of Tokyo, 11–13 July 2000 (Ochiai, Y., Asano, T., Endo, M., Hane, K., Hori, M., Ishibashi, K., Kanayama, T., Kinoshita, H., Komuro, M., Kotera, M., Matsui, S., Meguro, T., Niwa, O., Nomura, E., Ohki, S., Ohtsuka, H., Saito, K., Saitoh, K., Shiraishi, H., Sugioka, K., Shoji, S., Takakuwa, Y., and Terasawa, T., eds.). Japan. J. Appl. Phys. 39 (2000) No. 12B, 6771–7171. 14. Microprocesses and Nanotechnology. Kunibiki Messe, Matsue-shi, Shimane, 31 October–2 November 2001 (Ochiai, Y., Asano, T., Endo, M., Hane, K., Hori, M., Ishibashi, K., Kanayama, T., Kitamori, T., Komuro, M., Kotera, M., Matsui, S., Meguro, T., Murai, F., Nomura, E., Ohki, S., Ohtsuka, H., Shoji, S., Sugioka, K., Takakuwa, Y., and Terasawa, T., eds.). Japan. J. Appl. Phys. 41 (2002) No. 6B, 4033–4026.

346

HAWKES

15. Microprocesses and Nanotechnology. Tokyo Fashion Town, Tokyo, 6–8 November 2002. Japan. J. Appl. Phys. 42 (2003). I. Microanalysis 1. The Electron Probe Analysis Society of America (EPASA), subsequently the Microbeam Analysis Society (MAS) For an account of the prehistory of microbeam analysis, in the USA especially, and of the creation of the EPASA in 1966/7, formalized on 21 May 1968, see Wittry (1992). In 1974, the name was changed to Microbeam Analysis Society (MAS). Proceedings of early meetings were produced as spiral-bound books until in 1973, the San Francisco Press began publication as bound volumes. McKinley et al. (1966), which records the papers presented in Washington DC, 12–15 October 1964, is occasionally regarded as the ‘‘proceedings of the zeroth meeting.’’ In 1992, the Microbeam Analysis Society created its own journal, Microbeam Analysis, which amalgamated with the Journal of the Microscopy Society of America, later Microscopy and Microanalysis, in 1996. 1966: First National Conference on Electron Probe Microanalysis, Center of Adult Education, College Park MD, 4–6 May 1966 (Marton, L. L., Chairman). 1967: Second National Conference on Electron Probe Microanalysis, Somerset Hotel, Boston MA, 14–16 June 1967 (Ogilvie, R. E., Chairman). 1968: Third National Conference on Electron Microprobe Analysis, Chicago IL, 31 July–2 August 1968 (Smith, J. V., Knowles, C. R., and Beaman, D. R., Chairmen). 1969: Proceedings Fourth National Conference on Electron Microprobe Analysis, California Institute of Technology, Pasadena CA, 16–18 July 1969 (Wittry, D. B., Chodos, A. A., and Andersen, C. A., Chairmen). 1970: Proceedings Fifth National Conference on Electron Probe Analysis, New York City NY, 22–24 July 1970 (Lublin, P., Chairman). 1971: Proceedings Sixth National Conference on Electron Probe Analysis, Pittsburgh PA, 27–30 July 1971 (Vassamillet, L. F., Chairman). 1972: Seventh National Conference on Electron Probe Analysis, San Francisco Hilton, San Francisco CA, 17–21 July 1972 (Ruscica, R. J. and Kyser, D. F., Chairmen). 1973: Eighth National Conference on Electron Probe Analysis, Jung Hotel, New Orleans LA, 13–17 August 1973 (Lublin, P., Chairman). 1974: Tutorial and Proceedings Ninth Annual Conference of the Microbeam Analysis Society, Carleton University, Ottawa, 22–26 July 1974 (Plant, G., Chairman).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

347

1975: Proceedings Tenth Annual Conference of the Microbeam Analysis Society, MGM Hotel, Las Vegas NV, 11–15 August 1975 (Chodos, A. A., Chairman). 1976: Eleventh Annual Conference of the Microbeam Analysis Society, Fontainebleau Hotel, Miami Beach FL, 9–13 August 1976. 1977: Eighth International Conference on X-Ray Optics and Microanalysis. Papers presented at the Eighth International Conference on X-Ray Optics and Microanalysis, held in Boston, Massachusetts, August 18–24, 1977 and sponsored by the Microbeam Analysis Society (Beaman, D. R., Ogilvie, R. E., and D. B. Wittry, D. B., eds.; Pendell, Midland MI 1980). 1978: Proceedings Thirteenth Annual Conference of the Microbeam Analysis Society, Ann Arbor MI, 19–23 June 1978 (Bigelow, W. C., Bomback, J. L., and Kyser, D. F., Chairmen). 1979: Microbeam Analysis 1979. Proceedings of the 14th Annual Meeting of the Microbeam Analysis Society, San Antonio TX, 12–17 August 1979 (Newbury, D. E., ed.; San Francisco Press, San Francisco 1979). 1980: Microbeam Analysis 1980. Proceedings of the 15th Annual Meeting of the Microbeam Analysis Society, San Francisco CA, 4–8 August 1980 (Wittry, D. B., ed.; San Francisco Press, San Francisco 1980). 1981: Microbeam Analysis 1981. Proceedings of the 16th Annual Meeting of the Microbeam Analysis Society, Vail CO, 13–17 July 1981 (Geiss, R. H., ed.; San Francisco Press, San Francisco 1981). 1982: Microbeam Analysis 1982. Proceedings of the 17th Annual Meeting of the Microbeam Analysis Society, Washington DC, 9–13 August 1982 (Heinrich, K. F. J., ed.; San Francisco Press, San Francisco 1982). 1983: Microbeam Analysis 1983. Proceedings of the 18th Annual Meeting of the Microbeam Analysis Society, Phoenix AZ, 6–12 August 1983 (Gooley, R., ed.; San Francisco Press, San Francisco 1983). 1984: Microbeam Analysis 1984. Proceedings of the 19th Annual Meeting of the Microbeam Analysis Society, Lehigh University, Bethlehem PA, 16–20 July 1984 (Romig, A. D. and Goldstein, J. I., eds.; San Francisco Press, San Francisco 1984). 1985: Microbeam Analysis 1985. Proceedings of the 20th Annual Meeting of the Microbeam Analysis Society, Louisville KY, 5–9 August 1985 (Armstrong, J. T., ed.; San Francisco Press, San Francisco 1985). 1986: Microbeam Analysis 1986. Proceedings of the 21st Annual Meeting of the Microbeam Analysis Society, Albuquerque NM, 11–15 August 1986 (Romig, A. D. and Chambers, W. F., eds.; San Francisco Press, San Francisco 1986). 1987: Microbeam Analysis 1987. Proceedings of the 22nd Annual Meeting of the Microbeam Analysis Society, Kona HI, 13–17 July 1987 (Geiss, R. H., ed.; San Francisco Press, San Francisco 1987). Joint with the Japanese

348

HAWKES

Society for the Promotion of Science and the Australian Microbeam Analysis Society. 1988: Microbeam Analysis 1988. Proceedings of the 23rd Annual Meeting of the Microbeam Analysis Society, Milwaukee WI, 8–12 August 1988 (Newbury, D. E., ed.; San Francisco Press, San Francisco 1988). 1989: Microbeam Analysis 1989. Proceedings of the 24th Annual Meeting of the Microbeam Analysis Society, Asheville NC, 16–21 July 1979 (Russell, P. E., ed.; San Francisco Press, San Francisco 1989). 1990: Microbeam Analysis 1990. Proceedings of the 25th Annual Meeting of the Microbeam Analysis Society, Seattle WA, 12–18 August 1990 (Michael, J. R. and Ingrams, P., eds.; San Francisco Press, San Francisco 1990). 1991: Microbeam Analysis 1991. Proceedings of the 25th [sic] Annual Meeting of the Microbeam Analysis Society, San Jose CA, 4–9 August 1991 (Howitt, D. G., ed.; San Francisco Press, San Francisco 1991). 1992: Microbeam Analysis 1992. Proceedings of the 27th Annual Meeting of the Microbeam Analysis Society, Boston MA, 16–21 August 1992. See Microsc. Soc. Am. Proc. 50 (1992) Part 2. 1993: Microbeam Analysis 1993. Proceedings of the 28th Annual Meeting of the Microbeam Analysis Society, Los Angeles CA, 11–16 July 1993 (Armstrong, J. T. and Porter, J. R., eds.). Microbeam Analysis 2 (1993) S1–S298. 1994: Microbeam Analysis 1994. Proceedings of the 29th Annual Meeting of the Microbeam Analysis Society, New Orleans LA, 31 July–5 August 1994. See Microsc. Soc. Am. Proc. 52 (1994). 1995: Microbeam Analysis–1995. Proceedings of the 29th [sic] Annual Meeting of the Microbeam Analysis Society, Breckenridge CO, 6–11 August 1995 (Etz, E. S., ed.; VCH, New York, Weinheim and Cambridge 1995). 1996: Proceedings Microscopy and Microanalysis 1996. 54th Annual Meeting Microscopy Society of America, Twenty-third Annual Meeting Microscopical Society of Canada/Socie´te´ de Microscopie du Canada, 30th Annual Meeting Microbeam Analysis Society, Minneapolis MN, 11–15 August, 1996. (Bailey, G. W., Corbett, J. M., Dimlich, R. V. W., Michael, J. R. and Zaluzec, N. J., eds.; San Francisco Press, San Francisco 1996). From this date onwards, the Microbeam Analysis Society Proceedings are regularly included in those of the annual MSA Conference (see Section IV. B.1). In 2000, however, the MAS Conference was also part of IUMAS-2 (below). 2. International Union of Microbeam Analysis Societies In 1994, several Microbeam Analysis Societies formed the International Union of Microbeam Analysis Societies. In 2000, IUMAS comprises eight

ELECTRON OPTICS AND ELECTRON MICROSCOPY

349

member organizations: the Australian Microbeam Analysis Society, the Sociedade Brasileira de Microscopia e Microana´lise, the Microscopy Society of Canada, the China Joint Committee of Microbeam Analysis, the European Microbeam Analysis Society, Committee 141 (Microbeam Analysis) of Japan Society, Committee 141 (Microbeam Analysis) of Japan Society for the Promotion of Science, the Korean Society for Electron Microscopy and the Microbeam Analysis Society of the USA. 1996: IUMAS-1. First Meeting of the International Union of Microbeam Analysis Societies, held jointly with the Australian Microbeam Analysis Society and the 14th Australian Conference on Electron Microscopy (see Section IV.F.1), University of Sydney, 5–9 February 1996. 2000: IUMAS-2. Microbeam Analysis 2000, Proceedings of the Second Conference of the International Union of Microbeam Analysis Societies, Kailua-Kona HI, 9–14 July 2000. (Williams, D. B. and Shimizu, R., eds.; Institute of Physics Publishing, Bristol and Philadelphia 2000). Institute of Physics Conference Series 165. 3. Analytical Electron Microscopy 1976: Report on a Specialist Workshop in Analytical Electron Microscopy, Cornell University, Ithaca NY, 3–6 August 1976, by M. S. Isaacson and J. Silcox. Ultramicroscopy 2 (1976) 89–104. 1978:ProceedingsofaSpecialistWorkshop inAnalyticalElectronMicroscopy, Cornell University, Ithaca (NY), 25–28 July 1978 (Fejes, P. L., ed.). 1981: Analytical Electron Microscopy 1981. Proceedings of a Workshop held at Vail CO, 13–17 July 1981 (Geiss, R. H., ed.; San Francisco Press, San Francisco 1981). 1984: Analytical Electron Microscopy 1984. Proceedings of a Workshop held at Bethlehem PA, 16–20 July 1984 (Williams, D. B. and Joy, D. C., eds.; San Francisco Press, San Francisco 1984). 1987: Analytical Electron Microscopy 1987. Proceedings of a Workshop held at Kona HI, 13–17 June 1987 (Joy, D. C., ed.; San Francisco Press, San Francisco 1987). This was held in conjunction with the first meeting of the International Union of Microbeam Analysis Societies (IUMAS-1). 1990: See ICEM-XII, Seattle 1990. 4. European Microbeam Analysis Society This society (EMAS) organized its first European Workshop on Modern Developments and Applications in Microbeam Analysis in 1989 and such workshops have been held biennially ever since. The principal themes, dates and venues are as follows.

350

HAWKES

1. ‘‘Quantitative electron probe microanalysis.’’ Congress Centre ‘t Elzenfeld, Antwerp, 8–10 March 1989; organized by EMAS and the University of Antwerp. 2. ‘‘Analytic electron microbeam methods.’’ Dubrovnik, 14–18 May 1991; organized by EMAS and the Yugoslav Federal Society for Electron Microscopy (YUDEM). Mikrochim. Acta (1992) Supplement 12: Electron Microbeam Analysis (Boekestein, A. and Pavicˇevic´, M. K., eds.). 3. ‘‘Analytical electron and ion microbeam methods.’’ Rimini, 9–13 May 1993; organized by EMAS and Consiglio Nazionale delle Ricerche, PF ‘‘Materiali speciali per tecnologie avanzate,’’ SIME and CESEM. Mikrochim. Acta 114/115 (1994); (Armigliato, A., Dack, L. van’t, Werner, H. W., and Wermisch, J., guest eds.). 4. ‘‘Microbeam and nanobeam techniques.’’ St Malo, 14–19 May 1995; organized by EMAS and l’ Association Nationale de la Recherche Technique. Mikrochim. Acta (1996) Supplement 13: Microbeam and Nanobeam Analysis (Benoit, D., Bresse, J.-F., Dack, L. van’t, Werner, H., and Wermisch, J., eds.). 5. ‘‘Modern developments and applications in microbeam analysis.’’ Torquay (Devon), 11–15 May 1997; organized by EMAS and the Royal Microscopical Society. Mikrochim. Acta (1998) Supplement 15 (Love, G., Nicholson, W. A. P., and Armigliato, A., eds.). 6. ‘‘Modern developments and applications in microbeam analysis.’’ Konzilgeba¨ude, Konstanz, 3–7 May 1999. Mikrochim. Acta 132 (2000) Nos. 2–4 (Walker, C. T., Karduck, P., and Armigliato, A., eds.). 7. ‘‘Modern developments and applications in microbeam analysis.’’ Tampere Hall, Tampere, 6–10 May 2001. Mikrochim. Acta 138 (2002) Nos. 3–4 and 139 (2002) Nos. 1–4 (Heikinheimo, H., Walker, C. T., and Armigliato, A., eds.). 8. ‘‘Modern developments and applications in microbeam analysis.’’ Hotel Valentin Sancti Petri, Chiclana de la Frontera (Ca´diz), 18–22 May 2003. 9. ‘‘Modern developments and applications in microbeam analysis.’’ Firenze 2005. Regional workshops (held biennially since 1994) 1. ‘‘Electron Probe Microanalysis of Materials Today–Practical Aspects,’’ Kirkkonummi (Finland) 15–17 June 1994. Proceedings edited by E. Heikinheimo and J. Broomberg, distributed as report TKK-V-B96 by the Laboratory of Metallurgy, Helsinki University of Technology. 2. ‘‘Electron Probe Microanalysis of Materials Today–Practical Aspects,’’ Balatonfu¨red (Hungary) 19–22 May 1996. Proceedings edited by J. L. La´ba´r, E. Heikinheimo and P. Nicholson, distributed

ELECTRON OPTICS AND ELECTRON MICROSCOPY

351

by the Department of Solid-state Physics, Kossuth Lajos University, Debrecen, Helsinki University of Technology. 3. ‘‘Electron Probe Microanalysis Today, Practical Aspects,’’ Facultat de Geologia, Universitat de Barcelona, 13–16 May 1998. Proceedings edited by X. Llovet, C. Merlet and F. Salvat, published by Ediciones de la Universitat de Barcelona. 4. ‘‘Electron Probe Microanalysis of Materials Today–Practical Aspects,’’ Za´mecky´ Hotel, Trˇesˇt (Czech Republic), 17–20 May 2000. Proceedings edited by V. Stary´, K. Masˇek, and K. Hora´k, published by the Czech Technical University in Prague, Faculty of Mechanical Engineering. 5. ‘‘Electron Probe Microanalysis of Materials Today–Practical Aspects,’’ Szczyrk (Poland), 22–25 May 2002. Chairperson, A. Czyrska-Filemonowicz. 5. European Workshops on Electron Spectroscopic Imaging and Analysis Techniques 1. Proceedings of the [First] Workshop, Tu¨bingen, 8–9 June 1989. Ultramicroscopy 32 (1990) No. 1, 1–91 (Reimer, L., guest ed.). 2. Papers from the Second Workshop, Dortmund, 17–18 April 1990. J. Microscopy 162 (1991) No. 1, 1–200 (Reimer, L., guest ed.). 3. Papers from the Third Workshop, organized by Carl Zeiss Oberkochen and the Max-Planck-Institut fu¨r Biochemie, Martinsried, Martinsried-bei-Mu¨nchen, 27–28 June 1991. J. Microscopy 166 (1992) No. 3, 255–416. 4. Papers from the Fourth Workshop, organized by Carl Zeiss Oberkochen and the Institute of Physics at the University of Mu¨nster, Mu¨nster, 3–4 June 1993. J. Microscopy 174 (1994) No. 3, 131–238. 6. Workshops on Electron Energy-Loss Spectroscopy and Imaging (EELSI) These workshops are held shortly after the International Congresses on Electron Microscopy (ICEM) 1. Proceedings of the Lake Tahoe Workshop on Electron Energy Loss Spectroscopy, Granlibakken Conference Centre, Tahoe City CA, 18–22 August 1990. Microsc. Microanal. Microstruct. 2 (1991) Nos. 2/3, 143–411 (Krivanek, O. L., guest ed.). 2. Proceedings of the Second Workshop on Electron Energy-loss Spectroscopy and Imaging (EELSI), Leukerbad, 24–28 July 1994. Parts 1 and 2, Ultramicroscopy 59 (1995) Nos. 1–4 (Krivanek, O. L., guest ed.); Part 3, Microsc. Microanal. Microstruct. 6 (1995) No. 1, 1–157 (Krivanek, O. L., guest ed.).

352

HAWKES

3. ‘‘Towards atomic resolution analysis’’ (TARA-98), Port Ludlow Olympic Resort and Conference Center, Olympic Peninsula WA, 6–11 September 1998. Micron 30 (1999) No. 2 (Krivanek, O. Leapman, R. D. and Sarikaya, M., guest eds.); Ultramicroscopy 78 (1999) Nos. 1–4 (Leapman, R. D., guest ed.). 4. ‘‘Strategies and Advances in Atomic Level Spectroscopy’’ (SALSA2002), Hotel Salako, Gosier, Guadeloupe 5–9 May 2002. Ultramicroscopy (2003) and J. Microscopy (2003).

J. Frontiers of Electron Microscopy in Materials Science 1. Proceedings of the First Conference, Argonne National Laboratory, Argonne IL, 20–23 April 1986. Ultramicroscopy 22 (1987) Nos. 1–4 (Bradley, S. and King, W., guest eds.). 2. Proceedings of the Second Conference, Hyatt Oak Brook Hotel, Oak Brook IL, 16–19 May 1988. Ultramicroscopy 29 (1989) Nos. 1–4 (Bradley, S. A., King, W. E., and Allen, C. W., guest eds.). 3. Proceedings of the Third Conference, Hyatt Oak Brook Hotel, Oak Brook IL, 20–24 May 1990. Ultramicroscopy 37 (1991) Nos. 1–4 (Bradley, S. A., King, W. E., and Allen, C. W., guest eds.). 4. Proceedings of the Fourth Conference, Oakland CA, 21–24 April 1992. Ultramicroscopy 51 (1993) Nos. 1–4 (Bradley, S. A. and King, W. E., guest eds.). Ultramicroscopy 96 (2003) Nos. 3–4 and J. Microscopy 210 (2003) Part 1. 5. Proceedings of the Fifth Conference, Claremont Resort Spa and Tennis Club, Oakland CA, 21–24 June 1994. J. Microscopy 180 (1995) No. 1, 1–89. 6. Proceedings of the Sixth Conference, Hyatt Regency Oak Brook Hotel, Oak Brook IL, 4–7 June 1996. Ultramicroscopy 67 (1997) Nos. 1–4 (Bradley, S. A., Allen, C. W. and King, W. E., guest eds.) and Microsc. Microanal. 3 (1997), No. 2, 108–153. 7. Proceedings of the Seventh Conference, Kloster Irsee, Irsee (Germany), 19–24 April 1998. J. Microscopy 194 (1999) Part 1 (Ernst, F., Mayer, J. and Ru¨hle, M., guest eds.). 8. Proceedings of the Eighth Conference, ‘‘Bridge to 21st Century Electron Microscopy,’’ Kunibike Messe (Shimane Prefectural Convention Centre), Matsue City, 13–17 November 2000. J. Electron Microsc. 50 (2001) No. 6 and 51 (2002) Supplement (Furuya, K. and Ichinose, H., guest eds.). 9. Ninth Conference, Claremont Resort and Spa, Berkeley CA, 5–10 October 2003.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

353

K. COMPUMAG For an account of the genesis of the compumag conferences, the first of which was held in Oxford in 1976, see Trowbridge (1996); the information collected there is the basis of the details of the first ten meetings in the following list. From the third meeting onwards, the proceedings have been published in the IEEE Proceedings on Magnetics and for these we simply give the venue, dates, editor and journal reference. 1. St. Catherine’s College, Oxford, 31 March–2 April 1976 (Trowbridge, C. W., ed.; Rutherford Appleton Laboratory, Chilton 1976). 2. Laboratoire d’Electrotechnique, ENSEGP, Grenoble, 4–6 September 1978 (Sabonnadie`re, J. C., ed. ENSEGP, Grenoble 1978). 3. Americana-Congress Hotel, Chicago IL, 14–17 September 1981 (Turner, L. E., chairman). IEEE Trans. Magn. 18 (1982) No. 2, 309–700. 4. Genoa, 30 May–2 June 1983 (Molinari, G., ed.). IEEE Trans. Magn. 19 (1983) No. 6. 5. Fort Collins CO, 3–6 June 1985 (Lord, W., ed.). IEEE Trans. Magn. 21 (1985) No. 6. 6. Graz, 25–28 August 1987 (Richter, K., ed.). IEEE Trans. Magn. 24 (1988) No. 1. 7. Tokyo, 3–7 December 1989 (Miya, K., ed.). IEEE Trans. Magn. 26 (1990) No. 2. 8. Sorrento, 7–11 July 1991 (Martone, R., ed.). IEEE Trans. Magn. 28 (1992) No. 2. 9. Miami FL, 31 October–4 November 1993 (Lowther, D. A., ed.). IEEE Trans. Magn. 30 (1994) No. 5. 10. Berlin, 10–13 July 1995 (Biro, O., ed.). IEEE Trans. Magn. 32 (1996) No. 3, Part 1. 11. Rio de Janeiro, 3–6 November 1997 (Ida, N., Chairman ed. board). IEEE Trans. Magn. 34 (1998) No. 5, Parts 1 and 2. 12. Keio Plaza Hotel, Sapporo, 25–28 October 1999 (Takagi, T., ed.). IEEE Trans. Magn. 36 (2000) No. 4, Part I. 13. Lyon-Evian, 2–5 July 2001. IEEE Trans. Magn. 38 (2002) No. 2, Part 1, 14. Saratoga Springs NY, June 2003. IEEE Trans. Magn. 40 (2004). VI. Acknowledgments A. Individuals Haroon Ahmed (Cambridge) Aldo Armigliato (Bologna)

354

HAWKES

Marcia Attias (Rio de Janeiro) Vitalii Vasil’evich Aristov (Chernogolovka) Robert Bakish (Englewood NJ) Hieronim Bartel (Ło´dz´) Victor Benghiat (Weizmann Institute, Rehovoth) Susan Betteridge (Royal Microscopical Society, Oxford) Wiesława Biczysko (Poznan´) Walter Jose Botta Filho (Sa˜o Carlos) Pieter Brederoo (Leiden) Javier Bringas (Biblioteca Marcel Roche, IVIC, Caracas) Joe Britton (Cambridge) Paul Brown (Cambridge and Nottingham) Vladimir Bumbasˇirevic´ (Belgrade) Yurdagu¨l Canberk (Istanbul) Jose-Maria Carazo (Madrid) Barry Carter (Minneapolis) Raimond Castaing (Paris) Torranin Chairuangsri (Chiang Mai, Thailand) Edward Chan (Chinetek Corporation, Hong Kong) Lih J. Chen (National Tsing Hua University, Taiwan) Art Chodos (Monroeville CA) ˇ iampor (Bratislava) Fedor C John Cleaver (Cambridge) David Cockayne (Oxford) Marie Colbert (Hamilton, Ontario) Floriana Colombo (Biblioteca Central ‘‘Luis F. Leloir,’’ Universidad de Buenos Aires) David Cottell (Dublin) Stuart Craig (Canberra) Robin Cross (Grahamstown) ´ gnes Csana´dy (Budapest) A Anthony Cullis (Sheffield) Alexandre Lobo da Cunha (Porto) Aleksandra Czyrska–Filemonowicz (Krako´w) Luc van’t Dack (Antwerp) Steven Dale (Scientific Periodicals Library, Cambridge) Muhammad Dani (Jakarta) Jenny Denman (Delft) Dirk van Dyck (Antwerp) Ulrich Ehrenwerth (Mu¨nster) Carolyn Emerson (St Johns, Newfoundland) Tu¨rkaˆn Erbengi (Istanbul)

ELECTRON OPTICS AND ELECTRON MICROSCOPY

Viviana Falco´n Cama (La Habana) Robert M. Fisher (Seattle WA) Colette Fradin (Orsay) Ludeˇk Frank (Brno) Wolfgang Geymayer (Graz) Jon Gjønnes (Oslo) Audrey Glauert (Cambridge) Luz Stella Gomez (Bogota´) Peter Goodhew (Liverpool) Alan Hall (Pretoria) Ian Hallett (Auckland) Hatsujiro Hashimoto (Okayama) John Hawkes (Cristal, London) Werner Hax (Philips, Eindhoven) Geoffrey Hayward (Philips, Eindhoven) Hans Hebert (Stockholm) H. Henke (Aachen) Siebe Henstra (Wageningen and Nijmegen) Francisca Herna´ndez (Biblioteca Nacional, Madrid) Marı´a Inmaculada Herrera (Madrid) Walter Hert (Munich) Johannes Heydenreich (Halle) Michio Hibino (Nagoya and Toyota) Yasuhiro Horiike (Toyo University, Kujirai, Kawagoe) Karlen Hovnanyan (Yerevan) Archie Howie (Cambridge) Miguel Ipohorski (Buenos Aires) Wim Jacob (Antwerp) Stanisława Jasien´ska (Cracow) Luis Felipe Jimenez Garcia (Mexico City) M. S. Johal (Chandigarh) Om Johari (Chicago IL) Susana Beatriz Jurado (La Plata) Koichi Kanaya (Tokyo) Maria Kanellaki–Kyparissi (Thessaloniki) Janet Kershaw (Elsevier, Amsterdam) Elliot Kitajima (Sa˜o Paolo) Pedro Kiyohara (Sa˜o Paolo) Henk Koerten (Leiden) Pieter Kruit (Delft) Dong-wha Kum (Seoul) Hu-chul Lee (Seoul)

355

356

HAWKES

Francis Leggett (Microscopical Society of Canada) Bohumila Lencova´ (Brno) Hanna Levanony (Rehovot) Jill Lewis (London) Hing Hiang Lian (Malaysia) Marı´a Luisa Lo´pez Alvarez (Santiago de Chile) Dorothe´e Lotthe´ (Socie´te´ Franc¸aise des Microscopies, Paris) Francisco Lovey (Bariloche, Argentiana) Jirˇ´ı Ludvı´k (Prague) Bill MacLennan (Canadian Agriculture Library, Ottawa) Maria Conceia˜o Magalha˜es (Porto) Rui Malho (Porto) L. Malicsko´ (Budapest) Lukas Margaritis (Athens) Lawrence Maser (Microscopy Society of America, Pocasset MA) Arvid Maunsbach (Aarhus) Dennis McMullan (Cambridge) Ilaria Meliconi (Royal Microscopical Society, Oxford) Jan Mellema (The Hague) Ognjen Milat (Zagreb) Aramayis Mkrtchyan (Yerevan) Mikhail Monastyrskii (Moscow) Rebecca Morden (Royal Microscopical Society, Oxford) Alberto Moreira Jorge (Universidade Federal de Sa˜o Carlos, SP) Ilona Mu¨llerova´ (Brno) Tom Mulvey (Birmingham) Androula Nassiopoulos (Ag. Paraskevi Attikis) Sergei Nepijko (Kiev and Berlin) Mary (Mah-Lee) Ng (Singapore) Mitsuo Ogura (Caracas) Meliton Ordillas (Queson City, Philippines) Jon Orloff (College Park MD) Fabian Pease (Stanford CA) Monique Penochet (Toulouse) A. Petrossian (Yerevan) Kurt Pulfer (Basel) Sharon Rabesa (Microscopy Society of America, Pocasset MA) Juan F. Ramos Sanchez (Universidad Complutense, Madrid) Eduard Ivanovich Rau (Moscow) Ludwig Reimer (Mu¨nster) Jose´ Reyes–Gasca (Mexico City) Godfried Roomans (Uppsala)

ELECTRON OPTICS AND ELECTRON MICROSCOPY

Gustavo Rubio Coronel (Guayaquil) Manfred Ru¨hle (Stuttgart) Ashok Sahni (Chandigarh) Gerhard Schimmel (Reichelsheim) Dominique Schryvers (Antwerp) Jose´ Serrano (Me´rida) Madan Lal Sharma (Chandigarh) Surjeet Kumar Sharma (New Delhi) Makoto Shiojiri (Kyoto) Balgopal Shyamala (Library, University of Illinois at Urbana IL) Ge´rard Simon (Hamilton ON) Celia Snyman (Durban) Wanderley de Souza (Rio de Janeiro) John R. Stanley (Electrochemical Society, Princeton NJ) Martin Steer (Dublin) Leopold Stockinger (Vienna) Ei-ichi Sukedai (Okayama) Kabkaew Sukontason (Chian Mai) Tatiana Sukhanova (Saint Petersburg) Charles Su¨sskind (San Francisco CA) Paul Sutherland (Auckland) Susan Tai (Caracas and Hong Kong) Juan Takano Moro´n (Lima) Michiyoshi Tanaka (Tohoku University, Sendai) Minoru Tanaka (Voreppe) Julia Tancock (Institute of Physics Publishing, Bristol) Erdogˇan Tekin (Ankara) Don Tennant (Holmdel NJ) Bernd Tesche (Mu¨lheim an der Ruhr) Michel Thellier (Mont-Saint-Aignan, Rouen) Peggie Tilman (National Library of Medicine, Bethesda MD) Akira Tonomura (Hatoyama, Saitama) Jorge Troccoli (Montevideo) Peter Turner (Belmont, Victoria) Caribay Urbina (Caracas) Ugo Valdre` (Bologna) Ernesto Valiente Madriz (Me´rida) Pramote Vanittanakom (Chiang Mai) Patricia Vive`s (Socie´te´ Franc¸aise des Microscopies, Paris) Jovan Vukovic´ (Belgrade) Guangquan Wan (Wu Shan, Guangzhou) Jaang J. Wang (Taipei)

357

358

HAWKES

David Wild (Meat Industry Research Institute of New Zealand, Hamilton) David Williams (Bethlehem PA) John Williams (Bilbrough and Co., Hong Kong) Eddie Wisse (Brussels) David Wittry (Los Angeles, CA) Michael Witcomb (Johannesburg) Lambiek van den Wittenboer (Philips, Eindhoven) Miguel Jose´ Yacaman (Mexico City) Yevgeny Yakimov (Chernogolovka) Jun-en Yao (Beijing) Mikhail Yavor (St Petersburg) Stella Yavor (St Petersburg) Zul Azhar Zahid Jamal (Universiti Sains Malaysia, Pulau Pinang) Elmar Zeitler (Berlin) B. Libraries National Library of Medicine, Bethesda MD The British Library, Boston Spa Biblioteca Central ‘‘Luis F. Leloir,’’ Universidad de Buenos Aires University Library, Cambridge Scientific Periodicals Library, Cambridge Cambridge University Engineering Department Library Cavendish Laboratory Library, Cambridge IVIC Library, Caracas Centre de Documentation, INRA, Jouy-en-Josas Linda Hall Library, Kansas City MO Patent Office Library, London Biblioteca, Universidad Complutense de Madrid Biblioteca Nacional, Madrid Institut de l’Information Scientifique et Technique, Nancy–Vandœuvre Canadian Agriculture Library, Ottawa Bibliothe`que Universitaire, Toulouse Library, University of Illinois at Urbana IL

VII. Acronyms ACEM: Australian Conference on Electron Microscopy AEM: Analytical Electron Microscopy AMM: Asociatio´n Mexicana de Microscopı´a APEM: Asia-Pacific Conference on Electron Microscopy

ELECTRON OPTICS AND ELECTRON MICROSCOPY

359

APEMEL: Asociacio´n Peruana de Microscopı´a Electro´nica ASEAN: Association of South-East Asian Nations (currently Brunei Darussalam, Cambodia, Indonesia, Laos, Malaysia, Myanmar, The Philippines, Singapore, Thailand and Vietnam) BEDAO: Beitra¨ge zur elektronenmikroskopischen Direktabbildung und Analyse von Oberfla¨chen BEDO: Beitra¨ge zur elektronenmikroskopischen Direktabbildung von Oberfla¨chen BJCEM: British Joint Committee on Electron Microscopy BVEM (¼ SBME): Belgische Vereniging voor Electronenmicroscopie, Socie´te´ Belge de Microscopie Electronique BVM (¼ SBM): Belgische Vereniging voor Microscopie, Socie´te´ Belge de Microscopie CAPSEM: Committee of Asia–Pacific Societies for Electron Microscopy CESM: Committee of European Societies of Microscopy CESEM: Committee of European Societies of Electron Microscopy CIASEM: Comite´ de Sociedades Interamericanas para Microscopı´a Electro´nica, Committee of Inter-American Societies for Electron Microscopy CNRS: Centre National de la Recherche Scientifique CPO: Charged Particle Optics ˇ SMS: C ˇ eskoslovenska´ Mikroskopicka´ Spolecˇnost, Czechoslovak MicroC scopy Society DGE or DGEM: Deutsche Gesellschaft fu¨r Elektronenmikroskopie EELSI: Electron Energy-loss Spectroscopy and Imaging EMAG: Electron Microscopy and Analysis Group [of the Institute of Physics, London] EMAS: European Microbeam Analysis Society EMS: European Microscopy Society EMSA: Electron Microscopy Society of America EMSI: Electron Microscope Society of India EMSSA: Electron Microscopy Society of Southern Africa, Elektronenmikroskopievereniging van Suidelike Afrika EMST: Electron Microscopy Society of Thailand EPASA: Electron Probe Analysis Society of America EUREM: European Conference on Electron Microscopy FEMMS: Frontiers of Electron Microscopy in Materials Science HVEM: High Voltage Electron Microscopy ICEM: International Congress on Electron Microscopy ICXOM: International Congress on X-ray Optics and Microanalysis IFEMS: International Federation of Electron Microscopy Societies IFSEM: International Federation of Societies of Electron Microscopy

360

HAWKES

IFSM: International Federation of Societies for Microscopy IITRI: Illinois Institute of Technology Research Institute IoP: Institute of Physics [London] ISEM: Irish Society of Electron Microscopy ISEM: Israel Society of Electron Microscopy ISM: Israel Society for Microscopy ISMM: Indonesian Society of Microscopy and Microanalysis JSEM: Japanese Society of Electron Microscopy KSEM: Korean Society of Electron Microscopy MAS: Microbeam Analysis Society ME: Microelectronic Engineering MNE: Micro- and Nanoelectronic Engineering MSA: Microscopy Society of America MSC (¼ SMC): Microscopy Society of Canada, Socie´te´ de Microscopie du Canada MSI: Microscopical Society of Ireland MSS: Microscopy Society (Singapore) MSSA: Microscopy Society of Southern Africa, Mikroskopievereniging van Suidelike Afrika NVEM: Nederlandse Vereniging voor Elektronenmicroscopie NVvM: Nederlandse Vereniging voor Microscopie ¨ GE: O ¨ sterreichische Gesellschaft fu¨r Elektronenmikroskopie O RMS: Royal Microscopical Society SAMIC: Sociedad Argentina de Microscopı´a SBM (¼ BVM): Socie´te´ Belge de Microscopie, Belgische Vereniging voor Microscopie SBME: Sociedade Brasileira de Microscopia Eletroˆnica SBME (¼ BVEM): Socie´te´ Belge de Microscopie Electronique, Belgische Vereniging voor Electronenmicroscopie SBMM: Sociedade Brasileira de Microscopia e Microana´lise. SCANDEM: Scandinavian Society of Electron Microscopy SEME: Sociedad Espan˜ol de Microscopı´a Electro´nica SEME: Sociedad Ecuatoriana de Microscopı´a Electro´nica SFME: Socie´te´ Franc¸aise de Microscopie Electronique SF: Socie´te´ Franc¸aise des Microscopies SGOEM (¼ SSOME): Schweizerische Gesellschaft fu¨r Optik und Elektronenmikroskopie SIME: Societa` Italiana di Microscopia Elettronica SLAME: Sociedad Latino-Americana de Microscopı´a Electro´nica SMC (¼ MSC): Socie´te´ de Microscopie du Canada, Microscopy Society of Canada SME: Sociedad de Microscopı´a Espan˜ola

ELECTRON OPTICS AND ELECTRON MICROSCOPY

361

SPME: Sociedade Portuguesa de Microscopia Electro´nica e Biologia Celular SSOM: Swiss Society for Optics and Microscopy SSOME (¼ SGOEM): Socie´te´ Suisse d’Optique et de Microscopie Electronique SVME: Sociedad Venezolana de Microscopı´a Electro´nica.

References In addition to the references cited in the text, this list contains details of a number of bibliographies (Marton and Sass, 1943, 1944, 1945; Rathbun et al., 1946; Cosslett, 1950; Marton et al., 1950; von Borries and Ruska, 1951/2, 1952/3, 1956/8). A few journals continue to provide some bibliographic information, but the fullest sources are of course the major abstracting services (inspec, biosis, Referativnye Zhurnal). From 1972 to 1983, volumes 1–11 of Electron Microscopy Abstracts were published by the Science and Technology Agency, London; the original editor was J. G. Wilkinson, with the editorial assistance of I. L. F. Ray and the last editor was I. W. Drummond. An ‘‘International Bibliography of Electron Microscopy’’ was compiled on cards by the New York Society of Electron Microscopy and subsequently issued in book form (NYSEM 1959, 1962). Articles chronicling the progress of electron optics and microscopy were included for several years in VDI-Zeitschrift and Zeitschrift fu¨r Instrumentenkunde, see von Borries and Langner (1956), Ruska (1957b, 1959, 1962, and 1964) and Riecke (1967, 1968, 1971). Bibliographic listings of publications on computer processing of electron microscope images have been prepared by Hawkes (1978, 1982a, 1992, 1993) and Baker (1981); for a bibliography of publications on magnetic lens properties, see Hawkes (1982b). A bibliography of books on materials microscopy has been prepared by Mannheimer (1989), updated in 1996. A list of key papers in electron microscopy has been compiled by Haguenau et al. (2003). Numerous papers of historical interest are reproduced in Hawkes (1994) and a number of recondite papers from the early years are cited in Hawkes (1985b). Also see Peven and Grahn (1985) and Hirsch (1998). The electron microscope is beginning to attract the attention of historians and philosophers of science. I have included a few books and articles that have come to my attention, notably van Fraassen (1980), Hacking (1983), Kosso (1988), Rasmussen (1993, 1996, 1998a,b), Kunkle (1994) to which may be added the collections Hawkes (1985a) and Mulvey (1996). The ‘‘discovery’’ of the electron by J. J. Thomson in 1897 is the object of several books and articles, several of which are linked to the centenary (Arabatzis, 1996; Chayut, 1991; Dahl, 1997; Davis and Falconer, 1997; Falconer, 1987; Feffer, 1989; Hawkes, 1997b, 1998, 2000, 2001; Humphreys, 1997; Spence, 1997; Springford, 1997; Buchwald and Warwick, 2001). Note that references to the international (ICEM), European (EUREM), and Asia–Pacific (APEM) meetings are given in the short form [acronym]-[meeting number]; for full bibliographic details, see Sections II (ICEM), III.A (EUREM), and III.B (APEM). ¨ bermikroskop’’ (Petersen, W. and AEG: Jahrbuch der AEG-Forschung 7 (1940), Sonderheft ‘‘U Ramsauer, C., eds.), pp. 1–90 (Springer, Berlin 1940). See Bru¨che (1940) below. Afzelius, B. A.: Half a century of electron microscopy: the early years. Ultrastruct. Pathol. 3 (1981) 309–311. Afzelius, B. A.: Highlights of the IFSEM congresses, biology. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 385–391.

362

HAWKES

Agar, A. W.: Electron microscopy in the United Kingdom–the AEI contribution. Micron Microsc. Acta 17 (1986) 77–85. Agar, A. W.: Early performance claims for electron microscopes. Proc. Roy. Microsc. Soc. 29 (1994) 100–101 [comment on Watson 1992. 1993 and Doane et al. 1993]. Agar, A. W.: The story of European commercial electron microscopes. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 415–584. Agar, A. W., Brown, P. D., Mulvey, T., and Smith, K. C. A.: The curious history of the first SEM produced independently by AEI Manchester. In The Electron, Proceedings of the International Centennial Symposium on the Electron, Churchill College, Cambridge 15–17 September 1997 (Kirkland, A. and Brown, P. D., eds.) 266–269 (IoM Communications, London 1998). Allen, C. W.: High voltage electron microscopy: where it is and where it may be going. Proc. MSA 53 (1995) 80–81. Allen, C. W., and Dorignac, D.: Survey of high voltage electron microscopy worldwide in 1998. ICEM-14, Cancu´n 1 (1998) 275–276. Anderson, R. (ed.): Specimen Preparation for Transmission Electron Microscopy of Materials II (Materials Research Society, Pittsburgh PA 1990). Record of a Symposium, San Francisco CA, 19–20 April 1990. MRS Symp. Proc. vol. 199. Anderson, R. M., and Walck, S. D. (eds.): Specimen Preparation for Transmission Electron Microscopy of Materials IV (Materials Research Society, Pittsburgh PA 1997). Record of a Symposium, San Francisco CA, 2 April 1997. MRS Symp. Proc. vol. 480. Arabatzis, T.: Rethinking the ‘discovery’ of the electron. Stud. Hist. Philos. Mod. Phys. 27 (1996) 405–435. Ardenne, M. von: Ein Glu¨ckliches Leben fu¨r Technik und Forschung (Kindler, Munich 1972; reprinted by Suhrkamp, Taschenbuch 310, 1976). Ardenne, M. von: Zur Geschichte der Rasterelektronenmikroskopie und der Rasterelektronenmikrosonde. Optik 50 (1978) 177–188. Ardenne, M. von: On the history of scanning electron microscopy, of the electron microprobe and of early contributions to transmission electron microscopy. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 1–21. Ardenne, M. von: Reminiscences on the origins of the scanning electron microscope and the electron microprobe. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 635–652. Ardenne, M. von: Perso¨nliche Erinnerungen an das Entstehen des Elektronenmikroskopes und seiner Technik. Optik 105 (1997) 139–153. Bahr, G. F., and Zeitler, E. H. (eds.): Quantitative Electron Microscopy. Proceedings of a Symposium held at the Armed Forces Institute of Pathology, Washington DC, 30 March– 3 April 1964. Laboratory Investigation 14 (1965) 739–1340. Also published as a separate volume by Williams and Wilkins, Baltimore MD 1965. Baker, T. S.: Image processing of biological specimens: a bibliography. In J. D. Griffith (ed.), Electron Microscopy in Biology (Wiley–Interscience, New York and Chichester 1981), vol. 1, 189–290. Bakker, C. J. G.: Elektronenmicroscopie in Nederland. Overzicht en evaluatie van 40 jaar speuren ontwikkelingswerk in een technisch-fysisch specialisme. FOM Rapport Nr 43105, Utrecht 1977. Basu, S., and Millette, J. R. (eds.): Electron Microscopy in Forensic, Occupational, and Environmental Health Sciences (Plenum, New York and London 1986). Baumeister, W., and Vogell, (eds.): Electron Microscopy at Molecular Dimensions, State of the Art and Strategies for the Future, Proceedings of an International Workshop on Regular 2-D Arrays of Biomacromolecules: Structure Determination and Assembly, Burg Gemen, June 1979 (Springer, Berlin and New York 1980).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

363

Baumeister, W., and Zeitler, E. (guest ed. and ed.): Quantitative Electron Microscopy, Proceedings of a Symposium, Schloß Ringberg am Tegernsee, 2–7 December 1991. Ultramicroscopy 46 (1992) Nos. 1–4. Beer, M., Frank, J., Hanszen, K.-J., Kellenberger, E., and Williams, R.: The possibilities and ˚ ) on biological material using prospects of obtaining high-resolution information (below 30 A the electron microscope. Some comments and reports inspired by an EMBO workshop held at Gais, Switzerland, October 1973. Q. Rev. Biophys. 7 (1975) 211–238. Bentley, J., Allen, C., Dahmen, U., and Petrov, I. (eds.): Advances in Materials Problem Solving with the Electron Microscope (Materials Research Society, Warrendale PA 2001); MRS Proceedings vol. 589. Biegelsen, D. K., Smith, D. J., and Tong, S. Y. (eds.): Atomic-scale Imaging of Surfaces and Interfaces (Materials Research Society, Pittsburgh PA 1993). Record of a Symposium, Boston MA, 30 November–2 December 1992. Boekestein, A. (ed.) Past, Present, and Future of Electron Microscopy in Agricultural Research (Dienst Landbouwkundig Onderzoek, Technische en Fysische Dienst voor de Landbouw, Wageningen 1991). Borasky, R.: The electron microscope: priority of invention. Microscope 30 (1982), No. 1, 11–15. See Ruska (1985). Borries, H. von: Bodo von Borries, pioneer of electron microscopy. Adv. Electron. & Electron Phys. 81 (1991) 127–176. Borries, B. von, and Langner, G.: Elektronenmikroskopie. VDI-Z. 98 (1956) 622–624. Borries, B. von, and Ruska, E.: Neue Beitra¨ge zur Entwicklungsgeschichte der Elektronen¨ bermikroskopie. Phys. Z. 45 (1944) 314–326. mikroskopie und der U Borries, B. von, and Ruska, H.: Schrifttum der Elektronenmikroskopie I and II. Z. Wiss. Mikrosk. & Mikrosk. Tech. 60 (1951/2) 103–112 and 212–243. Borries, B. von, and Ruska, H.: Schrifttum der Elektronenmikroskopie III, IV, V, and VI. Z. Wiss. Mikrosk. & Mikrosk. Tech. 61 (1952/3) 106–120, 253–257, 306–319, and 448–460. Borries, B. von, and Ruska, H.: Schrifttum der Elektronenmikroskopie VII. Z. Wiss. Mikrosk. & Mikrosk. Tech. 63 (1956/8) 243–256. Bravman, J. C., Anderson, R., and McDonald, M. L. (eds.): Specimen Preparation for Transmission Electron Microscopy of Materials (Materials Research Society, Pittsburgh PA 1988). Record of a Symposium, Boston MA, 3 December 1987. MRS Symp. Proc. vol. 115. Brederoo, P. (ed.): EM Conferentie Leiden, Programma en Abstracts Najaarsvergadering van de Nederlandse Vereniging voor Electronenmicroscopie, Collegebouw, Gorlaeus Laboratoria, Rijksuniversiteit Leiden, 10–11 November 1977 (New Rhine Publishers, Leiden 1977). See in particular ‘‘Electronenmicroscopie in Nederland, en historische tentoonstelling’’ pp. 1–37 with contributions by W. van Iterson, B. J. Spit, E. M. Jan, J. E. Mellema, J. B. Le Poole, A. C. van Dorsten (two contributions), P. van Zuylen, P. F. Elbers, W. C. de Bruijn, H. B. Haanstra, E. F. J. van Bruggen, M. G. van Heel, H. F. Cohen, and G. A. C. Veeneman. Brederoo, P.: Introduction [to the historical contributions to the 25th Anniversary Meeting of the Netherlands Society for Electron Microscopy]. Ultramicroscopy 3 (1978) 112. Breton, B. C., McMullan, D. and Smith, K. C. A. (eds.): Sir Charles Oatley and the scanning electron microscope. Adv. Imaging and Electron Phys. 130 (2004). Broers, A. N.: High resolution electron beam fabrication: a brief review of experimental studies that began at Cambridge University in 1962. J. Microscopy 139 (1985) 139–152. Broglie, L. de (ed.): L’Optique Electronique, Re´unions d’Etudes et de Mises au Point tenues en mai–juin 1945 sous la pre´sidence de Louis de Broglie (Editions de la Revue d’Optique The´orique et Expe´rimentale, Paris 1946).

364

HAWKES

Brown, P. D., McMullan, D., Mulvey, T., and Smith, K. C. A.: On the origins of the first commercial transmission and scanning electron microscopes in the UK. Proc. Roy. Microsc. Soc. 31 (1996) 161–166. Bru´, L.: The early days of electron microscopy in Spain. Microscopy & Analysis (1992), No. 31, 31–33. Bruaux, A.: Un microscope e´lectronique de construction belge. Ses applications industrielles. In Vandermeerssche (1954) 327–336. Bru¨che, E.: 10 Jahre Entwicklung [des Elektronen-Mikroskops der AEG], Jahrbuch der AEGForschung 7 (1940) 2–8. ¨ bermikroskops mit elektrostatischen Linsen. Bru¨che, E.: Zur Entwicklung des Elektronen-U VDI-Z 85 (1941) 221–228. Bru¨che, E.: Zum Entstehen des Elektronenmikroskops. Phys. Z. 44 (1943a) 176–180. ¨ bermikroskop und seine Entwicklung. Die Kameradschaft, Bru¨che, E.: Das Elektronen-U Werkzeitschrift der Betriebsgemeinschaft AEG 9 (1943b) 166–167. Bru¨che, E.: 25 Jahre Elektronenmikroskop. Naturwissenschaften 44 (1957) 601–610. Bru¨che, E.: Forschungsreise durch das Gebiet der Elektronenoptik. Optik 24 (1966/7) 290–295. Bruggen, E. F. J. van: 25 years electron microscopy in molecular biology. Ultramicroscopy 3 (1978) 123–125. Buchwald, J. Z., and Warwick, A. (eds.): Histories of the Electron, the Birth of Microphysics (MIT Press, Cambridge MA and London 2001). Busch, H., and Bru¨che, E. (eds.): Beitra¨ge zur Elektronenoptik (Barth, Leipzig 1937). Buseck, P. R. (ed.): ‘‘Minerals and reactions at the atomic scale: transmission electron microscopy,’’ Proceedings of the 18th Short Course, sponsored by the Mineralogical Society of America, Hueston Woods State Park, College Corner OH, 23–25 October 1992. Reviews in Mineralogy 27 (1992). Calbick, C. J.: Historical background of electron optics. J. Appl. Phys. 15 (1944) 685–690. Calbick, C. J.: An autobiography. EMSA Bull. 18 (1988) No. 1, 49–56. With an introduction by J. H. Reisner on p. 49. Castillo de Maruenda, E., Rojas Moro´n, N., and Lo´pez Milla, A. (eds.): La Microscopi´a Electro´nica en el Peru´ (Consejo Nacional de Ciencia y Tecnologı´a [CONCYTEC], Lima 1996). Chapman, J. N., and Craven, A. J. (eds.): Quantitative Electron Microscopy, Proceedings of the 25th Scottish Universities Summer School in Physics, Glasgow, 7–27 August 1983 (Scottish Universities Summer Schools in Physics, Edinburgh 1984). Charles, D.: The art of survival, New Scientist 16 March 1996 [Re von Ardenne]. Chayut, M.: J. J. Thomson: the discovery of the electron and the chemists. Ann. Sci. 48 (1991) 527–544. Cohen, A. L.: Recollections of a rather late early electron microscopist. EMSA Bull. 17 (1987) No. 2, 46–56. With an introduction by J. H. Reisner on p. 46. Cohen, A.L, and Steever, R. G. E.: An early American electron microscope. Proc EMSA 29 (1971) 4–5. Colliex, C., and Isaacson, M. (guest eds.): Electron-beam-induced Spectroscopies with High Spatial Resolution, Proceedings of the NSF/CNRS Workshop, Centre Paul Langevin, Aussois, 28 February–5 March 1988. Ultramicroscopy 28 (1989) Nos. 1–4. Cosslett, V. E. (ed.): Bibliography of Electron Microscopy (Arnold, London 1950). Cosslett, V. E: The early history of the Electron Microscopy and Analysis Group. In Electron Microscopy and Analysis. Proceedings of the 25th Anniversary Meeting of the Electron Microscopy and Analysis Group of the Institute of Physics (Nixon, W. C., ed.; Institute of Physics, London 1971) 1–5. Reprinted in Phys. Bull. 22 (1971) 339–341. Cosslett, V. E.: Electron microscopy: retrospect and prospect. A personal history of the subject. Proc. Roy. Microsc. Soc. 14 (1979) 101–107.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

365

Cosslett, V. E.: The development of electron microscopy and related techniques at the Cavendish Laboratory, Cambridge, 1946–1979. Part 1, 1946–1960. Part 2, 1961–1979. Contemp. Phys. 22 (1981) 3–36 and 147–182. Cosslett, V. E.: Origins of electron microscopy in Britain. EMSA Bull. 12 (1982) No. 2, 11–15. With editorial comment by J. H. Reisner on p. 15. Cosslett, V. E.: Random recollections of the early days. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 23–61. Cosslett, V. E.: Early metallurgical applications of the electron microscope in Cambridge. Micron Microsc. Acta 17 (1986) 87–90. Cosslett, V. E.: Early history of the International Federation of Societies of Electron Microscopy. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 3–20. Couteaux, R.: La Socie´te´ Franc¸aise de Microscopie Electronique a trente ans. Bref rappel de ses de´buts et de son histoire. Bull. Soc. Franc¸aise Microsc. Electron. (1989) No. 4, 4–8. Couteaux, R.: The French Society of Electron Microscopy. Its beginnings and its history. Biol. Cell 80 (1994) 85–87. Dahl, P. F.: Flash of the Cathode Rays: a History of J. J. Thomson’s Electron (Institute of Physics Publishing, Bristol and Philadelphia 1997). Das Gupta, N. N. (ed.): Electron Microscopy in Life Science. Abstracts of papers read at an International Symposium, University of Calcutta, 1–5 December 1969 (Palit Laboratory in Physics, University of Calcutta 1969). Das Gupta, N. N., and De, M. L.: The electron microscope—its past and present. Indian J. Hist. Sci. 3 (1968) 25–41. Davis, E. A., and Falconer, I. J.: J. J. Thomson and the Discovery of the Electron (Taylor & Francis, London 1997). Delong, A.: Early history of electron microscopy in Czechoslovakia. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 63–101. Delong, A.: On the development of electron microscopes in Brno. EUREM-12, Brno (2000) vol. 3, I3–I6. Deutsch, K.: Some notes on the development of electron microscopy. NTM: Z. Geschichte Naturwiss. Technik Medizin 1 (1960) No. 4, 45–49. Disko, M. M., Ahn, C. C., and Fultz, B. (eds.): Transmission Electron Energy Loss Spectroscopy in Materials Science. Shaped by a Symposium held at the TMS Annual Meeting, New Orleans LA, 17 February 1991 (Minerals, Metals & Materials Society, Warrendale PA, 1992). Doane, F.: EM pioneer wives—Florence Hillier and Helen Prebus. Microsc. Soc. Canada Bull. 20 (1992) No. 4, 6–9. Doane, F. W., Simon, G. T., and Watson, J. H. L.: Canadian Contributions to Electron Microscopy. An historical account of the development of the first electron microscope in North America and the first 20 years of the Microscopical Society of Canada/Socie´te´ de Microscopie du Canada (Microscopical Society of Canada, Toronto 1993). Donelli, G., Merli, P. G., Pasquali Ronchetti, I., and Valdre`, U.: La microscopia elettronica in Italia: stato attuale e prospettive di sviluppo. Ann. Ist. Superiore di Sanita` 18 (1982) 153–161; Electron microscopy in Italy. Proc. Roy. Microsc. Soc. 20 (1982) 249–253. Dorsten, A. C. van: The beginning of electron microscopy at Philips. Some historical notes. Ultramicroscopy 3 (1978) 115–118. Drummond, D. G. (ed.): ‘‘The Practice of Electron Microscopy.’’ J. Roy. Microsc. Soc. 70 (1950) Part 1, 1–141. Dupouy, G.: Electron microscopy at very high voltages. Adv. Opt. & Electron Microsc. 2 (1968) 167–250.

366

HAWKES

Dupouy, G.: Megavolt electron microscopy. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 103–165. Dupouy, G., Perrier, F., and Durrieu, L.: L’observation de la matie`re vivante au moyen d’un microscope e´lectronique fonctionnant sous tre`s haute tension. C. R. Acad. Sci. Paris 251 (1960) 2836–2841. Dyck, D. van: Electron microscopy in Belgium. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 67–78. ECBO: Mikroskope und Zellbiologie in drei Jahrhunderten, Three Centuries of Microscopes and Cell Biology, Exhibition for the Second International Congress on Cell Biology, Berlin, 31 August– 5 September 1980. Organized by P. Giesbrecht, D. E. Olins, and E. Ruska; Catalogue by J. Lemmerich and H. Spring; Foreword by P. Giesbrecht, H.-G. Schweiger, and W. W. Franke. Elbers, P. F.: The development of ultramicrotomy. Ultramicroscopy 3 (1978) 119–120. Elder, H. Y. (ed.): MICRO 90. Proceedings of the Royal Microscopical Society Conference, London, 2–6 July 1990. Trans. Roy. Microsc. Soc. 1 (1990). Elder, H. Y.: Scottish Annual Microscopy Symposia Series. Proc. Roy. Microsc. Soc. 31 (1996) 179–180. Elektronenmikroskopie: Gru¨ndungsprotokoll der Deutschen Gesellschaft fu¨r Elektronenmikroskopie. 40 Jahre DGE. Elektronenmikroskopie Nr. 1 (1990) 31–34. Everhart, T. E.: Persistence pays off: Sir Charles Oatley and the scanning electron microscope. J. Vac. Sci. Technol. B14 (1996) 3620–3624. Falconer, I.: Corpuscles, electrons and cathode rays: J. J. Thomson and the ‘‘discovery’’ of the electron. Brit. J. Hist. Sci. 20 (1987) 241–276. Feffer, S. M.: Arthur Schuster, J. J. Thomson and the discovery of the electron. Hist. Stud. Phys. Biol. Sci. 20 (1989) 33–61. Ferna´ndez-Mora´n, H.: Cryo-electron microscopy and ultramicrotomy: reminiscences and recollections. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 167–223. Fisher, R. M.: Highlights in the development of electron microscopy in the United States: a bibliography and commentary of published accounts and EMSA records. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 347–382. Fisher, R. M., and Imura, T.: New applications and extensions of the unique advantages of HVEM for physical and materials research. Ultramicroscopy 3 (1978) 3–18. Fitzgerald, A. G., Storey, B. E., and Fabian, D. (eds.): Quantitative Microbeam Analysis, Proceedings of the 40th Scottish Universities Summer School in Physics, Glasgow, 16 August–4 September 1992 (Scottish Universities Summer Schools in Physics, Edinburgh and Institute of Physics Publishing, Bristol and Philadelphia 1993). Fraassen, B. van: The Scientific Image (Oxford University Press, Oxford 1980). [Frequently cited for the discussion on ‘‘observability.’’] Franklin, U. M., Weatherly, G. C., and Simon, G. T.: A history of the first north American electron microscope. ICEM-9, Toronto 3 (1978) 4–17. Freundlich, M. M.: Origin of the electron microscope. Science 142 (1963) 185–188. Freundlich, M. M.: Electron microscope. In The Encyclopedia of Physics (Besanc¸on, R., ed.), 354–362 (Van Nostrand–Reinhold, New York, 3rd ed. 1985). [Contains very brief historical notes.] Freundlich, M. M.: The history of the development of the first high-resolution electron microscope. Remembering the Knoll research team at the Technical University, Berlin, 1927–1934. EMSA Bull. 24 (1994) No. 1, 405–415. Fujita, H. (ed.): In Situ Experiments with High Voltage Electron Microscopes, Proceedings of the International Symposium on ‘‘Behavior of lattice imperfections in materials–in situ experiments with high voltage electron microscopes (HVEM),’’ Suita Campus, Osaka

ELECTRON OPTICS AND ELECTRON MICROSCOPY

367

University, 18–20 November 1985 (Research Center for Ultra-high Voltage Electron Microscopy, Osaka University 1985). Fujita, H.: History of Electron Microscopes (Issued by Business Center for Academic Societies Japan, Tokyo 1986). Fujita, H., Ura, K., and Mori, H. (guest eds.): ‘‘New Directions and Future Aspects of High Voltage Electron Microscopy,’’ Proceedings of the Second Osaka International Symposium on High-voltage Electron Microsocpy, Osaka, 8–10 November 1990. Ultramicroscopy 39 (1991) Nos. 1–4. Fullam, E. F.: Reminiscences. EMSA Bull. 16 (1986) No. 2, 35–43. With an introduction by J. H. Reisner. Gabor, D.: Die Entwicklungsgeschichte des Elektronenmikroskops. Elektrotech. Z. 78 (1957) 522–543. Gabor, D.: Ernst Ruskas grosses Lebenswerk. Optik 24 (1966/7) 370–374 [60th birthday tribute]. Gabor, D.: The history of the electron microscope, from ideas to achievements. ICEM-8, Canberra 1 (1974) 6–12. Giesbrecht, P.: Ernst Ruska and the reconstruction of the first electron microscope. Proc. Roy. Microsc. Soc. 16/1 (1981) 48–49. Gilroy, J., Hart, R. K., Weber, M., and Kopta, J. (eds.): Proceedings of the AMU–ANL Workshop on High-voltage Electron Microscopy, June 13–July 15 1966. Argonne National Laboratory Report ANL-7275 (1966). Gloede, W.: Vom Lesestein zum Elektronenmikroskop (Verlag Technik, Berlin 1986). Goodchild, D. J., and Dowell, W. C. T.: The beginnings of electron microscopy in Australia. Micron Microsc. Acta 17 (1986) 101–105. Goodman, P. (ed.): Fifty Years of Electron Diffraction (Reidel, Dordrecht, Boston and London 1981). Grivet, P.: The French electrostatic electron microscope (1941–1952). In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 225–274. Gru¨newald, H.: Zur Entstehungsgeschichte des Elektronenmikroskops mit elektromagnetischen Linsen. Technikgeschichte 43 (1976) 213–222. See Ruska (1977). Gru¨mm, H., and Schiske, P.: Reminiscences of Walter Glaser, In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 59–66. Gu¨nther, J. R.: History of the Swiss Society for Optics and Electron Microscopy, in History of Electron Microscopy in Switzerland (Gu¨nther, J. R., ed.) 205–213 (Birkha¨user, Basel, Boston and Berlin 1990). Hacking, I.: Do we see through a microscope? Pacific Philos. Quarterly 62 (1981) 305–322. Reproduced with little change as Chapter 11, ‘‘Microscopes,’’ of Representing and Intervening (University Press, Cambridge 1983) pp. 186–209. Haguenau, F.: Electron microscopy in France in the domain of biology. A brief survey of the early period. Biol. Cell 80 (1984) 89–90. Haguenau, F.: Electron microscopy in France, early findings in the life sciences. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 93–100. Haguenau, F., Hawkes, P. W., Hutchison, J., Satiat–Jeunemaıˆtre, B., Simon, G. T., and Williams, D. B.: Key events in the history of electron microscopy. Microsc. Microanal. 9 (2003) 96–138. Hall, C. E.: Recollections from the early years. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 275–296. Hansl, R.: A short history of Canalco. EMSA Bull. 17 (1987) No. 1, 47–49. Hashimoto, H.: Direct imaging of atomic processes in crystals: some personal steps towards this goal. J. Roy. Microsc. Soc. 18 (1983) 298–308. Hashimoto, H.: Atom images and IFSEM affairs in Kyoto, Osaka and Okayama. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 597–633.

368

HAWKES

Hashimoto, H.: Scenery in the World observed by a Researcher of the Microworld—sketches made during the past 23 years (1999). Hawkes, P. W. (ed.): Image Processing and Computer-aided Design in Electron Optics. (Academic Press, London and New York 1973). Hawkes, P. W.: Electron image processing: a survey. Computer Graphics & Image Proc. 8 (1978) 406–446. Hawkes, P. W.: An account of electron optics in the Cavendish laboratory and related developments. J. Microscopy 117 (1979) 39–53. Hawkes, P. W.: Electron image processing: 1978–1980. Computer Graphics & Image Proc. 18 (1982a) 58–96. Hawkes, P. W.: Bibliography of publications on magnetic electron lens properties. In Magnetic Electron Lenses (P. W. Hawkes, ed.) pp. 421–448 (Springer, Berlin & New York 1982b). Hawkes, P. W. (ed.): ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985a). Hawkes, P. W.: Complementary accounts of the history of electron microscopy. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985b) 589–618. Hawkes, P. W. (1992). Electron image processing: 1981–1990. J. Computer-Assisted Microsc. 4, 1–72. Hawkes, P. W. (1993). Electron image processing: 1981–1990. J. Computer-Assisted Microsc. 5, 255. Hawkes, P. W. (ed.) (1994). Selected Papers on Electron Optics. SPIE Milestone Series, vol. MS94 (SPIE Optical Engineering Press, Bellingham WA. Hawkes, P. W. (1996a). Highlights of the IFSEM congresses, electron optics. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96, 405–412. Hawkes, P. W. (1996b). Conference proceedings and conference abstracts. Appendix to ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96, 849–873. Hawkes, P. W. (1997a). La pre´histoire de la SFME. Bull. Soc. Franc¸aise Microsc. No. 2, 3–4. Hawkes, P. W. (1997b). Electron Microscopy and Analysis: The first 100 years. In Proceedings EMAG 97 (Rodenburg, J. M., ed.), pp. 1–8 (Institute of Physics Publishing, Bristol and Philadelphia). Hawkes, P. W.: A century of electrons and electron instruments. Microscopy & Analysis (1998), No. 1, pp. 9–12. Hawkes, P. W.: Electron microscopy at the millenium. EUREM-12, Brno (2000), vol. 3, I7–I12. Hawkes, P. W.: Formation and processing of the electron image—new attitudes to electron microscopy. In Proc. 5th Multinational Congress on Electron Microscopy (Dini, L. and Catalano, M., eds.) pp. 1–12 (Rinton Press, Princeton NJ 2001). Hawkes, P. W., and Kasper, E.: Principles of Electron Optics (Academic Press, London & San Diego, vols 1 and 2 1989, vol. 3 1994, paperback edition with addenda 1996). Henstra, S.: 35 jaar elektronenmicroscopisch onderzoek in Wageningen. Landbouw & Onderzoek Part 1 (1986) 20–24. Heydenreich, J., Luppa, H., Recknagel, A. and Stiller, D.: Electron microscopy in the former German Democratic Republic. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 171–180. Hibi, T. (1985). My recollections of the early history of our work on electron optics and the electron microscope, in ‘‘The Beginnings of Electron Microscopy.’’ Adv. Electron. & Electron Phys. Supplement 16, 297–315. Higashi, N: A brief survey covering the development of the electron microscopy in Japan. ICEM-6, Kyoto (1966) vol. 1, 1–2.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

369

Higashi, N.: A brief survey covering the development of electron microscopy in Japan. J. Electron Microsc. 16 (1967) 2–4. Hillebrand, R., Scheerschmidt, K., Neumann, W., Werner, P., and Pippel, A.: Bildinterpretation in der Hochauflo¨sungs-Elektronenmikroskopie. Numerische und optische Verfahren zur Interpretation elektronenmikroskopischer Hochauflo¨sungsabbildungen kristalliner Objekte. Beitra¨ge zur Forschungstechnologie 11 (1984). Hillier, J.: Remarks before the Burton Society of Electron Microscopists, Toronto, Canada, 28 October 1960. EMSA Bull. 16 (1986) No. 1, 45–54. Hillier, J.: Early EM research and a career offshoot. Proc. Microsc. Soc. Canada 22 (1995) 6–9. Hirsch, P., Howie, A., Nicholson, R. B., Pashley, D. W., and Whelan, M. J.: Electron Microscopy of Thin Crystals (Butterworth, London 1965, 1967). Inspired by the Summer School held in Cambridge in July 1963. Hirsch, P. B.: Electron microscopy in materials science—a historical perspective. In The Electron, Proceedings of the International Centennial Symposium on the Electron, Churchill College, Cambridge 15–17 September 1997 (Kirkland, A. and Brown, P. D., eds.) 64–81 (IoM Communications, London 1998). Hitachi: Publication on the Hitachi Electron Microscope (1959). Anthology of papers from Hitachi. Hobbs, L. W., Westmacott, K. H., and Williams, D. B. (eds.): Materials Problem Solving with the Electron Microscope, Proceedings of a Materials Research Society Symposium, Boston MA, 2–4 December 1985 (Materials Research Society, Pittsburgh PA 1986). Hoppe, W., and Mason, R. (eds.): ‘‘Unconventional Electron Microscopy for Molecular Structure Determination,’’ Proceedings of a Conference held in Alpbach (Austria), 28 March–2 April 1976. Advances in Structure Research by Diffraction Methods 7 (1979). Hoppe, W., Mo¨llenstedt, G., Perutz, M. F., and Ruska, E. (guest eds.): ‘‘Methoden zur Untersuchung der atomaren Struktur von Biogenen Makromoleku¨len,’’ Diskussionstagung der Deutschen Bunsen-Gesellschaft fu¨r Physikalische Chemie, Hirschegg, 15–18 March 1970. Ber. Bunsen-Ges Phys. Chem. 74 (1970), No. 11, 1089–1216. Ho¨rl, E. M.: The Austrian Society for Electron Microscopy. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 55–57. Howatson, A. F.: The Burton Society of Electron Microscopists. Microsc. Soc. Canada Bull. 10 (1982) No. 2, 9–11. Huang, L.-y.: The construction of commercial electron microscopes in China. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996a) 805–847. Huang, L.-y.: Recollections of early electron microscope construction in China [in Chinese]. J. Chinese Electron Microsc. Soc. 15 (1996b) 344–352. Humphreys, C. J.: Electron microscopy and analysis: the future. In Proceedings EMAG 97 (Rodenburg, J. M., ed.) pp. 31–34 (Institute of Physics Publishing, Bristol and Philadelphia 1997). Hutchison, J. L.: Highlights of the IFSEM congresses, materials science. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 393–404. Huxley, H. E., and Klug, A. (eds.): ‘‘A Discussion on new developments in electron microscopy with special emphasis on their application in biology,’’ Royal Society, London, 12–13 March 1970. Phil. Trans. Roy. Soc. London B261 (1971), No. 837, 1–230. Imura, T., and Fisher, R. M. (eds.): New Applications and Extensions of the Unique Advantages of HVEM for Physical and Materials Research. Proceedings of the US–Japan HVEM Seminar, Honolulu, 6–10 December 1976 (Laboratory of Metal Physics, Department of Metallurgy, Nagoya University, Nagoya 1977). Ipohorski, M.: Electron microscopy in Latin America. Ultramicroscopy 3 (1978) 97. Iterson, W. van: Electron microscopy in the Netherlands, earliest developments. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 271–285.

370

HAWKES

JEOL: The World through the Electron Microscope (1959–1973). A series of albums, published in three series: Metals (1959), which then became Metallurgy (1962, 1965, 1968, 1971); Chemistry (1960, 1963, 1966, 1969, 1972); Biology (1961, 1964, 1967, 1970, 1973). Jervis, P.: Innovation in electron-optical instruments—two British case histories. Res. Policy 1 (1971/2) 174–207. Johansen, B. V., and Høglund, S.: Report on a Symposium on contrast problems in transmission electron microscopy, organized by S. Høglund and B. V. Johansen, at Uppsala University, Sweden, 10 January 1975. Ultramicroscopy 1 (1975) 83–87. Jouffrey, B. (ed.): Me´thodes et Techniques Nouvelles d’Observation en Me´tallurgie Physique, PerrosGuirec, 5–17 September 1969 (Socie´te´ Franc¸aise de Microscopie Electronique, Paris 1972). Jouffrey, B.: Electron microscopy in France, the development for physics and materials sciences. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 101–130. Jouffrey, B., Bourret, A., and Colliex, C. (eds.): Cours de l’Ecole de Microscopie Electronique en Science des Mate´riaux, Bombannes, 21 September–3 October 1981 (Editions du CNRS, Paris 1983). Kanaya, K.: Reminiscences of the development of electron optics and electron microscope instrumentation in Japan. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 317–386. Kinder, E.: Professor Dr Ernst Kinder erinnert sich. . . Elektronenmikroskopie Nr.2 (1990) 31–33. Knoll, M.: Dankesworte fu¨r die Verleihung der Ehrenmitgliedschaft der DGE. Mikroskopie 23 (1968) 70. Kohl, W. H., Hall, C. E., Prebus, A. F., Hillier, J., and Watson, J. H. L.: Recollections. ICEM-9, Toronto 3 (1978) 18–25. Komoda, T., Ito, K., Shimadzu, S.-i., Kamogawa, H., Tashiro, Y., Oyama, A., Ichikawa, A., Watanabe, Y., Hosaka, Y., Hirano, T., Fujita, H., and Yada, K.: Electron microscopes and microscopy in Japan. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 653–789. Kontron Seminar-Bericht: ‘‘200-kV Elektronenmikroskopie,’’ Hoˆtel Bachmair, Rottach–Egern, November 1971 (Kontron, Eching 1971). Kontron Seminar-Bericht: ‘‘Hochspannungs-Elektronenmikroskopie’’ (Hampel, P. and Wilkens, M., eds.), Atlantis-Hoˆtel, Zu¨rich, 1972 (Kontron, Eching 1972). Kontron Seminar-Bericht: ‘‘Ho¨chstauflo¨sung in der Elektronenmikroskopie’’ (Mo¨llenstedt, G., ed.), Hilton Hoˆtel, Munich, July 1973 (Kontron, Eching 1973). Kontron Seminar-Bericht: ‘‘200 kV Elektronenmikroskopie im Rasterverfahren’’ (Mo¨llenstedt, G., ed.), Munich, 3–4 March 1975 (Kontron, Eching 1975a). Kontron Seminar-Bericht: ‘‘Biomolecular Electron Microscopy’’ (Kleinschmidt, A. K., ed.), Munich 1975 (Kontron, Eching 1975b). Kosso, P.: Dimensions of observability. Brit. J. Philos. Sci. 39 (1988) 449–467. Krakow, W., and O’Keefe, M. (eds.): Computer Simulation of Electron Microscope Diffraction and Images. Proceedings of a Topical Symposium held at the TMS Annual Meeting in Las Vegas, NV, 28 February–3 March 1989 (Minerals, Metals & Materials Society, Warrendale PA 1989). Krakow, W., Smith, D. A., and Hobbs, L. W. (eds.): Electron Microscopy of Materials, Proceedings of a Materials Research Society Symposium, Boston MA, 14–17 November 1983 (North-Holland, New York, Amsterdam and Oxford 1984). Kruger, D. H., Schneck, P., and Gelderblom, H. R.: Helmut Ruska and the visualisation of viruses. Lancet 355 (2000) 1713–1717; reprinted in Elektronenmikroskopie (2002) No. 18, 33–38.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

371

Kruit, P., Schapink, F. W., Geus, J. W., Verkleij, A. J., and Hulstaert, C. E.: Electron microscopy in the Netherlands, developments since the 1950s. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 287–299. Kunkle, G. C.: Technology in the seamless web: ‘‘success’’ and ‘‘failure’’ in the history of the electron microscope. Technology & Culture 36 (1994) 80–103. Ku¨pfmu¨ller, K.: Zur Geschichte des Elektronenmikroskops. Phys. Z. 45 (1944) 47–51. Lambert, L.: Some anecdotes around and with Colonel Ernst August Ruska. Ultramicroscopy 20 (1986) 337–340. Lambert, L., and Mulvey, T.: Ernst Ruska (1906–1988), designer extraordinaire of the electron microscope: a memoir. Adv. Imaging & Electron Phys. 95 (1996) 3–62. Langrish, J., Gibbons, M., Evans, W. G., and Jevons, F. R.: Wealth from Knowledge, Studies of Innovation in Industry (Macmillan, London & New York 1972). See Chapter 7: ‘‘Cambridge Scientific Instruments: Stereoscan electron microscope,’’ pp. 167–175. Lenz, F.: Erfindung des Elektronenmikroskops. Physik in unserer Zeit 17 (1986) 188–189. Lenz, F.: Bodo von Borries–ein Ru¨ckblick. Elektronenmikroskopie Nr. 12 (1995) 35–38. Lenz, F.: Towards atomic resolution. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 791–803. Lenz, F.: Preface to Memorial Issue for the late Professor Gottfried Mo¨llenstedt. J. Electron Microsc. 47 (1998) 367–369. Lenz, F.: 50 Jahre Deutsche Gesellschaft fu¨r Elektronenmikroskopie. Elektronenmikroskopie (1999) No. 18, 15–18. ¨ ber die Entwicklung der Elektronenmikroskopie in Holland. Optik 24 (1966/7) Le Poole, J. B.: U 375–382. Le Poole, J. B.: The early days of electron microscopy in Holland. Ultramicroscopy 3 (1978) 113–115. Le Poole, J. B.: Early electron microscopy in the Netherlands. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 387–416. Lichte, H.: Gottfried Mo¨llenstedt and his electron biprism: four decades of exciting and challenging electron physics. J. Electron Microsc. 47 (1998) 387–394. Lin, Z.-r. (ed.): Proceedings of International Sixth Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 24–27 October 1995, Part A, Electron Microscopy (Peking University Press, Beijing 1995). Lin, Z.-r. (ed.): Proceedings of Seventh International Beijing Conference and Exhibition on Instrumental Analysis, Shanghai, 14–17 October 1997, Part A, Electron Microscopy (Peking University Press, Beijing 1997). Lin, Z.-r. (ed.): Proceedings of Eighth International Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 25–28 October 1999, Part A, Electron Microscopy (Peking University Press, Beijing 1999). Lin, Z.-r. (ed.): Proceedings of Ninth International Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 17–20 October 2001, Part A, Electron Microscopy (Peking University Press, Beijing 2001). Mangclaviraj, V.: Electron microscopy in Thailand. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) xxiii–xxv. Mannheimer, W. A.: BIBMIC, a bibliography of books relating to materials microscopy. Metallography 22 (1989) 123–176. Mannheimer, W. A.: BibMic: a bibliography of books relating to materials microscopy. Materials Characterization 36 (1996) 105–149. Manton, I.: Recollections in the history of electron microscopy. Proc. Roy. Microsc. Soc. 13 (1978) 45–57.

372

HAWKES

Marton, L.: Note historique sur les premie`res anne´es de la microscopie e´lectronique. Bull. Cl. Sci. Acad. Roy. Belg. 46 (1960) 119–121. Marton, L.: Historical background of image formation. Lab. Invest. 6 (1965) 739–746. Marton, L.: Early History of the Electron Microscope (San Francisco Press, San Francisco 1968; 2nd ed., 1994). Marton, C., and Sass, S. (eds.): A bibliography of electron microscopy. I. J. Appl. Phys. 14 (1943) 522–531. Marton, C., and Sass, S. (eds.): A bibliography of electron microscopy. II. J. Appl. Phys. 15 (1944) 575–579. Marton, C., and Sass, S. (eds.): A bibliography of electron microscopy. III. J. Appl. Phys. 16 (1945) 373–378. Marton, C., Sass, S., Swerdlow, M., Bronkhorst, A. van, and Meryman, H. (eds.): Bibliography of Electron Microscopy. National Bureau of Standards Circular 502 (1950). Matthias, A.: Bemerkung zur Entstehung des Elektronemikroskops. Phys. Z. 43 (1942) 129–130. Maunsbach, A. B., and Afzelius, B. A.: The development of electron microscopy in Scandinavia. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 301–321. Maunsbach, A. B., and Thomas, G.: IFSEM 1995: objectives, organization, and functions. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 21–35. Maurice, F., Meny, L., and Tixier, R. (eds.): Microanalyse et Microscopie Electronique a` Balayage, Lectures delivered at a Summer School, St-Martin-d’He`res (Ise`re), 11–16 September 1978 (Editions de Physique, Orsay 1979); English edition, Microanalysis and Scanning Electron Microscopy (Editions de Physique, Orsay 1980). McKinley, T. D., Heinrich, K. F. J., and Wittry, D. B. (eds.): The Electron Microprobe. Proceedings of the Symposium sponsored by the Electrothermics and Metallurgical Division, the Electrochemical Society, Washington DC, 12–15 October 1964 (Wiley, New York and London 1966). McMullan, D.: Recollections of the early days of SEM in the Cambridge University Engineering Department, 1948–1953. J. Microscopy 139 (1985) 129–138. McMullan, D.: von Ardenne and the scanning electron microscope. Proc. Roy. Microsc. Soc. 23 (1988) 283–288. Reprinted in Hawkes (1994) pp. 171–176. McMullan, D.: SEM—past, present and future. J. Microscopy 155 (1989) 373–392. McMullan, D.: The prehistory of scanned image microscopy. Part 1, Scanned optical microscopes. Part 2, The scanning electron microscope. Proc. Roy. Microsc. Soc. 25 (1990) 127–131 and 189–194. McMullan, D.: Scanning electron microscopy 1928–1965. Proc. MSA 51 (1993) 762–763. McMullan, D.: Scanning electron microscopy 1928–1965. Scanning 17 (1995) 175–185. Merli, P. G., and Antisari, M. V. (eds.): Microscopia Elettronica in Trasmissione e Tecniche di Analisi di Superfici nella Scienza dei Materiali. Lectures delivered at a Seminar held at Castro Marina (Lecce), 16–21 September 1985 (ENEA, Rome 1985). Merli, P. G., and Antisari, M. V. (eds.): Electron Microscopy in Materials Science, Proceedings of the International School on Electron Microscopy in Materials Science, Mesagne (Brindisi), 7–19 October 1991 (World Scientific, Singapore, River Edge NJ, London and Hong Kong 1992). Merli, P. G., Calestani, G., and Vittori-Antisari, M. (eds.): ‘‘Microscopy, Spectroscopy, Holography and Crystallography with Electrons.’’ Adv. Imaging & Electron Phys. 123 (2002). Meryman, H. T.: A memoir of my early days in electron microscopy. EMSA Bull. 17 (1987) No. 1, 45–47. With an Introduction by J. H. Reisner. See Hansl (1987).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

373

Messerschmidt, U., Appel, F., Heydenreich, J., and Schmidt, V. (eds.): Electron Microscopy in Plasticity and Fracture Research of Materials, Proceedings of an International Symposium, Holzhau, 8–13 October 1989 (Akademie-Verlag, Berlin 1990). Mo¨llenstedt, G.: My early work on convergent-beam electron diffraction. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 585–595. Mo¨llenstedt, G.: Zum 90. Geburtstag von Manfred von Ardenne. Optik 105 (1997) 137–138. Mo¨llenstedt, G.: The history of the electron biprism. In Introduction to Electron Holography (Vo¨lkl, E., Allard, L. F. and Joy, D. C., eds.) 1–15 (Kluwer/Plenum, New York, Dordrecht, London and Moscow, 1999). Mo¨llenstedt, G.: Obituary of Gottfried Mo¨llenstedt (1913–1997) by A. Tonomura. J. Electron Microsc. 47 (1998) 363–364. Morniroli, J.-P.: Diffraction Electronique en Faisceau Convergent a` Grand Angle (LACBED); Applications aux De´fauts Cristallins (Socie´te´ Franc¸aise des Microscopies, Paris 1998); Largeangle Convergent-beam Electron Diffraction (LACBED). Applications to Crystal Defects (Socie´te´ Franc¸aise des Microscopies, Paris 2002). Morniroli, J.-P. (ed.): Microscopie des De´fauts Cristallins, cours de l’Ecole The´matique, SaintPierre d’Ole´ron, 7–18 May 2001 (Socie´te´ Franc¸aise des Microscopies, Paris 2001). Mulvey, T.: Origins and historical development of the electron microscope. Brit. J. Appl. Phys. 13 (1962) 197–207. Mulvey, T.: The history of the electron microscope. In Historical Aspects of Microscopy (Bradbury, S and Turner, G., eds.) 201–227 (Heffer, Cambridge); also published as Proc. Roy. Microsc. Soc. 2 (1967) 201–227. Mulvey, T.: The development of electron optics. Proc EMSA 29 (1971) 2–3. Mulvey, T.: Forty years of electron microscopy. Phys. Bull. 24 (1973) 147–154. Mulvey, T.: The industrial development of the electron microscope by the Metropolitan Vickers Electrical Company and AEI Limited. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 417–442. Mulvey, T.: Electron microscopy—the U.K. involvement. Micron Microsc. Acta 17 (1986) 67–75. Mulvey, T.: Ernst Ruska and his electron microscope. Proc. Roy. Microsc. Soc. 22/2 (1987) 103–109. Mulvey, T.: The electron microscope: the British contribution. J. Microscopy 155 (1989) 327–338. Mulvey, T. (ed.): ‘‘The Growth of Electron Microscopy.’’ Adv. Imaging & Electron Phys. 96 (1996). Mulvey, T., and Laak–Tijssen, D. J. J. van de: Jan Bart Le Poole (1917–1993). Pioneer of the electron microscope and particle optics. Adv. Imaging & Electron Phys. 115 (2000) 287–354. National Bureau of Standards: Electron Physics. Proceedings of the NBS Semicentennial Symposium on Electron Physics, Washington, 5–7 November 1951. National Bureau of Standards Circular 527 (1954). Newberry, S.: Electron microscopy, the early years. EMSA Bull. 15 (1985a) No. 1, 39–50. With an introduction by J. H. Reisner. Newberry, S.: Electron microscopy, the early years, Part II of a scientific memoir. EMSA Bull. 15 (1985b) No. 2, 39–47. Newberry, S.: Electron microscopy, the early years, Part III of a scientific memoir. EMSA Bull. 16 (1986) No. 1, 44–45. With an introduction by J. H. Reisner. Newberry, S. P.: EMSA and its People, the First Fifty Years (Electron Microscopy Society of America, 1992). Newberry, S. P.: Early electron microscopy remembered through the candlelight of history. EMSA Bull. 23 (1993) No. 2, 146–152.

374

HAWKES

Niedrig, H.: Die Anfa¨nge der Elektronenmikroskopie: Eine Berliner Entwicklung. Physik und Didaktik 15 (1987) No. 1, 52–64. Niedrig, H.: The early history of electron microscopy in Germany. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 131–147. Nyrikov, V. G., and Kuschnier, U. M.: Electron microscope and electron diffraction cameras developed in the U.S.S.R. APEM-1, Tokyo (1956) 30–34. NYSEM: The International Bibliography of Electron Microscopy, I: 1950–1955 (New York Society of Electron Microscopy, New York 1959). NYSEM: The International Bibliography of Electron Microscopy, II: 1956–1961 (New York Society of Electron Microscopy, New York 1962). Oatley, C. W.: The early history of the scanning electron microscope. J. Appl. Phys. 53 (1982) R1–R13. Oatley, C. W., Nixon, W. C., and Pease, R. F. W.: Scanning electron microscopy. Adv. Electron. & Electron Phys. 21 (1965) 181–247. Oatley, C. W., McMullan, D., and Smith, K. C. A.: The development of the scanning electron microscope. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 443–482. Ogura, M.: Historia de la microscopı´a electro´nica en Venezuela. Memorias del 1er. Congreso Atla´ntico de Microscopı´a Electro´nica/The Proceedings of the First Atlantic Congress of Electron Microscopy, edited by E. Valiente (Editorial Venezolana, Me´rida, 1992) p. 52. Ottensmeyer, P.: Ge´rard T. Simon: reflections on his retirement. Microsc. Soc. Canada Bull. 23 (1995) No. 4, 8–10. Palucka, T.: Seeing serendipity: James Hillier’s view on the invention of the electron microscope. MRS Bull. 27 (2002) 995–999. Pantic´, V. R. (ed.): Osnovi Elektronske Mikroskopije [Basic Electron Microscopy]. (Naucˇna Knjiga, Beograd 1962). Parsons, D. F. (ed.): ‘‘Short Wavelength Microscopy,’’ Papers presented at a Workshop held by the New York Academy of Sciences, The Roosevelt Hotel, New York, 23–25 February 1977. Ann. New York Acad. Sci. 306 (1978). Peven, D. R., and Gruhn, J. D.: The development of electron microscopy. Arch. Pathol. Lab. Med. 109 (1985) 683–691. Porter, K. R.: Electron microscopy and cell biology. EMSA Bull. 23 (1993) No. 2, 159–162. Prebus, A. F.: Events at the University of Alberta and the University of Toronto, leading to the first North American electron microscope. Micron 29 (1998) 289–292. Preston, R. D.: Early days in electron microscopy. Proc. Roy. Microsc. Soc. 18 (1983) 37–40. Qian, L.-z., and Kuo, K.-h.: The Chinese Electron Microscopy Society. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1998) xxix–xxxii. Qing, L.: Zur Fru¨hgeschichte des Elektronenmikroskops (Verlag fu¨r Geschichte der Naturwissenschaften und der Technik, Stuttgart 1995). Ramsauer, C. (ed.): Zehn Jahre Elektronenmikroskopie. Ein Selbstbericht des AEG ForschungsInstituts (Springer, Berlin 1941); 2nd ed., Elektronenmikroskopie, 1942; 3rd ed., 1943. Rang, O.: Zur Elektronenmikroskopie kurz nach 1945—Erinnerungen eines Zeitzeugen. Elektronenmikroskopie No. 5 (1992) 25–32. Rasmussen, N.: Facts, artifacts, and mesosomes: practicing epistemology with the electron microscope. Stud. Hist. Phil. Sci. 24 (1993) 227–265. Rasmussen, N.: Making a machine instrumental: RCA and the wartime origins of biological electron microscopy in America, 1940–1945. Stud. Hist. Phil. Sci. 27 (1996) 311–349. Rasmussen, N.: Picture Control: the Electron Microscope and the Transformation of Biology in America, 1940–1959 (Stanford University Press, Stanford 1998a).

ELECTRON OPTICS AND ELECTRON MICROSCOPY

375

Rasmussen, N.: Instruments, industrialists and the specificity of ‘‘influence.’’ In Lo¨wy, I. and Gaudillie`re, J.-P. (eds.), The Invisible Industrialist: Manufacturers and the Construction of Scientific Knowledge (Macmillan, London 1998b). Rasmussen, N.: What moves when technologies migrate? ‘‘Software’’ and hardware in the transfer of biological electron microscopy to postwar Australia. Technology & Culture 40 (1999) 47–73. Rathbun, M. E., Eastwood, M. J., and Arnold, O. M. (eds.): Supplementary bibliography of electron microscopy. J. Appl. Phys. 17 (1946) 759–762. Reed, R.: Some recollections of electron microscopy in Britain from 1943 to 1948. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 483–500. Reimann, B. E. F.: Post war experiences of an electron microscopist in the Old World (1947–1957). EMSA Bull. 19 (1989) No. 1, 46–50. With an introduction by J. H. Reisner. Reisner, J. H.: An historical perspective on the origins of the EMSA. EMSA Bull. 11 (1981a) No. 1, 13–15. Reisner, J. H.: The formation of the EMSA. EMSA Bull. 11 (1981b) No. 2, 16–19. Reisner, J. H.: The EMSA’s first year. EMSA Bull. 12 (1982) No. 1, 11–14. Reisner, J. H.: Early meetings of the EMSA. EMSA Bull. 13 (1983) No. 2, 12–18. Reisner, J. H.: EMSA’s growing pains. EMSA Bull. 14 (1984) No. 1, 14–17. Reisner, J. H.: An early history of the electron microscope in the United States. Adv. Electron. Electron Phys. 73 (1989a) 133–231. Reisner, J. H.: Reflections. EMSA Bull. 19 (1989b) No. 2, 47–51. Reisner, J. H.: Reflections. EMSA Bull. 20 (1990) No. 1, 42–43 and No. 2, 49–53. Reisner, J. H.: Reflections. EMSA Bull. 21 (1991) No. 1, 63–64 and No. 2, 51–54. Reisner, J. H.: Introduction—a glimpse at the Anderson papers. EMSA Bull. 22 (1992) No. 3, 50–58. Rempfer, G. F.: Electrostatic electron optics in the 1940s and today. EMSA Bull. 23 (1993) No. 2, 153–158. Rickerby, D. G., Valdre`, G., and Valdre`, U (eds.): Impact of Electron and Scanning Probe Microscopy on Materials Research (Kluwer, Dordrecht, Boston and London 1999) NATO ASI Series E, vol. 364. Riecke, W. D.: Neuere Entwicklungen in der Elektronenmikroskopie. Z. Instrumentenkde 75 (1967) 341–352 and 377–389. Riecke, W. D.: Elektronenmikroskopie. VDI-Z. 110 (1968) No. 7, 291–295. Riecke, W. D.: Elektronenmikroskopie. VDI-Z. 113 (1971) No. 3, 223–228. Rochow, T. G.: Significant early days in industrial electron microscopy. EMSA Bull. 13 (1983) No. 1, 10–18. With an introduction by J. H. Reisner, p.10. Ru¨denberg, R.: The early history of the electron microscope. J. Appl. Phys. 14 (1943) 434–436. Rudenberg, H. G.: The 50 years before the electron microscope: from electron to electron lens— Hans Busch and the ‘‘Go¨ttingen Group,’’ EMSA Proc. 50 (1992) 1084–1086. Rudenberg, F. H, and Rudenberg, H. G.: Early contributions to the electron microscope: the relevant work and patents of Reinhold Ru¨denberg, EMSA Proc. 50 (1992) 1086–1087. Rudenberg, H. G., and Rudenberg, F. H.: Reinhold Ru¨denberg as a physicist—his contribution and patent on the electron microscope, traced back to the ‘‘Go¨ttingen electron group.’’ EMSA Bull. 24 (1994) No. 4, 572–580. Ru¨hle, M., Phillipp, F., Seeger, A., and Heydenreich, J.: Selected papers from an International Workshop on High-voltage and High-resolution Electron Microscopy, Stuttgart, 21–24 February 1994. Ultramicroscopy 56 (1994) Nos. 1–3, 1–232. See also Interface Science 3 (1995) No. 2 (article by Dahmen et al., p. 125) and No. 3 (articles by Hofmann and Ernst, p. 201, Ruvimov et al., p. 211 and Hesse et al., p. 221).

376

HAWKES

Ruska, E.: Geschichte des Elektronenmikroskops. APEM-1, Tokyo (1956a) 19–29. Ruska, E.: Deutsche Forschungsstellen und Arbeiten zur Entwicklung der Elektronenoptik, des Elektronenmikroskops und der Pra¨parationstechnik. APEM-1, Tokyo (1956b) 35–49. Ruska, E.: 25 Jahre Elektronenmikroskopie. Elektrotech. Z. A78 (1957a) 531–543. Ruska, E.: Elektronenmikroskopie. VDI-Z. 99 (1957b) 604–606. Ruska, E.: Elektronenmikroskopie. VDI-Z. 101 (1959) 1106–1109. Ruska, E.: Elektronenmikroskopie. VDI-Z. 104 (1962) 836–839. Ruska, E.: Elektronenmikroskopie. VDI-Z. 106 (1964) 811–814. Ruska, E: Zur Vor- und Fru¨hgeschichte des Elektronenmikroskops. ICEM-8, Canberra, 1 (1974) 1–5. Ruska, E: Gegendarstellung zu dem Aufsatz von Dr.-Ing. Heinrich Gru¨nwald ‘‘Zur Entstehungsgeschichte des Elektronenmikroskops mit magnetischen Linsen,’’ In: Technikgeschichte 43 (1976) S. 213 bis 222. Technikgeschichte 44 (1977) 86–87. Ruska, E.: Die fru¨he Entwicklung der Elektronenlinsen und der Elektronenmikroskopie. Acta Historica Leopoldina, No. 12 (1979) 136 pp. English translation: The Early Development of Electron lenses and Electron Microscopy (Hirzel, Stuttgart 1980); also published as Supplement 5 to Microscopica Acta. Ruska, E.: Die Entstehung des Elektronenmikroskops. Archiv der Geschichte der Naturwissenschaften No. 11/12 (1984) 525–551; English translation: The emergence of the electron microscope. Connection between realization and first patent application, documents of an invention. J. Ultrastruct. Mol. Struct. Res. 95 (1986) 3–28. Ruska. E.: Supplement to the paper: ‘‘The electron microscope: priority of invention,’’ published in The Microscope 30: 1, 11–15 (1982) by R. Borasky. Microscope 32 (1985), No. 2, 154–156. Ruska, E.: The development of the electron microscope and of electron microscopy. In Les Prix Nobel 1986, pp. 58–83 (Almqvist and Wiksells International, Stockholm 1987). This is Ruska’s Nobel Prize address and is preceded by autobiographical notes (in German), pp. 54–57; the address is reproduced in Angewandte Chemie Int. Ed. in English 26 (1987) 595–605 and in EMSA Bull. 18 (1988), No. 2, 53–61, where it is preceded by tributes by J.-P. Revel, S. Short, J. H. Reisner and B. Tesche. For the German text, see Das Entstehen des Elektronenmikroskops und der Elektronenmikroskopie, Phys Bla¨tter 43 (1987), No. 7, 271–281. Schimmel, G.: The history of the German Society for Electron Microscopy. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 149–170. Schimmel, G.: Das physikalische Laboratorium Mosbach, Erinnerungen an Prof. Dr. Ernst Bru¨che. Elektronenmikroskopie Nr.16 (1998) 26–32. Schiske, P. (1968). Zur Frage der Bildrekonstruktion durch Fokusreihen. EUREM-4, Rome 1 (1968) 145–146. Republished as Image reconstruction by means of focus series. J. Microscopy 207 (2002) 154–155. Schulze, D.: Die Elektronenmikroskopie im Deutschen Museum. Elektronenmikroskopie No. 5 (1992a) 57–60. Schulze, D.: Elektronenmikroskope aus dem Osten. Komerzielle Entwicklungen in der DDR. Elektronenmikroskopie No. 6 (1992b) 32–40. Schulze, D.: Von Louis de Broglie zu Erwin Schro¨dinger. Elektronenmikroskopie No. 10 (1994) 23–26. Schulze, D.: Problemfall Elektron-Entdeckungsgeschichte. Elektronenmikroskopie No. 13 (1996) 22–23. Schulze, D.: Elektronemikroskopie diesseits und jenseits—Erinnerungen. Elektronenmikroskopie (1999) No. 18, 18–22.

ELECTRON OPTICS AND ELECTRON MICROSCOPY

377

Sharma, R., Gai, P. L., Gajdardziska–Josifovska, M., Sinclair, R., and Whitman, L. J. (eds.): In Situ Electron and Tunneling Microscopy of Dynamic Processes (Materials Research Society, Pittsburgh PA 1996). Record of a Symposium, Boston MA, 27–30 November 1995. MRS Symp. Proc. vol. 404. Simon, G. T., and Doane, F. W.: The history of electron microscopy in Canada. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 79–91. Sinclair, R., Smith, D. J., and Dahmen, U. (eds.): High Resolution Electron Microscopy of Defects in Materials (Materials Research Society, Pittsburgh PA 1990). Record of a Symposium, San Francisco CA, 16–18 April 1990. MRS Symp. Proc. vol. 183. Sitte, P.: Das Elektronenmikroskop im Biobereich—Ru¨ckblick und Ausblick. Elektronenmikroskopie (1999) No. 18, 22–23. Smith, T. A.: About the electron microscope. EMSA Bull. 14 (1984) No. 2, 19–22. With an introduction by J. H. Reisner, p. 19. Smith, D. J. (ed.): Atomic Resolution Microscopy of Surfaces and Interfaces (Materials Research Society, Pittsburgh PA 1997). Record of a Symposium, Boston MA, 3–5 December 1996. MRS Symp. Proc. vol. 466. Smith, K. C. A.: Charles Oatley: pioneer of scanning electron microscopy. In Proceedings EMAG 97 (Rodenburg, J. M., ed.) pp. 9–16 (Institute of Physics Publishing, Bristol and Philadelphia 1997). Smith, K. C. A.: The origins and history of a versatile electron optical instrument combining facilities for SEM, TEM and x-ray point projection microscopy. In The Electron, Proceedings of the International Centennial Symposium on the Electron, Churchill College, Cambridge 15–17 September 1997 (Kirkland, A. and Brown, P. D., eds.) 552–556 (IoM Communications, London 1998). Spence, J. C. H.: New electron microscopies–the state of the art. In Proceedings EMAG 97 (Rodenburg, J. M., ed.) pp. 23–30 (Institute of Physics Publishing, Bristol and Philadelphia 1997). Spit, B. J.: Twenty-fifth anniversary of the Netherlands Society for Electron Microscopy. Ultramicroscopy 3 (1978) 112–113. Springford, M. (ed.): Electron, a Centenary Volume (Cambridge University Press, Cambridge 1997). Stadhouders, A. M., and Haelst, U. J. G. M. van: 25 years electron microscopy and biomedical sciences. Ultramicroscopy 3 (1978) 120–123. Stewart, A. D. G.: The origins and development of scanning electron microscopy. J. Microscopy 139 (1985) 121–127. Stewart, A. D. G., and Snelling, M. A.: A new scanning electron microscope. EUREM-3, Prague A (1964) 55–56. Sugata, E.: Electron Microscopes of Japan 1936–1965, a Historical Survey (Maruzen, Tokyo 1968). Su¨sskind, C.: L. L. Marton, 1901–1979. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 501–523. Swann, P. R. (guest ed.): The High-voltage Symposium at emcon 1972. J. Microscopy 97 (1973) Nos. 1/2, 1–268. Tadano, B: Progress of electron microscopes in Hitachi. Hitachi Rev. 1 (1953), No. 1, 19–30. Takano Moro´n, J.: La microscopı´a en el Peru´. In Castillo de Maruenda et al. (1996), pp. 39–42. Tani, Y.: Welcome to the Tokyo Conference. APEM-1 (1956a) 8–9. Tani, Y.: General survey of development and present state of electron microscope in Japan. APEM-1, Tokyo (1956b) 50–57. Thomas, G. T., and Washburn, J. W. (eds.): Electron Microscopy and Strength of Crystals, Proceedings of the international conference on ‘‘The Impact of Transmission Electron

378

HAWKES

Microscopy on Theories of the Strength of Crystals,’’ Berkeley, 5–8 July 1961 (WileyInterscience, New York 1963). Thomas, G., Fulrath, R. M., and Fisher, R. M. (eds.): Proceedings of the Fifth International Materials Symposium ‘‘The Structure and Properties of Materials–Techniques and Applications of Electron Microscopy,’’ University of California at Berkeley, 13–17 September 1971 (University of California Press, Berkeley and London 1972). Thompson, M. N. (ed.): Proceedings of a Symposium on the Contribution of Electron Microscopy to Society, Eindhoven, 24–25 September 1987. Philips Electron Optics Bulletin No. 128 (n.d.). Tonomura, A., Allard, L. F., Pozzi, G., Joy, D. C., and Ono, Y. A. (eds.): Electron Holography (Elsevier, Amsterdam, Lausanne, New York, Oxford, Shannon and Tokyo 1995). Trowbridge, C. W.: Computing Electromagnetic Fields for Research and Industry: major achievements and future trends. IEEE Trans Magnetics 32 (1996) 627–630. Turner, P. S., and Vesk, M.: Electron microscopy in Australia. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 39–53. Ulrich’s International Periodicals Directory (published annually by Bowker, New Providence NJ; also on line from DIALOG). Valdre`, U. (ed.): Electron Microscopy in Material Science, Lectures delivered at the International School of Electron Microscopy, Erice, 4 –18 April 1970 (Academic Press, New York and London 1971). Valdre`, U.: Electron microscopy in Italy. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 193–215. Valdre`, U., and Ruedl, E. (eds.): Electron Microscopy in Materials Science, Lectures delivered at the Third Course of International School of Electron Microscopy, Erice, 7–21 April 1973 (Commission of the European Communities, Brussels and Luxemburg 1976), 4 vols (vols. 1 and 2 edited by UV and ER, vols. 3 and 4 by ER and UV). Vandermeerssche, G. (ed. and publisher): Rapport Europees Congres Toegepaste Electronenmicroscopie, Centrum Electronenmicroscopie, Rijksuniversiteit Gent, 7–10 April 1954 (Presses des Ateliers Publicitaires, Uccle–Bruxelles 1954). ´ .: The Hungarian Group for Electron Microscopy. In ‘‘The Growth Vira´gh, S., and Csana´dy, A of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 181–191. Vukovic´, J. B.: Twenty years of electron microscopy in Yugoslavia, 1969–1989. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) xxv–xxviii. Watson, J. H. L.: Microscopy and music–dimly related. Year Book New York Microsc. Soc. 10 (1964/5) 51–64. Watson, J. H. L.: Toronto ‘‘daze.’’ Microsc. Soc. Canada Bull. 2 (1974) 10–18. Watson, J. H. L.: The origins of electron microscopy on the North American continent at the University of Toronto. EMSA Bull. 22 (1992) No. 2, 97–106 and MSA Bull. 23 (1993) No. 2, 139–145. Watson, J. H. L.: Very early Canadian electron microscopists and their microscopy. Proc. Microsc. Soc. Canada 22 (1995) 2–5; Micron 26 (1995) 483–488. Weichan, C.: 1950–1960: a decade from the viewpoint of an applications laboratory. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 525–556. Wells, O. C.: Scanning Electron Microscopy (McGraw–Hill, New York 1974) pp. 13–19. Wilska, A. P.: Recollections and experiences from fifteen years of constructing electron microscopes. Ann. Acad. Sci. Fenn., Ser. A5, 106 (1964) 1–15. Wilska, A. P.: A series of articles on the life and work of Alvar Pietari Johannes Wilska (1911–1987), including extensive material on his activities in electron microscopy, is to be

ELECTRON OPTICS AND ELECTRON MICROSCOPY

379

found in the serial La¨a¨ka¨ri ja Vapaa Aika (1991) No. 2B (published by Culturamed, Klaukkala, Finland). Witcomb, M. J. and Wolfe-Coote, S. A.: Electron microscopy in Southern Africa. In ‘‘The Growth of Electron Microscopy,’’ Adv. Imaging & Electron Phys. 96 (1996) 323–345. Wittry, D. B.: The early history of the Microbeam Analysis Society (MAS). Microbeam Anal. 1 (1992) 1–2. Wolff, O.: From the cathode-ray oscillograph to the high-resolution electron microscope. In ‘‘The Beginnings of Electron Microscopy,’’ Adv. Electron. & Electron Phys. Supplement 16 (1985) 557–587. Wolpers, C.: Electron microscopy in Berlin 1928–1945. Adv. Electron. & Electron Phys. 81 (1991) 211–229. Wyckoff, R. W. G.: Entwicklungslinien in der amerikanischen Elektronenmikroskopie. Z. Wiss. Mikrosk. Mikrosk. Technik 62 (1954) 87–91. Ximen J.-y. and Lenz, F.: The two outstanding pioneers of electron microscope—Ruska and Borries ninetieth birthday anniversary [in Chinese]. J. Electron Devices 18 (1995) No. 1, 6–14; reproduced in J. Chinese Electron Microsc. Soc. 14 (1995) 235–242. Cf. Elektronenmikroskopie No. 11 (1995) 40. Yada, K., Tadano, B., Fukami, A., Adachi, K., Asakura, K., Kanaya, K., and Ura, K.: The growth of electron microscopy in Japan. Adv. Imaging & Electron Phys. 96 (1996) 217–269. Yao, J.-e.: The development of ultramicroscopy in China [in Chinese]. J. Chinese Electron Microsc. Soc. 15 (1996) 353–370. Yao J.-e. (ed.): Proceedings of Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 18–21 November 1985, Vol. 1, Electron Microscopy (BCEIA, Beijing 1985). Yao J.-e. (ed.): Proceedings of Second Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 20–23 October 1987, Part. 1, Electron Microscopy (BCEIA, Beijing 1987). Yao J.-e. (ed.): Proceedings of Third Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 27–30 October 1989, Vol. 1, Electron Microscopy (BCEIA, Beijing 1989). Yao J.-e. (ed.): Proceedings of International Fourth Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 18–24 October 1991, Vol. A, Electron Microscopy (Science Press, Beijing and New York 1991). Yao J.-e. (ed.): Proceedings of International Fifth Beijing Conference and Exhibition on Instrumental Analysis, Beijing, 9–12 October 1993, Vol. A, Electron Microscopy (Peking University Press, Beijing 1993). Yoshii, Z.: Pioneers of electron microscopy at Washington State University and their work. Bull. Yamaguchi Med. School 17 (1970) No. 3/4, 191–200. [Predates Cohen and Steever (1971).]

This Page Intentionally Left Blank

Index

Amplitude images, 9–10 Amplitude normalization, 166 Analytical Electron Microscopy conferences, 211, 345 Angular figures, segmentation of, 184 Anisotropic materials, in application of Maxwell’s equations, 60 Annealing, as effective means of obtaining high-performance QPM device, 49 Applied electric field general theorem for capacitance variation under, 11–14 polarization direction changed in PZT thin film, 39 Approximation, as issue in finite volume methods, 95 Arbeitskreis fu¨r Elektronenmikroskopischen Direktabbildung von Oberfla¨chen (Arbeitskreis EDO, AEDO), 239–241 Argentinian national congresses of electron microscopy, 292 Armenian Electron Microscopy Society, 221–222 Artificial neuron, 126–127 Asia-Pacific Conference on Electron Microscopy (APEM), 217–218 Asia-Pacific regional meetings, for electron microscopy, 207 Asian national congresses of electron microscopy, 298–317 Asociacio´n Mexicana de Microscopı´a, 296 Atomic force microscopy (AFM), 7 limiting read speed of cantilever, 50 Audio signal identification, 125, 156, 162–163

90-degree a-c domain, in BaTiO2 single crystal, 8 180-degree c-c domain boundary, one-dimensional calculated images of, 19

A Absorbing layer, 85 Acceleration signals, 185 as filtering parameter, 181 GCD filter offering best performance for, 182 Accuracy, advantages of higher-order methods for, 69, 90 Acoustics, fictitious and overlapping grid methods in, 80–84 Acronyms, relevant to electron microscopy, 354–357 Adaline neural network model, 125–126, 128–130 Adaptive linear element, 128–129 Admissibility condition, in wavelet transform functions, 146 Advances in Imaging & Electron Physics, 264 African national congresses of electron microscopy, 317–319 Algorithms, used in preprocessing techniques, 125 Alternating applied electric field, resonance frequency varying as function of, 28–29 Alternating capacitance variation, ratio of, to static value of capacitance without time dependence, 5 Alternating electric field, capacitance variation with, 3–6

381

382 Australian Academy of Science, 319–320 Australian national congresses of electron microscopy, 319–321 Australian Society of Electron Microscopy, 320 Autoassociative memories, 134–135 combining with wavelet transform in face recognition, 186–197 Autoassociators, 192–195 as pattern completion devices, 191 Automated sleep scoring, 168 Average power spectrum density (APSD) patterns, 165 Axial illumination, for missiles, 92

B Back-propagation algorithm, slower than radial basis function networks, 144 Back-propagation learning rule, 126, 137–141 Backward differentiation method, 113 Balkan meetings, for electron microscopy, 207 Band-pass filter, 170 Basis functions number equal to patterns in data set, 142 in RBF training, 144 BaTiO3 limitations as a recording medium, 50 as single-crystal material for nanodomain switching, 50 Beitra¨ge zur elektronenmikroskopischen Direktabbildung and Analyse von Oberfla¨chen (BEDAO), 233 Beitra¨ge zur Forschungstechnologie, 234 Belgische Comiteit voor Elektronenmicroscopie, 224–226

INDEX

Belgische Vereniging voor Microscopie, 224–226 Best basis selection algorithm, 150 Bethe, Heinz, 234 Biczysko, Wieslawa, 254 Biocell, 208, 219 Biological neuron, 126–127 Biological Research, 295 Biology, pattern recognition in, 124 Biorthogonal wavelet system, 147 Bit storage, in ultrahigh-density recording systems, 49 Blemishes, recognition and classification of, 172 Block mean values, 167 Body-conforming grids, 86 block-structured, 94 computing scattering by a PEC cylinder, 110 Bosnian national congresses of electron microscopy, 281–282 Boundaries difficulties with accurate representations of, 79 inability of global spectral methods to handle, 85 local modifications of staggered grid schemes close to, 74 need to approximate in Yee scheme, 60 stable and accurate treatment of metallic, 76 unsupported by commercial grid generation software, 116 Boundary conditions, 62, 67 along a dielectric interface, 66 global, 116 periodic, 67 vectorial nature of, in applying Maxwell’s equations, 60 Boundary fluxes, penalizing, 106 Boundary integration operator, 105 Brain disorder classification, 168, 169–170

INDEX

Brazilian national congresses of electron microscopy, 292–294 British Institute of Physics, Electron Microscopy Group, 268, 270–274 British Joint Committee for Electron Microscopy (BJCEM), 268 British national congresses of electron microscopy, 268–277 Broad-band signals applications of Maxwell’s equations to, 60 random effects complicating, 60 Brown, K. L., 326 Bulletin de Microscopie Applique´, 230 Bulletin of the Academy of Sciences of the USSR (Physical Series), 209–210 Burton Society of Electron Microscopists, 289 Business Center for Academic Societies Japan, 209

C c-c domain boundary, detectable by 333 imaging, 35 Calibration data, for tip sensitivity, 28 Canadian national congresses of electron microscopy, 289–292 Cantilever, 6 electroconductive AFM type, 19 electrolytic polishing of, 7 Cantilever-type SNDM difficulties of quantitative measurement with, 20 measured data of nonlinear dielectric constant by, 30 model of, 26, 27 for measuring nonlinear dielectric constant, 29 quantitative measurement of linear dielectric constant, 25–28 of nonlinear dielectric constant, 28–30

383 vs. needle-type, 17–20 Capacitance separating, in cantilever-type SNDM, 26 static, 5 Capacitance variation with alternating electric field, 3 caused by nonlinear dielectric response, 7 first-order, 13 general theorem for, under applied electric field, 11–14, 31–32 perpendicular to polarization direction, 37 relationship to frequency variation, 29 Capacitance variation susceptibility, 15 as function of linear relative dielectric constant, 16 Carl Zeiss, 235 Carrier frequency, 23 Carrier frequency shift, 21–22 Cartesian unit vectors, 95 Cavity problem, 74 Cell-averaged values, reconstructing local solutions using, 96 Centers determining in neural network training process, 142 setting as origin, 161 Central difference approximation, to spatial derivative, 68 Ceskoslovenska´ Spolecnost pro Elektronovou Mikroskopii (Czechoslovak Society for Electron Microscopy) (CSEM/CSMS), 228–230 Charge distribution, 63 Charged-Particle Optics (CPO) conferences, 326 Chebyshev Gauss Lobatto nodes, 87, 90 Chebyshev polynomials, 87

384 Chilean national congresses of electron microscopy, 294–295 Chinese Electron Microscopy Society, 299–301 Chinese-Japanese joint symposia on electron microscopy, 208–209, 300–301 Chinese national congresses of electron microscopy, 299–301 Chinese Society for Electron Microscopy (Taiwan), 315 Cieciura, Leszek, 254 Class discrimination, retaining information for, as step in preprocessing, 125 Classical fourth-order compact scheme, 77 Classical integral-based solution techniques, limitations of, 60 Classical phase-error analysis, 84 Classification applications eddy current signal analysis, 167 underwater signals, 164–165 wavelets in, 156 Classifier module in pattern recognition, 156 training invariance into, 161 Closed curves, Fourier and wavelet transforms for discrimination of, 161 Clustering method, in RBF neural network training, 143 Clustering parameter, 144 Coefficient of correlation, 189 Coiflet function, 145, 154 Colombian national congresses of electron microscopy, 295 Comite´ Belge de Microscopie Electronique, 224 Comite´ de Sociedades Interamericanas para Microscopı´a Electro´nica (CIASEM), 292 Commercial grid generation software, limitations of, 116

INDEX

Committee of Asia-Pacific Societies of Electron Microscopy (CAPSEM), 218, 298–299 Committee of European Societies of Electron Microscopy (CESEM), 215, 221 Committee of European Societies of Microscopy (CESM), 215 Committee of Interamerican Societies for Electron Microscopy (CIASEM), 208, 219 Compact schemes, 77–80 Compact support, as characteristics of wavelet basis functions, 147 Comparison tests, in filtering handwriting movement signals, 181–183 Complex function, in choice of wavelet function, 151 Complex geometries issues with using higher-order accurate schemes, 84 limitations of global methods with, 86 Complex orthogonal estimation algorithm, 172 COMPUMAG, 211 Computation time, for metallic cavity problem, 71 Computational domain nonoverlapping elements in, 86–90 truncating via Runge-Kutta scheme, 92, 115 Computational efficiencies, issues when using very wide stencils, 84 Computational electromagnetics, slow acceptance of high-order methods in, 61 Computational fluid dynamics, borrowing from, 95 Computational geometry, embedding of, in solving Maxwell’s equations, 60

385

INDEX

Computational resources, dramatic reductions using high-order methods, 61 Computational simplicity, of RBF classifiers, 143 Computed vs. exact solution, 72 Computer science, pattern recognition in, 124 Concurrence matrices, 174 Conductance, 106 Conductivity, of the media, 66 Conference Series of Institute of Physics Publishing, 209 Congruent lithium tantalate (CLT), 51–54 Conservation law, discretization of, 94 Constant of proportionality, 128, 129 Content addressable memory for face images, 187–188 Context gasdynamics, 90 Continuity, of tangential field components, 65 Continuous finite element techniques, 98–104 Continuous wavelet transform definition of, 145 in handwriting movement analysis, 177 Contour recognition, 158 Convergence fourth-order, 71 impact of smoothness of solutions around nonsmooth geometries, 115 Convex spaces, mappable by RBF, 144 Convolution operator, 176 Cost consideration, in production of recording media, 50 Cost function, in wavelet packet functions, 150 Counter propagating waves, 90 Covariance matrix, in principal component analysis, 157 Cross-derivatives, 83

Crossed electric field, used in 311-type measurement, 38 Crystal anisotropy, imaging in ferroelectric and piezoelectric materials, 2 Cuban national congresses of electron microscopy, 295 Cumann Leictreon Mhiocrasco´paithe na hE´ireann, 244–245 Curie point, importance in recording medium, 50 Curl-conforming elements, 100, 101 Nedelec, 102 Curl-curl equations, discretization of, 114 Curl operator, null of, 100 Curved interfaces, stable and accurate treatment of, 76 Curvilinear interfaces, 86, 88 mass lumping to diagonalize mass matrices accurately, 101 Czechoslovak Microscopy Society (Ceskoslovenska´ Mikroskopicka´ Spolecnost) (CSMS), 228–230 Czyrska-Filemonowicz, Aleksandra, 254

D Data compression in image processing, 156 in preprocessing for pattern recognition applications, 125 Data sets, in audio signal processing, 163 Daubechies function, 145, 152–154 in recognition and classification of blemishes, 172 Deferred correction technique, 113 Degraded images, 191–195 restoration of, 156 Degraded stimuli, role in autoassociative memory, 135

386 Degrees of freedom of curl-conforming Nedelec elements, 102 in dispersion-relation-preserving fourth-order explicit schemes, 77 global, in continuous finite element schemes, 101 local, 106 reduced in higher order schemes, 62 Delta rule, 129 Denshikenbikyo [Electron Microscopy], 209, 303 Depth information, estimation of using SNDM, 16 Depth sensitivity for eta SNDM imaging, 32–33 integration region for obtaining, 34 of SNDM, 17, 18 Derivatives classical fourth-order compact scheme for computing, 77 computing for individual elements to high orders, 86–87 evaluating with spectral accuracy, 88 spatial, 88 Deutsche Gesellschaft fu¨r Elektronmikroskopie (DGE, DGEM), 233–237 Dianzi Xianwei Xuebao, 299 Dielectric constant, 5, 22 decreasing after titanium diffusion, 22 Differentiation matrix, in continuous finite element schemes, 99 Diffractive optics problems, 93–94 Diffractive waveguide coupler, 89 Dilation factor, in wavelet transform functions, 146 Dimensional reference magnetic field strength, 64 Discontinuous element schemes, for solving Maxwell’s equations, 63, 104–113

INDEX

Discontinuous fields limitations of global spectral methods with, 86 problematic in multidimensional problems, 73 Discontinuous Galerkin method, 107 Discrete differentiation operator, 105 Discrete Fourier transform (DFT), 158 Discrete stability, 115 in higher-order methods for solving Maxwell’s equations, 63 Discrete wavelet transform (DWT), 145, 147–149 computational intensity of, 154 in eddy current signal analysis, 167 in image processing, 149 Disjoint spaces, mappable by RBF, 144 Dispersion errors addressed by fourth-order explicit schemes, 76–77 of compact schemes, 97 as complicating factor in Yee scheme, 61 Dispersion-relation-preserving (DRP) fourth-order explicit schemes, 76–77 Divergence conserved by compact schemes due to staggered grid, 77 preserving in regions of homogeneous materials, 80 Divergence constraint, 102 Domagala, Wenecjusz, 254 Domain detection device, importance of resolution for ultrahighdensity data storage applications, 50 Domain distribution, of PPLN in depth direction, 48 Domain patterns on PPLN substrate, 45–49 by SNDM, 8 Domain size, of SLT vs. CLT media, 51 Domain structure, depending on film growth conditions, 40

INDEX

Domain wall thickness, of ferroelectric vs. ferromagnetic materials, 49 Downwind positive eigenvalues, 95 Drawing signals segmentation in, 183 wavelet filtering technique in, 178–186 Dreila¨ndertagung, for electron microscopy, 207 Dyadic wavelet transform, 150 Dynamic measuring method of capacitance variation with alternating electric field, 24

E Earthquake prediction applications, 173 EC Lissajous patterns, 167 Ecuadorian national congresses of electron microscopy, 295 Eddy current signal identification, 125, 156 neural network classifier for, 167–168 Edge detection application of wavelets in, 156 compared with tonals in audio signals, 164 in handwriting movement analysis, 179 Edge effects, avoided by Haar wavelet function, 152 Edge elements, 100 Edinburgh Microscopical Society, 270 EEG signal processing, 125 Ehrenwerth, Christel, 233 Eigen and sigular value decomposition, PCA approach to, 19 Eigenvalue decomposition, 187–190 defined, 189 Eigenvectors, 157

387 defined, 189 Elasticity problems, fictitious and overlapping grid methods for solving, 80–84 Electric displacement calculating, 3 first-order variation of, 5 as function of electric field, 30 Electric field, 63 resolution of SNDM as function of, 31 Electric field vector, 39 Electric flux density, 63 Electrical load forecasting, short-term, 157 Electrochemical Society conferences, 330–332 Electroencephalogram classification using wavelet functions, 157 using wavelet transform and neural networks, 168–172 Electrolytic polishing, 7 of tungsten wire as probe tip, 17 Electromagnetic waves, propagation of, described by Maxwell’s equations, 59 Electromagnetics, fictitious and overlapping grid methods for solving problems in, 80–84 Electron, ion, and photon beam conferences, 330–335 Electron and Optical Beam Testing of Integrated Circuits [of Electronic Devices] Conferences, 211 Electron cyclotron resonance (ECR) sputtering, 43, 44 Electron Energy-Loss Spectroscopy and Imaging workshops, 211 Electron Microscope Society of India (EMSI), 301–302 Electron microscopy acronyms relating to, 354–357 African national congresses of, 317–319

388 Electron microscopy (Cont.) Argentinian national congresses of, 292 Armenian national congresses of, 221–222 Asia-Pacific regional congresses of, 217–218 Asian national congresses of, 298–317 Australian national congresses of, 319–321 Balkan regional congresses of, 218–219 Belgian national congresses of, 224–226 Brazilian national congresses of, 292–294 British national congresses of, 268–277 Canadian national congresses of, 289–292 Chilean national congresses of, 294–295 Chinese national congresses of, 299–301 Colombian national congresses of, 295 Croatian national congresses of, 226–228 Cuban national congresses of, 295 Czech, Slovak, and Czechoslovakian national congresses of, 228–230 Ecuadorian national congresses of, 295 European national congresses of, 220–282 European regional congresses of, 215–217 French national congresses of, 230–233 German national congresses of, 233–241 Greek national congresses of, 241–242

INDEX

Hungarian national congresses of, 242–244 Indian national congresses of, 301–302 Indonesian national congresses of, 302–303 international congresses of, 212–215 Irish national congresses of, 244–245 Israeli national congresses of, 245–247 Italian national congresses of, 247–250 Japanese national congresses of, 303–311 Korean national congresses of, 311–313 libraries devoted to, 354 Malaysian national congresses of, 313–314 Mexican national congresses of, 296 Netherlands national congresses of, 250–253 New Zealand national congresses of, 321–322 North American national congresses of, 282–292 Peruvian national congresses of, 296 Philippine national congresses of, 314 Portuguese national congresses of, 257–259 Prominent professionals in, 349–354 regional congresses of, 215–220 Scandinavian national congresses of, 259–263 Singapore national congresses of, 314 Slovenian national congresses of, 263–264 South African national congresses of, 317–319 South American national congresses of, 292–298 South American regional congresses of, 219–220

INDEX

Spanish national congresses of, 264–265 Swiss national congresses of, 265–266 Taiwanese national congresses of, 315 Thai national congresses of, 315–317 thematic meetings of, 322–349 Turkish national congresses of, 267–268 Uruguayan national congresses of, 296 U.S. national congresses of, 282–289 USSR and Russian national congresses of, 277–281 Venezuelan national congresses of, 297–298 Yugoslavian, Serbian, Bosnian, and Herzegovinan national congresses of, 281–282 Electron Microscopy and Analysis Group (EMAG) of the British Institute of Physics, 209 Electron Microscopy Council of the Academy of Sciences of the USSR, 277 Electron Microscopy Group of the Hellenic Society for Biological Sciences, 241–242 Electron Microscopy Section of the Physical Society of the German Democratic Republic, 233 Electron Microscopy Society of America (EMSA), 209, 282–289 Electron Microscopy Society of Malaysia, 313–314 Electron Microscopy Society of Southern Africa (EMSSA), 208–209, 317–319 Electron Microscopy Society of Thailand, 315–317 Electron Probe Analysis Society of America (EPASA), 342–344 Electronics, pattern recognition in, 124 Elektronenmikroskopie, 233–234

389 Elements, connecting in local schemes, 90–92 Embedding schemes, 74, 76 EMSA Bulletin, 282, 283 EMSA Proceedings, 283 EMSI Bulletin, 301 Energy density, 12 Engine vibrations, wavelet transform as best preprocessing technique for, 125 Equidistant grid, 68 abandoning for local schemes, 87–90 Eta311 type measurement, 38 Eta311 type probe, 3 schematic configuration of, 38 EUREM acronym, use of, for European Microscopy Society conferences, 215–216 European and American Microbeam Analysis Society meetings, 211 European Conferences on Electron and Optical Testing of Integrated Circuits/Electronic Devices, 338–340 European Journal of Cell Biology, 222 European Microbeam Analysis Society, 345–347 European Microscopy Society, 218 European regional meetings, for electron microscopy, 207 European Symposia on Reliability of Electron Devices, Failure Physics and Analysis (ESREF), 338–340 European Workshops on Electron Spectroscopic Imaging and Analysis Techniques, 211, 347 Expansion coefficients, computing, 108 Experimental results, of threedimensional measurement technique, 39

390

F Face recognition, 125, 186–197 Far field scattering, accurate prediction of, 113 Fast Fourier transform, 85, 86, 158 Fast linear supervise methods, 144 Feature extraction, 159 in brain disorder classification, 169–170 in eddy current signal analysis, 167 in industrial tool regocnition, 173 invariant, 161 in myoelectric signal processing, 171 in underwater signal classification, 164–165 using invariance properties, 160–162 Feature generator, in sleep stage classification, 168–169 Feature projection methods, 171 Feature vector, 156 Features extraction, 156 Feed-forward networks, 137–139 Ferroelectric data storage, 49–54 ideal characteristics of recording material, 50 superior to ferromagnetic materials, 49 Ferroelectric domain, microscopic observation of area distribution of, 7–10 Ferroelectric materials, microelectronics applications of, 2 Ferroelectric nanodomain engineering system, based on SNDM, 51 Ferroelectric random access memory, 41 Ferroelectric thin films, 41 Ferromagnetic materials, inferior to ferroelectric materials, 49 Fictitious grid methods, 80–84 Fifth Yugoslav Symposium on Electron Microscopy, 226 Film growth process, schematic illustration of, 44

INDEX

Film polarity, importance of determining for engineering purposes, 41 Filter bank, 145, 148–149 Filtering techniques choice of optimal parameters for, 180–181 in handwriting movement analysis and face recognition, 175–176 Financial data, wavelet transform as best preprocessing technique for, 125 Finite computational domains, issues with using higher-order accurate schemes, 84 Finite conductivity, 63 Finite difference schemes, 68 inhibiting acceptance of finite element methods, 97 for solving Maxwell’s equations, 62, 63 Finite-difference time-domain (FDTD) method, 70 Finite element schemes continuous, 98–104 discontinuous, 104–113 slow acceptance of, 97 success in solid and fluid dynamics, 97 Finite impulse response (FIR) filters, 175 Finite volume schemes, reconstructing local solutions using only cell-averaged values in, 96 First Croatian Symposium on Electron Microscopy, 226 Fixed window function, filtering by, 145 Flaw echo location, 157 Fluid dynamics, success of finite element methods in, 97 Fourier coefficients (FDs), 158 Fourier descriptors, 158 in eddy current signal analysis, 167 Fourier transform of Lemarie mother wavelet, 170

INDEX

limitations in time domain, 160 as preprocessing technique, 125 for signal representation and analysis, 145 Fourth-order scheme, 73 semidiscreet for one-dimensional Maxwell’s equations, 78 Fractal properties, wavelet applications in study of, 156 French multinational meetings, for electron microscopy, 207–208 Frequency analysis, 156 Fourier transform methods superior for, 158 Frequency deviation, 23 Frequency modulated (FM) signal, of probe/oscillator, 7 Frequency representation, compared with temporal representation, 176 Frequency variation, relationship to capacitance variation, 29 Frontiers of Electron Microscopy in Materials Science conferences, 211, 348–349

G Gabor transform, 145, 159 Galerkin form, semidiscrete, 102 Gas dynamics, 94–97 connection with electromagnetics, 95 Gauss’ law for change conservation, 64 Gauss’ theorem, 97 Gaussian basis function, 142, 159 Gaussian pulse, in one domain, 83 Gaussian random noise, in face recognition applications, 195–197 Generalization in neural networks, 126 robust with RBF classifiers, 143 Generalized Canny Deriche (GCD) filter, 175, 191

391 in face recognition applications, 187, 192 filtering handwriting movements, 182–183 in handwriting movement analysis, 176–178, 179 as smoothing filter, 185 Generalized Widrow-Hoff rule, 139–141 Geoacoustic signals, 162 Geometric complexity of real-world applications of Maxwell’s equations, 60 SBP methods inadequately addressing, 80 Geometric conformity, not required in cavity problem, 74 Geometric flexibility examples borrowed from gas dynamics, 94–97 in multidomain schemes, 114 as tradeoff for high-order accuracy, 62 Gesellschaft fu¨r Topochemie und Elektronenmikroskopie (GTE), 237–238 Ghost grids, 81 Ghost point approach to twodimensional problems, 83 use with smooth interfaces, 83 Glaser, Walter, 222 Global boundary conditions, 116 Global solutions, assembling from local elements, 90–92 Global spectral method, 82, 84–86 limitations with complex geometries, 86 Greek Society of Electron Microscopy, 241–242 Grid, spatially staggered equidistant, 70 Grid anisotropy, effects on wave propagation, 69 Grid-conforming geometries, 72, 86 Grid functions, vectors of, 78 Groniowski, Janusz, 253–254

392 Ground fault discrimination, 157 Ground/foliage penetrating radars, Maxwell’s equations and, 60

H Haar high-pass filter, 153 Haar scaling function, 151 Haar wavelet function, 145 advantages of, 152 in audio signals identification, 162–163 in recognition and classification of blemishes, 172 Halle Proceedings, 238–239 Handwriting movement analysis, 125, 175 process-oriented approach to, 179 task-oriented approach to, 179 wavelet filtering technique in, 178–186 Hawkes, P.W., 326 Hawkes & Kasper, major proceedings volumes for electron microscopy conferences, 211 Herzegovinan national congresses of electron microscopy, 281–282 Heteroassociative memories, 137 Hidden unit, activation of, 141 High-frequency components, suppressed by scaling function, 148 High-order filter, 94 High-order finite difference schemes, 70–84 accuracy of, 90 fictitious and overlapping grid methods and, 80–81 High-order finite volume schemes, 94–97 development of accurate, 95 High-order schemes, direct construction of, 79–80 High-order spatial discretization schemes, 113

INDEX

High-speed electronics/electro-optics, Maxwell’s equations and, 60 High-Voltage Electron Microscopy (HVEM), conferences, 211, 322–324 Higher-order accurate schemes, 84 Higher-order methods advantageous situations for, 69 in computational electromagnetics, case for, 67 degrees of freedom reduced in, 62 in solutions to Maxwell’s equations, 61 time-stepping and discrete stability in, 63 tradeoff between simplicity and accuracy in, 61 Higher-order neural networks (HONNs), 161 Homogeneous materials, preserving divergence in regions of, 80 Human movement planning and execution, 179 Human perceptual learning, modeling with autoassociative memories, 135 Human speech recognition, wavelet transform as best preprocessing technique for, 125 Hungarian national conferences on electron microscopy, 242–244

I Identification procedure, in audio signal identification, 162–163 Image classification, by content, 125, 157, 172–175 Image compression, 156 Image content, extracting from Landsat images, 173–175 Image processing demand for real-time operations in, 154

393

INDEX

enhanced edge detection in, 156 face recognition applications, 196–197 fractal properties in, 156 and handwriting movement analysis, 179 industrial tool shape recognition, 172–173 pattern recognition in, 124 by wavelet transform, 149 Incident field, 65 Independent component analysis (ICA), 158 Indian national congresses of electron microscopy, 301–302 Indonesian Society of Microscopy and Microanalysis, 302–303 Inductance, 20 Industrial tools, image recognition of, 172–173 Input neurons, 126 Input patterns, associating with responses, 126 Input vector, distance from prototype vector, 141 Instrument recognition, 163 Interfaces local modifications of staggered grid schemes close to, 74 necessity of approximating in Yee scheme, 60 problems with curvilinear, 86 Interior interfaces, inability of global spectral methods to handle, 85 International Conferences on Charged-particle Optics (CPO), 211 International Conferences on X-Ray Optics and Microanalysis (ICXOM), 324–326 International Congresses of Electron Microscopy (ICEM) proceedings of, 212–215 schedule of, 207

International Congresses on X-ray Optics and Microanalysis (ICXOM), 211 International Federation of Electron Microscope Societies (IFEMS), 212 International Federation of Societies for Microscopy (IFSM), 207 International Federation of Societies of Electron Microscopy (IFSEM), 212 congresses and proceedings of, 212–215 member countries of, 213 International Union of Microbeam Analysis Societies, 344–345 Interpolation, exact, 141–142 Invariance, building into classifiers, 161 Irish Electron Microscope Users’ Group, 244–245 Irish Society of Electron Microscopists, 244–245 Israel Society for Microscopy, 245–247 Israel Society of Electron Microscopy, 245–247 Italian national congresses of electron microscopy, 247–250 Iteration number, in Widrow-Hoff learning rule, 130 Izvestiya Akademiya Nauk SSSR (Seriya Fizika), 209

J Japanese Microprocesses and Nanotechnology conferences, 211 Japanese national congresses of electron microscopy, 303–311 Japanese Society of Electron Microscopy (JSEM), 209, 303–311 Jenaer Jahrbuch, 235

394 Johari, O., 328 Journal de Microscopie, 230 Journal of Applied Physics, 209 Journal of Electron Microscopy, 209, 211, 234, 303 Journal of Structural Biology, 210 Journal of the Chinese Electron Microscopy Society, 209, 299 Journal of the Electron Microscopy Society of Thailand, 209, 315 Journal of the Microscopy Society of America, 209, 283 Journal of the Royal Microscopy Society, 269 Journal of Ultrastructure Research, 210, 211 JPEG 2000 standard, based on wavelet lifting scheme, 156

K Karhunen-Loeve transform, 186 Kernel classifiers, RBF neural networks as, 142–143 Kernel estimates, 175, 180 in handwriting movement filters, 182 smoothing handwriting signals with, 179 Kernel regression, 156 Kilarski, Wincenty, 254 Kiselev, N. A., 277 Korean Journal of Electron Microscopy, 209, 311 Korean national congresses of electron microscopy, 311–313 Korean Society of Electron Microscopy, 311

L Lagrange polynomials, 82, 87, 108, 109 Landsat images, neural network classification of, 173–175

INDEX

Large problems continuous finite element scheme limitations for, 104 efficient discretizations of, with higher-order methods, 69 Latin American Society, for electron microscopy, 208 LC lumped constant resonator probe, 6 LC resonator, resonance frequency of, 6 Lead zirconate titanate, in ferroelectric thin films, 49 Leapfrog schemes, second-order, 113 Learned basis functions, 157 Learning rate, in perceptron neural network model, 128 Least mean square learning procedure, 130 Lemarie wavelet, 170 Light wavelength conversion applications, 44–49 LiNbO3 annealing to obtain highperformance QPM device, 49 negative area more easily damaged, 35 ZnO thin films on, 42 Linear activation function, 129 Linear associative memories, principles and architecture of, 134–136 Linear autoassociative memory, 126 training with Widrow-Hoff learning rule, 136 Linear autoassociators, 187–190 Linear dielectric constant distribution of, 27 experimental result of average value of, 27 quantitative measurement using cantilever-type SNDM, 25–28 quantitative measurement using needle-type SNDM, 20–21 Linear dielectric constant distribution, of TiO2-Bi2Ti4O11 ceramic, 29

395

INDEX

Linear discriminant analysis (LDA) classifier, 171 Linear discriminant function, 128 Linear filtering, 175 Linear heteroassociative memory, 126, 136–137 Linear separability, in neural networks, 126 Linear time-frequency representations, 162 LiTaO3 applications in electro-optical, integrated-optical, and piezoelectric devices, 50–51 as candidate for ultrahigh-density storage medium, 51 nonlinear dielectric constant of, 25 one-dimensional image of ZnO thin film on, 43 ZnO thin films on, 42 Local impedance, 106 Local modifications, 76 connecting to form a stable global solution, 86–87 of staggered grid schemes close to boundaries and interfaces, 74 Local schemes, 87–90 Local smooth mappings, limitations of global spectral methods for, 86 Localization in time, not provided by Fourier transform, 158 Localization property, in Gabor and STFT vs. Fourier transforms, 145 Long time integration advantages of higher-order schemes for, 69 correct treatment of material interface necessary for, 83 Lossy materials, in applications of Maxwell’s equations, 60 Low-pass filter, scaling function as, 148

M Magnetic flux density, 63 Magyar Elektronmikroszko´pos Konferencia, 242–243 Malaysian national congresses of electron microscopy, 313–314 Mapping function, costly for exact interpolation procedures, 142 Mass lumping, 101 Mass matrices converting global with finite element schemes, 104 in finite element schemes, 99 local operators for, 105 Material interface difficulty with accurate representations of, 79 field components tangential to, 66–67 inability of finite difference schemes to deal accurately with, 97 inability to enforce physical jump conditions at, 72 pulse undergoing multiple reflections at, 82 Materials, vectors of, 78 Maxwell, James Clerk, 59 Maxwell’s equations, 59 complicating factors in solution of, 60 dispersion-relation-preserving (DRP) fourth-order explicit schemes to solve, 76–77 fourth-order semidiscrete compact scheme for, 78 high-order accurate methods in, 61 high-order finite difference schemes for, 70 in one and two dimensions, 66–67 spatial discretization of, 113 as strongly hyperbolic system, 90 in the time domain, 63–67 Mean square errors (MSE), 181

396 Mechanical oscillations, causing noise in handwriting movement analysis, 180 Medical images, wavelet transform as best preprocessing technique for, 125 Medical signal processing applications, 125, 157 Memory requirements needed to store fields, 69 reducing via spectral methods, 84–94 Metallic boundaries, stable and accurate treatment of, 76 Metallic cavity problem, 71 Mexican hat function, 154, 174 Mexican national congresses of electron microscopy, 296 Meyer function, 14, 145 Micro- and Nanoengineering (MNE) conferences, 211, 335–338 Microanalysis conferences, 342–348 Microbeam Analysis Society (MAS) conferences, 342–344 Microcircuit engineering conferences, 335–338 Microelectronics, applications of ferroelectric materials in, 2 Micromachining applications, 41 Micron, 211 Micron and Microscopica Acta, 210 MicroProcesses and Nanotechnology conferences, 340–342 Microscopı´a Electro´nica y Biologia Celular, 208, 219 Microscopia Elettronica, 248 Microscopical Society of Canada (MSC), 289 Microscopical Society of Ireland, 244–245 Microscopy and Microanalysis, 209, 283, 302–303 Microscopy of Semiconducting Materials (MSM), 269 Microscopy Research and Technique, 211

INDEX

Microscopy Society of America, 209 Microscopy Society of the Philippines, 314 Microskopie, 222 MicroSoM, 313 Military aircraft, electromagnetic scattering by, 112, 113 MLP neural network, 167 in sleep stage classification, 169 Modeling effects of polarization, reflection/ refraction, and diffraction, 67 wave-matter interaction, in solving Maxwell’s equations, 60 Mo¨llenstedt, Gottfried, 234 Morlet function, 145, 154, 155 Mother wavelet, 145, 147 Multi-National Conferences on Electron Microscopy (MCEM), 208, 209, 218–219 Multidimensional problems advantages of higher-order schemes for, 69 limitations in high-order finite difference schemes, 72–73 radial basis function methods for, 141–142 Multidomain formulations, 86–94, 114 Multidomain spectral grid, 89 Multilayer feed-forward networks, architecture and notation of, 137–139 Multilayer neural networks, 137–141 Multilayer perceptron, 165 Multiple linear regressions, in neural networks, 126 Multiresolution analysis, 145, 148–149, 149, 178 adapted to segmentation in handwriting movements, 183 filtering and signal segmentation using, 180–185 Multiresolution wavelet decomposition, 174–175

INDEX

Multiscale-based shape recognition method, 172–173 in handwriting movement analysis, 176–178 Multiscale edges, in face recognition, 190–195 Multiscale GCD operator, 193–195 as edge detector, 191 in face recognition applications, 187 superior generalization performance in face recognition, 196–197 Music, using wavelet transform as preprocessing technique for, 125 Myoelectric signal processing, 125, 157 using wavelet transform and neural networks, 168, 171–172

N Nanodomain character examples, 53 Nanodomain dots inverted, 49 in LiTaO3 single-crystal, 51–54 typical shapes of, 52 Nanodomain engineering tool, SDNM system use as, 3 Nanodomain switching, BaTiO3 as single-crystal material for, 50 Nanoscale ferroelectric domain, on PZT thin film, 9 National Bureau of Standards Circular, 284 National meetings, for electron microscopy, 207 National societies of electron microscopy, 207 Nearest-neighbor values, in compact schemes, 77 Nedelec elements, 102 Nederlandse Vereniging voor Electronenmicroscopie (NVEM), 250–253

397 Nederlandse Vereniging voor Microscopie (NVvM), 250 Needle, of LC resonator, 6 Needle tip, model for calculating nonlinear dielectric signal, 14 Needle-type SNDM quantitative measurement using, 20–25 vs. cantilever-type, 17–20 Netherlands national conferences of electron microscopy, 250–253 Neural computing and learning problems, 125 Neural network classifiers for eddy curent signals, 167–168 in underwater signal classification, 165 Neural networks classical models of, 125–126 nonlinear models of, 126 O algorithm in, 163 in pattern recognition applications, 124 radial basis function (RBF), 141–144 single-layer, 125 used as secondary classifiers, 161 New faces, recognition of, 196 New Zealand national congresses of electron microscopy, 321–322 Nodal elements, advantage of, 111 Nodal points, 108 distribution of, 108, 109 Noise reduction limitations of linear autoassociators in, 186 via second-order Butterworth filters, 175 Wiener filter in, 192 Noisy patterns, completion of, 195–197 Nonlinear dielectric constants, 2, 38 detectability of, 7 of LiNbO3 and LiTaO3, 25 measured data by cantilever-type SNDM, 30

398 Nonlinear dielectric constants (Cont.) measured data for, 24 in optical region of PPLN substrate, 48 quantitative measurement using cantilever-type SNDM, 28–30 quantitative measurement using needle-type SNDM, 22 spatial variation of, 3 Nonlinear dielectric imaging with subnanometer resolution, 7–11 theory for, 11–17 Nonlinear dielectric microscopy, higher-order experimental details of, 33–36 theory for, 30–33 Nonlinear dielectric response, capacitance variation caused by, 7 Nonlinear materials, in applications of Maxwell’s equations, 60 Nonlinear neural network models, 126 Nonlinear spaces, mappable by RBF, 144 Nonlinear transfer function, 137–138 Nonoverlapping elements, in computational domain, 86–90 Nonparametric regression methods, 180 Nonperiodic signals, Fourier analysis in, 176 Nonpropagating characteristic waves, 91 Nonsmooth geometries, 115 Finite volume methods successful with, 94–97 issues in using fictitious/ghost points with, 84 Nonstaggered grid, fictitious and overlapping grid methods employing, 80 Nonstationary signals limitations of Fourier methods for, 176

INDEX

wavelet transform as best preprocessing technique for, 125, 145, 159 Nordic Microscopy Society, 260–263 Nordic multinational meetings, for electron microscopy, 207 Norelco Reporter, 250 North American regional societies, for electron microscopy, 211 Noticiero de Biologia, 295 Numerical wave speed, 68 Nystrom methods, 114

O O algorithm, in audio signal processing, 163 Obsessive compulsive disorder, EEG signals in, 169–170 Ohm’s law, 63 One-dimensional scanning, 42 One-sided derivatives, 81 Ontario Group of Electron Microscopists, 289 Open space problems, global spectral methods suitable for, 85 Operational research, pattern recognition in, 124 Optical parametric oscillation (OPO) devices, 44 Optik, 209, 222 Orthogonal wavelets, 147, 172 Orthogonality property, 147 and choice of wavelet function, 151 Oscillating frequency, of LC resonator, 6 ¨ sterreichische Arbeitsgemeinschaft O fu¨r Ultrastrukturforschung ¨ AU), 222–224 (O ¨ sterreichische Gesellschaft fu¨r O Elektronmikroskopie ¨ GE), 222–224 (O Output neurons, 126 Overlapping grid methods, 80–84

INDEX

P Paraelectric layer, 36 Parallel processing, required for realtime operations in image processing, 154–155 Particle shape classification, 157 Partition of unity approach, 83 Pattern completion, of noisy patterns, 195–197 Pattern recognition application of wavelets in, 156 autoassociative memories used in, 135 classifier module in, 156 defined in engineering context, 124 example of, 131–134 preprocessing module in, 156–157 preprocessing techniques for, 157–162 prosthetic limb control application, 171 RBF neural networks useful in, 142 PEC conesphere, 110 Pen-tip displacements, 179–180 Penetration problems, involving linear materials, 65 Perceptron learning rule, 128 Perceptron neural network model, 125, 126–128 Perceptual activity, 126 Perfect conductor, 65 as boundary, 66 conesphere, 112 Perfectly matched layer (PML), 85, 115–116 Periodic signals, detectable by eta3333 imaging, 35 Periodic solutions, 85 Periodically poled LinbO3 (PPLN) fabricated by high voltage, 45 Single domain surface layer on virgin, 46 SNDM image prior to annealing, 48

399 SNDM study for quasi-phase matching device, 44 Permeability tensor, 63 Permittivity tensor, 63, 71 Peruvian national congresses of electron microscopy, 296 Pfefferkorn Conferences, 211, 327–329 Phase error, maximal acceptable, 69 Phase profiles, limitations of, 9 Phase-sensitive components, Maxwell’s equations and, 60 Phase transition, importance in recording medium, 50 Philippine national congresses of electron microscopy, 314 Philips Electron Optics Bulletin, 250 Philips Research Reports/Journal, 250 Physical jump conditions, 74 enforced via locally modified schemes, 76 enforcing via local schemes, 91 inability of high-order finite difference schemes to enforce, 72 Physikalische Gesellschaft der DDR (PG), 237–238 Physiological tremor, causing noise in handwriting movement analysis, 180 Piecewise constant permittivity, 71 Piezo-imaging, limitations of read speed for ultrahighdensity data storage applications, 50 Piezoelectric materials, imaging polarizations in, 2 Piezoelectric thin films, 41 Piezoresponse imaging, comparison with SNDM imaging, 10–11 Plane waveguide test case, 91 Point defects, in SLT vs. CLT media, 51 Point-to-point variations, of nonlinear dielectric constants, 2

400 Points per wavelength few required using high-order methods, 61 in global spectral methods, 84, 85 number of, 68 Polar dielectric thin films, recently developed types of, 41 Polarity distribution, measurement using SNDM, 42 Polarization distribution of directions of, 40 modeling effects of, 67 parallel and perpendicular components of, 40 spontaneous, observing microscopic distribution of, 2 of ZnO thin films on polar substrate, 41–44 Polarization direction, measuring parallel to surface, 38 Polarization inversion, 31 Polish Commission for Electron Microscopy, 253–255 Polish conferences, electron microscopy of solids, 256 Polish national conferences of electron microscopy, 253–257 Polish Society for Microscopy, 254 Polish Society of Anatamopathologists, 253 Polish symposia on electron microscopy, 256–257 Polishing, effects on surface layer of PPLN substrate, 46 Polishing depth, effect on change of offset of SNDM signal, 47 Polskie Towarzystwo Mikroskopii, 254 Portuguese national congresses of electron microscopy, 257–259 Position signals, as filtering parameter, 181 Potential, relationship with stored charged in the metal, 12

INDEX

PPLN, imaged virgin, before, and after polishing, 37 Preprocessing techniques in audio signal identification, 162–163 in eddy current signal analysis, 167–168 in face recognition applications, 186–197 necessity in pattern recognition applications, 125 in pattern recognition, 156 in underwater signal classification, 164–165 using multiscale edges in face recognition, 190–195 Previously learned faces, 196 Primer Curso Internacional de Microscopia Electro´nica para Cientı´ficos Latinoamericanos, 208 Principal axes, 158 Principal component analysis (PCA), 157–158 in eddy current signal analysis, 167 in face recognition applications, 186 as preprocessing technique, 125 Probabilistic neural network (PNN), 175 Proceedings of the Royal Microscopy Society, 269 Prosthetic limbs, powered control of, 171 Prototype vector, distance from input vector, 141 Pulse application times, for SLT vs. CLT media, 54 Pyramidal algorithm, 145 Pyroelectric thin films, 41 PZT film images of, 39 nanoscale ferroelectric domain on, 9 on a Sf TiO3 substrate, 8

401

INDEX

Q Quadratic spline function, 177 Quadratic time-frequency representations, 162 Quadrature mirror filters, 148 Quadrature points, polynomials defined at, 107 Quantitative measurement of linear dielectric constant using cantilever-type SNDM, 25–30 using needletype SNDM, 20–21, 22–25 of nonlinear dielectric constant, using cantilever-type SNDM, 28–30 using needle-type SNDM, 20–25 Quasi-phase matching (QPM) devices, 44–49

R Radar classification, 162 Radial basis function (RBF) neural networks, 126, 141–144, 165 Random effects, in applications of broadband technology, 60 RBF classifiers, advantages of, 143 Read speed, of SNDM vs. piezo imaging methods, 50 Real-time operations, demand for, necessitating parallel processing, 154–155 Reflection and refraction, modeling, 67 Refractive index, 22 Reisner, J. H., 282 Relative permittivity, 66 Remote sensing imagery, 173 Resolution high for higher-order nonlinear dielectric imaging, 33 higher lateral for 33333 imaging, 33

of needle-type vs. cantilever-type SNDM, 19–20 of SNDM, 9 subnanometer, 10 for ultrahigh-density data storage applications, 50 Resonance frequency, varying as function of alternating applied electric field, 28–29 Revista de Microscopı´a Electro´nica, 208, 219 Revista de Microscopı´a Electro´nica y Biologia Celular, 208, 219 Rosenblatt, and perceptron model, 125 Rotated Y-cut substrates, 43 Rotation invariance, 162 Royal Microscopy Society (RMS), 269, 274–276 Runge-Kutta scheme, 111–112, 113, 115 advancing in time with, 71 fourth-order, 92 Russian national congresses of electron microscopy, 277–281

S Scaling function, 145, 148–149 Scaling invariance, 161 Scandinavian regional congresses of electron microscopy, 260–263 Scanning, the Microcircuit Engineering Conferences, 211 Scanning electron microscopy conferences, 327–330 Scanning force microscopy (SFM) technique, 2 Scanning microscope, history of, 207 Scanning nonlinear dielectric microscopy (SNDM) applications of technique, 3 for high-performance ferroelectric material and devices, 41–54 defined, 2

402 Scanning nonlinear dielectric microscopy (SNDM) (Cont.) for determining polarity of thin films, 42 image of virgin/polished PPLN, 46 imaging comparisons with piezoresponse imaging, 10–11 microscopic observation of ferroelectric domain distribution using, 7–10 model of needle tip, 14 offset change of signal, 47 on periodically poled LiNbO3 for high-performance quasi-phase matching device, 44–49 principles of, 3–7 quantitative measurement of, 17–30 resolution as function of electric field, 31 subnanometer resolution of, 9, 50 system setup of, 6–7 and Tbit/inch2 ferroelectric data storage, 49–54 theoretical calculation for image, 14–17 Scanning probe microscopy (SPM), in formation of smal inverted domain dots, 49 Scanning transmission electron microscope, history of, 207 Scattered field formation, 65–66 Scattering problems axisymmetric three-dimensional metallic scatterer, 93, 94 far field scattering, 113 finite-length dielectric cylinder, 111 involving linear materials, 65 military aircraft, 112 two-dimensional PEC cylinder, 110 Schizophrenic EEG signals, 169–171 Schweizerische Gesellschaft fu¨r Optik und Elektronenmikroskopie, 265–267 Second-order accuracy, 73 insufficient for many applications, 71

INDEX

Second-order Butterworth filters, 175, 180 in filtering handwriting movement signals, 182 Second-order schemes central finite difference scheme, 70 leapfrog schemes, 113 Yee scheme, 60 Section de Microscopie et Diffraction Electroniques, 230–233 Segmentation of angular figures, 184 in handwriting movement analysis, 183 Seismic tremors Haar wavelet application to study of, 145 wavelet transform preprocessing method for, 125 Semidiscrete Galerkin form, 102 Semidiscrete stability, 115 Sensitivity calibration, 20 Serbian national congresses of electron microscopy, 281–282 Serrano, Jose´ A., 297 Seto, S., 303 Shallow areas, SNDM sensitivity in, 33 Shape recognition, 125, 157 Fourier and wavelet transformed used in, 161 and images of industrial tools, 172–173 in industrial tool applications, 173 obstacle in, 161 recognition and classification of blemishes, 172 wavelet index of texture for Landsat images, 173–174 Shift invariance of distance function, 162 vectors with, 172 Short time Fourier transform (STFT), for signal local description, 145 Siemens, 235 Sigmoid function, 138

INDEX

Signal compression, 156 Signal noise, in handwriting movement analysis, 180 Signal processing, 125 downsampled smoothed version (Daubechies function), 153 frequency vs. temporal representation in, 176 pattern recognition in, 124 scaling function suppressing high-frequency components, 148 wavelet tools for analysis, encoding, compression, reconstruction, and modeling, 156 Signal-to-noise ratio eliminating need for lock-in amplifier, 51 in handwriting movement analysis, 177 of SNDM signal, 15 Signal variance estimation, 156 Simultaneous approximation term (SAT), 80 Singapore national congresses of electron microscopy, 314 Single-layer feed-forward networks, 126–128 Single-layer linear networks, 125 limitations of, 126 Singular value decomposition, 189–190 Skandinaviska Fo¨reningen fo¨r Elektronmikroskopi (SCANDEM), 260–263 Sleep stages, classification of, 125, 157, 168–169 Slovenian national congresses of electron microscopy, 263 Smooth curvilinear mapping, 108 Smooth interfaces, use of overlapping patch or grid with, 83 Smoothing function, in handwriting movement analysis, 176 Smoothness, 115 loss of, across material interfaces, 66

403 of solution, efficiency of high-order methods linked to, 86 Sobel operator, 191, 192, 193, 194 in face recognition, 187 Sociedad Argentina de Microscopı´a, 292 Sociedad de Microscopı´a de Espan˜a, 264 Sociedad de Morfologı´a (Cuba), 295 Sociedad Ecuatoriana de Microscopı´a Electro´nica, 295 Sociedad Espan˜ola de Microscopı´a Electro´nica (SEME), 264–265 Sociedad Latinoamericana de Microscopı´a Electro´nica (SLAME), 219, 292 Sociedad Venezolana de Microscopı´a Electro´nica, 297–298 Sociedade Brasileira de Microscopia e Microana´lise, 292–294 Sociedade Brasileira de Microscopia Electroˆnica, 292–294 Sociedade Brasileira de Microscopia e Microana´lise, 292–294 Sociedade Brasileira de Microscopia Electroˆnica, 292–294 Sociedade Portuguesa de Microscopia Electroˆnica, 257–259 Sociedade Portuguesa de Microscopia Electroˆnica e Biologia Celular, 257–259 Societa` Italiana di Microscopia Elettronica (SIME), 247–250 Socie´te´ Belge de Microscopie, 224–226 Socie´te´ Franc¸aise de Microscopie Electronique (SFME), 230–233 Socie´te´ Franc¸aise des Microscopies (SF), 230–233 Society of Electron Microscope Technology (SEMT), 269–270, 276–277 Society of Photo-Optical Instrumentation Engineers (SPIE) conferences, 327

404 Society of Topochemistry and Electron Microscopy, 233 Solid mechanics, success of finite element methods in, 97 South American regional meetings, for electron microscopy, 207 Soviet All-Union meetings (electron microscopy), 209 Space-frequency plane, comparison of transform methods for processing, 145 Spanish national congresses of electron microscopy, 264–265 Spatial convergence rate, and highorder methods for solving Maxwell’s equations, 61 Spatial derivatives, 88 Spatial variation, of nonlinear dielectric constants, 3 Spectral methods, 84–94 in EEG and myoelectric signal processing, 168 Spectral multidomain scheme, 82 Speech recognition, 162 Spline functions, 175 Spurious solutions, in finite element schemes, 103 Square symmetrical matrix, in autoassociative memory training, 135–136 SrTiO3 single crystal, 22 Staggered grid, 74 conserving divergence in compact schemes via, 77 improving accuracy of Maxwell’s equations, 60 Staggered-in-time leap-frog scheme, 70 Staircased curve in higher-order finite difference schemes, 72 overcoming via local modifications close to boundaries and interfaces, 74 in Yee scheme for solving Maxwell’s equations, 60

INDEX

Standard normalizations, of Maxwell’s equations in the time domain, 62 Standing waves, as solution to higher-order finite difference problem, 71 Static capacitance, 5 Stationary signals, Fourier transforms for processing, 145 Statistical methods in handwriting movement analysis, 175 nonparametric regression techniques, 175–176 Statistics pattern recognition in, 124 wavelet analysis applications in, 156 Stencils computational efficiency issues when using wide, 84 large one-sided, in global spectral methods, 85 Stoichiometric lithium tantalate (SLT), 51–54 Stored charge, 12 Stored energy, 12 Stray capacitance, 20 Stretch reflexes, causing noise in handwriting movement analysis, 180 Stretched grids, errors associated with, 97 Strip shape domain pattern, 10 Summation by parts (SBP) schemes, 77–80 SuperSTEM workshops, 269 Surface acoustic wave (SAW) devices, 41 Surface current density, 65 Surface morphology, by AFM, 8 Swiss national congresses of electron microscopy, 265–267 Swiss Society for Optics and Electron Microscopy, 265–267

INDEX

Switching speeds, of SLT vs. CLT media, 54 System evaluation, in audio signal processing, 163

T Taiwanese national congresses of electron microscopy, 315 Tangential fields, continuity of, 91 Tangential velocity profile, 183 Tbit/inch2 ferroelectric data storage, 49–54 TE-polarized illumination, 112–113 Temporal integration approximating in high-order finite difference schemes, 70 issues in, 113–115 Temporal representation, compared with frequency representation, 176 Tensors even rank, and states of spontaneous polarization, 4 interpolating basis, 107 in local schemes, 88 nonlinear dielectric constants of, 2–3 polarity reversal of odd-ranked, 31 Texas Society for Electron Microscopy Journal, 284 Texture information, in Landsat images, 174 Thai national congresses of electron microscopy, 315–317 Third-order accuracy, finite volume schemes limited to, 96 Third-order recursive filters, 177 Three-beams conferences, 332–335 Three-dimensional axisymmetric missile, 92 Three-dimensional measurement technique, 37–40 Three-dimensional waveguide problems, 93–94 Three-point median filtering, 166

405 Threshold activation functions, 126–128 Time, advancement in, 111–112 Time-domain methods in myoelectric signal processing, 171 for solving Maxwell’s equations, 60 Time-domain signal, successive convolutions of, 148 Time-domain solution, solved by high-order methods, 61 Time-frequency plane comparison of transform methods for processing, 145 wavelet packet transform providing adaptive tiling in, 160 Time-frequency representations, 162 Time-step restriction, uniformly bounded, 76 Time-stepping, 115 at expense of accuracy, 114 in higher-order methods for solving Maxwell’s equations, 63 TiO2-Bi2Ti4O11 ceramic, linear dielectric constant distribution of, 29 Tip, of LC resonator, 6 Tip radius, exact evaluation of, 24 Tip sensitivity, calibration data for, 28 Tonals, in underwater signal classification, 164 Total field, 65 Transfinite blending functions, 89 Transmission electron microscope, history of, 207 Transverse electric form, of Maxwell’s equations, 67 Transverse magnetic form, of Maxwell’s equations, 67 Tridiagonal matrix solution, 77 Tungsten wire needle, electrolytic polishing of, 7 Tu¨rk Elektron Mikroskopi Dernegi, 267–268 Turkish national congresses of electron microscopy, 267–268

406 Two-dimensional PEC cylinder, 110 Two-dimensional TM-polarized wave propagation, 92

U Ultrahigh-density recording systems, role of nanozied inverted domain dots in, 49 Ultramicroscopy, 210, 211 Underwater acoustics, 162 Underwater signal identification, 125, 156, 164–165 United Kingdom national congresses of electron microscopy, 268–277 Unsupervised dimensionality reduction, 157 Upwind negative eigenvalues, 95 USA national congresses of electron microscopy, 282–289 local societies affiliated with, 283–284 USSR national congresses of electron microscopy, 277–281

V Vacuum permittivity and permeability, 64 Vacuum plane wave, 65 Vainshtein, B. K., 277 Variational statements, 103–104 addressing for finite element schemes, 97 Vector nodes, spurious, in continuous finite element schemes, 99 Vehicles, very low observable, Maxwell’s equations and, 60 Velocity signals, 185 as filtering parameter, 181 Venezuelan national congresses of electron microscopy, 297–298 Volcanic eruption predictions, 173 Vorbrodt, Andrzej, 254

INDEX

W Wave-matter interaction, complicating solution of Maxwell’s equations, 60 Wave periods, number of, 68 Wave propagation accurately depicted by high-order methods, 61 degrees of freedom optimizing characteristics of, 77 effects of grid anisotropy on, 69 global spectral methods in modeling large-scale problems of, 85 in lossy media, 94 solved by global spectral methods, 84 two-dimensional TM-polarized, 92 two-way, 66 Wavelet analysis compared with Fourier analysis in sleep research, 168 criticism of, 150–151 parallelism and applicaations of, 154–156 in statistical applications, 156 Wavelet basis functions, 147 Wavelet coefficient, 147, 149 of Daubechies function, 153 Wavelet decomposition methodology, 147 Wavelet filtering technique, in handwriting and drawing, 178–186 Wavelet function, GCD filter as, 179 Wavelet function normalization, 173 Wavelet functions, 145, 150–154 arbitrary choice as limitation of, 150–151 defined, 147 factors in choice of, 151 Wavelet index of texture, 173–175

407

INDEX

Wavelet lifting scheme, 156 Wavelet matrix, in Haar transform, 152 Wavelet packet transform, 150, 159–160 myoelectric signal classification using, 171–172 Wavelet prefiltering technique, 175 Wavelet representation, computation of, 148 Wavelet theory, 145 Wavelet transform, 159, 194 combining with autoassociative memory, 186–197 compared with Fourier, STFT, and Gabor transforms, 145 and concurence matrices, 174 continuous, definition of, 145–147 in face recognition applications, 193–195 in handwriting movement analysis, 176–178 image processing by, 149 intrinsic parallelism of, 155 offering best performance for handwriting movement analysis, 186 as prefiltering technique for handwriting and face recognition, 125 as preprocessing technique for nonstationary signals, 125 theory of, 175 Wavelet vector, in Haar transform, 152 Welsh Microscopical Society, 270

Widrow and Hoff, and adaline model, 125 Widrow-Hoff learning rule, 126, 129–130, 133 linear autoassociative memory training with, 136 Wiener filter, 192, 194 Wilska, Alvar, 260 Wollnik, H., 326 Workshops on Electron Energy-Loss Spectroscopy and Imaging (EELSI), 347–348 World of Microstructure, 221

Y Y-cut substrates, 43 Yee scheme extensions of, 62, 73–77 fictitious and overlapping grid methods extending, 80–84 finite-difference time-domain (FDTD) method and, 70 fourth-order extension of, 71 limitations of, 60 fourth-order extensions of, 77 for solving Maxwell’s equations, 60 Yugoslav Symposium on Electron Microscopy, 281 Yugoslavian national congresses of electron microscopy, 281–282

Z Zeitschrift fu¨r technische Physik, 234 ZnO thin films, 41–44